Brain-derived neurotrophic factor inhibits osmotic swelling of rat retinal glial (Müller) and bipolar cells by activation of basic fibroblast growth factor signaling.

Water accumulation in retinal glial (Müller) and neuronal cells resulting in cellular swelling contributes to the development of retinal edema and neurodegeneration. Intravitreal administration of neurotrophins such as brain-derived neurotrophic factor (BDNF) is known to promote survival of retinal neurons. Here, we show that exogenous BDNF inhibits the osmotic swelling of Müller cell somata induced by superfusion of rat retinal slices or freshly isolated cells with a hypoosmotic solution containing barium ions. BDNF also inhibited the osmotic swelling of bipolar cell somata in retinal slices, but failed to inhibit the osmotic soma swelling of freshly isolated bipolar cells. The inhibitory effect of BDNF on Müller cell swelling was mediated by activation of tropomyosin-related kinase B (TrkB) and transactivation of fibroblast growth factor receptors. Exogenous basic fibroblast growth factor (bFGF) fully inhibited the osmotic swelling of Müller cell somata while it partially inhibited the osmotic swelling of bipolar cell somata. Isolated Müller cells displayed immunoreactivity of truncated TrkB, but not full-length TrkB. Isolated rod bipolar cells displayed immunoreactivities of both TrkB isoforms. Data suggest that the neuroprotective effect of exogenous BDNF in the retina is in part mediated by prevention of the cytotoxic swelling of retinal glial and bipolar cells. While BDNF directly acts on Müller cells by activation of TrkB, BDNF indirectly acts on bipolar cells by inducing glial release of factors like bFGF that inhibit bipolar cell swelling.

Morphometric analysis of the retina from horses infected with the Borna disease virus.

Borna disease (BD) is a fatal disorder of horses, often characterized by blindness. Although degeneration of retinal neurons has been demonstrated in a rat model, there are controversial data concerning whether a similar degeneration occurs in the retina of infected horses. To investigate whether BD may cause degeneration of photoreceptors and possibly of other neuronal cells at least at later stages of the disease, we performed a detailed quantitative morphologic study of retinal tissue from Borna-diseased horses. BD was diagnosed by detection of pathognomonic Joest-Degen inclusion bodies in the postmortem brains. Paraffin sections of paraformaldehyde-fixed retinae were used for histologic and immunohistochemical stainings. Numbers of neurons and Müller glial cells were counted, and neuron-to-Müller cell ratios were calculated. Among tissues from 9 horses with BD, we found retinae with strongly altered histologic appearance as well as retinae with only minor changes. The neuron-to-Müller cell ratio for the whole retina was significantly smaller in diseased animals (8.5 +/- 0.4; P < .01) as compared with controls (17.6 +/- 0.8). It can be concluded that BD in horses causes alterations of the retinal histology of a variable degree. The study provides new data about the pathogenesis of BD concerning the retina and demonstrates that a loss of photoreceptors may explain the observed blindness in infected horses.

Hypoxia: modulation of endothelial cell proliferation by soluble factors released by retinal cells.

A devastating complication of ischemic retinopathies is retinal neovascularization. We studied the impact on retinal endothelial cell proliferation of soluble factors released from cultured retinal glial (Müller) cells and from retinal explant cultures. Hypoxia strongly stimulated VEGF release by all types of cultures but endothelial cell growth was not further increased by the corresponding conditioned media if compared to supernatants obtained under normoxia. When the final concentration of the hypoxia-conditioned media was adjusted to the VEGF level of normoxia-conditioned media, they even inhibited endothelial cell proliferation. Inhibition may be exerted by TGF-beta 2 but TGF-beta 2 mRNA and protein expression in Müller cells were found to be down-regulated under hypoxia. We conclude that retinal endothelial cell proliferation is controlled by the balance of the amount and/or efficacy of several stimulatory and inhibitory factors.

Changes of the organotypic retinal organization in Borna virus-infected Lewis rats.

Retinae of Borna disease virus (BDV)-infected Lewis rats were investigated with emphasis on long-term changes in organotypic tissue organization and glia-neuron relationship. Virus inoculation was attained via intracerebral BDV injection. Following survival times ranging between two and eight months, the retinal thickness was reduced up to one third of that of controls. Photoreceptor segments were completely extinguished and the number of neurons was dramatically reduced. The typical laminar organization of the retina was largely dissolved. Electron microscopy revealed severe spongy degeneration. Large numbers of activated microglia and macrophages were found, both cell types performing very active phagocytosis. The microglial cells expressed an extraordinary phenotype as characterized by large numbers of processes, with some of them penetrating the endfeet of Müller cells and others establishing highly complex interdigitations with vacuolized swellings and endings of neuronal processes. Müller cells were not reduced in number but displayed clear indications of gliosis such as alterations in the immunoreactivity for filament proteins and glutamine synthetase, significantly thickened stem processes, and an altered pattern of K(+) currents in patch-clamp recordings. These findings demonstrate for the first time long-term neuron-glia interactions in the retina of BDV-infected rats. Moreover, the data contribute to our knowledge on structural and functional alterations accompanying persisting virus infection in the central nervous system.

VEGF release by retinal glia depends on both oxygen and glucose supply.

Isolated retinae or isolated Müller cells were cultured in vitro, and vascular endothelial growth factor (VEGF) was assayed as protein (by ELISA) and as mRNA (by semi-quantitative RT-PCR). In both types of cultures, hypoxia (5% O2) resulted in an upregulated VEGF release. While the unstimulated VEGF secretion was virtually independent of glucose (0.125 - 25 mM), elevated glucose concentrations (10 - 25 mM) blocked most of the stimulatory effect of hypoxia on VEGF mRNA synthesis (determined in Müller cell cultures) as well as on VEGF release (in both retina and Müller cell cultures). It is concluded that in retinal glial (Müller) cells, being responsible for retinal VEGF synthesis (and, thus, for undesirable neovascularization), the metabolic effects of hypoxia can be compensated by a surplus of glucose.

Neuron-glia interactions in the rat retina infected by Borna disease virus.

Neuron-glia interactions in the Borna disease virus (BDV)-infected rat retina were investigated with emphasis on the ultrastructural characterization of degenerative alterations in the ganglion cell and photoreceptor layer. Immuno- and cytochemical techniques were applied to label microglia, macrophages and Müller (macroglial) cells. Four weeks after intracerebral infection of adult rats, the total thickness of the retina was considerably diminished, primarily due to the loss of photoreceptor segments and ganglion cells. A gradual reduction of both plexiform layers was also observed. There was a remarkable increase in the number of microglial cells, predominantly in the ganglion cell and the inner plexiform layers. Ultrastructural analysis confirmed that microglia, but also macrophages, were involved in phagocytosis accompanying severe neuronal degeneration in the ganglion cell and the photoreceptor layer. In contrast, Müller cells showed moderate morphological and cytochemical alterations, indicating that Müller cells play only a minor role in early stages of BDV-induced retinitis. Monitoring neuron-glia interactions in BDV-induced retinopathy, combined with the application of different protocols of immunosuppression effecting the BDV virus and/or the microglia, might help to establish specific strategies to suppress BDV-induced neuronal degeneration.

Retinal gliopathy accompanying thioacetamide-induced liver insufficiency: light and electron microscopic observations.

A recent examination of retinae of patients who had died with symptoms of liver insufficiency (LI) including hepatic encephalopathy (HE) revealed morphological changes in retinal Müller glia similar to the astrocytic changes normally accompanying HE, and the term "hepatic retinopathy" (HR) was coined to define these changes. In the present study, the immunomorphology and ultrastructure of Müller cells were examined in rats in which LI with accompanying HE was induced with a hepatotoxin, thioacetamide (TAA). Light microscopically, retinae of rats with LI were characterized by swelling of the Muller cell cytoplasm. Immunostaining for glia-specific marker proteins in Müller cells from LI rats revealed a strongly enhanced expression of glial fibrillary acidic protein, and a considerable increase in glutamine synthetase immunoreactivity, as compared to control animals. Ultrastructurally, the Müller cells of LI rats showed swelling and vacuolization of cell processes. In particular, the endfeet contained many swollen mitochondria. By contrast, LI produced no morphologically demonstrable changes in retinal neurons and photoreceptor cells. Thus, the retinal changes induced by TAA in the rats strongly resembled those described in human HR, rendering the present rat model suitable for more detailed investigations of the pathomechanism(s) of HR.

Distribution of mitochondria within Müller cells--II. Post-natal development of the rabbit retinal periphery in vivo and in vitro: dependence on oxygen supply.

The occurrence and localization of mitochondria within glial (Müller) cells and neurons of the peripheral (avascular) rabbit retina was studied electron microscopically and by immunocytochemical demonstration of the mitochondrial enzyme GABA transaminase (GABA-T). Post-natal development in vivo was compared with development of organ cultures from neonatal rabbit retinae, grown over 2 weeks in vitro. The adult pattern of mitochondrial localization (restriction to the sclerad end of the cells) was observed from the beginning of enzyme expression at early post-natal stages. However, when neonatal retinal pieces were grown in vitro with their vitread surface exposed to the air, their Müller cells contained mitochondria along most of their length. When functionally developed retinae from postnatal day 14 were explanted in vitro, they retained their sclerad mitochondrial distribution for almost 24 h but thereafter the inner portions of their cytoplasm became occupied by mitochondria within a few hours. This was achieved mainly by mitochondrial migration rather than by formation of new mitochondria because it was not prevented by cycloheximide-induced inhibition of protein synthesis. These data support the following hypotheses: (1) the mitochondrial distribution in Müller cells is determined by the local cytoplasmic O2 pressure (pO2), (2) existing mitochondria move towards cytoplasmic regions of sufficient pO2 by rather rapid migration and (3) the start of this migration is delayed by almost 24 h due to the action of as yet unknown control mechanisms. In contrast, the mitochondrial content of retinal ganglion and amacrine cells in the vitread retinal layers was virtually independent of the source and level of oxygen supply.

Distribution of mitochondria within Müller cells--I. Correlation with retinal vascularization in different mammalian species.

The distribution of mitochondria within retinal glial (Müller) cells and neurons was studied by electron microscopy, by confocal microscopy of a mitochondrial dye and by immunocytochemical demonstration of the mitochondrial enzyme GABA transaminase (GABA-T). We studied sections and enzymatically dissociated cells from adult vascularized (human, pig and rat) and avascular or pseudangiotic (guinea-pig and rabbit) mammalian retinae. The following main observations were made. (1) Müller cells in adult euangiotic (totally vascularized) retinae contain mitochondria throughout their length. (2) Müller cells from the periphery of avascular retinae display mitochondria only within the sclerad-most end of Müller cell processes. (3) Müller cells from the vascularized retinal rim around the optic nerve head in guinea-pigs contain mitochondria throughout their length. (4) Müller cells from the peripapillar myelinated region ('medullary rays') of the pseudangiotic rabbit retina contain mitochondria up to their soma. In living dissociated Müller cells from guinea-pig retina, there was no indication of low intracellular pH where the mitochondria were clustered. These data support the hypothesis that Müller cells display mitochondria only at locations of their cytoplasm where the local O2 pressure (pO2) exceeds a certain threshold. In contrast, retinal ganglion cells of guinea-pig and rabbit retinae display many mitochondria although the local pO2 in the inner (vitread) retinal layers has been reported to be extremely low. It is probable that the alignment of mitochondria and the expression of mitochondrial enzymes are regulated by different mechanisms in various types of retinal neurons and glial cells.

Müller (glial) cell development in vivo and in retinal explant cultures: morphology and electrophysiology, and the effects of elevated ammonia.

Retinal explant cultures have been established as a useful tool to study both the normal development of the mammalian retina and the effects of pathogenic agents. We used such cultures as a model for the (ammonia-induced) hepatic retinopathy, earlier observed in humans with chronical liver failure, and ascribed to a breakdown of Müller (glial) cell function. In the explant cultures, one day exposure to elevated (7 mM) ammonia was sufficient to cause Müller cell reactivity as indicated by increasing immunopositivity for glial fibrillary acidic protein. After 4 days in elevated ammonia, the Müller cells were severely deformed, the layered structure of the retinae became disorganized, and significant neuronal cell death occurred. Using whole-cell voltage-clamp recordings, the expression of K+ channels was compared in Müller cells isolated from retinae of rabbits at postnatal days 9 to 12 and from neonatal explants cultured for 9 to 12 days, respectively. Müller glial cells grown both in vivo and in vitro express the same set of K+ channels in their membranes: (i) inwardly rectifying K+ (K(IR)) channels which were selectively blocked by Ba2+ ions; (ii) large-conductance, Ca2+-activated K+ (BK(Ca)) channels which were blocked by iberiotoxin and were activated by phloretin; and (iii) delayed rectifying voltage-gated K+ channels. The presence of K(IR) channels indicates successful differentiation of the Müller cells grown in vitro, as these channels are not expressed in cells from neonatal animals. Four days of elevated ammonia in the culture medium caused a complete loss of K(IR) channels in Müller cell membranes, and a significant decrease of the membrane potential. The results indicate that in hepatic retinopathy, the well-known morphological and enzymatical alterations of Müller glial cells may be accompanied by changes in their membrane permeability for K+.

Comparative studies on mammalian Müller (retinal glial) cells.

Müller cells from 22 mammalian species were subjected to morphological and electrophysiological studies. In the 'midperiphery' of retinae immunocytochemically labeled for vimentin, estimates of Müller cell densities per unit retinal surface area, and of neuron-to-(Müller) glia indices were performed. Müller cell densities were strikingly similar among the species studied (around 8000-11,000 mm-2) with the extremes of the horse (< or = 5000 mm-2) and the tree shrew (> or = 20,000 mm-2). By contrast, the number of neurons per Müller cell varied widely, being clustered at 6-8 (in retinae with many cones), at about 16, and at up to more than 30 (in strongly rod-dominated retinae). Isolated Müller cell volumes were estimated morphometrically, and cell surface areas were calculated from membrane capacities. Müller cells isolated from thick vascularized retinae (carnivores, rats, mice, ungulates) were longer and thinner, and had smaller volumes but higher surface-to-volume ratios than cells from thin paurangiotic (i.e. with blood vessels only near the optic disc) or avascular retinae (rabbits, guinea pigs, horses, zebras). In whole-cell voltage-clamp studies, Müller cells from all mammals studied displayed two dominant K+ conductances, inwardly rectifying currents and delayed rectifier currents. TTX-sensitive Na+ currents were recorded only in some species. Based on these data, the following hypotheses are presented, (a) neuron-to-(Müller) glia indices are determined by precursor cell proliferation rather than by metabolic demands; (b) Müller cell volumes depend on available space rather than on the number of supported neurons; and (c) it follows that, the specific metabolic activities of Müller cells must differ greatly between species, a difference that may contribute to distinct patterns of retinal vascularization.

Changes in CD44 and ApoE immunoreactivities due to retinal pathology of man and rat.

In cases of retinal light damage, glaucoma, or senile macula degeneration, the loss of retinal neurons is thought to cause alterations of glial cells. We performed immunocytochemical studies on retinae of (i) healthy rats and human donors, (ii) rats exposed to enhanced illumination for 24 months, a procedure which leads to complete loss of photoreceptor cells, (iii) a human donor who had suffered from senile macula (photoreceptor cell) degeneration, and (iv) human donors who had suffered from glaucoma, known to be accompanied by a loss of ganglion cells and other retinal neurons. Furthermore, Müller cells were enzymatically isolated from human glaucomatous retinae. All preparations were subjected to immunocytochemistry for CD44 antigen and Apolipoprotein E (ApoE). In normal rat and human retinae, CD44 immunoreactivity was observed in the microvillous sclerad processes of Müller cells: in human retinae, perivascular (astro-)glial cell processes were also CD44 immunopositive. ApoE immunoreactivity was only found in some perivascular (astro-)glial cell processes of human retinae. Both rat and human Müller cells respond to photoreceptor cell damage by increased, and ectopic, expression of the CD44 antigen. Increased ApoE immunoreactivity was found in Müller cells from degenerative human retinae, but rarely in light-damaged rat retinae. It is concluded that degeneration-related reorganization involves enhanced expression of the glial cell adhesion molecule CD44 as well as elevated activity of the glial lipid transport molecule ApoE.

The Müller (glial) cell in normal and diseased retina: a case for single-cell electrophysiology.

In the retina of most vertebrates there exists only one type of macroglia, the Müller cell. Müller cells express voltage-gated ion channels, neurotransmitter receptors and various uptake carrier systems. These properties enable the Müller cells to control the activity of retinal neurons by regulating the extracellular concentration of neuroactive substances such as K+, GABA and glutamate. We show here how electrophysiological recordings from enzymatically dissociated mammalian Müller cells can be used to study these mechanisms. Müller cells from various species have Na(+)-dependent GABA uptake carriers, but only cells from primates have additional GABA receptors that activate Cl- channels. Application of glutamate analogues causes enhanced membrane currents recorded from Müller cells in situ but not from isolated cells. We show that mammalian Müller cells have no ionotropic glutamate receptors but respond to increased K+ release from glutamate-stimulated retinal neurons. This response is involved in extracellular K+ clearance and is mediated by voltage-gated (inwardly rectifying) K+ channels which are abundantly expressed by healthy Müller cells. In various cases of human retinal pathology, currents through these channels are strongly reduced or even extinguished. Another type of voltage-gated ion channels, observed in Müller cells from many mammalian species, are Na+ channels. In Müller cells from diseased human retinae, voltage-dependent Na+ currents were significantly increased in comparison to cells from control donors. Thus, the expression of glial ion channels seems to be controlled by neuronal signals. This interaction may be involved in the pathogenesis of retinal gliosis which inevitably accompanies any degeneration of retinal neurons. In particular, Müller cell proliferation may be triggered by mechanisms requiring the activation of Ca(2+)-dependent K+ channels. Ca(2+)-dependent K+ currents are easily elicitable in Müller cells from degenerating retinae and can be blocked by 1 mM TEA (tetraethylammonium). In purified Müller cell cultures, the application of 1 mM TEA greatly reduces the proliferative activity of the cells. These data clearly show that Müller cells are altered in cases of neuronal degeneration and may be crucially involved in pathogenetic mechanisms of the retina.