The locations of mitochondria in mammalian photoreceptors: relation to retinal vasculature.

Adult mammalian photoreceptors are elongated cells, and their mitochondria are sequestered to the ends of the cell, to the inner segments and (in some species) to axon terminals in the outer plexiform layer (OPL). We hypothesised that mitochondria migrate to these locations towards sources of oxygen, from the choroid and (in some species) from the deep capillaries of the retinal circulation. Six mammalian species were surveyed, using electron and light microscopy, including immunohistochemistry for the mitochondrial enzyme cytochrome oxidase (CO). In all 6 species, mitochondria were absent from photoreceptor somas and were numerous in inner segments. Mitochondria were prominent in axon terminals in 3 species (mouse, rat, human) with a retinal circulation and were absent from those terminals in 3 species (wallaby, rat, guinea pig) with avascular retinas. Further, in a human developmental series, it was evident that mitochondria migrate within rods and cones, towards and eventually past the outer limiting membrane (OLM), into the inner segment. In Müller and RPE cells also, mitochondria concentrated at the external surface of the cells. Neurones located in the inner layers of avascular retinas have mitochondria, but their expression of CO is low. Mitochondrial locations in photoreceptors, Müller and RPE cells are economically explained as the result of migration within the cell towards sources of oxygen. In photoreceptors, this migration results in a separation of mitochondria from the nuclear genome; this separation may be a factor in the vulnerability of photoreceptors to mutations, toxins and environmental stresses, which other retinal neurones survive.

Multiple vulnerability of photoreceptors to mesopic ambient light in the P23H transgenic rat.

The P23H transgenic rat was engineered to mimic a human form of retinal degeneration caused by a mutation in rhodopsin. We have tested whether the P23H transgene influences the vulnerability of photoreceptors to modest variations in ambient light, well within the physiological range. P23H-3 (P23H line 3) and control Sprague-Dawley (SD) rats were raised in cyclic light (12 h light, 12 h dark), with the light phase set at either 5 lx ('scotopic-reared') or 40-60 lx ('mesopic-reared'). Mesopic rearing reduced the length of outer segments (OSs) in both SD and P23H-3 strains, but the shortening was more marked in the P23H-3 strain. Mesopic rearing was associated with thinning of the ONL, again more prominently in the P23H-3. Correspondingly, mesopic rearing increased the rate of photoreceptor death (assessed by TUNEL labelling), the increase occurring during early postnatal life. Mesopic rearing upregulated FGF-2 (basic fibroblast growth factor) levels in photoreceptors and glial fibrillary acidic protein (GFAP) in Müller cells in both SD and P23H-3 strains; again the changes were more marked in the P23H-3. Finally, mesopic rearing decreased the amplitude of the a-wave of the ERG in both strains; again the effect was greater in the P23H-3 strain. The ERG decline induced in both strains by mesopic-rearing can be explained by a reduction of functional OS membrane, due to a combination of photoreceptor death and OS shortening. The P23H-3 transgene makes photoreceptors abnormally vulnerable to modest levels of ambient light, their vulnerability being evident in multiple ways. In humans suffering photoreceptor degeneration from comparable genetic causes, light restriction may preserve the number and the function of photoreceptors.

Transendothelial electrical resistance of bovine retinal capillary endothelial cells is influenced by cell growth patterns: an ultrastructural study.

The purpose of this study was to describe the ultrastructural features of an in vitro capillary endothelial cell model of blood-retinal barrier permeability and to relate morphological features with transendothelial electrical resistance. The electrical resistance of endothelial cell monocultures on small and large pore size polycarbonate Transwell filters was measured and compared with cocultures of endothelial cells and Müller cells. There was a wide variation in electrical resistance measurements with many preparations not achieving a functional barrier. The ultrastructural features associated with barrier function in vitro were studied by comparing cultures that exhibited a 'tight' or 'leaky' barrier when measured immediately prior to processing for electron microscopy. Preparations with low transendothelial electrical resistance were associated with irregular cell growth when studied morphologically. It was concluded that parallel light and electron microscopic studies are important for validation of in vitro models of vascular endothelial permeability.

FGFR1 expression and FGFR1-FGF-2 colocalisation in rat retina: sites of FGF-2 action on rat photoreceptors.

AIM: To identify sites of FGF-2 action on photoreceptors of the rat retina, by localizing FGFR1 in the intact retina, and to assess the colocalisation of FGF-2 with FGFR1. METHODS: Immunohistochemistry and confocal microscopy were used to localise FGF-2 and FGFR1 in cryosections of the rat retina, both normal retina and retina stressed by exposure to bright continuous light (1000 lux, 24h). Antibodies to synaptophysin (SY), cytochrome oxidase (CO) and opsin were used to relate FGFR1-labelling to synaptic vesicles, mitochondria and the photoreceptor cell membrane. Electron microscopy was used to demonstrate the location of synapses within the outer plexiform layer (OPL). RESULTS: FGFR1 was most prominent in the outer nuclear layer (ONL), as diffuse labelling of photoreceptor cytoplasm and as granules between photoreceptor somas. FGFR1 labelling was also observed in the outer synapse-rich sublayer of the OPL where it colocalised with SY, but not with CO-labelled mitochondria. In stressed retina, both at the edge of normal retina and after light stress, FGFR1 expression was upregulated in both the ONL and the OPL. Colocalisation of FGFR1 with FGF-2 could not be demonstrated in unstressed retina, but was demonstrable in stressed retina, in both the ONL and OPL. CONCLUSIONS: FGFR1 is prominent in the cytoplasm of photoreceptors, and in their axon terminals, where it is closely associated with synaptic vesicles. Colocalisation of FGFR1 and FGF-2 could be demonstrated in stressed retina, in the cytoplasm and the axon terminals of photoreceptors. The known protective action of FGF-2 may be exerted at the photoreceptor soma. The action of FGF-2 in inhibiting the ERG b-wave may be exerted at the axon terminal.