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

Intracerebral injection of proinflammatory cytokines or leukocyte chemotaxins induces minimal myelomonocytic cell recruitment to the parenchyma of the central nervous system

Neither excitotoxic neurodegeneration nor lipopolysaccharide induces an acute myelomonocytic exudate in the murine central nervous system (CNS) parenchyma (Andersson, P.-B., V. H. Perry, and S. Gordon. 1991. Neuroscience, 42:201; Andersson, P.-B., V. H. Perry, and S. Gordon. 1992. Neuroscience 48:169). In this study formyl-methionyl-leucyl-phenylalanine, platelet-activating factor, interleukin 8 (IL-8), IL-1, or tumor necrosis factor alpha were injected into the hippocampus to assess whether these leukocyte chemotaxins and known mediators of recruitment could bypass this block. They induced morphologic activation of microglia and widespread leukocyte margination but little or no cell exudation into the CNS parenchyma. By contrast, there was acute myelomonocytic cell recruitment to the choroid plexus, meninges, and ventricular system, comparable to that in the skin after subcutaneous injection. The normal CNS parenchyma appears to be a tissue unique in its resistance to leukocyte diapedesis, which is shown here to be at a step beyond chemotactic cytokine secretion or induction of leukocyte adhesion to cerebral endothelium.

Specific transcellular staining of microglia in the adult rat after traumatic degeneration of carbocyanine-filled retinal ganglion cells

The present work was undertaken to assess the fate of ganglion cell debris in the axotomized retina of adult rats and employed a new technique to label phagocytosing microglia via the internalized material. In the main experiment, transection axotomy was performed on the intraorbital segment of the optic nerve, and a fast-transported, vital fluorescent styryl dye (4Di-10ASP) was deposited at the ocular stump of the nerve in order to pre-label retrogradely the ganglion cells destined to die because of the axotomy. Optic nerve transection resulted in progressive degradation of ganglion cell axons, perikarya, and dendrites within the retina and in release of fluorescent material, which was then incorporated into cells identified as microglia. No other retinal cells stained, although astrocytes and Müller's cells also responded to neuron degeneration by accumulating glial fibrillary acidic protein. Incorporation of labelled material into microglia topo-chronologically paralleled the ganglion cell degeneration starting within the optic fibre layer (OFL) and proceeding towards the ganglion cell layer (GCL) and the inner plexiform layer (IPL) of the affected retina. Long-term labelling of microglia monitored up to 3 months after optic nerve transection indicated that labelled microglial cells persisted within the retina. Microglia displayed a strong territorial arrangement within the GCL and IPL, and staggered, bilaminated distribution in both layers. These studies directly prove that microglia in the retina can be transcellularly labelled during traumatic degeneration of ganglion cells. The findings suggest that microglial cells play an important role in axotomy-induced wound healing and removal of cell debris.

Microglial cell response to neuronal degeneration in the brain of brindled mouse

Reactive changes of microglia in response to neuronal degeneration were investigated in the brains of brindled mottled mice with immunocytochemical technique. This mutant has a genetic defect in copper metabolism and spontaneous neuronal degeneration develops around postnatal day 10, in particular in the parasagittal regions of the cerebral cortex and thalamus. The antibodies to macrophage specific antigen, F4/80 and to type-three complement receptor, Mac-1 were used for the study. Reactive morphological changes of microglia, which are immuno-reactive to the antibodies to F4/80 and/or Mac-1, were demonstrated in areas corresponding to those of neuronal degeneration, coincident with the emergence of cells expressing major histocompatibility complex class II, Ia, antigen. Some of the Ia expressing cells had morphological features of ramified microglia, while others were rod shaped with few processes and were mostly located in the perivascular regions. The focal nature of such cellular changes suggests that signal(s) from the degenerating neurons may be responsible for microglial activation and cellular expression of the Ia antigen in the brain of the brindled mouse.

The acute inflammatory response to lipopolysaccharide in CNS parenchyma differs from that in other body tissues

Acute inflammation is important for defence against infection, wound repair and the mediation of auto-immune tissue destruction. Myelomonocytic recruitment in acute inflammation is a stereotyped and non-specific response to tissue insult which begins within 2 h. In this study, lipopolysaccharide was injected into the murine CNS and other body sites of mice to compare the inflammatory responses. Doses of lipopolysaccharide which induced typical myelomonocytic recruitment in skin and the choroid plexus had no effect in CNS parenchyma, apart from the morphological activation of local resident microglia. The CNS parenchymal response proceeded independently of that in the choroid plexus-cerebral ventricles and had three distinct and unique phases. Initially there was minimal neutrophil exudation and a two-day delay before any increase in macrophage-microglial cell number. Next, there was a rapid increase in macrophage-microglial cell numbers during the third day, mainly due to recruitment of blood monocytes. During this phase, leukocyte recruitment was restricted to monocytes which rapidly adopted the arborized microglial phenotype. Monocytes migrated through an intact blood-brain barrier independent of changes in solute permeability. Finally, there was a florid myelomonocytic reaction predominantly in the white matter, one week after intracerebral injection of 2 micrograms lipopolysaccharide. At this time, the leukocyte reaction disrupted the blood-brain barrier, mononuclear phagocytes expressed macrophage morphology and abundant major histocompatibility complex Class II antigen, and T lymphocytes were present. Myelomonocytic entry into the CNS was partially inhibited by prior blockade of the type 3 complement receptor, known to mediate leukocyte adhesion to endothelium elsewhere. The processes which lead to rapid myelomonocytic recruitment in other tissues are absent in CNS parenchyma. Understanding the molecular mechanisms responsible could have considerable significance both for CNS pathophysiology as well as possible anti-inflammatory therapeutic application elsewhere in the body.

Turnover of resident microglia in the normal adult mouse brain

We undertook this study to determine whether the microglia, the resident macrophages of the central nervous system, turn over in the steady-state. The turnover of brain macrophages would lend support to the "Trojan Horse" hypothesis of central nervous system infection, since one origin of replacement cells is the circulating monocyte pool. We combined the immunohistochemical detection of F4/80, a specific macrophage marker, with [3H]thymidine incorporation and autoradiography in normal adult mice. We could detect double-labelled cells in the brains of mice perfused 60 min after isotope administration. Such cells were few in number, randomly scattered throughout the brain and had the morphology of typical resident cells. The labelling index at this survival time was 0.052 +/- 0.003%. Thus resident microglia can synthesise DNA in situ. After longer survival times, we detected larger numbers of double-labelled cells. F4/80+ cells with resident morphology, mitotic figures, pairs of closely apposed (daughter) cells and cells with rounded macrophage-like morphology, all exhibited silver labelling. Twenty-four hours after isotope administration the labelling index was 0.192 +/- 0.052%. From morphologic evidence and comparison of labelling indices at different survival times, we concluded that: (i) resident microglia can synthesise DNA and go on to divide in situ; (ii) cells are recruited from the circulating monocyte pool through an intact blood-brain barrier and rapidly differentiate into resident microglia. We estimate that the two processes contribute almost equally to the steady-state turnover of resident microglia.

Treatment of inherited neurometabolic diseases: the future

The past 10 years' experience with bone marrow transplantation from normal, immunologically compatible donors indicates its possible use in various neurometabolic diseases, particularly in a patient who has not suffered irreparable brain damage. This experience may be a prelude to treatment by somatic gene therapy. This can be applied as an autologous bone marrow transplant, grafting the patient's own stem cells inserted with the normal gene. Although somatic gene therapy will be relatively easy for tissues with dividing cells, its application to target tissues with little or no cell division may pose difficulties. Meanwhile, techniques for the preservation, culture, and grafting of fetal neurons in humans have been developed and have been used in the treatment of Parkinson's disease. These procedures could readily be transferred to the treatment of other neurodegenerative diseases that cause significant morbidity, but ethical, legal, and religious considerations must be taken into account. All these efforts promise novel and improved management of inborn neurometabolic errors.

Identification of elastase as a secretory protease from cultured rat microglia

In the course of studying the secretory products of microglia, we detected protease activity in the conditioned medium. Various proteins (casein, histone, myelin basic protein, and extracellular matrix) were digested. The protease activity was characterized by using purified myelin basic protein as a substrate. Maximal activity was observed at neutral pH levels (7-8), which was different from the optimum pH level of proteolytic activity observed in the cell homogenate. The activity was inhibited approximately 60 and 50% by 1 mM phenylmethylsulfonyl fluoride and 40 microM elastatinal, respectively. In gel filtration, the major activity, which was inhibited in the presence of N-methoxysuccinyl-Ala-Ala-Pro-Val-methyl chloride, eluted at a position corresponding to a molecular mass of approximately 25 kDa. These results suggest that the major protease present in microglial conditioned medium is elastase or an elastase-like protease. This suggestion was confirmed by the finding that the 25-kDa protein band was stained with anti-elastase antiserum by western blotting. De novo synthesis of elastase in microglia was supported by [35S]methionine incorporation. In the presence of lipopolysaccharide, the secretory elastase decreased. These results demonstrate that microglia secrete proteases, one of which was identified as elastase. The significance of this enzyme production in physiological and pathological conditions is discussed.

A role for microglia in the maintenance of photoreceptors in retinal transplants lacking pigment epithelium

Studies on intact retina have pointed to a necessary role for retinal pigment epithelium in the maintenance of photoreceptor outer segments and for regeneration of visual pigment. However, it has been shown that when embryonic retinae are separated from the pigment epithelium and transplanted into the brain of neonatal rats, the transplanted photoreceptors develop outer segments and the retina responds to light in the apparent absence of pigment epithelial cells. We confirm that there are no retinal pigment epithelium cells associated with transplanted retinae in the present series of experiments and show that a row of cells, composed predominantly of microglia of host origin, border the graft. These cells can be seen to contain engulfed outer segments when they are apposed to the outer retina, suggesting that the microglia have assumed, at the least, the phagocytic function normally associated with retinal pigment epithelium. Microglial cells and their processes are also found within the transplant, but these cells are typically devoid of phagosomes, indicating an absence of phagocytic activity. The close physical association of these resting microglia with the transplant may facilitate their role in antigen presentation under specific conditions of immune provocation.

Brain macrophages: evaluation of microglia and their functions

There is now evidence approaching, if not having already surpassed, overwhelming in support of microglial cells as macrophages. Consistent with this cellular identity, they appear to arise from monocytes in developing brain where amoeboid microglia function in removing cell death-associated debris and in regulating gliogenesis. In normal adult tissue, ramified microglial cells with down-regulated macrophage functional properties may serve a constitutive role in cleansing the extracellular fluid. Under all conditions of brain injury, microglia appear to activate and convert into active macrophages. Activated and reactive microglia participate in inflammation, removal of cellular debris and wound-healing, the latter through regulation of gliosis in scar formation and a potential contribution to neural regeneration and neovascularization. In the activated state, microglia also express MHC's and, thus, may function in antigen presentation and lymphocyte activation for CNS immune responses. As uniquely adapted tissue resident macrophages within the CNS, microglia serve a variety of functional roles over the lifespan of this tissue. These cells may therefore be involved in or contribute to some disease states; such has been indicated in multiple sclerosis and AIDS dementia complex.

Determination of the origin and nature of brain macrophages and microglial cells in mouse central nervous system, using non-radioactive in situ hybridization and immunoperoxidase techniques

The origin and nature of brain macrophages and microglial cells in the mouse central nervous system (CNS) were investigated. First, the expression and localization of determinants recognized by the different monoclonal antibodies (mAbs) MOMA-1, Mac-1-alpha, and F4/80 (raised against cells of the mononuclear phagocyte system) were immunohistochemically studied in the developing and adult mouse brain. In order to clarify the origin of brain macrophages and microglial cells, we used bacteriophage lambda transgenic mice as donors for bone marrow transplantations in recipient mice of different ages. During ontogeny, numerous MOMA-1-, Mac-1-alpha-, and F4/80-positive blood monocyte-derived brain macrophages (amoeboid microglia) infiltrated the CNS parenchyma. These brain macrophages gradually disappeared from the brain parenchyma at postnatal day 7 (P7). From P17 on, Mac-1-alpha- and F4/80-positive cells were detected within the brain parenchyma with the morphology of resting microglial cells. Transitional forms between brain macrophages and "resting" microglia were not observed in the developing brain. Combined non-radioactive in situ hybridization and immunohistochemistry revealed many MOMA-1-positive bone marrow-derived brain macrophages that were located in the leptomeninges, the ventricles, and occasionally the blood vessel walls. These results show that brain macrophages are of bone marrow origin. Many "resting" microglial cells were detected in the brain, mainly in the white matter. It appeared that about 10% of these cells displayed the transgenic signal. This result indicates that the majority of "resting" microglial cells are of local, presumably neuroectodermal, origin.

Macrophage-like cells invading the suboptic necrotic centres of the avian embryo diencephalon originate from haemopoietic precursors

Macrophage-like cells have been previously shown within the suboptic necrotic centres of chick embryos during the period just previous to, and coinciding with, growth of the earliest optic axons through suboptic necrotic centres. In this paper, light and electron microscopy observations of chick embryos suggest that these macrophage-like cells originate from blood cells. Immunocytochemical techniques in chick-quail yolk sac chemeras, constituted of a chick embryo and a quail yolk sac, revealed that the macrophage-like cells within the suboptic necrotic centres are labelled with anti-MB1 antibody, which is specific for quail haemopoietic and endothelial cell lineage. These findings demonstrate that these phagocytic cells are of blood cell lineage, and originate in the extraembryonic tissues of the yolk sac. Diffuse staining around some suboptic necrotic centre macrophage-like cells suggests that they release MB1 antigens which may play a role in the growth of the optic axons through the suboptic necrotic centres.

Brain macrophages and microglia respond differently to lesions of the developing and adult visual system

Traumatic injury in the brain usually results in rapid degeneration of neuronal elements and a response by peripherally derived macrophages (brain macrophages, BMOs) and resident microglia. One intriguing result of lesions performed in the developing brain as compared to lesions of the mature brain is the faster resolution of the cellular debris and the absence of significant scarring. The purpose of this study was to examine the response of BMOs to induced cell death distant to the lesion site and to investigate possible differences in the responding phagocytic populations (BMOs versus microglia) following lesions in neonates and adults. Ablation of the visual cortex at birth results in very rapid retrograde degeneration and removal of neurons of the dorsal lateral geniculate nucleus (dLGN) within a few days. Lesions to the visual cortex of adult rats also induce neurons within the dLGN to die, but these cells do so over a much more protracted time course. Utilizing differences in morphology and immunocytochemical staining with the monoclonal antibodies ED1 and OX-42 to distinguish between BMOs and microglia, we found that in the developing CNS, BMOs are signalled rapidly and specifically to the location of induced cell death. Microglia are not involved in this response. As might be expected, the temporal response in the adult is much more protracted. In contrast to the developing brain, microglia and not macrophages are the predominant responding cell class after the adult lesion. The data suggest that these are distinct populations of phagocytic cells that respond to brain damage during development and in the adult, which may be critical in modulating the resolution and growth response after injury.

Increase of macrophage colony-stimulating factor and granulocyte-macrophage colony-stimulating factor receptors in the regenerating rat facial nucleus

Proliferation of microglial cells commonly occurs in the response of the central nervous system to injury, but little is known about how this process is regulated in vivo. Here we have studied the expression of receptors to macrophage colony-stimulating factor (MCSF) and granulocyte-macrophage colony-stimulating factor (GMCSF) in the normal and regenerating rat facial motor nucleus using receptor immunocytochemistry and in situ ligand binding methods. Under normal conditions, immunocytochemical staining with anti-MCSF receptor (MCSFR) antibody revealed a moderate but selective labelling of microglia-like cells of the facial motor nucleus. This immunostaining also colocalized with MUC102, a new monoclonal antibody raised against microglial cells in the rat central nervous system. Axotomy of the facial nerve led to a rapid increase in MCSFR-staining intensity 1 day after injury, which became maximal 7 days postoperatively and then decreased. A similar but somewhat slower increase was also observed for the specific [125I]MCSF binding with a maximum at 7 days. Specific [125I]GMCSF binding also increased, peaking at 4 days postoperatively and then rapidly decreasing to normal levels at 21 days after axotomy. In summary, axotomy of the facial nerve led to a rapid increase in receptors for MCSF and GMCSF, which coincided with the pattern of microglial proliferation in the regenerating facial motor nucleus. This apparent up-regulation of receptors for microglial growth factors may play an important role in preparing the microglia to participate in the cellular response to injury in the regenerating central nervous system.

Differential immunochemical markers reveal the normal distribution of brain macrophages and microglia in the developing rat brain

Brain macrophages and microglia play important roles in central nervous system (CNS) development, especially during regressive events in which particular neuronal and glial constituents are eliminated. The purpose of this study is to provide a complete map of brain macrophage and microglia distribution in all regions of the neuraxis from birth to sexual maturity. We have utilized morphology and immunostaining with the specific antibodies OX-42 and ED1 to distinguish between brain macrophages and microglia. Brain macrophages are large, round cells, 10-15 microns in diameter, with few or no cytoplasmic processes; these cells are ED1- and OX-42-immunopositive. Microglia have small cell bodies with numerous, ramified cytoplasmic processes. These cells are OX-42-positive, and ED1-negative. We found a specific pattern of distribution of brain macrophages, targeting specific cortical and subcortical areas transiently, including developing fiber tracts. These cells disappeared completely by the third postnatal week. In contrast, OX-42-positive microglia exhibited a gradual increase in number and were distributed uniformly throughout gray matter and within white matter tracts. These cells remain in the adult CNS, constituting the resident microglia population. We suggest that these two distinct phagocytic cell populations perform unique functions in the developing brain, including remodeling of restricted CNS areas by brain macrophages that is part of a normal morphological process.

Characterisation of two new monoclonal antibodies directed against rat microglia

With the aid of cultured rat microglial cells as immunogen, we raised two monoclonal antibodies, designated murine clone (MUC) 101 and 102, which recognised subsets of resident microglial cells in the normal central nervous system and cells of the mononuclear phagocyte system in peripheral organs. These antibodies were characterised by immunoperoxidase immunocytochemistry, immunoelectron microscopy, and immunoblotting. The immunostained cells were identified as microglial cells by double-immunofluorescence labelling with the B4-isolectin from Griffonia simplicifolia, an established microglial cell marker. Under normal conditions, both antibodies labeled resident microglia but with different distribution patterns. Under pathological conditions, e.g., after facial nerve transection, they labeled activated, perineuronal microglia in the operated facial nucleus. Immunoelectron microscopy demonstrated a membrane localisation of the antigen recognised by MUC 102. In peripheral organs, MUC 101 and 102 reacted with different cell populations of the mononuclear phagocyte system, particularly in thymus, spleen, and peripheral lymph node. Western blot experiments showed that MUC 101 recognised two proteins of 116 and 95 kD in fractions obtained from operated facial nucleus while MUC 102 reacted with two proteins of 62 and 70 kD molecular weight. These immunocytochemical results 1) confirm the antigenic similarity between microglia and cells of the monocyte-macrophage cell lineage, and 2) indicate that considerable antigen heterogeneity might exist among resident microglia. MUC 101 and 102 could thus become useful for studying the function of microglial cells both under normal and pathological conditions.

The proliferative response of S-100 protein-positive glial cells to injury in the neonatal rat brain

Astroglial proliferative response to unilateral injury of the cerebral hemisphere was studied in newborn rats using [3H]thymidine autoradiography combined with immunocytochemical staining for S-100 protein as an astroglial marker. The animals received [3H]thymidine at different intervals following injury and the distribution of double-labeled cells was recorded 4 h after each [3H]thymidine injection. The reactive proliferation of all cells within the investigated areas was detectable as early as about 2 h after injury. However, the quantitative increase in astroglial mitotic rate could be first detected on day 1 following the injury and the reactive changes were observed until day 8. They were regarded as evidence for the ability of astroglia to proliferate in response to injury in the neonatal brain. Since the astroglial divisions occurred mainly at the lesion site, the post-traumatic gliosis could be considered not only a result of astrocyte migration towards the lesion but also an effect of local mitotic activity.

Cytotoxic Effect of Brain Macrophages on Developing Neurons

Brain macrophages are transiently present in different regions of the central nervous system during development or in the course of tissue remodelling following various types of injuries. To investigate the influence of these phagocytes on neuronal growth and survival, brain macrophages stemming from the cerebral cortex of rat embryos were added to neuronal primary cultures. A neurotoxic effect of brain macrophages was demonstrated by the reduction of the number of neurons bearing neurites within two days of contact between the two cell types. Neuronal death and phagocytosis were also directly observed in video recordings of living cultures. This toxicity involved the production by brain macrophages of reactive oxygen intermediates, as shown by the protective effect of catalase, a scavenger of H2O2. In addition, the respiratory bursts of brain macrophages were stimulated in the presence of neurons. These results suggest that brain macrophages could favour the appearance of neuroregressive events which occur either during neurogenesis or in neurodegenerative diseases, implying intracerebral recruitment of mononuclear phagocytes.

An immunohistological study of macrophages in the human fetal brain

Brains from human fetuses of 13 to 27 weeks gestation have been examined immunohistologically for the presence of macrophages using the marker alpha-1-anti-chymotrypsin. A preliminary study demonstrated this to be a satisfactory marker of brain macrophages, although macrophages were also weakly positive for the more specific marker MAC-387. Macrophages were widely present within the cerebral hemispheres following their rapid accumulation between 14 and 16 weeks of gestation. They were identified in characteristic locations which, in the earliest gestation brains examined at 13 weeks, included the mid-line of the corpus callosum, around the optic tract and at the junction of the external and internal capsules near the apex of the putamen. Subsequently, macrophages were identified in abundance in the internal and external capsules and, by 22 weeks gestation, in the periventricular tissues. Their consistent presence and distribution indicate that at least the majority of these macrophages are a normal feature of the developing brain possibly related to remodelling processes.

The CNS acute inflammatory response to excitotoxic neuronal cell death

Acute inflammation is a stereotyped non-specific response to tissue injury which results in the recruitment of neutrophils and monocytes within minutes. In this study the myelomonocytic and microglial reaction to neuronal destruction following unilateral hippocampal injection of kainic acid neurotoxin was investigated. Despite extensive acute neuronal necrosis and notwithstanding a leaky blood-brain-barrier, there is no neutrophil recruitment and a 2-day delay before any increase in macrophage-microglial cell numbers. Resident microglia are capable of reversible upregulation to an activated morphology and the macrophage-microglial reaction is seen not only at the injection site, but also at distant sites related to the axonal pathways and synaptic terminals of the killed neurons.

Modulation of interleukin-1 and tumor necrosis factor expression by beta-adrenergic agonists in mouse ameboid microglial cells

Brain macrophages (ameboid microglial cells) purified to homogeneity and cultured in vitro synthesize and release IL-1 and TNF upon stimulation with lipopolysaccharide (LPS). This induction can be measured at the levels of transcription and translation. In the present study we have analysed whether certain compounds normally present in the nervous tissue could regulate cytokine production by brain macrophages. We demonstrate that the beta-adrenergic agonist isoproterenol, at a concentration of 10(-7) M; inhibits the LPS-induced transcription and release of TNF alpha. At the same concentration, isoproterenol increases the accumulation of IL-1 alpha and IL-1 beta mRNAs. In spite of its strong effect on IL-1 mRNA accumulation, the adrenergic agonist did not enhance IL-1 activity produced by microglial cells. On the contrary, as is the case for TNF, the LPS-induced production of IL-1 was inhibited by isoproterenol. The effects of isoproterenol on cytokine production specifically involve the beta 2 and not the beta 1 adrenergic receptor. It thus appears (i) that the accumulation of mRNAs coding for TNF alpha on one hand and IL-1 alpha and beta on the other is regulated in two opposite ways by the stimulation of the beta 2-adrenergic receptor and (ii) that mRNA accumulation and cytokine production and secretion are not necessarily coupled.

Murine retrovirus-induced spongiform encephalopathy: productive infection of microglia and cerebellar neurons in accelerated CNS disease

We have examined the pathological lesions and sites of infection in mice inoculated with a highly neurovirulent recombinant wild mouse ecotropic retrovirus (FrCasE). The spongiform lesions appeared initially as swollen postsynaptic neuronal processes, progressing to swelling in neuronal cell bodies, all in the absence of detectable gliosis. Infection of neurons in regions of vacuolation was not detected. However, high level infection of cerebellar granule neurons was observed in the absence of cytopathology, wherein viral protein was found associated with both axons and dendrites. Infection of ramified and amoeboid microglial cells was associated with cytopathology in the brain stem, and endothelial cell-pericyte infection was found throughout the CNS. No evidence of defective retroviral expression was observed. These results are consistent with an indirect mechanism of retrovirus-induced neuropathology.

Acidic fibroblast growth factor (aFGF) is expressed in the neuronal and glial spinal cord cells of adult mice

Fibroblast growth factors (FGFs) are known to be synthesized in the central nervous system (CNS) and to act on CNS cells in vitro, but less is known about their synthesis, expression, and role in vivo. In this work, using specific anti-acidic fibroblast growth factor (aFGF) antibodies, we have shown for the first time, by immunohistochemistry, that aFGF is expressed in spinal cord cells of young adult normal mice. This expression is predominant in the cell nucleus. Using immunohistochemical double staining procedures, we identified the cell type expressing aFGF as neurons, astrocytes, and oligodendrocytes, but for each type, cells were not all positively immunostained.

Leukotrienes in brain: natural occurrence and induced changes

Peptidoleukotrienes (SP-LTs) (both total product and individual LTC4 and LTE4 and LTB4 were measured by radioimmunoassay in cerebrospinal fluid (CSF) collected from the third ventricle of conscious cats. Total SP-LT was expressed as LTE4 after treating samples with crude gamma-glutamyltranspeptidase. Prostaglandin (PG) E2 and thromboxane (TX) B2, the stable metabolite of TXA2, were also assayed in part of the experiments. Under basal conditions, SP-LT and LTC4 were consistently measurable (respectively, 327 +/- 14 and 244 +/- 41 pg/ml), while native LTE4 was below the threshold of the assay (60-280 pg/ml) in most cases. LTB4 was barely detectable (30 +/- 2 pg/ml) or not detectable at all. PGE2 was normally less abundant than TXB2 (31 +/- 4 vs 281 +/- 47 pg/ml). Intracerebroventricular (i.c.v.) administration of arachidonic acid (40 microgram) caused a 4-fold increase in SP-LT levels which was relatively small and transient compared to PGE2 (76-fold) and TXB2 (23-fold), while there was no change in either native LTE4 or LTB4. A similar response was obtained with platelet-activating factor (PAF, 1 microgram i.c.v.), though SP-LT elevation (4-fold) was more persistent. A further rise in SP-LT (9-fold) was noted when PAF administration was preceded by indomethacin (500 microgram i.c.v.), whereas PAF effect was reversed by pretreatment with either the PAF antagonist, BN52021 (1 microgram i.c.v.), or the 5-lipoxygenase inhibitors, U-60,257 (75 micrograms i.c.v.) and L-651,392 (10 mg/kg p.o.). PAF was also effective in causing a 3-fold rise in LTC4. Unlike PAF, pyrogens (endotoxin i.c.v. or i.v.; interleukin-1 i.v.) at doses above threshold for fever had no effect on LT levels in CSF, both in the absence and presence of indomethacin pretreatment. We conclude that SP-LTs are a normal constituent of CSF, LTC4, being the major species. The response to PAF accords with a pathogenetic role of the compounds in inflammatory processes and the reactive changes to injury. No evidence was obtained for the involvement of SP-LTs in the central mechanism of fever.

Application of lectin and B-lymphocyte-specific monoclonal antibodies for the demonstration of human microglia in formalin-fixed, paraffin-embedded brain tissue

To evaluate the usefulness of microglial markers for routine neuropathological material, we studied formalin-fixed, paraffin-embedded human brain tissue with the immunoperoxidase method using the lectin Ricinus communis agglutinin (RCA-1) and four monoclonal antibodies (LN-1, LN-2, LN-3, anti-HLA-DR/alpha). RCA-1 stained resting microglia, but the staining intensity was mostly weak. LN-1 also stained resting microglia in paraffin sections first treated with protease. In contrast to LN-1, RCA-1 stained blood vessels heavily. LN-1 stained resting microglia more markedly than RCA-1 in brains fixed for a prolonged period of time. However, LN-1 recognized a small number of astrocytes in routine paraffin sections. LN-3 reactivity was detected on a few resting microglia, but was intensely expressed on large numbers of reactive microglia in many neurological diseases. Both LN-2 and anti-HLA-DR/alpha labelled microglia, but the reactions were inconsistent. This study suggests that the monoclonal antibodies LN-1 and LN-3 are useful for the demonstration of microglia in paraffin sections, and a combination of these antibodies and the antibody to glial fibrillary acidic protein is recommended in attempting to identify microglia.

Macrophages direct process elongation from adult frog motorneurons in culture

Motorneurons and macrophages have been isolated and identified in primary cultures from adult frog (Rana pipiens) spinal cord. Time-lapse video microscopy revealed that during the first two weeks migrating macrophages contact the growth cones of motorneurons. As they continue to migrate, the motorneuron processes elongate in close association with the moving macrophages. Elongating motorneuron processes are thereby brought into contact with other motorneurons and networks are formed. At later stages, the macrophages die but the motorneurons and the networks survive for at least another two weeks. These experiments show that macrophages can promote a directed elongation of motorneuron processes and suggest that they play a similar role during regeneration in vivo.

Evidence for motility and pinocytosis in ramified microglia in tissue culture

Ramified microglial cells were investigated in primary cultures of dissociated cerebral cortical tissue from rats. The identification of these cells was confirmed through immunohistochemical staining with 7 monoclonal antibodies selective for microglia. While there was significant variation in staining intensity with different antibodies, all stained the identified ramified cells; the antibodies OX-42 and ED1 yielded the most intense immunoreactivity. Based on distinctive morphological features, the microglia could be identified in living cultures where they were monitored using time-lapse video recording. This technique revealed extremely dynamic features of cellular plasticity and motility. Ramified microglia exhibited constant and rapid alterations in the size and shape of their cell body with an associated extension and retraction of processes; concomitantly, the cells moved about in a circumscribed area. These features of plasticity and motility were unique to this cell type, and correlated with OX-42 immunostaining. The microglia also possessed a differentially high level of pinocytotic activity; this too was correlated with OX-42 staining. From the nature of their morphological plasticity and motility, high pinocytosis, and cellular distribution, it is hypothesized that the ramified microglia specifically function as a system of fluid cleansing in normal brain tissue.

Characterization of ramified microglia in tissue culture: pinocytosis and motility

Functional properties of ramified microglia were investigated in primary cultures of rat cerebral cortical cells. These microglia could be readily identified in both fixed and living cultures through previously established features. Based on their destruction by 5 mM L-leucine methyl ester, a high level of intrinsic endocytotic activity was established. When cultures were incubated with fluorescent latex beads to assess phagocytosis, little or no such activity was exhibited by ramified cells. However, when cultures were incubated with dyes or other soluble tracer compounds, these cells always exhibited labeling. This labeling was selective for ramified microglia in the cultures and was demonstrated using a variety of compounds, including trypan blue, lucifer yellow, horseradish peroxidase (HRP), and India ink. Intracellular label could be observed in vesicular structures; this localization corresponded to an active cellular process. Also, cellular labeling was inhibited by the presence of colchicine. These features supported the inference that the labeling was attributable to pinocytosis, and this process appeared to account for the vast majority of endocytotic activity in the ramified microglia. Possible physiological significance of this pinocytotic activity was indicated by the accumulation of various neurotransmitters/modulators: gamma-aminobutyric acid and vasoactive intestinal polypeptide (VIP). Ramified cells in these cultures have been previously noted to exhibit a constant and rapid pattern of motility, which was consistently observed here through time-lapse video recording; pinocytosis and rapid motility were shown to concur in individual cells. Based on their high intrinsic pinocytotic activity and pattern of cellular motility, the ramified microglia specifically are suggested to serve a constitutive function of fluid cleansing within the interstitial spaces of brain tissue.

Neural differentiation-associated generation of microglia-like phagocytes in murine embryonal carcinoma cell line

We have characterized microglia-like cells which appeared at late stages of neural differentiation of the P19 line of embryonal carcinoma cells. The microglia-like cells arose as round cells on top of a bed layer in mixed cultures of retinoic acid-treated P19 cells. The period of initial appearance of these cells coincided with that of the disappearance of neurons, and the generation of these cells persisted thereafter. The P19-derived microglia-like cells morphologically resembled ameboid microglia, and some of these cells took the form of resting microglia. These cells, both in enriched cultures and in mixed cell cultures, showed phagocytic activity towards latex beads. Most of the P19-derived microglia-like cells in enriched cultures reacted with an antibody against Mac-1, a macrophage surface marker, whereas only the round cells on top of the bed layer showed positive Mac-1 immunoreactivity in a mixed P19 culture. Moreover, the microglia-like cells contained non-specific esterase, an enzyme marker for macrophages. These cells showed apparent morphological changes in response to macrophage colony stimulating factor secreted by cultured L929 cells. The above characteristics of the microglia-like cells derived from P19 cells are similar to those of brain microglia in primary culture. The generation of microglia-like cells from P19 cells during neural differentiation may provide insights into the origin and life cycle of brain microglia.

Retrovirus-induced spongiform myeloencephalopathy in mice: regional distribution of infected target cells and neuronal loss occurring in the absence of viral expression in neurons

The Cas-Br-E murine leukemia virus (MuLV) induces a spongiform myeloencephalopathy resulting in a progressive hindlimb paralysis. We have used in situ hybridization with a Cas-Br-E MuLV-specific probe to study viral expression in the central nervous system. Infected cells were concentrated in regions where spongiform lesions and gliosis are detected (lumbosacral spinal cord, brainstem, deep cerebellar regions), suggesting a causative link between the level of virus expression and the degree of pathological changes in this disease. However, viral expression was not in itself sufficient to cause disease, since significant viral expression was observed in regions that did not exhibit pathological changes (cerebellar cortex, hippocampus, corpus callosum, peripheral nervous system). In both diseased and nondiseased regions, endothelial and glial cells were identified as the main target cells. Neurons in diseased regions did not show viral expression. The regional distribution of the spongiform changes appears to be laid down very early following infection, since expression could be detected at 10 days postinfection in regions that become diseased. These results indicate that nonneuronal cells have distinct properties in various regions of the central nervous system and suggest an indirect mechanism of neuronal loss consequent to viral expression in nonneuronal cells.

Microglial progenitors with a high proliferative potential in the embryonic and adult mouse brain

Single cell suspensions, prepared from brain stem, cerebellum, and forebrain parenchyma of embryonic and adult mice, were plated on monolayers of an astroglial cell line derived from a spontaneously immortalized mouse cerebellar culture, the D19 clone. A few of the brain cells adhering to the D19 monolayers were immunoreactive to the Mac-1 antibody, which labels all cells of the monocytic and granulocytic lineages. The Mac-1-positive cells proliferated vigorously and later most of them acquired the F4/80 epitope specific for macrophages and microglia cells. Studies in clonal conditions allowed development of large colonies of about 2 x 10(5) cells that expressed typical microglia markers. Bone marrow Mac-1-positive cells cocultured on D19 monolayers were also induced to proliferate, whereas peritoneal macrophages were not. D19 astrocytes express macrophage colony-stimulating factor (CSF-1) activity at a high level, and their conditioned media induced the proliferation of brain and bone marrow Mac-1-positive cells. A specific anti-CSF-1 antiserum completely blocked bone marrow macrophage progenitor proliferation and significantly reduced the multiplication of microglial precursors induced by the D19-conditioned medium. These data indicate that the embryonic and adult mouse brain parenchyma contains potential progenitors for microglial cells.

Pinocytotic activity in ramified microglia

Pinocytotic activity was investigated in rat cerebral cortex using the soluble tracers horseradish peroxidase and Lucifer yellow. A subpopulation of cells selectively accumulated both compounds and the labelling was mainly present in pinocytotic vesicles associated with the cell body. Labelled cell bodies were small, round to oval, and distributed in an almost regular array throughout the tissue. Based on distinctive morphological features, some of the labelled cells could be determined as ramified microglia. This identification was confirmed by immunofluorescence staining with the monoclonal antibody OX-42, which specifically recognizes microglia; OX-42 staining consistently co-localized with pinocytotic labelling. The possibility that ramified microglial cells perform a normal function of continuous fluid exchange in brain tissue is discussed.

Macrophages, microglial cells, and HLA-DR antigens in fetal and infant brain

Immunohistochemical reactions for macrophages, microglia, and HLA-DR antigens were tested on frozen sections of necropsy brain tissue from 20 fetuses and infants ranging in age from 18 weeks' gestation to 8 months post term. No primary central nervous system disease was present but there were four cases of sudden infant death syndrome (SIDS). Macrophages were detected in all the samples studied and were located in the germinal matrix zone, in perivascular spaces throughout the brain, and in the leptomeninges and subependymal layer. Well differentiated microglia were present in all cases examined after 35 weeks' gestation and less well ramified forms were seen at earlier stages of gestation. HLA-DR antigens were detected on a small number of macrophages, chiefly in a perivascular location, in all but three cases. The fewest reactive cells and the weakest reactions occurred in the youngest fetuses. One case of SIDS showed increased foci of microglia in perivascular white matter: this case and one other case of SIDS were the only cases with well ramified microglia that expressed HLA-DR antigens. These findings may be relevant to an understanding of local immune responses in fetal brain infections, including human immunodeficiency virus infection.

Human microglial cells: characterization in cerebral tissue and in primary culture, and study of their susceptibility to HIV-1 infection

Neuropathological studies have shown that human immunodeficiency virus type 1-infected cells within the brain express several markers characteristic of macrophages and could either be microglial cells, or monocytes invading the CNS, or both. To better define the target cells of human immunodeficiency virus type 1 within the brain, we have studied human microglial cells, both in vivo and in vitro, and compared them to monocytes for their antigenic markers and their susceptibility to human immunodeficiency virus type 1 infection. Brain-derived macrophages were isolated from primary cortical and spinal cord cultures obtained from 8 to 12-week-old human embryos. The isolated cells presented esterase activity, phagocyted zymosan particles, expressed several (Fc receptors, and CD68/Ki-M7 and CD11b/CR3 receptors) of the macrophagic antigenic markers, and appeared to be resident microglial cells from human embryonic brain. Conversely, brain-derived macrophages did not express antigens CD4, CD14, or CD68/Ki-M6, which are easily detected on freshly isolated monocytes. Using these antigenic differences between isolated microglial cells and monocytes, we have observed that two populations of macrophages could be individualized. In the normal adult brain, microglial cells were numerous in both the gray and the white matter. The infrequent cells sharing antigens with monocytes were found almost exclusively around vessels. In 8 to 12-week-old human embryos, microglial cells were found in both the parenchyma and the germinative layer. Cells sharing antigens with monocytes were only found at the top of and inside the germinative layer. In brain tissue from patients with human immunodeficiency virus type 1 encephalitis, cells sharing antigens with monocytes are abundant not only around the vessels but also in the parenchyma. In double-labeling experiments, human immunodeficiency virus type 1-infected cells showed monocyte antigens. Finally, microglial cells also differ from monocytes in their in vitro susceptibility to human immunodeficiency virus type 1 infection; after stimulation by r-TNF alpha or GmCSF, monocytes but not microglial cells can replicate human immunodeficiency virus type 1. This in vitro difference in human immunodeficiency virus type 1 susceptibility between monocytes and microglial cells together with the presence of monocytic antigens within the brain tissue of human immunodeficiency virus type 1-infected patients suggest that human immunodeficiency virus type 1-infected cells within the brain are either monocytes that have crossed the blood-brain barrier and spread through the tissue or perivascular microglial cells that, after phagocyting infected blood lymphocytes, subsequently contain viral antigen and migrate to brain tissue.

Microglial cells are a component of the perivascular glia limitans

The ultrastructural relation between microglial cells and cerebral blood vessels was studied in rat brains by immune electron microscopy using antibodies against the common leukocyte antigen (Ox1), the complement receptor 3 (Ox42), and against class I and class II histocompatibility antigens (MHC antigens; Ox3, Ox6, Ox18, and I1-69). Microglial cell processes were found incorporated between the astrocytic foot processes of the glia limitans in 4-13% of cerebral microvessels. After intravenous injection of gamma-interferon, either alone or in combination with tumor necrosis factor, these microglial cell processes expressed classes I and II MHC antigens. Studies in (Lewis X DA)F1-DA bone marrow chimeras demonstrated that these cell processes belonged to resident microglia. This study suggests that microglial cells may play an important role in antigen recognition at the blood-brain barrier.

The distribution of microglia and cell death in the fetal rat forebrain

The appearance and distribution of microglia in the fetal and early postnatal rat forebrain have been examined with the aid of a peroxidase-conjugated lectin derived from Griffonia simplicifolia. This distribution has in turn been correlated with that of pyknotic figures in the same Nissl-counterstained sections. Round and ameboid microglia may be recognised in the fetal forebrain as early as E11, at a stage when the telencephalic vesicles are beginning to develop. By E13, concentrations of round microglia are found at the dorsal and rostral limits of the diencephalic vesicle (dorsal lamina terminalis) and in the adjacent medial walls of the telencephalic vesicles. These cells are often seen to have pyknotic material within their cytoplasm. Microglia remain concentrated in this region until E17. From E15, blood vessels and round and ameboid microglia concentrate in the region of the future hippocampus and appear to be drawn into the hippocampal fissure as the cortical plate folds to form Ammon's horn. At E15, ameboid microglia are also concentrated in the developing fornix, which first becomes apparent at this age. Microglia remain concentrated in the septomesocortical junction area, and may contribute to the concentrations of microglia previously reported in the region of the developing corpus callosum and cavum septi pellucidi. Microglia probably concentrate in the dorsal lamina terminalis and medial telencephalon at E13 in response to the cell death noted in this region, but other concentrations of microglia in the forebrain are not accompanied by similar aggregations of cell death. These findings indicate that the junction of the telencephalon and rostral diencephalon attracts concentrations of microglia from E13 throughout fetal and early postnatal life, coincident with the infolding of the hippocampus (E13-E19) and several days before the development of the corpus callosum (from E19 onwards).

Persistence of fluoro-gold following degeneration of labeled motoneurons is due to phagocytosis by microglia and macrophages

When the neural tracer Fluoro-Gold is used to retrogradely label a population of axotomized neurons, cellular labeling can persist in the axotomized nucleus even when Nissl staining indicates that the injured neurons have degenerated. In order to determine the identity of the labeled cells that remain, this study combines retrograde transport of Fluoro-Gold with immunocytochemical methods for identification of specific non-neuronal cell types following peripheral axotomy and Fluoro-Gold labeling of motoneurons in the dorsal motor nucleus of the vagus in neonatal and adult rats. Fourteen days following cervical vagotomy in neonatal rats, Nissl staining revealed a virtually complete loss of vagal motoneurons. Fourteen days after cervical vagotomy in adult rats, vagal motoneuronal loss was not yet extensive but chromatolysis had clearly begun. Injection of Fluoro-Gold into the vagus nerve just prior to the vagotomy led to Fluoro-Gold labeling of remaining vagal motoneurons. In addition, many other small, brightly labeled cells were present in the lesioned vagal nuclei of all rats. Immunofluorescent identification of astrocytes with anti-glial fibrillary acidic protein and microglia and macrophages with OX42 (anti-C3bi complement receptor) and ED1 (anti-monocyte/macrophage cytoplasmic antigen) demonstrated that the small, bright Fluoro-Gold-labeled cells were non-neuronal, non-astrocytic phagocytes, including microglia. These results indicate that phagocytic microglia and other macrophages sequester Fluoro-Gold in the axotomized dorsal motor nucleus of the vagus of neonatal and adult rats, leading to persistence of fluorescent cellular labeling following the loss of retrogradely labeled axotomized neurons.

Receptors for interleukin-1 (alpha and beta) in mouse brain: mapping and neuronal localization in hippocampus

Interleukin-I receptors were mapped and characterized in mouse brain by quantitative autoradiography using human recombinant [125I]interleukin-I alpha and [125I]interleukin-1 beta as ligands. Both ligands provide identical receptor mapping. In terms of specificity, interleukin-1 alpha and interleukin-1 beta were equally potent in binding competitions assays with either [125I]interleukin-1 alpha or [125I]interleukin-1 beta (EC50 11 pM). These receptors were shown to be highly concentrated in the dentate gyrus, in the choroid plexus at various levels of the brain, in the pituitary and in the meninges. They were also present at low concentrations in the cortex but undetectable in other brain structures. In the dentate gyrus, interleukin-1 receptors were localized on the granular and molecular layers (granule cells) when visualized on slides dipped in nuclear emulsion. Cellular localization of interleukin-1 receptors was assessed using selective lesion by colchicine. The complete loss of [125I]interleukin-1 binding in hippocampal areas where neurons were destroyed by colchicine demonstrates that interleukin-1 receptors are located on granule cells. Following lesion, sparse undestroyed cells, with glial cell morphology, also showed significant labelling. In conclusion, interleukin-1 receptors are located on the granule cells in the mouse dentate gyrus. These neurons may therefore be targets for neuromodulation by interleukin-1 and they may play a key role in the central effect of interleukin-1 as well as in the control of the immune response by the brain.

Fibronectin and laminin regulate the in vitro differentiation of microglial cells

During development, the differentiation of ameboid microglia (brain macrophages) into ramified microglia is marked by a loss of macrophage-like properties and the extension of thin cytoplasmic projections. We have studied the influence of two extracellular matrix proteins, laminin and fibronectin, on microglia differentiation, using cell cultures. Brain macrophages were isolated from primary glial layers derived from embryonic rat brain and further cultured in serum-free medium. The addition of fibronectin induced the transformation of round or spindle-shaped brain macrophages into cells displaying a reduced cell body and extending thin and long processes. This morphological transformation was associated with a reorganization of the vimentin network, including a condensation of dispersed filaments into thick bundles and a modification of the phosphorylation state of vimentin monomers. In addition, compared to brain macrophages, the process-bearing microglia lost the ability to engulf zymosan particles, and showed reduction in non-specific esterase activity and superoxide anion generation. In contrast, laminin reduced the spontaneous transformation of brain macrophages into process-bearing cells. Moreover, laminin and serum induced a reverse transformation of process-bearing cells when added to cultures pretreated with fibronectin. Altogether these results demonstrate antagonist effects of fibronectin and laminin on the in vitro differentiation of brain macrophages towards a "resting" phenotype, which shares several properties with the ramified microglia present in the adult brain. We suggest that fibronectin and laminin regulate the differentiation of microglial cells, which takes place during development or following various types of lesions in the adult brain.

The kinetics and morphological characteristics of the macrophage-microglial response to kainic acid-induced neuronal degeneration

Outside the nervous system myelomonocytic cells are known to play an important role in the inflammatory response and tissue repair after injury. In this study we have examined the myelomonocytic response to neuronal destruction following unilateral injection of the excitotoxin kainic acid into the mouse hippocampus. Intrahippocampal injection of kainate induces rapid, synchronous neuronal death. There is no neutrophil recruitment and a delay of at least 48 h before macrophage-microglial cell numbers increase. The microglial reaction in the injected hippocampus consists of altered morphology, a 6-9-fold increase in mononuclear phagocyte cell numbers and enhanced expression of the macrophage-specific plasma membrane antigen, F4/80, assessed immunohistochemically and by Western blotting. Microglia also respond at distant sites related to the projection pathway and terminals of killed pyramidal cells but the reaction varies in cell numbers, kinetics and morphology. The absence of neutrophil recruitment and the delay in an increase in macrophage or microglial cells shows that the CNS differs from other sites in the body with regard to the kinetics and nature of the myelomonocytic cell inflammatory response. The role of mononuclear phagocytes in tissue repair in the CNS remains to be defined.

Macrophage-like cells originate from neuroepithelium in culture: characterization and properties of the macrophage-like cells

Cultures of astroglia from C3H/HeJ mice, which are resistant to bacterial cell wall polysaccharide (LPS), initiated from embryos of Theiler stage 14 (9 days of gestation) up to Theiler stage 25 (17 days of gestation) as well as newborn animals, when subjected to nutritional deprivation, i.e. non-feeding of cultures, form large numbers of macrophage-like cells. These cells express Mac-1, Mac-3, F4/80 and Fc antigens. The cells are negative for GFAP, positive for vimentin, express Ia antigen and take up DiL-Ac-LDL. They are positive to non-specific esterase, secrete lysozyme and are phagocytic. Their morphology and ultrastructure closely resemble those of macrophages. Cultures initiated from neuroepithelium of Theiler stage 13 (8.5 days of gestation), before vascularization, when subjected to nutritional deprivation, also produce macrophage-like cells. Using spleen colony assay and methyl cellulose cultures, we were unable to detect the presence of hemopoietic (macrophage) precursor cells in astroglia cultures. This supports the hypothesis that the macrophage-like cells are of neuroectodermal origin and probably correspond to resident microglia of the CNS. Using nutritionally deprived astroglia cultures, a procedure was developed for isolation of macrophage-like cells and production of highly enriched macrophage-like (microglia) cultures.

Diversity amongst the microglia in growing and regenerating fish CNS: immunohistochemical characterization using FL.1, an anti-macrophage monoclonal antibody

We have immunohistochemically characterized the forms and distribution of microglia--the macrophages of the CNS--in fish, using a new monoclonal antibody (mAb), FL.1. This mAb specifically reacts with resident macrophages throughout the body in Oreochromine fish, including Kuppfer cells, gut-associated myeloid cells, and peritoneal macrophages, as well as with microglia, but circulating monocytes are not labelled with FL.1. The FL.1-epitope, which is lost following treatment with reducing agents, has an extracellular location and is associated with three integral membrane glycoprotein variants. FL.1-staining shows that microglia are extremely abundant throughout the fish CNS. For example, they comprise a third of the glia in the optic nerve, and 30% of all cells, including neurons, in the spinal cord, i.e., fish have about tenfold more microglia than mammals. Two forms of FL.1-positive microglia are predominant in fish, one resembling their mammalian counterparts, but less ramified, and the other comprising smaller rounded cells with very little cytoplasm, which are most numerous in the ependymal region of the optic tectum. Apart from the conventional microglia, the optic nerves also contain large lipid-laden macrophages which comprise a third form of FL.1-positive cell in the CNS. Fish optic nerves contain astrocytes of a distinct type which form reticular networks, but lack connections to capillaries (Maggs and Scholes, J. Neurosci. 1990;10:1600-1614). The co-distribution of foamy macrophages may have a metabolic role that is performed by ordinary astrocytes elsewhere in the CNS. An antiserum against the beta 2 subunit of the human leukocyte integrins (Kishimoto et al., Cell 1987a; 50:193-202) was found selectively to recognize the foamy macrophages in Oreochromis. Following lesion to the optic nerve, FL.1-labelling shows that microglia proliferate throughout the visual pathway. In the optic tectum, the additional FL.1-positive cells are concentrated in the vicinity of degenerating retinal axons and their terminals. Most of the microglia in the injured optic nerve have amoeboid morphologies, and the foamy macrophages become depleted.

Microglia in tadpoles of Xenopus laevis: normal distribution and the response to optic nerve injury

We have studied the distribution of microglia in normal Xenopus tadpoles and after an optic nerve lesion, using a monoclonal antibody (5F4) raised against Xenopus retinas of which the optic nerves had been cut 10 days previously. The antibody 5F4 selectively recognizes macrophages and microglia in Xenopus. In normal animals microglia are sparsely but widely distributed throughout the retina, optic nerve, diencephalon and mesencephalon (other regions were not examined). After crush or cut of an optic nerve, or eye removal, there occurs an extensive microglial response along the affected optic pathway. Within 18 h an increase in the number of microglial cells in the optic tract and tectum can be detected. This response increases to peak at around 5 days after the lesion. At this time the nerve distal to the lesion contains many microglial cells; the entire optic tract is outlined by microglia, extended along the degenerating fibres; and the affected tectum shows a heavy concentration of microglia. This microglial response thereafter decreases and has mostly gone by 34 days. We conclude that the microglial response to optic nerve injury in Xenopus tadpoles starts early, peaks just before the regenerating optic nerve axons enter the brain, and is much diminished by the time the retinotectal projection is re-established. The timing is such that the microglial response could play a major role in facilitating regeneration.

Isolation and characterization of human fetal brain-derived microglia in in vitro culture

Human brain microglia may play a central role in immunopathogenesis of CNS diseases including HIV infection, multiple sclerosis and Alzheimer's disease. In order to investigate the possible relationship between microglia and the mononuclear phagocyte system, human brain microglia were isolated from 14-18-week-old fetal brains, and maintained in in vitro culture. Enriched fetal brain microglia were stained for different monocyte/macrophage and glial cell markers. Fresh dissociated brain cells lacked macrophage surface markers. Isolated microglial cells stained positive for complement receptor C3bi, Class II [human leukocyte antigen-DR (HLA-DR)] antigen and with the lectin Ricinus communis. Microglia also share several functional properties with monocyte/macrophages, which include generation of superoxide anion and histochemically demonstrable intracellular acid phosphatase and non-specific esterase. Primary human dissociated brain cultures were maintained in culture for at least 28 weeks. Although microglia were not observed above the astrocyte cell layer after 5 weeks in culture, microglia-like cells appear below the astrocyte layer after 12 weeks in culture. These cells stained positive for non-specific esterase and displayed oxidative burst activity upon activation with phorbol myristate acetate. Thus, we have successfully isolated an enriched population of microglia from human fetal brain and have demonstrated that these cells possess markers and properties which are characteristics of mononuclear phagocytes.

Separate blood and brain origins of proliferating cells during gliosis in adult brains

The response of the brain to injury involves the accumulation of a large number of proliferating cells at the site of damage. Neither the identity nor the origin of these cells is unequivocally established. We have investigated this proliferative response after unilateral kainic acid lesions in the striatum of adult mice by labeling with tritiated thymidine (3H-thy) or bromodeoxyuridine (Brdu) to identify cells passing through S-phase. Labeled cells were seen only ipsilaterally in coronal section and extended laterally from the subependymal zone lining the lateral ventricle, through the striatal kainic acid injection site and into the cortex. The maximum proliferative response, after a single pulse of 3H-thy administered 4 h before sacrifice, was seen 6 days post-lesion close to the injection site. The proliferating cells were not astrocytes, as neither 3H-thy- nor Brdu-labeled cells were double-labeled with antisera to glial fibrillary acidic protein after the lesion. Animals given 3H-thy on day 3 post-lesion and then sacrificed on days 4, 5 or 6 post-lesion showed cumulative increases in the number of proliferating cells at the injection site with no increases in the surrounding tissue. We hypothesized that this reflected the presence of 2 sources of labeled cells: (1) an exogenous population of blood cells coming in through the broken blood-brain barrier and accumulating at the injection site and (2) endogenous cells (microglia) which are normally quiescent in the adult but proliferate in response to injury. By irradiating adult mice (900 rads) we attempted to selectively remove the blood stem cell precursors which gave rise to the proposed exogenous source of cells.(ABSTRACT TRUNCATED AT 250 WORDS)

Immunological and neuropathological significance of the Virchow-Robin space

The anatomy of the Virchow-Robin space is reviewed and attention is drawn to its importance as a compartment which is in communication with lymphatic channels of the head and neck and in which local immunological reactions take place. Macrophages in the Virchow-Robin spaces express MHC class II antigens and are well placed to interact with lymphocytes derived from the blood in initiating and promoting immune response to foreign antigens in the brain. The immunological reactions taking place in the Virchow-Robin spaces in encephalitis, multiple sclerosis and human immunodeficiency virus encephalitis are examined for the light they may throw on the pathogenesis of these conditions.

The complex of microglial cells and amyloid star in three-dimensional reconstruction

Ultrastructural, three-dimensional reconstruction and morphometric studies of classical plaques from the cortex of a patient with Alzheimer's disease showed five or six microglial cells, which form, together with the amyloid star, the central complex of the classical plaque. Microglial cells associated with the amyloid star show marked polymorphism, but all forms possess an amyloid making pole. The surface of the cell membrane at this pole is extended by apparent connection with membranes of cytoplasmic channels filled with amyloid fibers. The amyloid pole also shows other features of local activation with nuclei translocation, expansion of Golgi apparatus and endoplasmic reticulum, and multiplication of vacuoles and coated vesicles that are in close proximity to channels filled with new polymerized amyloid fibers. On the basis of ultrastructural studies, three forms of microglial cells can be distinguished: macrophage-like, cap-like, and octopus-like cells. The most effective in production of amyloid fibers seem to be cap-like microglial cells, which have the greatest interface with the amyloid star. Octopus-like cells have the least contact with the amyloid star. The size of the surface of the interface with the amyloid star appears to be an indicator of the extent of cell engagement in amyloid fiber formation.

Tetanus toxin-induced seizures cause microglial activation in rat hippocampus

Tetanus toxin (about 20 mouse LD50) injected into the ventral hippocampus of rats leads to brief seizures occurring intermittently over a period of weeks. Toxin injection leads to the appearance of activated microglia (detected with OX42 immunohistochemistry) in the hippocampus. After 7-14 days, many activated microglia are visible in CA1 area of dorsal hippocampus aligned with the pyramidal cell dendrites and having the morphology characteristic of 'rod cells'. Extensive cell loss is found in dorsal CA1, but not at the injection site, in about one third of injected rats.