Autoradiographic studies of aged primate macular retinal pigment epithelium.

The authors used 35S-sulfate and 3H-proline to trace labeled molecules in autoradiograms of aged monkey and human macular retina to detect the synthesis of extracellular matrix (ECM) components by retinal pigment epithelial (RPE) cells. Quantitative analysis of silver grains 6 hr and 3 d after intravitreal injection of 35S-sulfate in the monkey showed that secretion from the basal pole of the RPE occurs at a slower rate than from the apical pole. In vitro incubation of human maculas produced poor autoradiographs with 35S-sulfate. Good autoradiographs were obtained using 3H-proline. Human macular RPE showed uneven labeling, but densely labeled cells did not correlate with sites of basal linear deposits, ECM though to be basement membrane material, and a hallmark of age-related maculopathy. The time course of labeling in adult primate tissue showed a fairly high turnover rate for these molecules. Scant labeling of ECM at drusen sites and no labeling in basal linear deposits suggested that either (1) these structures have a slow turnover or (2) their components contain scant sulfate and proline. Alternatively, faulty degradative processes rather than enhanced synthesis may account for the accumulation of abnormal ECM at the RPE-Bruch's membrane interface in aged maculas.

Lipids in human lipofuscin-enriched subcellular fractions of two age populations. Comparison with rod outer segments and neural retina.

The fatty acid composition and content of total phospholipids, free fatty acids (FFA), diacylglycerols (DG), phosphatidylcholine (PC), phosphatidylserine (PS), and phosphatidylethanolamine (PE) were studied in lipofuscin granules of human donors in two age groups, young (less than 40 yr old) and old (more than 47 yr old), and compared with lipids of the photoreceptor rod outer segments (ROS). Neural retina (NR) and retinal pigment epithelium (RPE) also were studied. In both age groups, the lipid composition of the lipofuscin granules differed from that of the ROS, with a decrease in the proportion of phospholipid and an increase in FFA, suggesting very high phospholipase activity in the lipofuscin granules. In ROS, docosahexaenoic acid (22:6) was the predominant FFA, whereas palmitic acid (16:0), arachidonic acid (20:4) and oleic acid (18:1) were the major fatty acids in the lipofuscin granules. The fatty acid compositions of PC, PE, and PS of lipofuscin granules were different from those of the retina. There was proportionally less 22:6 in lipofuscin, and the amounts of saturated and monounsaturated fatty acyl chains such as 16:0, stearate (18:0), and 18:1 were greater than in retina. Compared to ROS, the lipofuscin granules showed a significant decrease in DG containing 20:4 but not 22:6. With aging, there was a decrease in the amount of total polyunsaturated fatty acyl chains (22:6 and 20:4) in the lipofuscin granules. These results show that the lipid composition of lipofuscin is different from that of ingested ROS, probably because of increased phospholipase and peroxidative activities in lipofuscin, directed toward ingested ROS as well as toward other materials from the RPE and blood.

Age-related macular changes in humans over 90 years old.

The macula lutea of 23 donors aged 90 to 101 years were examined by light and electron microscopy and compared to maculas from a 49- to 68-year-old age group. The number of foveal photoreceptors and retinal pigment epithelial cells, the presence of macular pigment, and lipofuscin fluorescence were assessed. Pathologic characteristics typical of age-related macular degeneration occurred in nine of the 90- to 101-year-old group with changes ranging from early neovascularization to fully developed disciform scars, geographic atrophy, and macular holes. Several retinas had pigment epithelial and photoreceptor cell numbers equal to those of the younger group, but most showed cell loss. Thickened, debris-filled Bruch's membrane and choriocapillary atrophy, although common, were not an invariable accompaniment to old age. Clinicians should advise elderly patients that their chances of maintaining macular structure, and hopefully function, are better than 50%.

Stimulation of cell division by argon and Nd:YAG laser trabeculoplasty in cynomolgus monkeys.

Although laser treatment of the trabecular meshwork is the most common form of surgery for glaucoma, the tissue response to this therapy is still incompletely understood. We applied argon or Nd:YAG laser to the trabecular meshwork of six monkeys. Cell division was identified by injecting tritiated thymidine into the anterior chamber 24 hr after laser application. Autoradiography of tissue sections revealed significantly more labelled cells in eyes treated with laser than in the untreated controls. In addition, cells in neighboring tissues such as iris, ciliary body and sclera showed labelling in association with laser application. Furthermore, comparison of argon-induced lesions with those caused by pulsed Nd:YAG suggests that there are quantitative and qualitative differences in the response of trabecular meshwork and surrounding tissues to these two forms of laser energy.

The fate of immunoreactive opsin following phagocytosis by pigment epithelium in human and monkey retinas.

Polyclonal and monoclonal antibodies to human rhodopsin were used to identify and localize this principal glycoprotein of the photoreceptor outer segment discs on thin sections of human and monkey retinal pigment epithelium (RPE) and on immunoblots of RPE subcellular fractions following gel electrophoresis. Antiopsin was visualized with protein A-gold labeling by electron microscopy or peroxidase-linked second antibody on immunoblots. In immunocytochemical studies using polyclonal antibodies, the rod outer segments (ROS) were heavily labeled whereas cone outer segments labeling was variable and more sparse. Phagosomes and other small bodies in the RPE, interpreted as secondary lysosomes, were labeled. In contrast, lipofuscin granules, osmiophilic residual bodies of the lysosomal system of the RPE, were negative. No reactive sites were found in Bruch's membrane or in drusen. A monoclonal antibody (MAB) specific for the amino terminus of human opsin and another MAB specific for the carboxy terminal region of bovine opsin produced labeling patterns similar to, but about one-half the density obtained with polyclonal antibodies. In the immunoblot analyses, the lipofuscin granule fraction from a sucrose density gradient of human RPE homogenates was positive for rhodopsin only in those specimens that were found, upon ultrastructural examination, to contain recognizable phagosomes. When phagosomes were lacking and therefore did not contaminate the lipofuscin granule fraction, the immunoblots were negative for opsin. Melanolipofuscin granule fractions were uniformly negative for opsin. We conclude that the superficial hydrophilic antigen binding sites on the opsin molecule for which the antibodies are specific have been altered or destroyed by lysosomal enzyme digestion within the phagolysosomal system of the RPE prior to formation of definitive lipofuscin granules. Thus, these antibodies are of limited value in revealing the ultimate fate of the whole rhodopsin molecule, eg, the hydrophobic sequences that are the most likely residues in lipofuscin granules.

Ocular pathology in melanomatous Sinclair miniature swine.

Eyes of cutaneous melanoma-bearing miniature Sinclair swine were examined by light and electron microscopy during growth and maturation of animals. The histology of ocular depigmentation, which occurs during melanoma arrest and skin depigmentation, was studied in eyes of animals that had successful and unsuccessful tumor regression. In 12 animals one eye was removed at an early stage and the second eye at a later stage of disease or maturation. The histologic and clinical course of the ocular changes was correlated with changes in the cutaneous tumors. Animals showing rapid cutaneous tumor regression developed acute uveitis that was characterized by an influx of mononuclear cells into the stroma of the ciliary body; later this spread to the iris and choroid. In late stages the cornea sometimes developed a band of calcium precipitates beneath the basal epithelial cells (band keratopathy). Uveal melanocytes developed watery cytoplasm typical of cells with ruptured plasma membranes. Released melanin granules were taken up initially into the lysosomal compartment of mononuclear cells having the various morphologic features of lymphocytes and monocytes; later, melanin also appeared in fibrocyte lysosomes. The relationship of these processes to various cell-mediated immunity phenomena is being studied.

Vitamin E in human neural retina and retinal pigment epithelium: effect of age.

Vitamin E levels were measured in retina and retinal pigment epithelium from human eye bank donors of from 12-82 years of age. In comparison to an age group of 12-45 years, humans 59-82 years of age had a higher concentration of vitamin E in both retina and retinal pigment epithelium. Depending on age, the concentration of vitamin E in retinal pigment epithelium was from 4-7 times higher than in retina. Vitamin E accumulated in the human retinal pigment epithelium in an age dependent fashion, so that by 80 years it was from 3-4 times higher than in those 20 years old. The level of vitamin E in young human retinal epithelium, however, was higher than in comparable bovine tissue. The age-related increase in human tissue vitamin E levels does not appear to be affected by postmortem time.

Lysosomal enzyme cytochemistry of human RPE, Bruch's membrane and drusen.

Twenty-five human eyes of various ages from eye bank donors and surgical enucleations were obtained for ultrastructural cytochemical demonstration of acid phosphatase (AcPase) and arylsulfatase B (ASB) in the retinal pigment epithelium (RPE) and Bruch's membrane. Results with post-mortem (less than 10 hr) tissues were comparable to those of fresh specimens. Vigorous reactivity was demonstrated in lysosomes of RPE and choriocapillary endothelium but no reactive sites were found in Bruch's membrane, although many lysosome-like dense bodies occurred in eyes greater than 20 yr of age. Granular drusen of 30-70-yr-olds contained no reactive bodies. In eyes greater than 80 years old blebs of RPE basal cytoplasm protruding into Bruch's membrane contained reactive lysosomes. We conclude that the RPE ordinarily does not extrude or exocytose active lysosomes (ie, phagolysosomes, other secondary lysosomes, residual bodies, lipofuscin) or lysosomal enzymes. Aged RPE, however, extrudes cytoplasm with active lysosomes into Bruch's membrane. The possible impact of this process on the extracellular connective tissue is discussed, particularly with regard to age-related deterioration of Bruch's membrane and neovascularization.

Age-related changes in the ultrastructure of Bruch's membrane.

An ultrastructural study of Bruch's membrane in 68 human eyes including at least five individuals from each of ten decades of life showed that changes began in the second decade in macular specimens but were delayed in nonmacular specimens. All specimens from 20-year-old donors had altered macular Bruch's membranes. The severity of changes increased with age. The most common change for all ages was debris on both sides of the elastic layer. Drusen sites often showed no debris in the underlying outer collagenous zone, suggesting that drusen may result from failure of dispersal of the accumulated material at these sites. Secretion by pigment epithelium of a "linear basal deposit" is a late (after the age of 50 years) change.

Uveitis in melanomatous swine: lack of evidence for humoral immune melanocyte destruction.

A strain of miniature Sinclair swine that have cutaneous malignant melanomas destroy their tumors via an immunologic process. Destruction of apparently normal melanocytes of the uveal tract occurs concurrently with tumor cell lysis. The authors investigated possible involvement of humoral antibodies in uveal melanocyte lysis. Uveal antigens solubilized by a variety of reagents were tested by immunodiffusion against autologous and homologous sera; no precipitation lines were found. Western blots of electrophoresed uveal proteins when tested against serum from an animal with a great tumor load showed no antigen-antibody reactions. No specific humoral antibodies were detected. Sera tested for immune complexes also were negative. Invasion of the uveal tract by mononuclear cells was quantified in 1-micron-thick sections of tissue taken at various stages of ocular depigmentation. Disappearance of melanocytes was preceded by an increase in mononuclear cells and followed by an increase in melanin-containing macrophages. The ultrastructural features of the invading cells are presented. The authors conclude that cell-mediated uveal melanocyte destruction is the most likely basis for the uveitis and other damaging sequelae.

Aging human RPE: morphometric analysis of macular, equatorial, and peripheral cells.

Retinal pigment epithelium (RPE) of 50 human eyes, five from each 10 decades of life, were analyzed using ultrastructural morphometric techniques. Content of three types of pigments, lipofuscin, melanin, and complex granules, (melanolipofuscin, melanolysosomes) were recorded for cells from macular, equatorial, and peripheral retinal specimens. Areas occupied by pigments, nucleus, and cytoplasmic space were calculated. Data were analyzed by a computer for age-related changes and effects of fixation delay time. The largest increase in lipofuscin granules occurred between the first and second decade of life, and further increases occurred with age. The content of "pure" melanin declined with age, whereas the number of complex melanin granules increased. Macular RPE contained more complex granules than nonmacular RPE, particularly in young eyes. The volume of RPE cytoplasm not occupied by pigments ("free space") decreased with age. No significant effects of fixation delays between 2 and 9 hours postmortem were found on the parameters studied here. These findings may serve as a baseline for estimating normalcy of human RPE specimens.

Uveitis caused by cytotoxic immune response to cutaneous malignant melanoma in swine: destruction of uveal melanocytes during tumor regression.

In a substrain of Sinclair miniature black swine, bred for increasing incidence of cutaneous malignant melanomas, tumor regression occurs spontaneously and is accompanied by depigmentation of the skin, hair, and eyes. We conducted a 12-month longitudinal study of the ocular phenomena in 30 swine beginning at 3 weeks of age. The clinically observed sequence of depigmentation of the fundus and iris was correlated with histopathologic changes in selected enucleated eyes. Normal melanocytes of the uveal tract are destroyed between the 4th and 16th week of life. Melanocyte destruction is preceded by an invasion of the uveal tract by mononuclear cells having the ultrastructural features of lymphocytes and monocytes. Melanin and other cellular debris of ruptured melanocytes are ingested by macrophages which then migrate to the walls of blood vessels. Cataracts and band keratopathy develop secondary to the uveitis in some animals. Pilot electroretinograms show diminished electrical activity in photoreceptors of totally depigmented eyes possibly indicating ischemic or toxic damage to the retina. The retinal pigment epithelium remains essentially normal during the acute stages of uveal inflammation; later some damage and reparative hyperplasia may occur. The death of normal uveal melanocytes that occurs during the systemic attack on the cutaneous malignant melanomas appears to be an "innocent bystander" error in the immune recognition mechanism. The antigenic basis of this immunologic cross reaction is under investigation.

The fate of the phagosome: conversion to 'age pigment' and impact in human retinal pigment epithelium.

Phagosomes are converted to phagolysosomes and then to residual bodies (also known as lipofuscin granules or age pigment). Lipofuscin granules of retinal pigment epithelial (RPE) cells of single human eyes were isolated and analysed for enzyme content and fluorescence spectra. The granules are low in lysosomal enzymes and they fluoresce yellow-gold. Thin layer chromatography reveals several separable constituents. Lipofuscin granules accumulate with age, the largest increase occurring after the first decade of life in humans. Macular RPE cells accumulate granules more than non-macular cells.

Lipofuscin: resolution of discrepant fluorescence data.

Lipofuscin granules (age pigments) emit yellow light under ultraviolet excitation in the fluorescence microscope. The reported blue emission maximum of extracts of lipofuscin-laden cells may result from instrumental bias. The major fluorescent components that accumulate with age in these lysosomal residual bodies of human retinal pigment epithelium are yellow-emitting fluorophores. Different age-related fluorophores, which do emit blue light, are derived from other intracellular sources. A reevaluation of the connection between blue-emitting lipid peroxidation products and the age-related lipofuscin granules of classical pathology is necessary.

Maculopathy in cynomolgus monkeys. A correlated fluorescein angiographic and ultrastructural study.

Maculae of seven cynomolgus macaque monkeys (Macaca fascicularis) showing abnormalities in color fundus photographs and fluorescein angiograms were studied in serial sections by light and electron microscopy and compared with three eyes without clinically visible defects in the macula. Maculae that showed hyperfluorescent nonleaking window defects showed no drusen or interruptions in the retinal pigment epithelium (RPE). Six of these monkeys had mis-shapen foveal depressions, all showed some degree of photoreceptor degeneration, and one had cells in Bruch's membrane. Bright yellow spots correlated with scattered RPE filled with lipid vacuoles. Shallow RPE elevations correlated with diffuse nonleaking window defects. Patches of RPE deficient in melanin occurred at sites of hyperfluorescence. Quantitative studies showed that maculae with window defects had more lipofuscin and less melanin per RPE cell. Maculae deemed normal by photography showed degenerating photoreceptors.