Human retinal microglia: Expression of immune markers and relationship to the Glia limitans

The immunoreactivity, morphology and relationship to the glia limitans of microglia were investigated in flatmounts and sections of normal human retina, using immunogold histochemistry, electron microscopy (EM), and antibodies directed against CD45, major histocompatability complex class I (MHC-I), MHC-II, and human macrophage antigens. Immunoreactivity was evident for all antibodies tested, including MHC-I, which labeled both microglia and retinal vascular endothelium. Most consistent labeling was obtained using antibodies to CD45, MHC-II, and anti-human macrophage (S22) antigen. Immunoreactive cells were seen in the perivascular space (perivascular cells), where they were closely adherent to the vessel profile, and in the retinal parenchyma (microglia). Some parenchymal microglia were also vessel associated and by EM were seen to be closely related to the glia limitans (paravascular microglia). Paravascular microglia were shown by optical densitometry, to express higher levels of MHC antigens than neighboring, non-vessel associated, parenchymal microglia. In addition, paravascular microglia were macrophage (S22) antigen positive, while other parenchymal microglia did not express macrophage antigens. Quantitative data indicate that similar populations of microglia are immunoreactive to CD45, MHC-I, and MHC-II, while relatively few microglia (approximately 10%) are immunoreactive for human macrophage (S22) antigens, supporting previous suggestions that microglia are a heterogeneous population. © 1995 Wiley-Liss, Inc.     

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. © 1995 Wiley-Liss, Inc.     

Characterization of microglia induced from mouse embryonic stem cells and their migration into the brain parenchyma

We derived microglia from mouse embryonic stem cells (ES cells) at very high density. Using the markers Mac1⁺/CD45^low and Mac1⁺/CD45^high to define microglia and macrophages, respectively, we show that Mac1⁺ cells are induced by GM-CSF stimulation following neuronal differentiation of mouse ES cells using a five-step method. CD45^low expression was high and CD45^high expression was low on induced cells. We used a density gradient method to obtain a large amount of microglia-like cells, approximately 90% of Mac1⁺ cells. Microglia-like cells expressed MHC class I, class II, CD40, CD80, CD86, and IFN-γR. The expression level of these molecules on microglia-like cells was barely enhanced by IFN-γ. Intravenously transferred GFP⁺ microglia derived from GFP⁺ ES cells selectively accumulated in brain but not in peripheral tissues such as spleen and lymph node. GFP⁺ cells were detected mainly in corpus callosum and hippocampus but were rarely seen in cerebral cortex, where Iba1, another marker of microglia, is primarily expressed. Furthermore, both GFP⁺ and Iba1⁺ cells exhibited a ramified morphology characteristic of mature microglia. These studies suggest that ES cell-derived microglia-like cells obtained using our protocol are functional and migrate selectively into the brain but not into peripheral tissues after intravenous transplantation.     

Distribution and Temporal Regulation of the Immune Response in the Rat Brain to Intracerebroventricular Injection of Interferon-γ

The response to intracerebroventricular administration of interferon (IFN)-γ was examined in the adult Wistar rat brain: major histocompatibility complex (MHC) antigens class I and II, CD8 and CD4 antigens, and the macrophage/microglia antigen OX42 were analyzed in respect to saline-injected cases over 1 week. The glial cell type expressing MHC antigens was characterized with double labeling. IFN-γ was thus found to induce MHC class I and II expression in microglia, identified by tomato lectin histochemistry, and not in GFAP-immunostained astrocytes. MHC antigen-expressing microglia was detected in the periventricular parenchyma, several fields of the cerebral cortex, cerebellum, major fiber tracts, and brainstem superficial parenchyma. Different gradients of density and staining intensity of the MHC-immunopositive elements were observed in these regions, in which MHC class I antigens persisted up to 1 week, when MHC class II induction had declined. Quantitative analysis pointed out the proliferation of OX42-immunoreactive cells in periventricular and basal brain regions. CD8+ T cells were observed in periventricular regions, basal forebrain, and fiber tracts 3 days after IFN-γ injection and their density markedly increased by 7 days. CD4+ T cells were also seen and they were fewer than CD8+ ones. However, numerous CD4+ microglial cells were observed in periventricular and subpial regions, especially 1 week after IFN-γ injection. Our data indicate that this proinflammatory cytokine mediatesin vivomicroglia activation and T cell infiltration in the brain and that the cells involved in this immune response display a regional selectivity and a different temporal regulation of antigen expression.     

Distribution in ocular structures and optic pathways of immunocompetent and glial cells in an experimental allergic encephalomyelitis (EAE) relapsing model

Relapsing experimental allergic encephalomyelitis (EAE) was induced in DA rats and the ocular pathologic events were examined at the various phases of the illness. About 80% of EAE rats presented anterior uveitis (AU), even after complete EAE recovery. We studied the phenotype and localization of immunocompetent cells, the major histocompatibility complex (MHC) class I and II antigen expression, as well as the chemokine monocyte chemoattractant protein-1 (MCP-1) appearance. In control animals, there were many glial fibrillary acidic protein (GFAP)(+) cells and OX42(+) cells in the ciliary body, retina, optic nerve and chiasma. Except in retina, we observed constitutive MHC class I and II expression. During the EAE acute phase, there was up-regulation of MHC class II and GFAP antigens in iris, ciliary body, limbus, and optic pathways. MHC class I and ED2 antigens were expressed in meninges and in the prechiasmatic cisterna, by cells which could have a role in immune surveillance. MCP-1 mRNA was highly expressed in optic pathways during the acute phase and the protein was expressed by astrocytes, macrophages, and lymphocytes. During the relapsing phase, MCP-1 was weakly expressed to disappear almost completely during the final recovery phase. The expression of MHC class II on astrocytes was increased during the relapsing and final recovery phase in which the inflammatory lesions persisted. These findings suggest that ocular areas and optic pathways, mainly optic chiasma, are important targets in the relapsing EAE.     

Optic nerve degeneration induces the expression of MHC antigens in the rat visual system

The brain has long been considered to be an immunologically privileged site. However, privilege is not absolute, as has been shown by the inability of foreign tissue grafts to survive indefinitely in the brain. The rejection of this tissue is accompanied by the upregulation of major histocompatibility complex (MHC) antigen expression. Therefore it is essential to define conditions that influence the expression of these antigens in the brain, especially since such a definition may further the understanding of disease processes that lead to the autoimmune destruction of the central nervous system. Here we show that both MHC class l and class II antigens are expressed within 1 or 2 days of eye removal by cells showing the morphological characteristics of microglia. Expression is seen along the optic pathway and within the brainstem centers to which optic axons project. In the early stages of the reaction, MHC class I antigen expression is seen throughout the optic pathway, including the terminal distribution areas of the subcortical visual centers, while MHC cells class II are localised mainly to degenerating myelinated fiber systems. These changes are not accompanied by any alteration in the integrity ofthe blood-brain barrier. During the second week postlesion, class l positive cells are found beyond the confines of the degenerating pathways, while class II positive cells are seen within regions such as the stratum griseum superficiale of the superior colliculus, where few myelinated axons are present. There is subsequent diminution of MHC positive cells, but a small number of cells are still seen 60 days post-lesion. Focal lesions within the eye show that at early survival times, while class l MHC positive cells are distributed throughout the nerve, class II positive cells are largely absent from the unmyelinated segment of the nerve. Retrograde changes in the retina after nerve section are accompanied only by MHC class l antigen expression. These observations show that neural degeneration is accompanied by a rigid sequence of events involving expression of MHC antigens by microglia. If foreign antigens were present in the brain while these events were taking place, it is possible that such antigens would be recognised and destroyed by the host immune system. © 1993 Wiley-Liss, Inc.     

Immunophenotypic analysis of infiltrating leukocytes and microglia in an experimental rat glioma

Summary The appearance and cellular distribution of major histocompatibility complex (MHC), as well as lymphocytic and macrophage antigens has been studied in a fully developed experimental rat forebrain glioma. Activated microglial cells and microglia-derived macrophages expressing CR3 complement receptor molecules and MHC class II (Ia) antigen were found throughout the tumor, and with increased density along the tumor's periphery. MHC class I antigen expression was entirely absent from tumor cells, and found only occasionally on microglia. The expression of leukocyte common antigen, and CD4 and CD8 antigens was conspicuous throughout the tumor, and associated with lymphocytes, perivascular cells, and microglia. Cells expressing the ED2 macrophage epitope were almost exclusively of the perivascular type and revealed a distribution dissimilar to that of cells positive for Ia antigen. The ED2 epitope was found sporadically on ramified microglial cells. The results show that despite heavy infiltration with blood mononuclear and CNS microglial cells, the tumor showed no evidence of destruction caused by inflammatory cells. Possible mechanisms of tumor immunosuppressive activity preventing the full immunological activation of microglia and blood mononuclear cells are discussed.Supported in part by an American Cancer Society Institutional Research Grant at the University of Florida     

Absence of donor-type major histocompatibility complex class I antigen-bearing microglia in the rat central nervous system of radiation bone marrow chimeras

Localization of bone marrow-originated cells in the central nervous system (CNS) of the rat was investigated by using bone marrow chimeras. In order to do this, Lewis rats which carry major histocompatibility complex (MHC) class I antigens haplotype 1 (RT1.Al) were reconstituted with (Lew X PVG)F1 (RT1.Al/c) bone marrow cells after lethal irradiation. Transferred bone marrow cells were detected by immunohistochemical staining using a monoclonal antibody, OX27, specific for haplotype c of rat MHC class I antigens (RT1.Ac). The spleen and thymus of chimeric rats were fully reconstituted with transferred F1 cells 4 weeks after bone marrow transplantation. At this stage, mononuclear cells in the subarachnoid space of the CNS expressed OX27 antigen indicating that they were of bone marrow origin. A few OX27-positive blood cells were scattered in the CNS parenchyma 4-12 weeks after reconstitution. Ramified microglia, however, remained OX27-negative. Bone marrow-derived microglia were not observed throughout the period of examination until 24 weeks. In addition, experimental allergic encephalomyelitis (EAE) was induced in chimeric rats in order to augment the expression of MHC class I antigens on microglia. Even under this condition, no OX27-positive microglia were observed. Taken together, ramified microglia might be of neuroectodermal origin and there is little possibility that the microglia are derived from the bone marrow. However, if the ramified microglia are derived from blood cells, the microglia may be expected to have characteristic cell kinetics from the following points: (1) the precursor cells of the microglia may enter the CNS only at the perinatal stage; and (2) even under the condition in which lymphocytes and macrophages enter the CNS as observed in EAE, the precursor cells of the microglia are not supplied from the blood.     

Expression of Major Histocompatibility Complex Antigens and Induction of Human T-Lymphocyte Proliferation by Astrocytes and Macrophages from Porcine Fetal Brain

Porcine fetal brain cells are of potential use as donor cells for transplantation therapies of neurodegenerative diseases in humans. Our aim was to determine the immunestimulatory properties of astrocytes and macrophages from porcine fetal brain in vitro. By flow cytometry, freshly isolated porcine fetal brain cells were nonautofluorescent, while primary cultures of these cells, prepared to favor growth of astrocytes and macrophages/microglia, consisted of both an autofluorescent and a nonautofluorescent cell population. The cultured autofluorescent cells had qualities typical of macrophages: CD18 (beta(2) integrin subunit) expression, high granularity, and phagocytic activity. The cultured nonautofluorescent cells stained positive for the astrocyte marker glial fibrillary acidic protein and CD56 (NCAM isoform). While freshly isolated porcine fetal brain cells expressed very low levels of major histocompatibility complex (MHC) class I and no MHC class II antigens, primary culture of these cells resulted in upregulation of MHC class I antigens on astrocytes and macrophages and MHC class II antigens on a subpopulation of the macrophages. Single-cell suspensions prepared from the primary cultures were flow sorted into astrocyte and macrophage populations on the basis of cell granularity and autofluorescence or on the basis of CD56 expression. Pure suspensions (>98%) of astrocytes induced a low proliferative response in human T lymphocytes, as determined by [(3)H]thymidine incorporation after 4 days of coculture. A suspension of 91% macrophages was a strong inducer of human T-cell proliferation, even stronger than allogeneic mononuclear blood cells. For neural xenotransplantation, our findings suggest that depletion of macrophages from the donor-cell suspensions may enhance graft survival by reducing cell-mediated rejection.     

Differential expression of class I and class II major histocompatibility complex antigen in early postnatal rats

Cells expressing major histocompatibility complex (MHC) antigens are rarely found in normal mature brains, but cells resembling microglia can be induced to express these antigens following the onset of neural degeneration. In young rats, these cells show spontaneous expression of class I MHC antigens, which is further enhanced in the superior colliculus by the degeneration resulting from eye removal. By contrast, class II MHC antigen expression does not occur spontaneously and can only be induced by eye removal when the lesion is performed after the first postnatal week, when the optic tract begins to myelinate. We suggest that different signals are responsible for induction of class I and of class II MHC antigen expression.     

Turnover of resident retinal microglia in the normal adult mouse

The retina contains two distinct populations of monocyte-derived cells: perivascular cells (macrophages) and parenchymal cells (microglia), important in homeostasis, neuroinflammation, degeneration, and injury. The turnover of these cells in the retina and their repopulation in normal physiological conditions have not been clarified. Bone marrow (BM) cells from EGFP-transgenic mice were adoptively transferred into lethally irradiated normal adult C57BL/6 mice. Eight, 14, and 26 weeks later mice were sacrificed and retinal flatmounts were prepared. Retinal microglia were identified by F4/80, CD45, and Iba-1 immunostaining. BrdU was injected into normal mice for 3-14 days and cell proliferation was examined by confocal microscopy of retinal flatmounts. Few (6.15 +/- 2.02 cells/retina) BrdU(+) cells were detected and of these some coexpressed CD11b (1.67 +/- 0.62 cells/retina) or F4/80 (0.57 +/- 0.30 cells/retina). BM-derived EGFP(+) cells were detected by 8-weeks post-transplantation. By 6 months, all retinal myeloid cells were EGFP(+). Consecutively, donor BM-EGFP(+) cells were demonstrated within the: (1) peripheral and juxtapapillary retina, (2) ganglion cell layer, (3) inner and outer plexiform layers, and (4) photoreceptor layer. EGFP(+) cells within the ganglion layer were amoeboid in shape and F4/80(high)CD45(high)Iba-1(high), whereas cells in the inner and outer plexiform layers were ramified and F4/80(low) CD45(low)Iba-1(low). Perivascular macrophages expressed less F4/80, CD45, and Iba-1 compared with parenchymal microglia. Our results suggest that BM-derived monocyte precursor cells are able to migrate across the BRB and replace retinal microglia/macrophages. The complete replacement of retinal microglia/macrophages takes about 6 months. In situ proliferation was predominantly of nonhemopoetic retinal cells.     

Constitutive and visna virus induced expression of class I and II major histocompatibility complex antigens in the central nervous system of sheep and their role in the pathogenesis of visna lesions

Expression of major histocompatibility complex (MHC) antigens was studied in the brains of 10 healthy sheep 2 months to 5 years old and 13 sheep infected with visna virus by intracerebral inoculation and killed one and 6 months post infection (p.i.). In healthy sheep there was prominent expression of class I, mainly on endothelial cells but also detected on ependyma, choroid plexus and in the leptomeninges. Class II expression was sparse. It was observed on perivascular cells, in choroid plexus, leptomeninges and on microglial cells in the white matter. No definite increase with age in the constitutive expression of class I and II was observed, confirming that we are dealing with a true constitutive expression. In visna-infected sheep a considerable induction of MHC antigens on microglia was observed, which correlated with severity of lesions and was mainly found in or adjacent to inflammatory infiltrates of the white matter. Increase in class II antigen expression was detected in all sheep but class I only in sheep with the most severe lesions 6 months p.i., an indication of a higher threshold for induction of class I than class II antigens on microglia. Few cells expressed viral antigens, indicating that direct immune-mediated destruction of infected cells plays a minor role in evolution of lesions. Since the preferential induction of MHC antigens on microglia in the white matter correlated with the lesion pattern, activated microglia may play a considerable role in the pathogenesis of lesions.     

Isolation of retinal progenitor cells from post-mortem human tissue and comparison with autologous brain progenitors

The goal of the present study was threefold: to determine whether viable human retinal progenitor cells (hRPCs) could be obtained from cadaveric retinal tissue, to evaluate marker expression by these cells, and to compare hRPCs to human brain progenitor cells (hBPCs). Retinas were dissected from post-mortem premature infants, enzymatically dissociated, and grown in the presence of epidermal growth factor and basic fibroblast growth factor. The cells grew as suspended spheres or adherent monolayers, depending on culture conditions. Expanded populations were banked or harvested for analysis by RT-PCR, immunocytochemistry, and flow cytometry. hBPCs derived from forebrain specimens from the same donors were grown and used for RT-PCR. Post-mortem human retinal specimens yielded viable cultures that grew to confluence repeatedly, although not beyond 3 months. Cultured hRPCs expressed a range of markers consistent with CNS progenitor cells, including nestin, vimentin, Sox2, Ki-67, GD2 ganglioside, and CD15 (Lewis X), as well as the tetraspanins CD9 and CD81, CD95 (Fas), and MHC class I antigens. No MHC class II expression was detected. hRPCs, but not hBPCs, expressed Dach1, Pax6, Six3, Six6, and recoverin. Minority subpopulations of hRPCs and hBPCs expressed doublecortin, beta-III tubulin, and glial fibrillary acidic protein, which is consistent with increased lineage restriction in subsets of cultured cells. Viable progenitor cells can be cultured from the post-mortem retina of premature infants and exhibit a gene expression profile consistent with immature neuroepithelial cells. hRPCs can be distinguished from hBPC cultures by the expression of retinal specification genes and recoverin.     

Differential expression of MHC class II and B7 costimulatory molecules by microglia in rodent gliomas

To assess the immune function of microglia and macrophages in brain tumors, the expression of MHC class II and B7 costimulatory molecules in three rodent glioma models was examined. Microglia and macrophages, which accounted for 5-12% of total cells, expressed B7.1 and MHC class II molecules in the C6 and 9L tumors, but not RG2 gliomas. Interestingly, the expression of B7.1 and MHC class II molecules by microglia and macrophage was associated with an increase in the number of tumor-infiltrating lymphocytes in C6 and 9L tumors. B7.2 expression, which was present at low levels on microglia and macrophages in normal brain, did not significantly change in tumors. Interestingly, the expression of all three surface antigens increased after microglia were isolated from intracranial C6 tumors and cultured for a short period of time. We conclude that microglia immune activity may be suppressed in gliomas and directly correlates to the immunogenecity of experimental brain tumors.     

Survival of intrastriatal xenografts of ventral mesencephalic dopamine neurons from MHC-deficient mice to adult rats

Previous studies of neural xenografts have used immunosuppressive agents to prevent graft rejection. In the present study we have examined the survival of mouse dopamine neurons lacking either MHC class I or MHC class II molecules transplanted into rat brains and the host immune and inflammatory responses against the xenografts. Survival of neural grafts was immunocytochemically determined at 4 days, 2 weeks, and 6 weeks after transplantation by counting tyrosine hydroxylase (TH)-positive cells in the graft areas. In addition, the host immune and inflammatory responses against neural xenografts were evaluated by semiquantitatively rating MHC class I and class II antigen expression, accumulation of macrophages and activated microglia, and infiltration of CD4- and CD8-positive T-lymphocytes. For the negative controls, the mean number of TH-positive cells in rats that received wild-type mouse tissue progressively decreased at various time periods following transplantation. In contrast, intrastriatal grafting of either MHC class I or MHC class II antigen-depleted neural xenografts resulted in a prolonged survival and were comparable to cyclosporin A-treated rats that had received wild-type mouse tissue. These results indicate that genetically modified donor tissue lacking MHC molecules can be used to prevent neural xenograft rejection.     

Up-regulation of major histocompatibility complex class II antigen expression in the central nervous system of dogs with spontaneous canine distemper virus encephalitis

Major histocompatibility complex class II (MHC II) and canine distemper virus (CDV) antigen expression were compared by immunohistochemistry in the cerebellar white matter of ten dogs with naturally occurring canine distemper encephalitis. In addition, infiltrating mononuclear cells were characterized by employing poly- and monoclonal antibodies directed against human CD3, canine MHC II, CD5, B cell antigen and CDV-specific nucleoprotein. Positive antigen-antibody reaction was visualized by the avidin-biotin-peroxidase complex method on frozen sections. Histologically, neuropathological changes were categorized into acute, subacute, and chronic. In control brains, MHC II expression was weak and predominantly detected on resident microglia of the white matter and on endothelial, perivascular and intravascular cells. In CDV antigen-positive brains, MHC II was mainly found on microglia and to a lesser extent on endothelial, meningeal, choroid plexus epithelial, ependymal and intravascular cells. In addition, virtually all of the perivascular cells expressed MHC II antigen. CDV antigen was demonstrated most frequently in astrocytes. Of the perivascular lymphocytes, the majority were CD3-positive cells, followed by B cells. Only a small proportion of perivascular cells expressed the CD5 antigen. In addition, B cells and CD3 and CD5 antigen-positive cells were found occasionally in subacute and frequently in chronic demyelinating plaques. In acute encephalitis, CDV antigen exhibited a multifocal or diffuse distribution, and MHC II was moderately up-regulated throughout the white matter and accentuated in CDV antigen-positive plaques. In subacute encephalitis, moderate multifocal CDV antigen and moderate to strong diffuse MHC II-specific staining, especially prominent in CDV antigen-positive lesions, were observed. In chronic encephalitis, CDV antigen expression was restricted to single astrocytes at the edge of the lesions or was absent, while MHC II expression, especially prominent on microglia, was strongly up-regulated throughout the white matter, most pronounced in demyelinated plaques. In summary, in acute and subacute lesions without perivascular cuffs, MHC II expression correlated with the presence of CDV antigen. In contrast, in chronic lesions, MHC II expression on microglial cells was the most prominent despite a few CDV antigen-positive astrocytes, indicating that nonviral antigens may play an important role as triggering molecules for the process of demyelination. Received: 13 September 1995 / Revised: 26 February 1996 / Accepted: 1 April 1996     

Differential expression of MHC class II molecules by microglia and neoplastic astroglia: relevance for the escape of astrocytoma cells from immune surveillance

There is increasing evidence that microglia serve as antigen presenters in the human CNS. Although the occurrence of MHC class II immunoreactive cells has been reported in astrocytic gliomas, the relative contribution of microglia to this cell population has not been studied in detail. Using computer-assisted image analysis, we have investigated the expression of MHC class II molecules and of the microglia/macrophage markers Ki-M1P, RCA-1, KP1 and iba1 , in 97 astrocytic gliomas comprising all WHO grades to answer the question whether there is a correlation between tumour grade and the number of MHC class II positive microglia/macrophage profiles. Microglia expressing MHC class II were common in astrocytomas and anaplastic astrocytomas but rare in pilocytic tumours although there was significant variation within each group. MHC class II immunoreactivity was reduced in highly cellular areas of glioblastomas where large numbers of cells expressing macrophage markers were still present. Thus, there was no simple relationship between tumour grade and microglial/macrophage MHC class II expression. In addition, up to 55% of astrocytic gliomas contained MHC class II immunoreactive tumour cells. Microglia but not tumour cells were found to express the BB1/B7 costimulator. We conclude that microglia in astrocytic gliomas are well equipped to function as antigen presenting cells. Yet, neoplastic astroglia appear to acquire the capacity to downregulate microglial MHC class II expression and, at the same time, may induce T-cell clonal anergy through aberrant expression of MHC class II molecules.