Developmental alterations in membrane organization of rat subpial astrocytes

The morphology and spatial arrangement of astrocytes in the optic nerve head of the mouse

We evaluated the shapes, numbers, and spatial distribution of astrocytes within the glial lamina, an astrocyte-rich region at the junction of the retina and optic nerve. A primary aim was to determine how the population of astrocytes, collectively, partitions the axonal space in this region. Astrocyte processes labeled with glial fibrillary acidic protein (GFAP) compartmentalize ganglion cell axons into bundles, forming "glial tubes," and giving the glial architecture of the optic nerve head in transverse section a honeycomb appearance. The shapes of individual astrocytes were studied by using transgenic mice that express enhanced green fluorescent protein in isolated astrocytes (hGFAPpr-EGFP). Within the glial lamina the astrocytes were transverse in orientation, with thick, smooth primary processes emanating from a cytoplasmic expansion of the soma. Spaces between the processes of neighboring astrocytes were spatially aligned, to form the apertures through which the bundles of optic axons pass. The processes of individual astrocytes were far-reaching-they could span most of the width of the nerve-and overlapped the anatomical domains of other near and distant astrocytes. Thus, astrocytes in the glial lamina do not tile: each astrocyte participates in ensheathing approximately one-quarter of all of the axon bundles in the nerve, and each glial tube contains the processes of about nine astrocytes. This raises the mechanistic question of how, in glaucoma or other cases of nerve damage, the glial response can be confined to a circumscribed region where damage to axons has occurred.

Agrin defines polarized distribution of orthogonal arrays of particles in astrocytes

Accumulating evidence indicates that agrin, a heparan sulphate proteoglycan of the extracellular matrix, plays a role in the organization and maintenance of the blood-brain barrier. This evidence is based on the differential effects of agrin isoforms on the expression and distribution of the water channel protein, aquaporin-4 (AQP4), on the swelling capacity of cultured astrocytes of neonatal mice and on freeze-fracture data revealing an agrin-dependent clustering of orthogonal arrays of particles (OAPs), the structural equivalent of AQP4. Here, we show that the OAP density in agrin-null mice is dramatically decreased in comparison with wild-types, by using quantitative freeze-fracture analysis of astrocytic membranes. In contrast, anti-AQP4 immunohistochemistry has revealed that the immunoreactivity of the superficial astrocytic endfeet of the agrin-null mouse is comparable with that in wild-type mice. Moreover, in vitro, wild-type and agrin-null astrocytes cultured from mouse embryos at embryonic day 19.5 differ neither in AQP4 immunoreactivity, nor in OAP density in freeze-fracture replicas. Analyses of brain tissue samples and cultured astrocytes by reverse transcription with the polymerase chain reaction have not demonstrated any difference in the level of AQP4 mRNA between wild-type astrocytes and astrocytes from agrin-null mice. Furthermore, we have been unable to detect any difference in the swelling capacity between wild-type and agrin-null astrocytes. These results clearly demonstrate, for the first time, that agrin plays a pivotal role for the clustering of OAPs in the endfoot membranes of astrocytes, whereas the mere presence of AQP4 is not sufficient for OAP clustering.

The astrocyte response to gamma-aminobutyric acid attenuates with age in the rat optic nerve

There is increasing evidence that glial cells respond to the inhibitory neurotransmitter gamma-aminobutyric acid (GABA), and astrocytes have been shown to possess GABAA receptors both in vivo and in vitro. A recent study by Sakatani et al. (Proc. R. Soc. Lond. B247, 155 (1992)) demonstrated the transient expression of functional GABAA receptors in the developing rat optic nerve, but axonal and glial components of the response were not distinguished. To help address this problem, we have determined the electrophysiological response to GABA in astrocytes of the isolated intact optic nerves from neonatal rats, identified morphologically following intracellular injection of horseradish peroxidase. Astrocytes responded to GABA by a GABAA receptor-mediated depolarization which attenuated gradually during post-natal development; astrocytes in 21-day-old nerves were not observed to respond to GABA. The results indicate the transient presence of functional GABAA receptors in developing rat optic nerve astrocytes in situ, and we speculate upon a role for GABA in glial signalling and the organization of axonglial interrelations during development.

Astrocytes alter their polarity in organotypic slice cultures of rat visual cortex

The ultrastructure of astrocytes in an organotypic slice culture of the rat visual cortex was investigated using ultrathin sections and freeze-fracture replicas. After a culture period of 9-15 days, a glial scaffold formed that separated the bulk of the slice neuropil from the medium and the underlying plasma clot. However, the glial cells and processes did not build a dense barrier but allowed the outgrowth of neurites. A basal lamina covering the medium-oriented surface of the astrocytes was not found. In freeze-fracture replicas, orthogonal arrays of particles (OAP) were characteristic components of astrocytic membranes. The OAP density in membranes bordering the medium was 35 +/- 13 OAP/microns 2, corresponding to 2.5% of this membrane area; the OAP density in membranes within the slice neuropil was 22 +/- 12 OAP/microns 2, corresponding to 1.4% of this membrane area. Although the difference was significant, it was greatly reduced when comparing OAP densities in endfoot and non-endfoot membranes in vivo. Another node of polarity was recognized in astrocytes of the organotypic slice culture. In membranes of astrocytes bordering upon the medium, the density of non-OAP intramembranous particles (IMP) was clearly higher (1130 +/- 136 IMP/microns 2) than in membranes of astrocytes in the center of the slice (700 +/- 172 IMP/microns 2). This pronounced IMP-related polarity was observed neither in vivo nor in cultured astrocytes. The present study suggests, together with data from the literature, that the distribution of astrocytic OAP across the cell surface is influenced by the existence of a basal lamina and neuronal activity, and that astrocytes possess a more remarkable plasticity of membrane structure than previously suspected.

Three-dimensional morphology of astrocytes and oligodendrocytes in the intact mouse optic nerve

The three-dimensional morphology of astrocytes and oligodendrocytes was analysed in the isolated intact mature mouse optic nerve, by correlating laser scanning confocal microscopy and camera lucida drawings of single cells, dye-filled with lysinated rhodamine dextran or horseradish peroxidase, respectively. These techniques enabled the entire process field of single dye-filled cells to be visualized in all planes and resolved the fine details of glial morphology. Morphometric analysis showed that the processes of all astrocytes had branches ending at the pial surface, on blood vessels, and freely in the nerve; branches ending in the nerve were described to end at nodes of Ranvier in the accompanying paper. Astrocytes were classified into a single morphological population in which each cell subserved multiple functions. The results of this study do not support the contention that astrocytes can be subdivided into two morphological and functional subtypes, namely type-1 and type-2, which have process ending either at the glia limitans or at nodes, respectively. Three-dimensional analysis of oligodendrocyte units, defined as the oligodendrocyte, its processes and the axons it ensheaths, showed the provision of single myelin segments for an average of 19 nearby axons (range 12-35) with a mean internodal length of 138 microns (range 50-350 microns). Mouse optic nerve oligodendrocytes were a homogeneous population and were markedly similar to those in the rat optic nerve. The results of our analysis of oligodendrocyte morphology are consistent with the view that the number and internodal length of myelin sheaths supported by a single oligodendrocyte are related to the diameter of the ensheathed axons.

Morphology of astrocytes and oligodendrocytes during development in the intact rat optic nerve

The detailed three-dimensional morphology of macroglial cells was determined throughout postnatal development in the intact rat optic nerve, a central nervous system white matter tract. Over 750 cells were analyzed by intracellular injection of horseradish peroxidase or Lucifer Yellow to provide a new perspective of glial differentiation in situ. Retrograde analysis of changes in glial morphology allowed us to identify developmental timetables for three morphological subclasses of astrocytes and oligodendrocytes, and to estimate their time of emergence from undifferentiated glial progenitors. Glial progenitors were recognised throughout postnatal development and persisted in 35-day-old nerves, where we suggest they represent adult progenitor cells. Astrocytes were present at birth, but the majority of these cells developed over the first week as three morphological classes emerged having either transverse, random, or longitudinal process orientation. Several lines of evidence led us to believe that the majority of astrocytes in the rat optic nerve were morphological variations of a single cell type. Young oligodendrocytes were first observed 2 days after birth, indicating that they diverged from progenitors at or near this time. During early development these cells extended a large number of fine processes, which then bifurcated and extended along axons. Later, as myelination proceeded, oligodendrocytes exhibited fewer processes which grew symmetrically and uniformly along the axons, resulting in a highly stereotypic mature oligodendrocyte form. Our analysis of oligodendrocyte growth suggests that these cells did not myelinate axons in a random manner and that axons may influence the myelinating processes of nearby oligodendrocytes.