Immunohistochemical localization of macrophages and microglia in the adult and developing mouse brain

Hypoxic-ischemic brain injury induces an acute microglial reaction in perinatal rats

Activated microglia may contribute to the progression of neuronal injury after a wide range of CNS insults. In this study, we used two complementary methods to evaluate acute changes in the morphology and regional distribution of microglia induced by a focal hypoxic-ischemic insult in 7-d-old (P7) rats. To elicit injury, P7 rats underwent right carotid ligation followed by 3 h of 8% O2 exposure; rats were killed 10 min to 5 d later (n > or = 3/group). A histochemical assay using Griffonia simplicifolia B4-isolectin enabled detection of both resting and activated microglia in tissue sections; vascular cells were also reactive. Activated microglia were also identified immunocytochemically using a macrophage-specific MAb, ED-1. In normal P7-12 brain, lectin, and ED-1 immunoreactive-activated microglia were concentrated in white matter; lectin-positive resting, ramified microglia were also detected throughout the gray and white matter. Subtle morphologic evidence of microglial activation was noted 10 min posthypoxia-ischemia in the lesioned right cerebral hemisphere; activated microglia began to accumulate within the next 4 h. Accumulation of lectin-positive activated microglia peaked at 2-4 d posthypoxia-ischemia. ED-1 immunoreactive-microglia were first noted 4 h after hypoxic-ischemic injury in the lesioned right hemisphere, and there was a corresponding increase in accumulation over the first 48 h posthypoxia-ischemia. In the left hemisphere, contralateral to the ligation, no increase in activated microglia were detected with either method. In brain sections where no neuronal injury was evident, activated microglia did not accumulate. These data demonstrate that perinatal hypoxic-ischemic brain injury induced rapid accumulation of activated microglia in hypoxic-ischemic forebrain.

Mouse microglial cell lines differing in constitutive and interferon-gamma-inducible antigen-presenting activities for naive and memory CD4+ and CD8+ T cells

We developed a panel of non-virus transformed cell lines derived from individual microglial precursors residing in the brains of normal mice. These colony stimulating factor-1-dependent cell lines are B7-1+ (CD80), Mac-1+, Mac-2+, Mac-3+, CD45+, MHC class I+, colony stimulating factor-1 receptor+, and they ingest antibody-coated particles. However, the cell lines differ in their expression of B7-2 (CD86), F4/80, Ly-6C and MHC class II molecules. They also differ in their ability to constitutively process and present antigens to naive CD4+ and CD8+ T cells, memory CD4+ and CD8+, and in the manner by which interferon gamma modulates their antigen-presenting activities. These cell lines should be valuable as models for studies on the immunobiology of the microglia.

Microglia in normal and regenerating visual pathways of the tench (Tinca tinca L., 1758; Teleost): a study with tomato lectin

We have studied the microglial cells in the normal and regenerating visual pathways of Tinca tinca (Cyprinid, Teleost) by using the lectin from Lycopersicum esculentum (tomato), which, in our case, has been demonstrated as a specific marker for teleost microglia. In the normal fish, there are tomato lectin positive microglial cells in the retina, optic nerve, and optic tectum. Following optic nerve crush, we observed a more extensive labeling of the microglia in the crushed optic nerve and in the contralateral optic tectum affecting the stratum opticum and stratum fibrosum et griseum superficiale. In both cases, there was an increase of rounded and less ramified microglial cells, and granular cells. This response of a more extensive labeling of microglial cells increases to a maximum at 2-3 weeks after the crush; the density of labeled microglial cells decreases after 3 months after crushing. However, in the retina no changes were observed after optic nerve crush. These results suggest that the microglial cells could play an important role in regeneration of fish optic pathway, as other neuroglial cells do.

Development of microglial topography in human retina

The development of microglial topography in wholemounts of human retina has been examined in the age range 10-25 weeks gestation (WG) using histochemistry and immunohistochemistry for CD45 and major histocompatibility complex class II antigens. Microglia were present in three planes corresponding to the developing nerve fibre layer/ganglion cell layer, the inner plexiform layer and the outer plexiform layer. Distribution patterns of cells through the retinal thickness and across the retinal surface area varied with gestational age. Microglia were elongated in superficial retina, large and ramified in the middle plane, and small, rounded and less ramified in deep retina. Intensely labeled, rounded profiles seen at the pars caeca of the ciliary processes, the retinal margin and at the optic disc may represent precursors of some retinal microglia. At 10 WG, the highest densities of microglia were present in middle and deep retina in the far periphery and at the retinal margin, with few superficial microglia evident centrally at the optic disc. At 14 WG, high densities of microglia were apparent superficially at the optic disc; microglia of middle and deep retina were distributed at more central locations although continuing to concentrate in the retinal periphery. Microglia appear to migrate into the developing human retina from two mains sources, the retinal margin and the optic disc, most likely originating from the blood vessels of the ciliary body and iris, and the retinal vasculature, respectively. The data suggest that the development of microglial topography occurs in two phases, an early phase occurring prior to vascularization, and a late phase associated with the development of the retinal vasculature.

The microglial cell. A historical review

Effectively, modern research has confirmed Hortega's view of the origin of the microgliacyte from circulating monocytes of the monocyte-macrophage series that invade the brain during embryonic and early postnatal life. Their phagocytic capacity is exercised during the brain remodelling that marks brain maturation. They then convert to the ramified resting microglial cell visualized in the silver carbonate staining technique of Hortega and by modern lectin-binding methods. In response to injury, reactive microglia exhibit hypertrophy and hyperplasia, and may or may not go on to form typical lipid-laden phagocytes. Activated microglia show upregulation of the many marker antigens they share with circulating monocytes, including the major histocompatibility class (MHC) class II antigens that bespeak their immunocompetent nature. However, MHC class I and II expression and development of immunohistochemical positivity for cytoplasmic and plasma membrane antigens that characterize the monocyte-macrophage do not necessarily indicate an immunological response though there is ample evidence that microglia can serve as antigen-presenting cells. Rather, microglia are extraordinarily sensitive to changes in the brain microenvironment, whatever the nature of the exciting mechanism or substance. They may be considered to serve an ever alert, protective and supportive function that can be assembled rapidly to deal with infections, physical injuries, physiologic changes and systemic influences. In addition to elaboration and secretion of cytokines with varied actions, e.g., suppression of astrogliosis, they secrete factors, including nerve growth factor, which are supportive of neurons. They have an important role in iron metabolism and the storage of iron and ferritin. They may promote central nervous system regeneration. They are prominently involved in such pathologic processes as the acquired immunodeficiency syndrome, multiple sclerosis, prion diseases and the degenerative disorders, e.g., Alzheimer's disease and Parkinson's disease. With aging, they grow more numerous, become richer in iron and ferritin and exhibit phenotypic alteration, e.g., the expression of MHC class II antigens that are not ordinarily demonstrable immunohistochemically in the resting state. The rate of growth of our knowledge of microglia during the last decade has been exponential and continues.

The AMOG/beta 2 subunit of Na,K-ATPase is not necessary for long-term survival of telencephalic grafts

Adhesion molecule on glia (AMOG) represents the beta 2-subunit of murine Na,K-ATPase. Mice carrying a targeted deletion of the AMOG/beta 2 gene exhibit tremor and limb paralysis at postnatal day (P) 15 and die 2 days after the onset of symptoms. The brains of these mice show edema and swelling of astrocytic end feet. However, the cause of death has remained unclear. To identify long-term consequences of AMOG/beta 2 deficiency, we have grafted parts of the embryonic telencephalic anlage of AMOG/beta 2-deficient mice into the caudoputamen of wild-type mice and analyzed the grafts up to 500 days after transplantation. Histological, immunocytochemical, and in situ hybridization techniques were applied to examine histoarchitecture, proliferation, differentiation, and long-term survival of grafts. AMOG/beta 2-deficient telencephalic grafts develop normally and form solid neural tissue that cannot be distinguished from control grafts by morphological features or with immunocytochemical stains for neuronal and glial markers. No signs of degeneration can be found. Expression analysis, however, revealed that no AMOG/beta 2 protein of possible host origin can be detected in AMOG/beta 2-deficient grafts. Graft-borne astrocytes express neither the AMOG/beta 1 nor the AMOG/beta 2 subunit of Na,K-ATPase as examined with immunocytochemistry and in situ hybridization. These findings indicate that AMOG/beta 2 is not necessary for long-term survival of telencephalic graft tissue.

Immunohistochemical detection of thrombospondin in microglia in the developing rat brain

The development of microglia involves the expression of a phenotype displaying phagocytic behaviour termed brain macrophage or amoeboid microglial cell. We have previously shown that rat brain macrophages purified in vitro secrete thrombospondin, an extracellular matrix protein, which acts on cultured neuronal cells by promoting neurite growth. In the present study, the expression of thrombospondin was investigated in tissue sections of the developing rat forebrain in relation to the distribution of microglia. These cells were identified using anti-macrophage antibodies and the isolectin B4 from Bandeiraea simplicifolia. Immunocytochemical detection of thrombospondin clearly outlined a cell population displaying the morphologies and distribution of brain macrophages, from the 17th day of embryonic life up to the end of the second postnatal week. These cells were most numerous in cortical and subcortical regions of developing fibre tracts such as the corpus callosum or the internal capsule. The localization of thrombospondin in brain macrophages was confirmed by double immunostaining using ED1 monoclonal anti-macrophage antibodies. Ramified microglial cells were also labelled transiently by anti-thrombospondin antibodies during early postnatal life. These results provide in situ evidence supporting the notion that microglial cells could favour axonal growth by producing thrombospondin during development.

Mouse model for the lysosomal disorder galactosialidosis and correction of the phenotype with overexpressing erythroid precursor cells

The lysosomal storage disorder galactosialidosis results from a primary deficiency of the protective protein/cathepsin A (PPCA), which in turn affects the activities of beta-galactosidase and neuraminidase. Mice homozygous for a null mutation at the PPCA locus present with signs of the disease shortly after birth and develop a phenotype closely resembling human patients with galactosialidosis. Most of their tissues show characteristic vacuolation of specific cells, attributable to lysosomal storage. Excessive excretion of sialyloligosaccharides in urine is diagnostic of the disease. Affected mice progressively deteriorate as a consequence of severe organ dysfunction, especially of the kidney. The deficient phenotype can be corrected by transplanting null mutants with bone marrow from a transgenic line overexpressing human PPCA in erythroid precursor cells. The transgenic bone marrow gives a more efficient and complete correction of the visceral organs than normal bone marrow. Our data demonstrate the usefulness of this animal model, very similar to the human disease, for experimenting therapeutic strategies aimed to deliver the functional protein or gene to affected organs. Furthermore, they suggest the feasibility of gene therapy for galactosialidosis and other disorders, using bone marrow cells engineered to overexpress and secrete the correcting lysosomal protein.

Inflammation in the nervous system

Earlier studies on inflammation in the CNS have largely focused on conditions with an immune component. Recent evidence has emerged, however, that the innate, acute inflammatory response in the CNS parenchyma is quite unlike that in other tissues. The meninges and ventricular compartments show more typical responses, as does the parenchyma of the brain in immature animals. It is becoming apparent that the cells of the mononuclear phagocyte lineage dominate inflammatory responses in the CNS parenchyma.

Macrophages and glia participate in the removal of apoptotic neurons from the Drosophila embryonic nervous system

Cell death in the Drosophila embryonic central nervous system (CNS) proceeds by apoptosis, which is revealed ultrastructurally by nuclear condensation, shrinkage of cytoplasmic volume, and preservation of intracellular organelles. Apoptotic cells do not accumulate in the CNS but are continuously removed and engulfed by phagocytic haemocytes. To determine whether embryonic glia can function as phagocytes, we studied serial electronic microscopic sections of the Drosophila CNS. Apoptotic cells in the nervous system are engulfed by a variety of glia including midline glia, interface (or longitudinal tract) glia, and nerve root glia. However, the majority of apoptotic cells in the CNS are engulfed by subperineurial glia in a fashion similar to the microglia of the vertebrate CNS. A close proximity between macrophages and subperineurial glia suggests that glia may transfer apoptotic profiles to the macrophages. Embryos affected by the maternal-effect mutation Bicaudal-D have no macrophages. In the absence of macrophages, most apoptotic cells are retained at the outer surfaces of the CNS, and subperineurial glia contain an abundance of apoptotic cells. Some apoptotic cells are expelled from the CNS, which suggests that the removal of apoptotic cells can occur in the absence of macrophages. The number of subperineurial glia is unaffected by changes in the rate of neuronal apoptosis.

Reaction of microglial cells and macrophages after cortical incision in rats: effect of a synthesized free radical scavenger, (+/-)-N,N'-propylenedinicotinamide (AVS)

Reactive microglial cells and macrophages appear after trauma to the brain. To investigate the accumulation patterns of reactive microglial cells and macrophages after cortical incision, these cells were stained immunohistochemically with anti-ED1 antibody in the brain sections before and 1, 3, 5, and 7 days after incision. And to ascertain the participation of oxygen free radicals in these cellular reactions, a synthesized free radical scavenger, (+/-)-N,N'-propylenedinicotinamide (AVS) was administered in this model. Rats were administered AVS (300 mg/kg, i.p.) 30 min before, 2.5 h and every 24 h after incision (AVS group), while only saline was administered in the same manner as a control (saline group). In the saline group, both reactive microglial cells and macrophages had already appeared on day 1 post-incision. The former continued to increase in number during the following days, whereas the latter increased in number up to day 3 and thereafter decreased. Both the numbers of reactive microglial cells and macrophages were significantly decreased (P < 0.05) in the AVS group on days 5 and 7. The results suggest the participation of oxygen free radicals in the reaction of microglial cells and macrophages in traumatic brain injury.

Resident microglia and hematogenous macrophages as phagocytes in adoptively transferred experimental autoimmune encephalomyelitis: an investigation using rat radiation bone marrow chimeras

Hematogenous macrophages are known to be involved in the induction of tissue damage in the central nervous system (CNS) as well as of clinical symptoms in experimental autoimmune encephalomyelitis (EAE). Although resident microglia can become phagocytic under certain circumstances, little is known about the role of these cells in brain inflammation in vivo. We thus studied EAE in the model of radiation bone marrow chimeras that allows us to distinguish donor-derived hematogenous cells from resident effector cells. Inflammation in the CNS was qualitatively and quantitatively similar in chimeras compared to fully histocompatible Lewis rats. Although activated resident microglial cells were outnumbered four- to sevenfold in EAE lesions by hematogenous macrophages, the number of resident microglia with ingested myelin was equal to that of macrophages containing myelin debris. Phagocytic resident microglia, expressing the macrophage activation marker ED1, showed ramified as well as amoeboid morphology. From our studies the following conclusions can be drawn. First, a considerable proportion of resident microglia upregulated ED1. Second, resident microglia provide a small but substantial source of brain macrophages in EAE as compared to the large influx of macrophages. Third, our results suggest that microglia, due to their strategic position within the CNS, are more effective in removal of myelin debris compared to hematogenous macrophages.

Monocyte recruitment into the scrapie-affected brain

The recruitment of monocytes into the scrapie-affected brain was investigated in female mice reconstituted with male bone marrow, using a Y-chromosome-specific probe and F4/80 immunocytochemistry. Recruitment of monocytes could be demonstrated in six out of eight animals and the number of recruited cells correlated with the severity of vacuolation in most, but not all, animals. The proportion of microglia derived from recruited monocytes varied between individual animals, did not correlate with the increase in cellularity (glia) in affected areas of brain and did not affect the length of incubation period. Thus, it is unlikely that the recruitment of monocytes is a pivotal event in the development of early pathological changes in scrapie. The morphology of recruited cells in scrapie lesions, as revealed by F4/80 immunoreactivity, was indistinguishable from that of activated resident microglia.

Occurrence of the opiate alkaloid-selective mu3 receptor in mammalian microglia, astrocytes and Kupffer cells

Evidence is presented for occurrence of opiate alkaloid-selective, opioid-peptide-insensitive receptor binding sites, labeled with [3H]morphine, in primary cultures of cat microglia and cat astrocytes, as well as on highly purified preparations of rat Kupffer cells. These receptors have been designated mu3 on the basis of their close similarity to receptors first found to be present on human peripheral blood monocytes. Exposure of the microglia to morphine and etorphine caused marked quantifiable changes in cellular morphology, including assumption of a more rounded shape and retraction of cytoplasmic processes; in contrast, several opioid peptides were without effect on morphology. The effects of morphine on microglial morphology were blocked by the opiate antagonist naloxone. These effects of drugs on morphology were as predicted for action via the mu3 receptor. Opiate alkaloid binding sites previously detected on the rat C6 glioma cell line were also characterized here as of the mu3 receptor subtype. It is proposed that mu3 receptors have broad distribution in different macrophage cell types of bone marrow lineage, including microglia and Kupffer cells. Furthermore, these receptors are not restricted to cells of bone marrow lineage, since they are also present on astrocytes.

Developmental profile of non-heme iron distribution in the rat brain during ontogenesis

The entry of iron from blood into the developing rat brain was studied by means of non-heme iron-histochemistry. The content of non-heme iron in the endothelial cells was manifest already from E14, declined from P3 to P5, and was almost absent on P10-P15. The choroid plexus epithelial cells of either ventricle was non-heme iron-containing from E14. Non-heme iron-containing macrophages situated in the stroma of the choroid plexus were also observed from E14. From E19, the macrophage-like cells tended to invade into (a) regions with transitory structures like the intermediate zone of the cerebral hemisphere, (b) developing axonal tracts like corpus callosum and internal capsule, and (c) deep layers of the tectum, a region with an extensive degree of naturally occurring cell death. The amoeboid macrophage-like cells observed in the brain parenchyma gradually acquired prolonged extensions and apparently differentiated into ramified microglia-like cells, which later lost their non-heme iron-content. Thus, at P70, non-heme iron-positive microglia-like cells were hardly seen reflecting the transitory event of non-heme iron in microglia-like cells. At P200, non-heme iron-containing microglia cells and oligodendrocytes appeared in manifestly higher number than at P70, a phenomenon probably related to aging. These results delineate for the first time the appearance of iron in the developing brain. The results are of relevance for understanding the potential of iron-deficiency for harming the developing central nervous system, generally by decreased transport of iron through brain capillaries and choroid plexus, and specifically by an impaired modulation of the developing brain parenchyma by iron-containing macrophages.

The unique characteristics of inflammatory responses in mouse brain are acquired during postnatal development

The kinetics of leukocyte recruitment during acute inflammation in adult mouse brain differ from the stereotyped response occurring in non-CNS tissues; neutrophil recruitment is minimal and monocyte recruitment occurs after a 48 h delay. One aspect of the CNS microenvironment which may contribute to restricted leukocyte recruitment is the highly differentiated nature of resident CNS macrophages, the microglia. Thus we studied the inflammatory response to intracerebral injections of endotoxin in neonates in which microglia are less differentiated and resemble more closely macrophages of non-CNS tissues. Mice injected with endotoxin on the day of birth exhibited both neutrophil and monocyte recruitment to the parenchyma, but the response differed from that occurring in non-CNS tissues such as skin. Leukocyte recruitment was very slow, the mononuclear phagocyte response peaking 14 days after endotoxin injection. This sluggish inflammatory response was reminiscent of that previously described in fetal wounds. However, when endotoxin was injected into brains of 7-day-old neonates the inflammatory response resembled that seen in non-CNS tissues; i.e. prolific neutrophil recruitment and a brisk mononuclear phagocyte response. Thus the unusual inflammatory cell kinetics are a property of the mature CNS microenvironment; all signals necessary to support typical leukocyte recruitment are present in the brain by 7 days of age but the brain becomes able to restrict leukocyte immigration during subsequent postnatal development. Developmental changes in the host response to identical inflammatory challenges suggest a window during which the brain may be particularly vulnerable to inflammatory bystander damage.

A role for ciliary neurotrophic factor as an inducer of reactive gliosis, the glial response to central nervous system injury

Within the central nervous system (CNS) ciliary neurotrophic factor (CNTF) is expressed by astrocytes where it remains stored as an intracellular protein; its release and function as an extracellular ligand are thought to occur in the event of cellular injury. We find that overexpression of CNTF in transgenic mice recapitulates the glial response to CNS lesion, as does its injection into the uninjured brain. These results demonstrate that CNTF functions as an inducer of reactive gliosis, a condition associated with a number of neurological diseases of the CNS.

Immunophenotypic evidence for distinct populations of microglia in the rat hypothalamo-neurohypophysial system

The morphology, distribution and immunophenotype of microglia throughout the adult rat hypothalamo-neurohypophysial system was examined. Four macrophage-associated antibodies (OX-42, F4/80, ED1 and ED2) were used; the expression of major histocompatibility complex antigens was investigated by use of antibodies against OX-6, OX-17 (MHC class II) and OX-18 (MHC class I). Three distinct types of microglia were identified. The first was located in the magnocellular nuclei; these 'radially branched' ('ramified') microglia had round cell bodies and long branched processes, and were strongly immunoreactive only for OX-42. The second was located outside the blood-brain barrier in the median eminence, pituitary stalk and neurohypophysis often close to blood vessels; these 'compact' microglia had irregular cell bodies and shorter processes, and were strongly labelled by OX-42 and F4/80, weakly labelled by OX-18, and generally unlabelled by ED1, ED2, OX-6 and OX-17. The third type was found in small numbers throughout the system at the surface of the nervous tissue or around blood vessels; these 'perivascular' microglia were elongated cells with no branching processes, and were strongly labelled by ED1, ED2, OX-18, OX-6, OX-17 and F4/80 antibodies but showed variable OX-42 immunoreactivity. Cells in a perivascular location were heterogeneous with respect to their immunophenotype. The presence in the normal adult rat hypothalamo-neurohypophysial system of MHC class-II molecules (OX-6 and OX-17) on a sub-set of perivascular microglia suggests that these cells are capable of presenting antigen to T lymphocytes.(ABSTRACT TRUNCATED AT 250 WORDS)

Keratan sulphate is a marker of differentiation of ramified microglia

Recently we reported that the keratan sulphate epitope recognised by the monoclonal antibody 5D4 is expressed by a population of ramified microglia in adult rats. As ramified microglia is believed to differentiate from ameboid microglia during postnatal development, we studied the rat brain from birth to 90 postnatal days of life with the monoclonal antibody 5D4. Contrary to all the other microglia markers until now described, keratan sulphate is not expressed by ameboid microglia and by macrophages but appears on the surface of microglia only when the cells are differentiated and show ramified processes. The keratan sulphate positive cells become evident at different times in different central nervous system areas; the first were localised in the pyriform cortex and brainstem from the end of the second postnatal week. These observations suggest that keratan sulphate expression on microglia cells is induced by differentiation and by a resting functional state. Moreover the 5D4 monoclonal antibody showed a strong diffuse positive staining of some cortical, thalamic and white matter areas during the first two postnatal weeks. This staining was transient and it does not seem biologically correlated with the expression of the keratan sulphate on differentiated microglia.

Macrophages, microglia, and astrocytes are rapidly activated after crush injury of the goldfish optic nerve: a light and electron microscopic analysis

Several matrix and adhesion molecules in fish optic nerve, which are constitutively expressed, are increased during axonal regeneration and are primarily associated with nonneuronal cells (W.P. Battisti, Y. Shinar, M. Schwartz, P. Levitt, and M. Murray [1992] J. Neurocytol. 21:557-573). The current study examines the reactions of specific cell types to optic nerve crush and axonal regeneration. The goldfish optic nerve contains macroglia and microglia as well as a population of monocyte-derived cells (granular macrophages) unique to goldfish. Two cell types were OX-42 positive (granular macrophages and microglia), indicating monocyte lineage, each with a distinct morphology and distribution within the nerve. Within hours of the optic nerve crush, the number of OX-42-labeled cell profiles increased near the crush site, remained elevated during the time axons were elongating, and then declined. Microglia, but not granular macrophages, were phagocytically active. Astrocytes are readily identified in the normal optic nerve, but they exhibited marked morphologic changes within hours of injury, which is consistent with the contribution these cells make to the altered environment. Oligodendroglia could not be reliably identified in regenerating optic nerves until myelin was formed. A comparison of the distribution of OX-42-labeled cells with that of transforming growth factor beta-1 (TGF-beta 1) and tenascin suggests that these molecules are expressed by granular macrophages. Tenascin staining may be additionally associated with astrocytes and/or microglia. The rapid response of these nonneuronal cells to injury, their rapid phagocytic activity, and the secretion of growth-promoting factors by these cells likely contributes to the environment that supports robust regeneration by optic axons in the goldfish.

Origin of microglia in the quail retina: central-to-peripheral and vitreal-to-scleral migration of microglial precursors during development

The origin, migration, and differentiation of microglial precursors in the avascular quail retina during embryonic and posthatching development were examined in this study. Microglial precursors and developing microglia were immunocytochemically labeled with QH1 antibody in retinal whole mounts and sections. The retina was free of QH1+ macrophages at embryonic day 5 (E5). Ameboid QH1+ macrophages from the pecten entered the retina from E7 on. These macrophages spread from central to peripheral areas in the retina by migrating on the endfeet of the Müller cells and reached the periphery of the retina at E12. While earlier macrophages were migrating along the inner limiting membrane, other macrophages continued to enter the retina from the pecten until hatching (E16). From E9 on, macrophages were seen to colonize progressively more scleral retinal layers as development advanced. Macrophages first appeared in the ganglion cell layer at E9, in the inner plexiform layer at E12, and in the outer plexiform layer at E14. Therefore, it seems that macrophages first migrated tangentially along the inner retinal surface and then migrated from vitreal to scleral levels to gain access to the plexiform layers, where they differentiated into ramified microglia. Macrophages appeared to differentiate shortly after arrival in the plexiform layers, as poorly ramified QH1+ cells were seen as early as E12 in the inner plexiform layer and at E14 in the outer plexiform layer. Radial migration of macrophages toward the outer plexiform layer continued until posthatching day 3, after which retinal microglia showed an adult distribution pattern. We also observed numerous vitreal macrophages intimately adhered to the surface of the pecten during embryonic development, when macrophages migrated into the retina. These vitreal macrophages were not seen from hatching onwards, when no further macrophages entered the retina.

Complement biosynthesis in the central nervous system

Complement is an important effector arm of the human immune response. Binding of proteolytic fragments derived from activation of complement by specific receptors leads to responses as diverse as inflammation, opsonization, and B-cell activation. The importance of characterizing the expression and regulation of complement in the CNS is highlighted by growing evidence that complement plays a significant role in the pathogenesis of a variety of neurological diseases, such as multiple sclerosis and Alzheimer's disease. In vitro studies have demonstrated that astrocytes, the predominant glial cell type in the brain, are capable of expressing or producing a majority of the components of the complement system. Expression of many complement proteins synthesized by astrocytes is regulated by both pro- and anti-inflammatory cytokines, many of which are also produced by several cell types in the CNS. In addition to astrocytes, ependymal cells, endothelial cells, microglia, and neurons have recently been shown to synthesize various complement proteins or express complement receptors on their cell surfaces. Together, these studies demonstrate that several cell types throughout the brain have the potential to express complement and, in many cases, increase expression in response to mediators of the acute phase response. These studies suggest that complement may play a greater role in CNS immune responses than previously thought, and pave the way for better understanding of the dynamics of complement expression and regulation in vivo. Such understanding may lead to therapeutic manipulation of complement host defense functions in a variety of inflammatory and degenerative diseases in the CNS.

Strongly GD3+ cells in the developing and adult rat cerebellum belong to the microglial lineage rather than to the oligodendrocyte lineage

A recent study has shown that ramified microglia in the adult rat optic nerve express the ganglioside GD3 [Wolswijk Glia 10:244-249, 1994], thereby raising the possibility that some GD3+ in the developing rat central nervous system (CNS) belong to the microglial lineage rather than to the oligodendrocyte lineage, as previously thought. To examine this possibility, sections of postnatal and adult cerebellum were double-labelled with markers for rat microglia [the B4 isolectin derived from Griffonia simplicifolia (GSI-B4), the ED1 monoclonal antibody (mAb), and the OX-42 mAb] and anti-GD3 mAbs (the mAbs R24 and LB1). These immunolabellings showed that ramified microglia as well as amoeboid microglia are strongly GD3+ in vivo. Moreover, most, if not all, cells that express high levels of GD3 in sections of developing cerebellum appear to belong to the microglial lineage. These observations contradict previous suggestions that the strongly GD3+ cells in the putative white matter regions of the developing brain are oligodendrocyte-type-2 astrocyte (O-2A) progenitor cells; the cells that give rise to oligodendrocytes in the CNS. The present study did, however, confirm that some O-2A progenitor cells in sections of postnatal cerebellum are weakly GD3+ in vivo. Amoeboid microglia are present in areas of the developing cerebellum where newly generated oligodendrocytes are found, suggesting that these cells play a role in the phagocytosis of the large numbers of oligodendrocytes that die as part of CNS development.

Neuroglial responses to the dopaminergic neurotoxicant 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine in mouse striatum

We have studied the reactive responses of both astrocytes and microglia to dopaminergic denervation of the striatum by MPTP. Following MPTP treatment, increased GFAP immunoreactivity reached a peak at 2 days and persisted for at least 6 weeks. Immunoreactivity to vimentin was also markedly increased in astrocytes 48 h after MPTP treatment. Striatal laminin immunoreactivity, however, appeared to be unaffected by drug treatment. GFAP protein levels increased to 196% and 321% of control 24 and 48 hours after MPTP treatment, respectively. Concomitantly, GFAP mRNA levels increased to 560% and 1620% of control, respectively. These reactive changes in striatal astrocytes in response to MPTP treatment were also accompanied by a reactive microglial response as evidenced by increased immunohistochemical visualization of striatal microglia using antibodies to Mac-1. Our results and those reported previously by O'Callaghan et al., strongly suggest that MPTP-induced reactive gliosis in mouse striatum is associated with reactive microglia, albeit without increased interleukin-1 beta.

The cranial meninges: anatomic considerations

The anatomy of the cranial dura and leptomeninges is both intricate and complex. A thorough discussion of the protective covering of the brain including the dura, arachnoid, and pia is provided on both gross and microscopic levels. An attempt to include issues of clinical relevance is made, highlighting the Virchow-Robin spaces and the optic sheath. In addition, the normal appearance of the dura and leptomeninges on MRI is presented to establish a framework for the discussion of leptomeningeal pathology.

Induction of MHC class II antigen expression on murine microglia by interleukin-3

The effects of various cytokines on MHC class II antigen expression were examined in murine microglia. Interleukin-3 (IL-3), as well as interferon-gamma (IFN-gamma), induced MHC class II antigen expression on these cells. IL-3 additionally enhanced MHC class II antigen expression induced by IFN-gamma. The induction of MHC class II antigen expression by IL-3 was not mediated via IFN-gamma production, because the effect was not blocked by antibodies to IFN-gamma. In contrast, granulocyte-macrophage colony-stimulating factor (GM-CSF) did not affect the expression of MHC class II antigen on naive cells and down-regulated IFN-gamma-mediated induction of MHC class II antigen expression on microglia. Because IL-3 and GM-CSF are apparently produced in the central nervous system, MHC class II antigen expression on microglia may be regulated by these cytokines synthesized in the central nervous system.

Quantification of the mononuclear phagocyte response to Wallerian degeneration of the optic nerve

We investigated the numbers, origin and phenotype of mononuclear phagocytes (macrophages/microglia) responding to Wallerian degeneration of the mouse optic nerve in order to compare it with the response to Wallerian degeneration in the PNS, already described. We found macrophage/microglial numbers elevated nearly four fold in the distal segments of crushed optic nerves and their projection areas in the contralateral superior colliculus 1 week after unilateral optic nerve crush. This relative increase in mononuclear phagocyte numbers compared well with the four-to-five-fold increases reported in the distal segments of transected saphenous or sciatic nerves. Moreover, maximum numbers are reached at 3, 5 and 7 days in the saphenous, sciatic and optic nerves respectively, suggesting that the very slow clearance of axonal debris and myelin in CNS undergoing Wallerian degeneration is not simply due to a slow or small mononuclear phagocyte response. The apparent delay in the response in the CNS occurs because the mononuclear phagocytes respond to the Wallerian degeneration of axons, which is slightly slower in the CNS than the PNS, rather than to events associated with the crush itself, such as the abolition of normal electrical activity in the distal segment. This was demonstrated by the protracted time course of the mononuclear phagocyte response in the distal segment following optic nerve crush in mice carrying the Wlds mutation which dramatically slows the rate at which the axons undergo Wallerian degeneration. By [3H]-Thymidine labelling or by blocking microglial proliferation by X-irradiation of the head prior to optic nerve crush, we showed that the majority of macrophages/microglia initiating the response to Wallerian degeneration were of local, CNS origin but these cells rapidly (from 3 days post crush) upregulate endocytic and phagocytic functional markers although they do not resemble rounded myelin-phagocytosing macrophages observed in degenerating peripheral nerves. We speculate that the poor clearance of myelin in CNS fibre tracts undergoing Wallerian degeneration compared to the PNS, in the face of a mononuclear phagocyte response which is similar in relative magnitude and time course, is because Schwann cells in degenerating peripheral nerves promptly modify their myelin sheaths such that they can be recognized and phagocytosed by macrophages, whilst in the CNS oligodendrocytes do not.

The immunological potential of apoptotic debris produced by tumor cells and during HIV infection

Apoptosis is a major cause of cell death in health and disease. In contrast to necrosis, apoptosis does not induce an inflammatory response and the cellular debris produced by apoptosis has been assumed to be biologically inert. This review challenges this assumption by suggesting that apoptotic debris (especially in the context of growing tumors or during HIV infection) may have immunological activities, mainly immunosuppressive but perhaps also immunostimulatory. In many cases, the surface of apoptotic cells differs from normal cells in that phosphatidylserine (PS) is aberrantly exposed on the external face of the cell membrane. Liposomes composed of PS may down-modulate macrophage anti-leishmanial activities, suppress macrophage TNF production, suppress lymphocyte proliferation, and increase macrophage proliferation. "Membrane shedding" has been described in certain malignancies where apoptosis may be occurring, and the shed tumor membrane vesicles have been shown to reduce MHC class II expression on macrophages and decrease lymphocyte responsiveness, perhaps because of their ganglioside content. Finally, the apoptotic debris from HIV-infected cells may bear on its surface viral proteins which contain immunosuppressive peptide sequences. This debris may also use viral envelope proteins to fuse into macrophages and thereby avoid phagocytosis and lysosomal destruction. These considerations suggest that the flux of apoptosing cells and debris through the immune system that occurs during tumor growth and HIV infection should not be assumed to be immunologically neutral. In particular, HIV-related apoptosis may have immunosuppressive effects in addition to the numerical depletion of lymphocytes.

Development of microglia in the quail optic tectum

The development of microglia in the quail optic tectum from embryonic day 6 to adulthood was studied by using the QH1 monoclonal antibody. In youngest tecta, microglial cells were scarcely present, but their number rose in subsequent stages. A clear pattern of microglial cell distribution was observable in embryos of 9-16 days. (1) Round cells appeared close to the ventricular layer. (2) Large numbers of ameboid and round labeled cells were seen in the stratum album centrale during development. A gradient of cell density was observable in this layer, as fewer labeled cells appeared in medial regions of the tectum than in lateral regions. (3) Maturing ramified cells were found in layers external to the stratum album centrale, where they increased in number and in branching complexity during development. In adult tecta, almost all microglial cells were of the mature ramified type and were distributed homogeneously in the different tectal layers, although in some layers they had particular morphological features. The distribution of microglia in the developing tectum and in adjacent regions provided insight into the routes of microglial cell invasion of the tectum during development. Apparently, a proportion of microglial cells reached the tectal parenchyma from the meninges and from the ventricular lumen, but the majority of them migrated along nerve fiber tracts from their entry point at the pial surface of the ventromedial caudal tectum. After they reached the stratum album centrale, microglial cells continued their migration toward more external layers, where they differentiated into ramified microglia.

Upregulation of the macrophage scavenger receptor in response to different forms of injury in the CNS

The monoclonal antibody 2F8 was used to localize the macrophage scavenger receptor by immunohistochemistry. In control adult mice, macrophage scavenger receptor expression in the brain was restricted to stromal and epiplexus macrophages of the choroid plexus, meningeal macrophages and to perivascular sites. Microglia did not express the receptor. In the developing mouse brain, macrophage scavenger receptor expression was high on meningeal macrophages and detectable on immature microglia in the supraventricular corpus callosum, cingulum, cavum septum and the periaqueductal area. In the aged mouse brain, the pattern of macrophage scavenger receptor expression was no different from that in the young adult brain. Macrophage scavenger receptor expression on resident microglia and recruited macrophages was detected 24 h after an intrahippocampal injection of either lipopolysaccharide or kainic acid. Macrophage scavenger receptor expression was also detected in microglia 3 days after optic nerve crush both in the nerve segment distal to the crush site and in the superior colliculus. These studies indicate a potential role for the macrophage scavenger receptor in the CNS in the clearance of debris during acute neuronal degeneration.

Selective expression of clusterin (SGP-2) and complement C1qB and C4 during responses to neurotoxins in vivo and in vitro

This study concerns expression of the genes encoding three multifunctional proteins: clusterin and two complement cascade components, C1q and C4. Previous work from this and other laboratories has established that clusterin, Clq and C4 messenger RNAs are elevated during Alzheimer's disease, and in response to deafferenting and excitotoxic brain lesion. This study addresses hippocampal clusterin, ClqB and C4 expression in response to neurotoxins that caused selective neuron death. Kainate, which preferentially kills hippocampal CA3 pyramidal neurons but not dentate gyrus granule neurons induced clusterin immunoreactivity in CA1 and CA3 pyramidal neurons and adjacent astrocytes, but not in dentate gyrus granule neurons. In contrast, colchicine, which preferentially kills the dentate gyrus granule neurons, induced clusterin immunoreactivity in the local neuropil as punctate deposits, but not in the surviving or degenerating dentate gyrus granule neurons. Clusterin messenger RNA was increased in astrocytes. ClqB and C4 messenger RNAs increased within 48 h after kainate injections, particularly in the CA3 pyramidal layer, less in the dentate gyrus-CA4, and less in CA1. Clq immunoreactivity was detected in CA1 pyramidal neurons and also as small punctate deposits in the CA1 region at eight and 14 days after kainate. The increase of both clusterin and ClqB messenger RNAs after kainate injections was blocked by barbiturates that prevented seizures and neurodegeneration. In primary hippocampal neuronal cultures treated with glutamate, a subpopulation of cultured neurons that survived glutamate toxicity also had parallel elevations of clusterin and ClqB messenger RNA. In conclusion, cytotoxins that target selective hippocampal neurons increase the expression of both clusterin and ClqB in vivo and in vitro. These results show that elevations of clusterin messenger RNA or protein can be dissociated from each other and from cell death. These increased messenger RNAs were associated with immunoreactive deposits that differed by cell type and intra- versus extracellular locations. These results suggest that the complement system is involved in brain responses to injury.

Interleukin-4 induces proliferation and activation of microglia but suppresses their induction of class II major histocompatibility complex antigen expression

We recently found that microglia, brain macrophages, express interleukin-4 (IL-4) receptor mRNA in vitro. Since IL-4 exhibits a variety of functions on the cells of monocyte-macrophage lineage, we examined the effects of IL-4 on the functions of microglia. Recombinant IL-4 induced the proliferation of microglia in a dose- and time-dependent manner as determined by MTT colorimetric assay, [3H]thymidine uptake and bromodeoxyuridine (BrdU) incorporation. IL-4 also synergistically enhanced the proliferation of microglia with such colony-stimulating factors as IL-3, granulocyte-macrophage colony-stimulating factor (GM-CSF) and macrophage colony-stimulating factor (M-CSF). It also increased acid phosphatase activity and superoxide anion formation by these cells. Despite these positive effects on proliferation and activation, IL-4 suppressed the IFN gamma-induced class II MHC antigen expression in these cells. Since these effects of recombinant IL-4 were inhibited by the addition of monoclonal antibody against IL-4 receptors, the effects of IL-4 on microglia appear to be a specific function via IL-4 receptors. Although microglia and astrocytes produce a variety of immunoregulatory cytokines, neither cell produced IL-4 as determined by bioassay or detection of IL-4 mRNA by RT-PCR method. Thus, the exogenous IL-4 may contribute to the accumulation of microglia in or around inflammatory lesions in the central nervous system, and may be involved in the regulatory mechanisms of microglia.

Cytokine induction during T-cell-mediated clearance of mouse hepatitis virus from neurons in vivo

To investigate the mechanism by which viruses are cleared from neurons in the central nervous system, we have utilized a mouse model involving infection with a neurotropic variant of mouse hepatitis virus (OBLV60). After intranasal inoculation, OBLV60 grew preferentially in the olfactory bulbs of BALB/c mice. Using in situ hybridization, we found that viral RNA localized primarily in the outer layers of the olfactory bulb, including neurons of the mitral cell layer. Virus was cleared rapidly from the olfactory bulb between 5 and 11 days. Athymic nude mice failed to eliminate the virus, demonstrating a requirement for T lymphocytes. Immunosuppression of normal mice with cyclophosphamide also prevented clearance. Both CD4+ and CD8+ T-cell subsets were important, as depletion of either of these subsets delayed viral clearance. Gliosis and infiltrates of CD4+ and CD8+ cells were detected by immunohistochemical analysis at 6 days. The role of cytokines in clearance was investigated by using an RNase protection assay for interleukin-1 alpha (IL-1 alpha), IL-1 beta, IL-2, IL-3, IL-4, IL-5, IL-6, tumor necrosis factor alpha (TNF-alpha), TNF-beta, and gamma interferon (IFN-gamma). In immunocompetent mice there was upregulation of RNA for IL-1 alpha, IL-1 beta, IL-6, TNF-alpha, and IFN-gamma at the time of clearance. Nude mice had comparable increases in these cytokine messages, with the exception of IFN-gamma. Induction of major histocompatibility complex class I (MHC-I) molecules on cells in infected brains was demonstrated by immunohistochemical analyses in normal and nude mice, suggesting that IFN-gamma may not be necessary for induction of MHC-I on neural cells in vivo.

A comparative Mac-1 immunocytochemical and lectin histochemical study of microglial cells in the normal and athymic mice

The number of microglial cells in the supraventricular part of the corpus callosum stained with Mac-1 antibody (against CR3 antigens) and the intensity of staining were studied in both the homozygous athymic nude mouse (nu/nu) and normal BALB/c mouse (+/+). For quantitative analysis, the mean microglial cell counts (expressed in terms of packing density) from 40 microns thick immunostained sections were obtained and tested by analysis of variance. The Mac-1 positive cells in neonatal nude mice were slightly less intensely stained than those of their normal littermates. Such was not noticeable in the 13-week- and 1-year-old animals. The mean number of immunopositive microglial cells per 0.0324 mm2 was significantly less in the 5-day-old (P < 0.001) and 13-week-old (P < 0.05) nude mice when compared to normal mice of corresponding ages. The difference was insignificant in the 1-year-old nude and normal mice. The distribution of Mac-1 labelled microglia in different areas of the brain of the postnatal nude and normal mouse was also examined. In the brain areas examined, e.g., the olfactory bulb, cerebral and cerebellar cortex, the number of microglia in the nude mouse was considerably reduced. The study of lectin labelled sections also showed a much smaller number of labelled microglial cells in the athymic mouse. This was especially obvious in the 5-day-old nude mouse when compared to the normal BALB/c mouse (P < 0.01).(ABSTRACT TRUNCATED AT 250 WORDS)

Fluid-phase endocytosis of horseradish peroxidase by cerebral endothelial cells in primary culture: characterization and kinetic analysis

Endocytosis of horseradish peroxidase (HRP) was studied in primary cultures of cerebral endothelial cells prepared from 2-week-old rats. These cultures were considered "endothelial-like" on the basis of their ability to internalize acetylated low density lipoprotein. Cellular localization of HRP protein was examined at the light and ultrastructural levels and endocytosis of the protein was evaluated by a colorimetric assay. HRP was localized in discrete cytoplasmic granules by light microscopy. At the ultrastructural level these granules corresponded to pleomorphic membrane-bound structures that were present throughout the cytoplasm. The amount of internalized HRP was directly related to the concentration of the protein in the medium, was not saturable at high concentrations of HRP, and increased with time. Endocytosis proceeded at 37 degrees C, but was abolished at 4 degrees C. In pulse-chase experiments, the quantity of internalized protein in the cells did not significantly change during the 2 hr chase period. Taken together, these findings suggest that internalization of HRP occurs by fluid-phase endocytosis, a non-receptor-mediated process, and that the protein is stable within an intracellular compartment for at least several hours.

Motoneurons deprived of trophic support in vitro require new gene expression to undergo programmed cell death

During normal development, large numbers of neurons die by programmed cell death. This phenomena has been extensively studied in the lateral motor column of chick embryos, where approximately 50% of the motoneurons that are initially produced, subsequently die due in part to competition for a limited supply of target-derived trophic support. Inhibitors of RNA and protein synthesis block this cell loss in vivo, indicating a requirement for new gene expression (Oppenheim et al., 1990). Prior to their commitment to death, motoneurons can be isolated as a relatively pure population from chick spinal cord for in vitro study. Cells plated with muscle extract, a potent source of target-derived trophic support, survive, and have large, phase-bright cell bodies and extensive neurite outgrowth. In contrast, motoneurons cultured in the absence of muscle extract die within 48 h. This death can be blocked by the RNA synthesis inhibitor actinomycin D, at the time when the cells become committed to die, suggesting that new gene expression is required for cell death. DNA fragmentation and nuclear condensation indicate that some of these cells die by apoptosis. Therefore, it appears that many aspects of motoneuron development observed in vivo can be reconstituted in vitro. These cultures can be used as a model system for studying neuronal death and may contribute to an understanding of the molecular mechanisms that mediate programmed cell death during neuronal development.

Expression of complement C1qB and C4 mRNAs during rat brain development

This study examined the distribution of complement C1qB and C4 mRNAs during rat brain development by northern blot and in situ hybridization. Both C1q and C4 mRNAs were already present at embryonic day 14 (E14) and showed little change in abundance through six weeks postnatal. At E16, C1qB mRNA was present at high abundance in putative microglia/macrophages in cortical marginal and intermediate zones, and hippocampal analge, but not in the neurogenic ventricular or sub-ventricular zones. C4 mRNA had a broadly similar regional distribution, but was present at lower abundance in a larger number of cells, putatively neurons. The distribution pattern for C1qB and C4 mRNAs did not change appreciably as brain development proceeded. The lower prevalence of C mRNAs in neuroepithelial or subventricular zones suggests an inverse relationship of C mRNA to cell proliferation. The frequency of apoptotic nuclear profiles, which was as much as ten-fold higher at P7 vs. E17, did not correlate anatomically with C1qB or C4 mRNA levels. Thus, the widespread distribution and consistent presence of each C mRNA during development argues against a role for C in programmed cell death during brain development. We suggest that C1q and C4 components have novel roles during brain development that may be unrelated to normal cytotoxic actions of the activated classical C cascade.

Infection of murine fetal brain cell cultures with Listeria monocytogenes

The uptake of Listeria monocytogenes by different brain cells was studied in primary dissociated brain cell cultures derived from murine fetuses. In respect to the supposed intraaxonal migration of Listeria monocytogenes in the pathogenesis of listeric focal brain stem encephalitis, it was examined whether the bacterium was internalized by neurons. Infection rates of distinct cell types were determined by double immunofluorescence with antibodies against cell type-specific markers and the bacterial pathogen. Because of the changing composition of the cultures and time-dependent expression of the oligodendrocyte marker galactocerebroside (GC), infections were carried out on day 4, 6, 8, and 15 in vitro. Listeria monocytogenes was detected predominantly within macrophages. Astrocytes, oligodendrocytes, and fibronectin-expressing cells were infected to a lesser extent. The lowest rates of infection were observed in neurons. A tropism of Listeria monocytogenes for neurons was not detected in vitro.

Phagocytic activity of macrophages and microglial cells during the course of acute and chronic relapsing experimental autoimmune encephalomyelitis

The ED1 monoclonal antibody recognizes an antigen in lysosomal membranes of phagocytes. The expression of this antigen in cells increases during phagocytic activity. Here we describe the expression of ED1-immunoreactivity during the various stages of both acute (monophasic) and chronic relapsing experimental autoimmune encephalomyelitis (EAE) in the Lewis rat. During the first attack of acute and chronic relapsing EAE, ED1-immunoreactivity was present in macrophages and in cells which displayed morphologic features of activated microglial cells (i.e., cells with thick short processes). At the ultrastructural level these cells were seen to contain phagocytosed myelin structures in lysosomes. ED1-immunoreactivity in these cells was present in the cytoplasm near lysosomes. During the remission phase of acute EAE and the relapse phase of chronic relapsing EAE, ED1-positive cells with dendritic morphology not only were present in or nearby lesions, but were also found at sites distant from lesions throughout large parts of the brain. These cells had a morphology comparable to microglial cells in normal brain. A major difference between animals which were in remission and animals which on day 25 were suffering from a relapse, was that the latter showed the presence of lesions with darkly stained round ED1-positive macrophages and activated microglial cells. These results indicate that during a relapse, newly recruited blood-borne macrophages infiltrate the brain and, together with activated lymphocytes and microglial cells, recommence a new demyelination process.

Expression of Bcl-2 protein in murine neural cells in culture

To explore the role of the protooncogene bcl-2 in the prevention of programmed cell death in the nervous system, we investigated its expression in mouse neural cells in primary culture. The 26 kDa protein product, Bcl-2, was detected by immunocytochemistry and immunoblotting in cultured neurons, astrocytes and oligodendrocytes, but the immunoreactivity of microglial cells was not detectable by immunoblotting. The subcellular distribution of Bcl-2 was similar between in vivo (brain) and in vitro (culture) and between cultured neurons and astrocytes, while the content was higher in astrocytes than in neurons. The substantial expression of bcl-2 in primary cultured brain cells suggests that it has some physiological control in the brain over programmed cell death, which may be exerted not only in neurons but also in some glial cells such as astrocytes.

Pathophysiology of glial growth factor receptors

Activation and proliferation of glial cells are common events in the pathology of the nervous system. Although we are only beginning to understand the molecular signals leading to glial activation in vivo, there is increasing evidence that growth factors and their receptors may play an important part. In this paper we summarize the data on the pathophysiology of glial growth factor receptors and their ligands in the central and peripheral nervous systems.

Brain macrophages stimulate neurite growth and regeneration by secreting thrombospondin

The presence of macrophages in the developing or lesioned central nervous system (CNS) led us to study the influence of these cells on neuronal growth. Macrophages were isolated from embryonic rat brain and we observed that factors released in vitro by these cells stimulate neurite growth and regeneration of cultured CNS neurons. This effect was inhibited by antibodies directed against thrombospondin, an extracellular matrix protein that we found to be synthesized and released by brain macrophages. Immunodetection of thrombospondin in the adult rat brain lesioned by kainic acid confirmed the production of this protein by brain macrophages and indicated an early intraparenchymal accumulation of thrombospondin following injury. These results suggest that brain macrophages contribute actively to neurite growth or regeneration during the development or in pathological contexts.

Macrophage recruitment during limb development and wound healing in the embryonic and foetal mouse

Macrophages play a pivotal role in the adult inflammatory response to wounding. They are directly responsible for cellular debridement and, by providing a source of growth factors and cytokines, they recruit other inflammatory and fibroblastic cells and influence cell proliferation and tissue remodelling. In this paper we investigate the role of macrophages in clearing areas of programmed cell death in the developing embryo and also their role in embryonic and foetal wound healing. Immunocytochemistry using the monocyte/macrophage-specific monoclonal antibody, F4/80, reveals a close association between areas of programmed cell death in the remodelling interdigital regions of the mouse footplate and of F4/80-positive cells, suggesting that monocyte-derived macrophages, and not locally recruited fibroblastic cells, as previously believed, are responsible for phagocytosing and clearing areas of interdigital apoptosis. Our studies of wound healing reveal that macrophages are not recruited to, and therefore cannot be playing an active role in the healing of, excisional wounds made in the mouse embryo at any stage up until E14.5. Beyond this transition stage we see a significant recruitment of macrophages within 12 hours of wounding. We find that macrophages can be attracted to wounds in earlier embryos if the wound results in significant cell death such as after burning.

Old dyes for new scopes: the phagocytosis-dependent long-term fluorescence labelling of microglial cells in vivo

The nature of the interactions between dying neurons and microglial cells within the developing and injured CNS remains controversial. A new technique for labelling microglial cells is available, which enables further studies of such interactions in a direct way. The value of the method relies on retrograde filling of neurons with vital fluorescent dye, subsequent degeneration of the neurons due to either naturally occurring cell death or as the result of axotomy, and phagocytotic removal of the fluorescent cell debris by microglial cells, which thus become identifiable. The fluorescent dye can be visualized in whole-mounted tissue or after sectioning. Photoconversion of the dye into electron-dense material permits examination of the microglial and dying ganglion-cell interactions at the ultrastructural level. This new principle of the function-dependent, selective fluorescent labelling of phagocytosing microglial cells, which might now be extended to other dyes and to other neurodegenerative models, promises to shed light onto the function of microglial cells within the brain.

Immunocytochemical localization of immunoglobulins in the rat brain: relationship to the blood-brain barrier

The central nervous system has been traditionally regarded as an immunologically privileged area. This feature has been in part attributed to the blood-brain barrier, which provides a restrictive interface to circulating immunoglobulins (IgG). Recent kinetic studies suggest that the barrier to immune proteins is not absolute, but rather may be regulated by a specific transfer mechanism. In this study, we confirm the presence of IgG in the central nervous system by immunocytochemistry and demonstrate a close anatomical relationship between the distribution of this protein and the blood-brain barrier. IgG was immunolocalized in the normal rat brain by using monoclonal and polyclonal antibodies to IgG and its subclasses. On the basis of an initial evaluation, the most appropriate antibodies and dilutions were selected for subsequent analyses. In the first study, IgG and albumin were immunolocalized in adjacent sections. In the second study, horseradish peroxidase (HRP) was given intravenously prior to sacrifice, in order to examine artifacts related to perfusion fixation. The distribution of HRP and IgG was then examined in adjacent sections. In the third study, IgG was immunolocalized in sections of brain after mild traumatic head injury. A monoclonal antibody to IgG2a and a polyclonal antibody to IgG were selected on the basis of specificity and consistent, mutual localization. Distinct, patchy, perivascular staining, infrequently associated with labeled neurons, was noted throughout the brain. Electron microscopy confirmed the perivascular localization; IgG was localized along the basal lamina of microvasculature and within the adjacent parenchyma. Albumin and HRP did not exhibit a similar pattern of perivascular immunostaining. After head injury, prominent immunostaining for IgG was observed in the injured hemisphere. In summary, these data indicate that the normal rat brain contains IgG, which dramatically increases after head injury. The distinct perivascular distribution in the normal brain suggests local microvascular permeability. This permeability is selective for IgG, since albumin does not share a similar perivascular localization. The neuronal staining which is closely associated with perivascular label may reflect one intracellular route for extravasated IgG.

Bone marrow transplantation corrects the enzyme defect in neurons of the central nervous system in a lysosomal storage disease

Neuronal storage disorders are fatal neurodegenerative diseases of humans and animals that are caused by inherited deficiencies of lysosomal hydrolase activity. Affected individuals often appear normal at birth but eventually develop progressive neurologic symptoms including sensory and motor deficits, mental retardation, and seizures. We have examined efficacy of bone marrow transplantation as a means of enzyme replacement, using cats with the lysosomal storage disease alpha-mannosidosis. Treated animals showed little or no progression of neurologic signs 1-2 years after transplant, whereas untreated cats became severely impaired and reached endstage disease by 6 months of age. Increased lysosomal alpha-mannosidase activity was found in brain tissue of the treated animals, and electron microscopy revealed no evidence of lysosomal storage within most neurons. Histochemical localization of acidic alpha-D-mannoside mannohydrolase (EC 3.2. 1.24), using 5-bromo-4-chloro-3-indolyl alpha-D-mannopyranoside, showed that functional enzyme was present in neurons, glial cells, and cells associated with blood vessels. This study provides direct evidence that bone marrow transplantation as treatment for a neuronal storage disease can lead to significant levels of a missing lysosomal hydrolase within neurons of the central nervous system and to compensation for the genetic metabolic defect.