Expression of reverse cholesterol transport proteins ATP-binding cassette A1 (ABCA1) and scavenger receptor BI (SR-BI) in the retina and retinal pigment epithelium.

AIMS: Excessive lipid accumulation in Bruch's membrane (BrM) is a hallmark of ageing, the major risk factor for age-related macular degeneration (AMD). Retinal pigment epithelial (RPE) cells may utilise reverse cholesterol transport (RCT) activity to move lipid into BrM, mediated through ATP-binding cassette A1 (ABCA1) and scavenger receptor BI (SR-BI). METHODS: ABCA1 expression was assessed by reverse transcription polymerase chain reaction (RT-PCR) and western blotting of human RPE cell extracts. Lipid transport assays were performed using radiolabelled photoreceptor outer segments (POS). ABCA1 and SR-BI expression was examined in normal mouse eyes by immunofluorescence staining. BrMs of ABCA1 and SR-BI heterozygous mice were examined microscopically. RESULTS: Human RPE cells expressed ABCA1 mRNA and protein. The ABCA1 and SR-BI inhibitor glyburide (also known as glibenclamide) abolished basal transport of POS-derived lipids in RPE cells in the presence of high-density lipoprotein. Mouse retina and RPE expressed ABCA1 and SR-BI. SR-BI was highly expressed in RPE. BrMs were significantly thickened in SR-BI heterozygous mice, but not in ABCA1 heterozygous mice. CONCLUSION: RPE cells express ABCA1 and SR-BI. This implies a significant role for SR-BI and ABCA1 in lipid transport and RCT in the retina and RPE.

Quantitative genetics of age-related retinal degeneration: a second F1 intercross between the A/J and C57BL/6 strains.

PURPOSE: Previously, several quantitative trait loci (QTL) that influence age-related retinal degeneration (ageRD) were demonstrated in a cross between the C57BL/6J-c(2J) and BALB/cByJ strains (B x C). In this study, as a complementary approach to ongoing recombinant progeny testing for the purpose of identifying candidate quantitative trait genes (QTG), a second test cross using the A/J and the pigmented C57BL/6J strains (A x B) was carried out. The albino A/J strain was selected because it had the most amount of ageRD among several inbred strains tested, and the pigmented C57BL/6J strain was selected because along with its coisogenic counterpart C57BL/6J-c(2J) it had the least amount of ageRD. Thus, the effect of pigment on ageRD could be tested at the same time that the C57BL/6 genetic background was kept in common between the crosses from the two studies for the purpose of comparison. METHODS: A non-reciprocal F1 intercross between the A/J and C57BL/6J strains produced 170 F2 progeny. At 8 months of age after being maintained in relatively dim light, F2 mice, control mice and mice of other strains were evaluated for retinal degeneration by measurement of the thickness of the outer nuclear layer of the retina. The F2 mice were genotyped with dinucleotide repeat markers spanning the genome. Correlation of genotype with phenotype was made with Map Manager QTX software. RESULTS: Comparison of several strains of mice including the pigmented strains 129S1/SvImJ and C57BL/6J and the albino strains A/J, NZW/LacJ, BALB/cByJ and C57BL/6J-c(2J), showed significant differences in ageRD. The greatest difference was between the albino A/J strain and the pigmented C57BL/6J strain. However, there was no significant difference between the pigmented C57BL/6J and its albino coisogenic counterpart C57BL/6J-c(2J). Neither was there significant difference between the pigmented and albino F2 mice from the A x B cross. On the other hand, F2 males had a small but significantly lower amount of ageRD than females. Several QTL were identified in the A x B cross but surprisingly none of the 3 major QTL present in the original B x C cross (Chrs 6, 10, and 16) was present. There were minor QTL on proximal Chr 12 and proximal Chr 14 in common between the two crosses, and the proximal Chr 12 QTL was present in a previous light damage study involving the B and C strains. At least one sex-limited QTL was present on the X chromosome with a peak in a different location from that of a sex-limited QTL in the previous B x C study. In addition, the protective X allele was from the BALB/cByJ strain in the B x C cross and from C57BL/6J in the A x B cross. In both crosses, the C57BL/6J X-chromosome allele was recessive. CONCLUSIONS: Significant differences were observed in ageRD among several inbred strains of mice maintained in relatively dim light. AgeRD was not influenced by pigment but was influenced by gender, albeit to a small degree. The presence of the same QTL in one light-induced and two ageRD studies suggests at least partial commonality in retinal degeneration pathways of different primary cause. However, the three main QTL present in the B x C cross were absent from the A x B cross. This suggests that the genetic determinants responsible for the greater sensitivity to ageRD of BALB/cByJ and A/J relative to C57BL/6J are not the same. This is supported by the presence of sex-limited X-chromosome QTL in the two crosses in which the C57BL/6J allele is protective relative to the A allele and sensitive relative to the C allele. The findings in the two studies of differing allelic relationships of QTG, and differing QTL aid the identification of candidate genes mapping to critical QTL. Identifying natural modifying genes that influence retinal degeneration resulting from any causal pathway, especially those QTG that are protective, will open avenues of study that may lead to broad based therapies for people suffering retinal degenerative diseases.

Lack of p75 receptor does not protect photoreceptors from light-induced cell death.

Rod photoreceptors are susceptible to light-induced cell death. Previous results have suggested that the neurotrophin receptor p75 in Müller cells controls photoreceptor cell death during light-exposure by suppressing trophic factor release; and consequently, if p75 is blocked or eliminated during light-exposure, apoptosis is delayed. We explored this question by examining photoreceptor cell survival in albino p75(-/-) mice as well as their heterozygous and homozygous littermates. Photoreceptor cell death was examined in semi-thin sections by counting the remaining rows of photoreceptors. No difference in the amount of cell death was found between p75(+/+) and p75(-/-) animals, whereas the single copy of p75 in the heterozygous p75(+/-) mice provided significant neuroprotection. Cell death in the wild-type animals may indeed be mediated by p75, whereas other known apoptosis pathways may be activated in the p75(-/-) mice. The pro-apoptotic activity of the p75 receptor may have been partially suppressed in the heterozygous p75(+/-) mice by the silencing effect of the Trk receptor. Thus, our results suggest that p75 signaling does not mediate the main apoptosis pathway activated during light-damage.

Correction of the retinal dystrophy phenotype of the RCS rat by viral gene transfer of Mertk.

The Royal College of Surgeons (RCS) rat is a widely studied animal model of retinal degeneration in which the inability of the retinal pigment epithelium (RPE) to phagocytize shed photoreceptor outer segments leads to a progressive loss of rod and cone photoreceptors. We recently used positional cloning to demonstrate that the gene Mertk likely corresponds to the retinal dystrophy (rdy) locus of the RCS rat. In the present study, we sought to determine whether gene transfer of Mertk to a RCS rat retina would result in correction of the RPE phagocytosis defect and preservation of photoreceptors. We used subretinal injection of a recombinant replication-deficient adenovirus encoding rat Mertk to deliver the gene to the eyes of young RCS rats. Electrophysiological assessment of animals 30 days after injection revealed an increased sensitivity of treated eyes to low-intensity light. Histologic and ultrastructural assessment demonstrated substantial sparing of photoreceptors, preservation of outer segment structure, and correction of the RPE phagocytosis defect in areas surrounding the injection site. Our results provide definitive evidence that mutation of Mertk underlies the RCS retinal dystrophy phenotype, and that the phenotype can be corrected by treatment of juvenile animals. To our knowledge, this is the first demonstration of complementation of both a functional cellular defect (phagocytosis) and a photoreceptor degeneration by gene transfer to the RPE. These results, together with the recent discovery of MERTK mutations in individuals with retinitis pigmentosa, emphasize the importance of the RCS rat as a model for gene therapy of diseases that arise from RPE dysfunction.

Retinal degeneration in the nervous mutant mouse. IV. Inner retinal changes.

We have recently noted that the inner nuclear layer (INL) and the inner plexiform layer (IPL) were significantly thinner in mice homozygous for the nervous defect (nr / nr) than in control (nr /+ or +/+) littermates. Here, we have carried out a series of anatomical studies to further understand these inner retinal changes. At postnatal day (P) 13, there was no difference in the inner retina between nervous and control mice, while a significant difference was observed at P30. Similar changes were not seen in other mouse models of photoreceptor degeneration. There was a significant reduction in the density of cells in the INL that were stained by antibodies against the inhibitory neurotransmitters GABA and glycine. These results indicate that the nervous defect causes a degeneration of one or more sub-types of amacrine cells, in addition to the loss of cerebellar Purkinje cells and retinal photoreceptors that is known to occur in these mutant animals. Finally, evidence is provided that photoreceptors die by an apoptotic pathway in nervous mice.

Ribozyme rescue of photoreceptor cells in P23H transgenic rats: long-term survival and late-stage therapy.

Ribozyme-directed cleavage of mutant mRNAs appears to be a potentially effective therapeutic measure for dominantly inherited diseases. We previously demonstrated that two ribozymes targeted to the P23H mutation in rhodopsin slow photoreceptor degeneration in transgenic rats for up to 3 months of age when injected before significant degeneration at postnatal day (P) 15. We now have explored whether ribozyme rescue persists at older ages, and whether ribozymes are effective when injected later in the degeneration after significant photoreceptor cell loss. Recombinant adeno-associated virus (rAAV) vectors incorporating a proximal bovine rod opsin promoter were used to transfer either hairpin or hammerhead ribozyme genes to photoreceptors. For the study of long-term survival, rAAV was administered by subretinal injection at P15, and the rats were allowed to live up to 8 months of age. For the study of late-stage gene transfer, rAAV was administered at P30 or P45, when 40-45% of the photoreceptors already had degenerated. Eyes were examined functionally by the electroretinogram and structurally by morphometric analysis. When injected at P15, expression of either ribozyme markedly slowed the rate of photoreceptor degeneration for at least 8 months and resulted in significantly greater electroretinogram amplitudes at least up to P180. When injected at P30 or P45, virtually the same number of photoreceptors survived at P130 as when injected at P15. Ribozyme rescue appears to be a potentially effective, long-term therapy for autosomal dominant retinal degeneration and is highly effective even when the gene transfer is done after significant photoreceptor cell loss.

A QTL on distal chromosome 3 that influences the severity of light-induced damage to mouse photoreceptors.

C57BL/6J-c(2J) (c2J) albino mice showed much less damage to their photoreceptors after exposure to prolonged light than BALB/c mice and seven other albino strains tested. There were no gender differences, and preliminary studies suggested that the c2J relative protective effect was a complex trait. A genome-wide scan using dinucleotide repeat markers was carried out for the analysis of 194 progeny of the backcross (c2J x BALB/c)F(1) x c2J and the thickness of the outer nuclear layer (ONL) of the retina was the quantitative trait reflecting retinal damage. Our results revealed a strong and highly significant quantitative trait locus (QTL) on mouse Chromosome (Chr) 3 that contributes almost 50% of the c2J protective effect, and three other very weak but significant QTLs on Chrs 9, 12, and 14. Interestingly, the Chrs 9 and 12 QTLs corresponded to relative susceptibility alleles in c2J (or relative protection alleles in BALB/c), the opposite of the relative protective effect of the QTLs on Chrs 3 and 14. We mapped the Rpe65 gene to the apex of the Chr 3 QTL (LOD score = 19.3). Northern analysis showed no difference in retinal expression of Rpe65 message between c2J and BALB/c mice. However, sequencing of the Rpe65 message revealed a single base change in codon 450, predicting a methionine in c2J and a leucine in BALB/c. When the retinas of aging BALB/c and c2J mice reared in normal cyclic light were compared, the BALB/c retinas showed a small but significant loss of photoreceptor cells, while the c2J retinas did not. Finding light damage-modifying genes in the mouse may open avenues of study for understanding age-related macular degeneration and other retinal degenerations, since light exposures may contribute to the course of these diseases.

Mutation of the receptor tyrosine kinase gene Mertk in the retinal dystrophic RCS rat.

Vertebrate photoreceptor cells are the basic sensory apparatus of the retina, capable of converting the energy of absorbed photons into neuronal signals. The proximal portions of mammalian photoreceptor outer segments are synthesized daily by cell bodies, and outer segment tips are shed with a circadian rhythm, resulting in a complete turnover of outer segments about every 9 days. The shed outer segments are phagocytosed by adjacent retinal pigment epithelial (RPE) cells, and metabolites are recycled to photoreceptors. The Royal College of Surgeons (RCS) rat is a widely studied, classic model of recessively inherited retinal degeneration in which the RPE fails to phagocytose shed outer segments, and photoreceptor cells subsequently die. We have used a positional cloning approach to study the rdy (retinal dystrophy) locus of the RCS rat. Within a 0.3 cM genetic inclusion interval, we have discovered a small deletion of RCS DNA that disrupts the gene encoding the receptor tyrosine kinase Mertk. The deletion includes the splice acceptor site upstream of the second coding exon of Mertk and results in a shortened transcript that lacks this exon. The aberrant transcript joins the first and third coding exons, leading to a frameshift and a translation termination signal 20 codons after the AUG. The concordance of these and other data indicate that Mertk is probably the gene for rdy. Our results provide genetic evidence for an essential role of a receptor tyrosine kinase in a specialized form of phagocytosis and suggest a molecular model for ingestion of outer segments by RPE cells.

Increased susceptibility to light damage in an arrestin knockout mouse model of Oguchi disease (stationary night blindness)

PURPOSE: To determine whether constitutive signal flow arising from defective rhodopsin shut-off causes photoreceptor cell death in arrestin knockout mice. METHODS: The retinas of cyclic-light-reared, pigmented arrestin knockout mice and wild-type littermate control mice were examined histologically for photoreceptor cell loss from 100 days to 1 year of age. In separate experiments, to determine whether constant light would accelerate the degeneration in arrestin knockout mice, these animals and wild-type control mice were exposed for 1, 2, or 3 weeks to fluorescent light at an intensity of 115 to 150 fc. The degree of photoreceptor cell loss was quantified histologically by obtaining a mean outer nuclear layer thickness for each animal. RESULTS: In arrestin knockout mice maintained in cyclic light, photoreceptor loss was evident at 100 days of age, and it became progressively more severe, with less than 50% of photoreceptors surviving at 1 year of age. The photoreceptor degeneration appeared to be caused by light, because when these mice were reared in the dark, the retinal structure was indistinguishable from normal. When exposed to constant light, the retinas of wild-type pigmented mice showed no light-induced damage, regardless of exposure duration. By contrast, the retinas of arrestin knockout mice showed rapid degeneration in constant light, with a loss of 30% of photoreceptors after 1 week of exposure and greater than 60% after 3 weeks of exposure. CONCLUSIONS: The results indicate that constitutive signal flow due to arrestin knockout leads to photoreceptor degeneration. Excessive light accelerates the cell death process in pigmented arrestin knockout mice. Human patients with naturally occurring mutations that lead to nonfunctional arrestin and rhodopsin kinase have Oguchi disease, a form of stationary night blindness. The present findings suggest that such patients may be at greater risk of the damaging effects of light than those with other forms of retinal degeneration, and they provide an impetus to restrict excessive light exposure as a protective measure in patients with constitutive signal flow in phototransduction.

Increased susceptibility to constant light in nr and pcd mice with inherited retinal degenerations.

PURPOSE: To determine whether the degenerating photoreceptors in nervous (nr/nr) and Purkinje cell degeneration (pcd/pcd) mutant mice are more susceptible to the damaging effects of constant light than those in age-matched normal mice. METHODS: Beginning at two ages for each mutant, albino nr/nr and pcd/pcd mice were placed into constant fluorescent light at an illuminance of 115 foot-candles to 130 foot-candles for a period of 1 week. Age-matched (usually littermate) normal (+/-) mice were exposed at the same time. The degree of photoreceptor cell loss was quantified histologically by obtaining a mean outer nuclear layer thickness for each animal. The light-exposed mice were compared with age-matched mutant and normal mice that were maintained in cyclic light. RESULTS: The homozygous mutants at each age showed a significantly greater loss of photoreceptor cells caused by constant light exposure than did the normal +/- mice in the same period of light exposure. The nr/nr and pcd/pcd mutants lost two to three times the number of photoreceptor cells than did the +/- mice during the constant light exposure. CONCLUSIONS: It has long been thought that excessive light may be harmful to patients with inherited or age-related photoreceptor degenerations. The present data add to other experimental evidence suggesting that photoreceptors already undergoing inherited or other forms of degeneration may be particularly susceptible to the damaging effects of excessive light.

Continuous exposure to bright light upregulates bFGF and CNTF expression in the rat retina.

PURPOSE: To examine mRNA expression of neurotrophic factors in the retina after exposure to bright light. METHODS: Male adult Sprague-Dawley rats were exposed to light of 115-130 ft-c. Retinas were collected after 1, 2, 4 or 7 days of exposure. Northern blot analysis was performed to determine mRNA levels for the following factors and their receptors: basic fibroblast growth factor (bFGF), ciliary neurotrophic factor (CNTF), acidic fibroblast growth factor (aFGF), brain-derived neurotrophic factor (BDNF), insulin-like growth factor 1 (IGF-1) and glial fibrillary acidic protein (GFAP). Expression of bFGF, CNTF and GFAP was localized by in situ hybridization. RESULTS: Exposure to light of 115-130 ft-c resulted in a substantial increase in bFGF and CNTF expression that persisted during the entire 7-day period of exposure. The peak expression of bFGF was almost 9-fold at day 2. The increase in CNTF mRNA reached a maximum of 6-fold at day 4. A small increase (50%) in IGF-1 mRNA was also seen at day 4. Among the receptors, an elevation of 3-fold in FGF receptor 1 (FGFR-1) was observed at day 2. There was also a small increase (70%) in IGF-1 receptor (IGF-1R) at day 2. In addition, the expression of GFAP showed a rapid elevation of about 8-fold by day 1 and 9-fold by day 2, and 18-fold by day 4. There was, however, no significant alteration in the expression of aFGF and BDNF. In situ hybridizations showed that the elevation of bFGF, CNTF and GFAP occurred across the entire retina with especially prominent increases over specific layers for each gene. CONCLUSIONS: Continuous exposure to bright light upregulates bFGF, CNTF, FGFR-1 and GFAP expression in the rat retina. The pattern of induced expression closely resembles that induced by mechanical injury, implying a common underlying mechanism.

Protection of mouse photoreceptors by survival factors in retinal degenerations.

PURPOSE: To examine the protective effect of a number of survival factors on degenerating photoreceptors in mutant mice with naturally occurring inherited retinal degenerations, including retinal degeneration (rd/rd), retinal degeneration slow (rds/rds), nervous (nr/nr), and Purkinje cell degeneration (pcd/pcd), in three different forms of mutant rhodopsin transgenic mice and in light damage in albino mice. METHODS: Various survival factors were injected intravitreally into one eye of mice at or soon after the beginning of photoreceptor degeneration, with the opposite eye serving as the control, and the eyes were examined histologically at later ages. The survival factors included brain-derived neurotrophic factor (BDNF), neurotrophin-3, neurotrophin-4, ciliary neurotrophic factor (CNTF), Axokine (a mutein of CNTF), leukemia inhibitory factor, basic fibroblast growth factor, and nerve growth factor and insulin-like growth factor II, either alone or in various combinations. RESULTS: Photoreceptor degeneration was slowed in rd/rd and nr/nr mutant mice and in Q344ter mutant rhodopsin mice by certain forms of CNTF; the degeneration in Q344ter mice was slowed by Axokine and by leukemia inhibitory factor; and the degeneration in a few nr/nr mice was slowed by BDNF. The other agents were ineffective in these mice, and none of the agents were effective in the other mutants and other mutant rhodopsin transgenic mice. However, light damage experiments that compared agent effectiveness in albino mice versus rats suggested a significant delivery problem with the very small mouse eye, thereby making the interpretation of negative findings equivocal in mutant mice. Basic fibroblast growth factor failed to protect the mouse retina from the damaging effects of constant light, whereas it showed a strong protective effect in the rat, indicating an important species difference. CONCLUSIONS: The slowing of degeneration in the rd/rd and Q344ter mutant mice demonstrated that intraocularly injected survival factors can protect photoreceptors from degenerating in animal models with the same or similar genetic defects as those in human inherited retinal degenerations.

Variability in rate of cone degeneration in the retinal degeneration (rd/rd) mouse.

The retinas of rd/rd mice with inherited retinal degeneration were examined histologically at postnatal days 60-66, an age when most rod cells already have degenerated and disappeared but when a significant number of cones are still present. We observed an unexpected hemispheric asymmetry and large variability in the number of surviving cones. Significantly more cones survived in the inferior than in the superior hemisphere in most retinas, although in about 15% of animals the hemispheric asymmetry was absent or was reversed. The number of surviving cones was highly variable from animal to animal, ranging from 3-30, a factor of 10, within the superior hemisphere, and from 7-51, a factor greater than 7, in the inferior hemisphere. If the specific hemisphere was ignored, the number ranged from 3-51, a factor of 17. These findings have significance for the examination of cone survival in the late stages of degeneration in this widely studied mutant, including therapeutic studies using transplantation, gene therapy or survival factors, as well as for the identification of surviving cells using cone-specific markers.

Injury-induced upregulation of bFGF and CNTF mRNAS in the rat retina.

Focal mechanical injury to the retina has been shown to slow or prevent photoreceptor degeneration near the lesion site in two animal models of retinal degeneration, inherited retinal dystrophy in the Royal College of Surgeons (RCS) and light damage in albino rats. Thus, when injured, the rat retina activates a self-protective mechanism to minimize damage. To identify injury responsive factors and cells, we examined the mRNAs for the following factors and some of their receptors: basic and acidic fibroblast growth factors (bFGF, aFGF) and FGF receptor-1 (FGFR1); ciliary neurotrophic factor (CNTF) and CNTF receptor alpha (CNTFR alpha); brain-derived neurotrophic factor (BDNF) and its receptor trkB; and insulin-like growth factor-1 (IGF-1) and IGFR-1 receptor (IGF-1R). After a single mechanical lesion to the subretinal space and retina, there was a substantial increase in bFGF and CNTF expression that persisted for the entire 10 d period of study. The increase in bFGF mRNA after injury was prompt and great in amplitude, while the elevation of CNTF mRNA was relatively slower. In addition, there was a transient increase in FGFR1 mRNA. In situ hybridizations showed that the elevation of bFGF and CNTF was localized to the vicinity of the lesion. The expression of GFAP (glial fibrillary acidic protein) mRNA also increased in response to injury. These findings strongly suggest that increases in endogenous bFGF and/or CNTF play key roles in injury-induced photoreceptor rescue.

Multiple growth factors, cytokines, and neurotrophins rescue photoreceptors from the damaging effects of constant light.

Recent demonstrations of survival-promoting activity by neurotrophic agents in diverse neuronal systems have raised the possibility of pharmacological therapy for inherited and degenerative disorders of the central nervous system. We have shown previously that, in the retina, basic fibroblast growth factor delays photoreceptor degeneration in Royal College of Surgeons rats with inherited retinal dystrophy and that the growth factor reduces or prevents the rapid photoreceptor degeneration produced by constant light in the rat. This light-damage model now provides an efficient way to assess quantitatively the survival-promoting activity in vivo of a number of growth factors and other molecules. We report here that photoreceptors can be significantly protected from the damaging effects of light by intravitreal injection of eight different growth factors, cytokines, and neurotrophins that typically act through several distinct receptor families. In addition to basic fibroblast growth factor, those factors providing a high degree of photoreceptor rescue include brain-derived neurotrophic factor, ciliary neurotrophic factor, interleukin 1 beta, and acidic fibroblast growth factor; those with less activity include neurotrophin 3, insulin-like growth factor II, and tumor necrosis factor alpha; those showing little or no protective effect are nerve growth factor, epidermal growth factor, platelet-derived growth factor, insulin, insulin-like growth factor I, heparin, and laminin. Although we used at least one relatively high concentration of each agent (the highest available), it is still possible that other concentrations or factor combinations might be more protective. Injecting heparin along with acidic fibroblast growth factor or basic fibroblast growth factor further enhanced the degree of photoreceptor survival and also suppressed the increased incidence of macrophages produced by either factor, especially basic fibroblast growth factor. These results now provide the impetus for determining the normal function in the retina, mechanism(s) of rescue, and therapeutic potential in human eye diseases for each agent.

Basic fibroblast growth factor and local injury protect photoreceptors from light damage in the rat.

Injection of basic fibroblast growth factor (bFGF) into the eye, intravitreally or subretinally, delays photoreceptor degeneration in inherited retinal dystrophy in the rat, as does local injury to the retina (Faktorovich et al., 1990). To determine whether this heparin-binding peptide or local injury is effective in any other form of photoreceptor degeneration, we examined their protective roles in light damage. Albino rats of the F344 strain were exposed to 1 or 2 weeks of constant fluorescent light (115-200 footcandles), either with or without 1 microliter of bFGF solution (1150 ng/microliters in PBS) injected intravitreally or subretinally 2 d before the start of light exposure. Uninjected and intravitreally PBS-injected controls showed the loss of a majority of photoreceptor nuclei and the loss of most inner and outer segments after 1 week of light exposure, while intravitreal injection of bFGF resulted in significant photoreceptor rescue. The outer nuclear layer in bFGF-injected eyes was two to three times thicker than in controls, and the inner and outer segments showed a much greater degree of integrity. Following recovery in cyclic light for 10 d after 1 week of constant light exposure, bFGF-injected eyes showed much greater regeneration of photoreceptor inner and outer segments than did the controls. bFGF also increased the incidence of presumptive macrophages, located predominantly in the inner retina, but the evidence suggests they are not directly involved in photoreceptor rescue. Subretinal injection of bFGF resulted in photoreceptor rescue throughout most of the superior hemisphere in which the injection was made, with rescue extending into the inferior hemisphere in many of the eyes. Remarkably, the insertion of a dry needle or injection of PBS into the subretinal space also resulted in widespread photoreceptor rescue, extending through 70% or more of the superior hemisphere, and sometimes into the inferior hemispheres. This implicates the release and widespread diffusion of some endogenous survival-promoting factor from the site of injury in the retina. Our findings indicate that the photoreceptor rescue activity of bFGF is not restricted to inherited retinal dystrophy in the rat, and that light damage is an excellent model for studying the cellular site(s), kinetics, and molecular mechanisms of both the normal function of bFGF and its survival-promoting activity. Moreover, the injury-related rescue suggests that survival-promoting factors are readily available to provide a protective role in case of injury to the retina, presumably comparable to those that mediate the "conditioning lesion" effect in other neuronal systems.

Photoreceptor degeneration in inherited retinal dystrophy delayed by basic fibroblast growth factor.

Numerous inherited retinal degenerations exist in animals and humans, in which photoreceptors inexplicably degenerate and disappear. In RCS rats with inherited retinal dystrophy, the mutant gene is expressed in the retinal pigment epithelial (RPE) cell, and leads to the loss of photoreceptor cells. Photoreceptors can be rescued from degeneration if they are juxtaposed to wild-type RPE cells in experimental chimaeras or by the transplantation of RPE cells from normal rats. In both cases, the rescue effect extends beyond the immediate boundaries of the normal RPE cells, suggesting trophic action of a diffusible factor(s) from the normal RPE cells. We considered that the fibroblast growth factors, aFGF and bFGF, might have such a trophic role as they are found in the retina and RPE cells; bFGF acts as a neurotrophic agent after axonal injury in several regions of the central nervous system, and bFGF induces retinal regeneration from developing RPE cells. Here we report that subretinal injection of bFGF results in extensive rescue of photoreceptors in RCS rats for at least two months after the injection, and that intravitreal injection of bFGF results in even more widespread rescue, across almost the entire retina. The findings demonstrate for the first time that bFGF can act as a survival-promoting neurotrophic factor in a hereditary neuronal degeneration of the central nervous system.

Light-evoked changes in the interphotoreceptor matrix.

The normal function of vertebrate photoreceptor cells depends on multiple interactions and transfer of substances between the photoreceptors and the retinal pigment epithelium (RPE), but the mechanisms of these interactions are poorly understood. Many are thought to be mediated by the interphotoreceptor matrix (IPM), a complex extracellular matrix that surrounds the photoreceptors and lies between them and the RPE. Histochemical, immunocytochemical, and lectin probes for several IPM constituents revealed that components of the IPM in the rat undergo a major shift in distribution or molecular conformation after the transition between light and dark. In the light, various IPM constituents concentrated in bands at the apical and basal regions of the outer segment zone; in the dark, they distributed much more uniformly throughout the zone. The change in IPM distribution was triggered by the light-dark transition; it was not a circadian event, and it was not driven by a systemic factor. The light-evoked change in IPM distribution may facilitate the transfer of substances between the photoreceptors and the RPE.