Vascular endothelial growth factor and fibroblast growth factor 5 are colocalized in vascular and avascular epiretinal membranes.

PURPOSE: To determine by immunocytochemical analysis of epiretinal membranes whether vascular endothelial growth factor and the fibroblast growth factor FGF-5 are present in patients with proliferative diabetic retinopathy or proliferative vitreoretinopathy. METHODS: Human surgical specimens of epiretinal membranes were obtained from 11 eyes with proliferative diabetic retinopathy and five eyes with proliferative vitreoretinopathy. Sections were immunostained with an affinity-purified antibody against an internal sequence of human FGF-5 and with a commercially available affinity-purified antibody corresponding to the first 20 residues of human vascular endothelial growth factor. Slides were visualized using avidin-biotin-peroxidase complex. Control studies were performed with nonimmune immunoglobulin G and preabsorbed vascular endothelial growth factor and FGF-5 antibody, respectively. RESULTS: Immunoreactive FGF-5 is present in most cells, including endothelial cells of vascular and avascular epiretinal membranes, but seems to be absent from the extracellular matrix. A similar staining pattern was observed for vascular endothelial growth factor. CONCLUSIONS: Vascular endothelial growth factor and FGF-5 are remarkably colocalized in both vascular and avascular epiretinal membranes arising from proliferative diabetic retinopathy and proliferative vitreoretinopathy, respectively. This result questions the concept that the presence of a single angiogenic factor determines the vascular status of an epiretinal proliferation.

Expression of fibroblast growth factor-5 by bovine choroidal endothelial cells in vitro.

PURPOSE: To demonstrate the expression of fibroblast growth factor-5 (FGF-5) by bovine choroidal microvascular endothelial (BCME) cells and to investigate its possible role as an autocrine mitogen in these cells. METHODS: Expression of FGF-5 by BCME cells was studied by a combination of Northern and Western blot analyses. Total RNA was isolated from BCME cultures at passages 5 through 8 and analyzed by Northern blot analysis for the presence of FGF-5 transcripts, using a 1-kb human complemetary DNA. Slot-blot analysis was performed to determine possible cross-reactivity between this probe and acidic and basic FGFs of human and bovine species. A previously characterized antibody directed against the aminoterminus of the human FGF-5 sequence was used in Western blot analyses to identify immunoreactive proteins released by BCME cells into the medium. Finally, the mitogenic activity of human recombinant FGF-5 on a variety of cell types was evaluated, using a cellular proliferation assay. RESULTS: Northern blot analysis provided evidence for the expression of two major FGF-5 transcripts at 4 kb and 3 kb and two minor transcripts at 2.2 kb and 1.7 kb. A single immunoreactive protein with a molecular weight of 34 kDa was identified by Western blot analysis of conditioned media. In cellular proliferation assays, human recombinant FGF-5 was not mitogenic in BCME cells but exhibited an approximate ED50 of 1.8 to 3.7 nM in BALB/c3T3 fibroblasts. This ED50 was within the range reported by the manufacturer, using a thymidine incorporation assay and a similar embryonic fibroblast cell line. Fibroblast growth factor-5 also stimulated proliferation of human retinal pigment epithelial cells. CONCLUSIONS: Bovine choroidal microvascular endothelial cells exhibit expression in vitro of FGF-5 at the messenger RNA and protein levels. Perivascular and endothelial cell staining for FGF-5 seen previously in choroidal neovascular membranes may therefore arise from expression by choroidal endothelial cells. Because nonglycosylated recombinant FGF-5 does not appear to be a mitogen in BCME cells in vitro, it is reasonable to question its role as an autocrine mitogen in vivo. Fibroblast growth factor-5 may instead be serving paracrine roles in the stimulation of fibroblasts and retinal pigment epithelial cells during the formation of choroidal neovascular membranes. Studies with fully glycosylated recombinant FGF-5 will be required, however, to assess the biologic activity of this member of the FGF gene family.

Apolipoprotein E in the subretinal fluid of rhegmatogenous and exudative retinal detachments.

BACKGROUND: Apolipoprotein E plays a key role in lipoprotein metabolism and is believed to be an important protein in other biologic functions such as wound healing processes of nerve tissue. Because the neurosensory retina is rich in lipids, we analyzed the subretinal fluid from patients with acute and chronic retinal detachments to determine whether apolipoprotein was present and whether it was involved in lipid metabolism during wound repair of the detached retina. METHODS: The subretinal fluid collected from eyes with rhegmatogenous and exudative retinal detachments first was analyzed by agarose and sodium dodecyl sulfate gel electrophoresis to identify the lipoprotein profiles. Western blot analysis confirmed the presence of apolipoprotein E, and semiquantitation was performed by densitometry of the corresponding immunodot-blot. RESULTS: Apolipoprotein E was present in the subretinal fluid of eyes with rhegmatogenous and exudative retinal detachments. Its concentration increased in eyes with chronic retinal detachments. CONCLUSIONS: We implicate apolipoprotein E found in the subretinal fluid in the wound-healing process of the detached retina.

Wagner vitreoretinal degeneration. Follow-up of the original pedigree.

PURPOSE: Wagner disease belongs to a heterogeneous group of hereditary vitreoretinal degenerations. The authors have observed complications of this disorder that have not been reported before and therefore re-examined Wagner's original pedigree to further delineate the spectrum of the associated findings and its prognosis. METHODS: Sixty members of the family agreed to be examined. All had complete clinical eye examinations, 40 had dark adaptation studies as well as single-flash and Ganzfeld rod and cone electroretinography. Fluorescein angiograms were performed in selected patients. RESULTS: Twenty-eight family members were affected. The most consistent finding was an empty vitreous cavity with avascular strands or veils. Chorioretinal atrophy and cataract increased with the patients' age and had occurred in all patients older than 45 years of age. Four patients had a history of a rhegmatogenous retinal detachment in one eye at a median age of 20 years. The authors observed peripheral tractional retinal detachments in 55% of eyes among patients older than 45 years. Glaucoma was present in ten eyes (18%), four of which showed neovascular glaucoma. Of all patients, 63% showed elevated rod and cone thresholds on dark adaptation, and 87% showed subnormal b-wave amplitudes of the rod- and of the cone system on the electroretinography. CONCLUSIONS: Clinical expressivity of Wagner disease varies from unaffected carriers to bilateral blindness. Rhegmatogenous retinal detachment is observed infrequently, whereas peripheral traction retinal detachment, chorioretinal atrophy, and cataracts are present in most of the elderly affected individuals. Progression of the chorioretinal pathology is paralleled by electrophysiologic abnormalities.

Genetic linkage of Wagner disease and erosive vitreoretinopathy to chromosome 5q13-14.

BACKGROUND: Wagner disease and erosive vitreoretinopathy are potentially blinding autosomal dominant diseases that share some similarities with Stickler syndrome. However, both disorders have associated retinal pigment epithelial changes, poor night vision, visual field defects, and abnormal electroretinographic findings, which are not found in families with COL2A1-associated Stickler syndrome. In addition, rhegmatogenous retinal detachments are uncommon in Wagner disease but occur in approximately 50% of patients with either Stickler syndrome or erosive vitreoretinopathy. OBJECTIVES: To identify the chromosomal location of the genes involved in Wagner disease and erosive vitreoretinopathy and to distinguish these conditions genetically from Stickler syndrome. METHODS: Fifteen affected members of a family affected with erosive vitreoretinopathy and 24 affected descendants of the pedigree described by Wagner were genotyped with a set of short tandem repeat polymorphisms distributed across the genome. RESULTS: Significant linkage was observed in each family between the disease phenotype and markers that map to chromosome 5q13-14. The highest lod score for the family affected with erosive vitreoretinopathy was 4.2 and was obtained with marker GATA3H06 (theta = 0). The highest lod score for the family affected with Wagner disease was 5.8 and was obtained with marker D5S815 (theta = 0). A candidate gene (cartilage link protein) that is known to lie near the linked interval was screened for mutations, but none was found in either family. CONCLUSIONS: These data suggest that erosive vitreoretinopathy and Wagner disease are allelic disorders and demonstrate that they are genetically distinct from COL2A1-associated Stickler syndrome.