Glia cells of the monkey retina--II. Müller cells.

In this paper, for the first time a quantitative description of the morphology and distribution of Müller cells in the macaque monkey retina using immunohistochemistry and high resolution confocal laser scanning microscopy is given. By their morphological features Müller cells are ideally adapted to their neuronal environment in the various retinal layers, with a dense network of horizontal processes, especially in the inner plexiform layer, and close contacts to neuronal somata especially in the outer nuclear layer and ganglion cell layer. Morphology varies with retinal eccentricity. The thickness of the inner trunk increases significantly with increasing retinal eccentricity. According to the overall thickness of the retina, Müller cells in central retina are longer than in peripheral regions. In the parafoveal region, the outer trunks of Müller cells in the outer plexiform layer are immensely elongated. These Müller fibres can reach lengths of several hundred micrometers as they travel through the outer plexiform layer from the foveal centre towards the foveal border where they enter the inner nuclear layer. Müller cell density varies between 6000 cells/mm2 in far peripheral and peak densities of > 30,000 cells/mm2 in the parafoveal retina. There is a close spatial relationship between Müller cells and blood vessels in the monkey retina, suggesting a role of Müller cells in the formation of the blood-retinal barrier, in the uptake of nutrients and the disposal of metabolites.

Glial reactivity in the retina of adult rats.

The present study aimed to characterize the reaction of mammalian (rat) retinal macroglia (Müller cells and astrocytes) to disturbances of their environment in the form of intraorbital section of the optic nerve, intraocular insertion of a thin glass capillary (without damage to the retina) or a combination of both. Glial reactivity was assessed through the use of a battery of antibodies which recognise four different proteins--glial fibrillary protein (GFAP) and three other proteins designated respectively MA1, 4D6 and 4H11. Retinal astrocytes did not exhibit any changes in normally expressed GFAP or MA1. By contrast, the expression of GFAP and MA1 in Müller cells increased 14 days following section of the optic nerve and/or intravitreal insertions of a glass capillary. Three days postoperatively, the expression of GFAP, but not MA1, had already increased significantly in Müller cells. 4D6 and 4H11 proteins were not expressed in astrocytes. In Müller cells, the levels of these proteins increased significantly following combined optic nerve section and intraocular insertion of a glass capillary. Thus, a mechanical disturbance of the intraocular environment constitutes a more effective stimulus in increasing the expression of some Müllerian proteins than damage to the axons of retinal ganglion cells. Such changes have important implications for various ocular treatments that involve intraocular administration of drugs, as well as for the survival/regeneration potential of retinal ganglion cells undergoing Wallerian degeneration.

Vitread proliferation of filamentous processes in avian Müller cells and its putative functional correlates.

In order to examine to what extent the neuronal and metabolic activities of avascular vertebrate retinae are reflected in the morphology of their Müller cells we have studied (by using several monoclonal antibodies) the morphology of Müller cells in two species of diurnal birds (chicken, Gallus domesticus, and pigeon, Columba livia) and one species of nocturnal saltwater crocodiles (Crocodylus porosi). In all three species, the outer nuclear layer is relatively thin and the Müller cell trunks divide into rootlets that wrap around the photoreceptors. In both diurnal birds studied, the trunks of Müller cells in the inner plexiform layers invariably divide into numerous fine filamentous processes that terminate in small expansions covering most of the vitreal surface of the retina. Furthermore, the networks of filamentous processes of birds' Müller cells exhibit conspicuous horizontal lamination in the inner plexiform layer. In contrast, the filamentous processes arising from the individual Müller cell trunks of the crocodile, if present, are much less numerous and less widely spread than those of diurnal birds. It is proposed that the splitting of the Müller cell trunks into numerous filamentous processes terminating in small vitreal expansions represents a morphological adaptation for: 1) effective spatial buffering of K+ ions in thick and presumably metabolically highly active, yet avascular, avian retinae, and 2) effective absorption and distribution of nutrients leaking from the vitreally located supplemental nutritive organ, the pecten.

The pattern of expression of a 13-kDa protein (CS-32 antigen) in the retina and its target nuclei in developing and mature rats.

A polyclonal antibody (designated CS-32) immunopurifies a 13-kDa protein from the superior colliculus (SC) of neonatal rat. There is a strikingly inverse temporal relationship in expression of the 13-kDa protein in the developing retina and its principal target nuclei, SC and the dorsal lateral geniculate nucleus. The 13-kDa is a cell-surface protein expressed exclusively by neurones and is clearly associated with synapses.

Interactions of living astrocytes in vitro: evidence of the development of contact spacing.

We have studied the behaviour of living, process-bearing astrocytes in vitro, observing groups of cells at daily intervals for up to 7 days. Each cell initially formed two processes, appearing bipolar in shape, and with further time in culture, grew additional processes and appeared stellate. As their processes grew, the interactions between astrocytes underwent characteristic changes. While bipolar, the cells appeared to avoid making contact, lying parallel to each other. As they became stellate, the astrocytes made extensive contact with neighbours, gradually forming extended, contacting networks in which their somas were regularly spaced (as previously described). The interactions which led to the establishing of such arrays were also evident. If two cells were initially close or adjacent, they extended short processes to contact each other; then, as their processes grew, their somas moved apart, until they were separated by 60-120 microns. If two cells were initially well separated, each directed processes towards the other until contact was made, often with striking precision, and their somas then moved together, until they were separated by 60-120 microns. These behaviours of contact, separation, and approach caused astrocytes to form clusters, within which their somas appeared regularly spaced, and may represent the interactions which occur among astrocytes during normal development to produce the regularly spaced arrays of astrocytes described in earlier studies of intact central nervous tissue.

Contact-spacing among astrocytes is independent of neighbouring structures: in vivo and in vitro evidence.

We have examined the morphology of astrocytes and the arrays they form in two situations, in retinas from which ganglion cells and blood vessels have been caused to degenerate, and in vitro. These observations were made to test whether the regularity of the spacing of astrocytes within normal central nervous tissue results from interaction among astrocytes, or from interaction between astrocytes and other elements of that tissue. Both in the partially degenerated cat retina, and in cultures of astrocytes from neonatal rat cortex, astrocytes make and maintain contact with neighbouring astrocytes, yet space their somas apart, giving regularity to the arrays. These results support the hypothesis that the regularity observed in arrays of astrocytes in intact tissue results from an interaction among astrocytes, independent of neighbouring structures, and lead us to suggest that the cell-cell interactions involved in contact spacing serve to distribute astrocytes through the central nervous system, and may, in other tissues, underlie the formation of epithelia.

Müller cells in vascular and avascular retinae: a survey of seven mammals.

Eight monoclonal antibodies were used to label Müller cells in four mammals that have vascular retinae (cats, dogs, humans, and rats) and in three with avascular retinae (echidnas, guinea pigs, and rabbits). Müller cells were found to have a fairly uniform retinal distribution in seven species, with a mean density of 8,000-13,000 cells mm-2. Müller cells in avascular retinae differ from their vascular counterparts in four respects. First, they are shorter than those in vascular retinae. This difference is mainly due to a reduction in the thickness of the outer nuclear layer. Second, the trunks of Müller cells in avascular retinae tend to be thicker, although those in echidnas are an exception to this trend. Third, Müller cell rootlets in avascular retinae follow a more tortuous course than those in vascular retinae, reflecting the fact that photoreceptor nuclei in the two types of retina have different shapes and stacking patterns. Fourth, due to a reduction in the density of photoreceptors in avascular retinae, there are fewer neurones per Müller cell. Although these four features may enable Müller cells to assist the nutrition of neurones in the inner layers of avascular retinae, they are unlikely to be morphological specializations that have evolved for that purpose. Rather, these features appear to be a direct consequence of the fact that avascular retinae are thinner and have a differently organised outer nuclear layer. These features aside, Müller cells in avascular retinae closely resemble their counterparts in vascular retinae.

Structure of the macroglia of the retina: sharing and division of labour between astrocytes and Müller cells.

A detailed comparison is made between astrocytes and Müller cells of the cat's retina, with emphasis on their structural specialisations. Evidence is presented that astrocytes and Müller cells both contribute to the formation of the inner glia limitans of the retina, the glia limitans of vessels, and the glial sheaths of neurones. In particular, it was noted that both astrocytes and Müller cells wrap bundles of ganglion cells axons, that both contribute processes to the glial convergence on the initial segments and node-like structures of axons, and that both wrap the somas of neurones in the ganglion cell layer. Further, it was noted that adherent junctions form between astrocytes, between Müller cells, and between astrocytes and Müller cells, but not between these cells and neurones, or among neurones. These similarities suggest that astrocytes and Müller cells function interchangeably in many respects, and we suggest that they be regarded as variants of macroglia. Quantitative differences between astrocytes and Müller cells were noted in their ensheathment of neurones. In particular, the glial sheaths around the somas of ganglion cells are formed predominantly by Müller cells, and the glial processes attached to node-like specialisations of their axons are formed mainly by astrocytes. One qualitative difference was noted between the two cell classes. The gap junctions which form between astrocytes do not form between Müller cells or between cells of the two classes. From these differences, and previously established features of their shape, orientation, distribution and origin, a hypothesis is developed of the specialisation of macroglia represented by Müller cells.

4B2: a monoclonal antibody to ganglion and bipolar cells in the retina of the cat, generated by intrasplenic immunization.

Using tissue from the inner layers of the retina as the immunogen, we have prepared a monoclonal antibody which is selective for two classes of neuron in the cat retina. The antibody, termed 4B2, was generated by intrasplenic injection, demonstrating that neuron-specific antibodies can be elicited by this technique, which requires only micrograms of immunogen. The 4B2 binds to the somas, dendrites and axons of ganglion cells. Double labelling with 4B2 and with a retrograde tracer injected into the retino-recipient nuclei of the brain demonstrates that, of the cell classes in the ganglion cell layer, 4B2 labels only ganglion cells, and that, amongst ganglion cells, 4B2 is strongly selective for large (alpha-) and medium-sized (beta- and perhaps gamma-) cells. Double labelling with 4B2 and with markers for amacrine cells confirms tht 4B2 does not label amacrine cells. Double labelling with 4B2 and with anti-GFAP confirms that 4B2 does not label the macroglia of the retina. The 4B2 also labels bipolar cells, showing their somas, dendrites and axons. The axons of the labelled cells terminate in large boutons, in the innermost part of the inner plexiform layer and in the ganglion cell layer. This level of termination suggests that 4B2 labels rod bipolars, and perhaps a subgroup of cone bipolars. Double labelling with 4B2 and with markers for amacrine, horizontal and Müller cells indicates that other cell classes with somas in the inner nuclear layer remain unlabelled. In addition, we have examined the ontogeny of 4B2 labelling in the cat retina. The developmental sequence of labelling with 4B2 follows the sequence in which the cells are born, first ganglion and then bipolar cells.

Contact spacing among astrocytes in the central nervous system: an hypothesis of their structural role.

Astrocytes are found throughout the central nervous system. They interact closely with surrounding structures, their processes contributing to the glia limitans of the neural tube, and to the glial investment of blood vessels, and of the somas, axons, and synaptic structures of neurones. This paper presents evidence that astrocytes in the central nervous system also interact with each other in a dual way, adhering to their neighbours via their processes, and repelling the somas of those neighbours. We suggest that this interaction, which has been termed contact spacing, distributes astrocytes through the central nervous system, and forms the basis of their structural role.

"Sunbursts" in the inner plexiform layer: a spectacular feature of Müller cells in the retina of the cat.

A previously unrecognised structure in the cat retina is described. Seen in Golgi-impregnated, wholemounted retinas, each such structure comprises processes radiating across the inner plexiform layer from a dense, vellate core. The processes are numerous, and largely unbranched, and give the impression of rays radiating from a point source; the structure is therefore termed a "sunburst." Evidence is presented from Golgi-impregnated retinas, and from retinas labelled with monoclonal antibodies to Müller cells, that the core of each sunburst is the inner process of a Müller cell. The sunbursts are numerous and overlap extensively, so that when neighbouring sunbursts are impregnated, they are seen to form a dense mat of processes extending across the IPL. It is suggested that each Müller cell forms a sunburst and that sunbursts form a major glial component of the neuropil of the inner plexiform layer.

Müller cells in adult rabbit retinae: morphology, distribution and implications for function and development.

We describe the morphology and distribution of Müller cells in wholemounts of rabbit retinae labelled with either monoclonal antibodies (anti-Vimentin, 3H3, 4D6, and 4H11), or intracellular horseradish peroxidase. Several new features of Müller cell organization are noted. First, Müller cells appear to compose a single morphological class and their morphology varies systematically with retinal thickness. Second, in contrast to other retinal glia, Müller cells have a neuronlike distribution, with a peak density of 10,700-15,000 cells per mm2 at the visual streak and a minimum density of 4,400-6,000 per mm2 at both the superior and inferior retinal edges. There are 4.2 +/- 0.5 x 10(6) Müller cells per retina. Third, unlike in other species, rabbit Müller cells do not contact blood vessels, suggesting that they do not participate in the transfer of metabolites or in the blood:retinal barrier. Fourth, each Müller cell has a vitread endfoot about 20-40 microns in diameter composed of numerous fimbriae. The fimbriae from a single Müller cell generally contact several axon fascicles in the nerve fibre layer, and at each point along its length each fascicle is enclosed by the overlapping fimbriae from several Müller cells. Fifth, in the inner and outer plexiform layers, numerous filamentous branchlets extend 20 microns or more from the radial trunk, interweaving with branchlets from nearby Müller cells to form dense and continuous strata. In the ganglion cell layer and outer nuclear layer, Müller cell processes completely wrap neuronal somata, whereas in the inner nuclear layer they partially wrap somata. We discuss the functional and developmental implications of these observations.

Evidence for three morphological classes of astrocyte in the adult rabbit retina: functional and developmental implications.

Commercially available antibodies to glial fibrillary acidic protein (GFAP) and vimentin were used in conjunction with our own antibodies (3F11, MA1 and 4D6) to label astrocytes in the adult albino rabbit retina. Anti-GFAP labels the entire astrocyte population and shows them to be morphologically diverse. By contrast, the remaining four antibodies label different subpopulations of astrocytes. Comparisons of the shapes and distributions of cells composing these subpopulations led us to distinguish three morphological classes: Class A are predominantly perivascular astrocytes, Class B are territorial and rarely contact vessels or axons, Class C (which seems to contain four subclasses) are astrocytes that are predominantly associated with axons.

Müller cell endfeet at the inner surface of the retina: light microscopy.

Using fractions of the protein spectrum of the cat retina as immunogens, we have generated antibodies with substantial specificity for the Müller cells of the retina of cat, rabbit, guinea pig, and rat. The antibodies appear to bind to the filamentous components of the Müller cells and allow demonstration of the pattern of Müller cell endfeet at the inner surface of the retina, best seen in wholemount preparations. In sections and at the edge of wholemount preparations the somas and processes of the cells can be observed. Müller cells are more evenly distributed over the retina than ganglion cells, indicating that their proliferation continues during the differential growth of retina which continues into postnatal life. The morphology and distribution of the endfeet varies with the structures present at the inner surface of the retina. Where the axon bundles are thick, the endfeet are relatively small and are confined to narrow rows between bundles. Müller cell endfeet are also separated widely by large blood vessels. In both situations, it seems likely that Müller cells and astrocytes both contribute, perhaps competitively, to form the glia limitans of the inner surface of the retina. Where the somas of neurones are densely packed in the ganglion cell layer, the endfeet are small and numerous, forming rings around the somas. Where axon bundles, vessels, and somas are sparse, the endfeet appear largest and form a regular array.

Relationship between astrocytes, ganglion cells and vasculature of the retina.

We have studied the distribution of astrocytes in the ganglion cell and nerve fibre layers of the retina in cat, rat, rabbit, and possum using anti-serum and a monoclonal antibody against glial fibrillary acidic protein (GFAP) and our own monoclonal antibody against glial filaments. The distribution of retinal astrocytes appears to be strongly determined by the vasculature of the retina; astrocytes are absent from almost all the retina of the possum and from the avascular regions of the rabbit retina. In the cat and rabbit, retinal astrocytes also show a strong affinity for the bundles of ganglion cell axons found at the inner surface of the retina. Retinal astrocytes do not invest the somas of ganglion cells, and even in areas of retina in which they are numerous, they are sharply confined to the layer of ganglion cell axons. It is suggested that retinal astrocytes are "immigrant" fibrous astrocytes that enter the retina with its vasculature.

Preparation of monospecific antiserum to lupin nodule glutamate dehydrogenase.

Immunization of rabbit, using biochemically homogeneous glutamate dehydrogenase, proved to be unsuitable to produce monospecific antiserum. The presence of traces of contaminating immunogen (undetected by physiochemical methods) induced the production of other antibodies. Procedures for rigorously establishing monospecificity of antisera and a technique for preparation of monospecific antiserum, using immunologically impure antigens, are described.