Fluorescent pigment accumulation in retinal pigment epithelium of antioxidant-deficient rats

M. L. Katz, W. L. Stone, and E. A. Dratz*

A yellow autofluorescent pigment, generally thought to be indicative of membrane autoxida- tion, was found to accumulate in the retinal pigment epithelium (RPE) of rats maintained for 32 weeks on diets producing physiological antioxidant deficiency. The largest build-up of fluores- cent pigment occurred in rats fed a diet high in polyunsaturated fatty acids (PUFAs) and. deficient in a-tocopherol (vitamin E), selenium, sulfur-containing amino acids, and chromium. These latter four nutrients have all been implicated in maintaining the antioxidant status of tissues, whereas PUFAs are pro-oxidants. Dietary supplementation with methionine and chromium significantly reduced the amount of fluorescent pigment accumulated in the RPE. Supple7iientation with all four nutrients further reduced the amount of fluorescent pigment to a very low level. Rats maintained on a normal laboratory diet, relatively low in PUFAs and presumably adequate in other nutrients, accumulated relatively small amounts of fluorescent pigment in the RPE. Of all tissues in the retina and choroid, the autofluorescent pigment was found to he almost entirely restricted to the RPE. The auto fluorescence produced, in the RPE by antioxidant deficiency was more concentrated than that produced in the testes, kidney, intes- tine, and heart. This suggests that the RPE is particularly sensitive to physiological antioxidant deficiencies. The increased fluorescent pigment build-up in the RPE of antioxidant-deficient rats appears to correlate with a decreased RPE melanin content. Similar changes in pigmenta- tion have been reported to occur in human RPE with age and in dominantly inherited, retinitis pigmentosa. Thus, with respect to its effect on RPE pigmentation, antioxidant deficiency appears to mimic aging and possibly some aspects of one type of retinitis pigmentosa.

Key words: retinal pigment epithelium, antioxidants, a-tocopherol, selenium, glutathione peroxidase, chromium, lipid autoxidation, fluorescent pigment, methionine, aging

A yellow autofluorescence pigment, usu- many tissues,1 including the RPE.2 4 A ally called lipofuscin or ceroid, has been build-up of lipofuscin in the RPE has also found to accumulate with advanced age in been correlated with retinal-choroidal de- generation in some cases of retinitis pigmen- tosa.3' 5 The mechanism of lipofuscin forma- From the Division of Natural Sciences, University of California, Santa Cruz. tion in vivo is somewhat controversial but is This investigation was supported by Research Grant correlated with increased autoxidation of EY01521 and by an Institutional Biomedical Grant polyunsaturated fatty acid (PUFA).2 Dietary 5SORR0713507 from the National Institutes factors required for the functioning of physio- of Health, U.S. Public Health Service. logical antioxidant mechanisms seem to play Submitted for publication Jan. 1, 1978. an important role in preventing the accumu- Reprint requests: E. A. Dratz, Division of Natural Sci- ences II, University of California, Santa Cruz, Calif. lation of autofluorescent pigment. An accel- 95064. erated build-up of these pigments has been

0146-0404/78/111049+10$01.00/0 © 1978 Assoc. for Res. in Vis. and Ophthal., Inc. 1049

Downloaded from iovs.arvojournals.org on 09/28/2021 Invest. Ophthalmol. Visual Sci. 1050 Katz, Stone, and Dratz November 1978

Table I. Composition of antioxidant-deficient chain reactions.l4 Selenium is an essential diet* component of glutathione peroxidase, which detoxifies hydrogen peroxide and fatty acid Weight percent Ingredient of diet peroxides. Glutathione, which serves as a re- ducing agent in this detoxification reaction, is Torula yeast 36.1 maintained in the reduced state by metabolic Corn starch 5.0 Sucrose 37.7 coupling to the pentose shunt pathway for Vitamin mixt 2.2 glucose catabolism.14 Since glutathione con- Corn oil, tocopherol stripped 15.0 tains the amino acid cysteine, at least part of Mineral mixj 4.0 the antioxidant function of sulfur amino acids * Dietary components which were varied in this study were may be attributed to their role in forming this present in the basal diet at the following levels: a-tocopherol, less than 3 mg/kg diet; selenium, 0.03 ppm; chromium, 0.14 tripeptide. Chromium is required for effi- ppm; methionine, 0.27%. cient transport of glucose across membranes t Milligrams per kilogram of diet: ascorbic acid, 990; inositol, 110; choline chloride, 1650; p-aminobenzoic acid, 110; niacin, of some cell types; rats deficient in this ele- 99; riboflavin, 22; pyridoxine HC1, 22; thiamin HC1, 22; calcium ment exhibit a decreased responsiveness of pantothenate, 66; biotin, 0.5; folic acid, 2.0; vitamin Bi2) 30; 15 vitamin A palmitate in corn oil (200 IU/mg), 99; vitamin D2 in blood glucose levels to insulin. Chromium corn oil (400 IU/mg), 5.5; menadione, 50; dextrose hydrate, may therefore function as an essential com- 18,722. (Grams per kilogram of diet: calcium carbonate, 6.54; cupric ponent of the glutathione peroxidase system sulfate • 5H2O, 0.0072; clacium phosphate • 2H2O, 14.22; through its link with glucose uptake, which ferric citrate • 3H2O, 0.64; manganese sulfate • H2O, 0.055; potassium citrate • H2O, 9.46; potassium iodide, 0.0016; po- supplies the pentose shunt pathway with re- tassium phosphate dibasic, 3.094; sodium chloride, 4.324; zinc ducing equivalents. carbonate, 0.018; magnesium carbonate, 1.460. In this investigation, we fed rats a basal observed in the tissues of animals fed diets diet known to be deficient in vitamin E, deficient in the antioxidant a-tocopherol6 and selenium, sulfur amino acids, and chromium in selenium,7 which is a necessary compo- and containing a high level of PUFA in or- nent of an antioxidant enzyme.8 der to augment the physiological oxidative The vertebrate rod outer segment (ROS) stress normally present. Using fluorescence disk membranes contain an exceptionally microscopy, we have examined the effects of high concentration of PUFA; nearly half prolonged dietary deficiency or supplemen- of the ROS fatty acids contain six double tation of these compounds on the retina and bonds.9* 10 Since the ease with which fatty choroidal tissues, testes, renal tubules, in- acids undergo autoxidation is directly pro- testine, heart, liver, aorta, skeletal muscle, portional to the degree of unsaturation,11 lung, and brain of rats. ROS lipids are very susceptible to autoxida- tion.12 The ROS disk membranes undergo a Materials and methods continuous process of turnover. The oldest Rat diets. Five-week-old inbred agouti rats disks are shed intermittently in packets from (Ratus rat us; ACI/fMai, Microbiological Associ- the apices of the rods and are phagocytized ates, Inc., Bethesda, Md.) were divided into four 13 and catabolized by the RPE. In view of the dietary groups designated lab chow (CHOW), de- large amounts of easily oxidized ROS mate- ficient (DEF), deficient plus chromium and me- rial phagocytized by the RPE, we decided to thionine (DEF + CrMet), and fully supple- investigate the sensitivity of the RPE to di- mented (SUP). Animals were maintained on their etary deficiency in compounds thought to respective diets for 32 weeks and then sacrificed play a role in inhibiting in vivo autoxidation for histological and biochemical determinations. reactions. The composition of the DEF diet is given in Table I and is modeled after the torula yeast diet used by Protection against lipid autoxidation in Schwarz and Foltz16; however, the vitamin and vivo is believed to be provided by a number mineral mixes were modified to meet the current of mechanisms. Vitamin E (a-tocopherol) is nutritional standards proposed by the National generally thought to act as a free radical Research Council for the laboratory rat.17 In the scavenger and thus to quench autoxidation DEF + CrMet diet, a supplement of L-me-

Downloaded from iovs.arvojournals.org on 09/28/2021 Volume 17 Number 11 Fluorescent pigment in antioxidant-deficient rats 1051

Table II. Antioxidant status of rats after 32 weeks on experimental diets* a-Tocopherol Glutathione peroxidase Retina activity (e.u.)\ Plasma (fig/gm Liver selenium Dietary group (fig/ml) phospholipid) RBC Liver concentration (ppm) CHOW 4.2 ± 1.0 560 ± 30 340 ± 50 750 ± 50 3.0 ± 0.3 DEF 0.4 ± 0.2 120 ± 30 8 ± 4 60 ± 30 N.D.| DEF + CrMet 0.7 ± 0.1 120 ± 30 12 ± 3 100 ± 30 < 0.1 SUP 9.7 ± 0.5 760 ± 30 450 ± 50 750 ± 50 3.5 ± 1.6 *A minimum of three independent determinations were made for each value in the table, t Enzyme units (e.u.) are defined as nmoles NAPDH oxidized/min-mg protein. JN.D. = Not determined.

thionine (2.5 gm/kg of diet) was added to the basal mately 5 X 1012 neutrons/cm2-sec for 2 hr. A DEF diet, and 3 ppm chromium (as CrCl3) was lithium drifted germanium detector was used for added to the distilled drinking water. In addition the analysis of the 136.00 kev energy line of Se75 to these same levels of L-methionine and chro- (with a half-life of 120 days) and the 320.03 kev mium, the SUP diet also had 0.4 ppm selenium energy line of Cr51 (with a half-life of 27.72 days). (added as DL-selenomethionine) and DL-a-tocoph- The neutron activation analysis was done at the erol (500 mg/kg of diet) added to the basal DEF Department of Nuclear Engineering at the Uni- diet. For each diet described above, the con- versity of California at Berkeley. stituents were mixed slowly, to avoid heating, in a Tissue preparation and microscopy. Animals Hobart food mixer with a Teflon-coated paddle from each dietary group were anesthetized with and bowl. The diets were prepared frequently in diethyl ether, and their tissues were immediately small batches and stored frozen at —20° C. The fixed by transcardiac perfusion with 1% formalde- feeders were filled eveiy 2 days, and any uneaten hyde-1.25%glutaraldehyde buffered to pH 7.4 food was discarded to minimize rancidity of the with 0.1M phosphate. The rats were decapitated, chow. The CHOW diet consisted of an unpurified the corneas were slit, and the heads and other laboratory rat and mouse feed (Feedstuffs Process- tissues to be examined were placed in the same ing Co.) containing 25 mg of a-tocopherol per ki- fixative overnight. The eyes and brains were dis- logram of diet. Tap water was also provided for sected out and returned to the fixative for an addi- animals receiving the CHOW diet. All four diets tional 48 hr. The eyes were then bisected dorso- and drinking water were provided ad libitum, and ventrally, and the lens halves were removed. All coprophagy was hindered by the use of %-inch tissue samples were dehydrated through alcohols wire mesh floors elevated 1.5 inches above the to xylene and embedded overnight in low-fluores- cage bottoms. All animals were maintained in cence Histowax (Matheson, Coleman & Bell, translucent plastic cages on a 12 hr light, 12 hr Cincinnati, Ohio), and 5 fim sections were cut dark cycle under illumination of about 25 foot- with an American Optical 820 microtome. Tissue candles from cool-white fluorescent bulbs (Gen- sections were mounted on glass slides, deparaf- eral Electric Corp.). finated, mounted in glycerol-0.5M sodium carbo- Antioxidant status of animals. In order to nate (1:1, v/v), and covered with glass coverslips. evaluate the physiological a-tocopherol status of Some slides were treated for 7 min with 0.25% rats fed the various diets, the plasma and retina potassium permanganate and, after-washing, for 3 a-tocopherol levels were determined by the min with 2.5% oxalic acid in order to selectively method of Dugan.18 Glutathione peroxidase activ- bleach the melanin pigment. Tissue sections were ity was measured in erythrocytes and in the examined with a Leitz Dialux microscope in both postmicrosomal supernatant from liver homoge- phase and fluorescence modes. A Leitz epifluores- nates by the method of Little and O'Brien.19 Neu- cence accessory was used with a No. 1 dichroic tron activation analysis was used to measure the mirror, a 50 W ultrahigh-pressure mercury lamp, selenium and chromium levels in 50 mg samples of and excitation filters BG12 and UG1. The UG1 lyophilized postmicrosomal liver supernatant. filter (2 mm thickness) has a peak light trans- Each lyophilized liver sample and a selenium mission near 360 nm, at which lipofuscin fluores- standard of seleno-DL-methionine was sealed in a cence is excited. The BG12 filter (3 mm thickness) quartz tube and irradiated in a flux of approxi- blocks a slight red transmission in the UG1 filter,

Downloaded from iovs.arvojournals.org on 09/28/2021 Invest. Ophthalmol. Visual Sci. 1052 Katz, Stone, and Dratz November 1978

but its omission does not detectably alter the ap- activation analysis. Our preliminary data in- pearance of the pigment. Fluorescence was ob- dicate that liver supernatants from the SUP served with 430 or 460 nm emission barrier filters. dietary group had significantly higher levels All photographs were taken through a 0.32:1 re- of chromium than the DEF + CrMet group, ducing lens with an oil-immersion (numerical despite the fact that both groups received the aperture, 1.3) X40 objective and X10 ocular same amount of dietary chromium (Stone and lenses on Kodak ED-135 high-speed Ektachrome film. Film exposure time and illumination were Dratz, unpublished data). The neutron acti- the same for tissue sections from all animals. vation analysis for chromium is, however, in- terfered with by trace levels of selenium. We Results are attempting to resolve these difficulties by The antioxidant status of a number of dif- use of alternative analytical methods for ferent rat tissues was examined to confirm chromium detection. that in vivo antioxidant levels reflected diet- Fig. 1 shows unstained paraffin sections of ary intake. As shown in Table II, the plasma posterior ocular tissue from rats maintained and retina a-tocopherol levels were sharply on the experimental diets for 32 weeks from lower in rats receiving the DEF or DEF the time of weaning. Some separation be- + CrMet diets than in those on the SUP tween adjacent RPE cells occurred in all sec- or CHOW diets. Plasma and retinal a- tions, apparently due to excessive perfusion tocopherol levels in our animals were simi- pressures. When the sections were examined lar to values previously reported for rats by fluorescence microscopy, tissues from the either supplemented or deficient in dietary different groups were found to vary greatly in a-tocopherol.20' 21 Since tissue levels of a- their content of yellow autofluorescent ma- tocopherol for both the DEF and DEF + terial. In posterior ocular tissues from all CrMet dietary groups were comparable, it is dietary groups, the major portion of the yel- unlikely that chromium and methionine sig- low fluorescent material was clearly localized nificantly affected the level of a-tocopherol in the RPE, as can be seen by comparing the deficiency. phase contrast and fluorescence photomicro- Erythrocyte and liver glutathione peroxi- graphs of identical sections in Fig. 1 (A vs. B, dase activity were found to vary in relation to C vs. D, E vs. F, and G vs. H). A small the dietary levels of selenium; this is in amount of fluorescent pigment was some- agreement with previous reports.22' 23 The times visible in the choroidal capillary en- enzyme activities for the DEF group were dothelium and the ROS tips, especially in the somewhat lower than for the DEF + CrMet, DEF animals. Any observed fluorescence in but this difference was not statistically sig- the ROS tips was diffuse. The effect of dietary nificant (p > 0.05). The selenium concen- antioxidants on the RPE fluorescence was trations in the liver supernatants (also used quite dramatic. A brilliant yellow fluores- for the glutathione peroxidase activity mea- cence was found in the posterior RPE of rats surements) indicate a direct correspondence maintained on the DEF diet (Fig. 1, B). between liver glutathione peroxidase activity Supplementation of the DEF diet with chro- and liver selenium content (Table II). These mium and methionine (i.e., the DEF + data are in agreement with the observation CrMet diet) produced a significant decrease that selenium is required for glutathione in the RPE autofluorescence (Fig. 1, D). A peroxidase activity and that the activity is further reduction in RPE autofluorescence to sensitive to either selenium depletion or much lower levels was seen in animals raised supplementation.22' 23 on the DEF diet supplemented with a- The possible role of chromium in protect- tocopherol and selenium in addition to chro- ing against in vivo lipid autoxidation has mium and methionine (i.e., the SUP diet, been given little attention. We have at- Fig. 1, F). The posterior RPE from animals tempted to measure the chromium concen- maintained on the CHOW diet appeared to tration in the liver supernatants by neutron contain very little yellow autofluorescent ma-

Downloaded from iovs.arvojournals.org on 09/28/2021 Fig. 1. Photomicrographs of ocular tissue sections from rats maintained for 32 weeks on diets containing varying levels of antioxidants. A, C, E, and G are phase contrast micrographs from DEF, DEF + CrMET, SUP, and CHOW rats, respectively. To the right of each of these phase contrast pictures is a fluorescence photomicrograph (B, D, F, and H) of the same tissue section. The bar in A represents a length of 25 yu.ni.

Downloaded from iovs.arvojournals.org on 09/28/2021 Downloaded from iovs.arvojournals.org on 09/28/2021 Volume 17 Number 11 Fluorescent pigment in antioxidant-dejicient rats 1055

terial (Figure 1, H). However, there were pigment in the liver, skeletal muscle, or occasional areas in the RPE of these animals, brain (cerebrum and hypothalamus), each of involving one to several cells, that contained which had occasional cells containing fluores- an amount of fluorescent pigment compara- cent pigment in all dietary groups. No yellow ble in intensity to that in the RPE of fluorescent pigment was detected in the aor- DEF + CrMet animals. tas or lung tissues from animals on any of the The data in Fig. 1 clearly indicate an in- diets used in this experiment. verse correlation between dietary levels of antioxidants and RPE autofluorescence in Discussion animals fed the purified diets. In recent ex- In vitro lipid autoxidation of subcellular periments we found that the yellow auto- organelles generates fluorescent products fluorescence was clearly visible in en face with spectral properties similar to the au- views of live, freshly dissected, unfixed eye tofluorescent pigments that accumulate in cups. A similar relationship between antiox- animal tissues as a function of age or dietary idant status and the amount of RPE fluores- antioxidant deficiencies.23 In this investiga- cence was also observed in these fresh, tion we have observed the accumulation in unfixed tissues. The RPE autofluorescence is vivo of a yellow autofluorescent pigment in not, therefore, an artifact of tissue prepara- the RPE of rats maintained on diets deficient tion. Vitamin A has fluorescence properties in various nutrients which play a role in phys- similar to those of lipofuscin, except that vi- iological antioxidant mechanisms. Hayes26 tamin A fluorescence rapidly faces under ul- has similarly observed an accumulation of traviolet (UV) light,6 whereas lipofuscin lipofuscin granules in the RPE of monkeys fluorescence does not. In fresh RPE prep- maintained for over 2 years on diets deficient arations, some of the yellow autofluorescence in a-tocopherol, and it is probable that these faded rapidly in the UV light of the micro- lipofuscin granules, identified by electron scope, but the remainder was stable. Since microscopy, are identical to the yellow au- the fluorescence differences we observed be- tofluorescent pigments that we observe. tween the fresh RPEs of the various dietary When eye sections were examined by groups were in the UV light-stable compo- phase contrast microscopy, it was found that nent, it is unlikely that these fluorescence the central RPEs of animals maintained on differences can be accounted for by a varia- the CHOW diet (Fig. 1, G) contain more tion in RPE vitamin A content. It is possible, melanin pigment than the RPE's of rats fed however, that some chemically altered form the defined diets Fig. I, A, C, and E). We of vitamin A contributes to stable lipofuscin noted an inverse correlation between the fluorescence.24 amounts of melanin in the central posterior In addition to the eye, we examined par- RPE and the antioxidant status of the ani- affin sections of various other tissues by mals; i.e., the animals with the least melanin fluorescence microscopy. In terms of the had the most intense RPE fluorescence. proportion of tissue volume occupied by fluo- Feeney et al.,2 using electron microscopy, rescent pigment not extracted in ethanol or noted that the melanin content of human xylene, the RPE appeared to be most sensi- RPE appeared to decrease with age, while tive to antioxidant deficiency. Other tissues the amount of lipofuscin in these cells in- examined in which antioxidant deficiency creased. Thus, with respect to its effect on produced significant increases in the content these two types of pigment, antioxidant of fluorescent pigment were the seminiferous deficiency seems to mimic aging of the RPE. epithelium of the testes, renal tubule epithe- In light of this finding, a reported age-related lial cells, the lamina propria of the small in- decline in human blood selenium concentra- testine, and cardiac muscle. Antioxidant de- tion and glutathione peroxidase activity27 ficiency did not produce apparent increases may be of some importance in relation to the in the amounts of yellow autofluorescent aging of the RPE in man.

Downloaded from iovs.arvojournals.org on 09/28/2021 Invest. Ophthalmol. Visual Sci. 1056 Katz, Stone, and Dratz November 1978

It seemed possible that the melanin pig- further toward the periphery heavy melanin ment may have been masking some of the pigmentation was restricted. The pattern of fluorescence in the RPE, thus creating an er- RPE pigmentation in the antioxidant-de- roneous impression of differences in amounts ficient animals is similar to that seen by Kolb of fluorescent pigment. In order to test this and Gouras2 in the RPE of a 68-year-old possibility, tissue sections were treated with human patient suffering from dominantly in- potassium permanganate, which selectively herited retinitis pigmentosa. The RPE of the bleaches melanin before it significantly af- CHOW rats appeared to be relatively uni- fects the fluorescent pigment. Permanganate formly pigmented with melanin. bleaching clearly removed melanin, but it In this initial study, nutrients with an- did not reveal any more fluorescence in the tioxidant activity were varied in pairs rather RPE of animals from any of the dietary than factorially. Thus, it is not yet possible to groups than was seen before bleaching (data estimate the relative importance of the indi- not shown). Thus the observed effect of anti- vidual nutrients studied in preventing the oxidant status on RPE autofluorescence in build-up of autoxidation products in the melanin-containing sections is an accurate RPE. In light of our present findings, we reflection of differences in fluorescent pig- have initiated experiments to separate the ef- ment content. Further confirmation of this fects of a-tocopherol, selenium, chromium, conclusion comes from our recent dietary ex- and methionine. periments with albino rats, which lack mela- The high PUFA content of ROS membrane nin pigment. In the albino, as in the pig- lipids makes them particularly susceptible to mented animals, antioxidant deficiency led to autoxidation and suggests that in vivo an- a dramatic increase in RPE autofluorescence. tioxidant mechanisms might be especially Permangate bleaching of eye sections from important in protecting the RPE and ROS the agouti rats did produce a yellow-orange from oxidative damage. If the RPE were un- fluorescence in melanophores of the choroid, able to degrade or otherwise rid itself of which had contained large amounts of mela- cross-linked oxidation products, these would nin before bleaching. Since no such fluores- tend to accumulate with time and could ac- cent material appeared in the choroid of an- count for the fluorescent pigment seen in the tioxidant deficient albino rats after bleaching, RPE of antioxidant-deficient animals. The ac- it is possible that the choroidal fluorescence cumulation of significant amounts of this in the agoutis is formed from melanin, either pigment could interfere with RPE function in vivo or as an artifact of the permanganate sufficiently to produce retinal-choroidal de- treatment. We are presently investigating generation as observed in vitamin E deficient the origin of this yellow-orange choroidal monkeys26 and in some human cases of domi- fluorescence. nantly inherited retinitis pigmentosa.3'5 One The amount of melanin pigment in the symptom in many cases of the Batten-Vogt RPE of agouti rats on the defined diets was syndrome is retinal degeneration accompa- least in the central, posterior area of the eye nied by the massive build-up of autofluores- and gradually increased toward the periphery cent lipopigment in nervous tissue.28 Evi- in sections examined in the dorsoventral dence has been presented that the Batten- plane. With the increase in melanin pigment Vogt pigment is made up of a complex of vi- toward the periphery, there was a concomi- tamin A derivatives not extractable with most tant decrease in fluorescent material. Again, organic solvents.25 The possibility should be there was a correlation between the antioxid- considered that the fluorescent pigment we ant status of the animal and the pattern of observed in the RPE in antioxidant-deficient pigmentation. The greater the antioxidant animals has a similar origin. These observa- deficiency, the further toward the periphery tions imply that the retina might be particu- the fluorescent pigment extended, and the larly sensitive to hereditary defects in an-

Downloaded from iovs.arvojournals.org on 09/28/2021 Volume 17 Number 11 Fluorescent pigment in antioxidant-deficient rats 1057

tioxidant protective mechanisms which lead A. I.: Measurement and spectral characteristics of to lipofuscin deposition in other tissues. The fluorescent pigments in tissues of rats as a function similarity of the effects of age and antioxidant of dietary polyunsaturated fats and vitamin E, J. Nutr. 103:908, 1973. deficiency on the RPE suggest that aging in 7. Tappel, A., Fletcher, B., and Deamer, D.: Effect of this tissue may be at least partially the result antioxidants and nutrients on lipid peroxidation of a gradual accumulation of autoxidative fluorescent products and aging parameters in the damage. mouse, J. Gerontol. 28:415, 1973. The possible occurrence of ROS lipid au- 8. Rotruck, J. T., Pope, A. L., Ganther, M. E., Swan- son, A. B., Hafeman, D. G., and Hoekstra, W. G.: toxidation suggests a reason for the rapid rate Selenium: biochemical role as a component of of ROS disk membrane renewal. We have glutathione peroxidase, Science 179:588, 1973. found that the rate of ROS renewal is 9. Stone, W. L., Famsworth, C. C, and Dratz, E. A.: identical in rats that are either deficient in or A reinvestigation of the fatty acid content of bovine, supplemented with dietary antioxidants (un- rat, and frog retinal rod outer segments, Exp. Eye Res. (in press). published observations of Bok, Farnsworth, 10. Daemen, F. J. M.: Vertebrate rod outer segment and Dratz). This observation suggests that membranes, Biochim. Biophys. Acta 300:255, 1973. any increase in the rate of ROS oxidative 11. Witting, L. A.: Lipid peroxidation in vivo. J. Am. damage caused by dietary antioxidant de- Oil Chem. Soc. 42:908, 1965. ficiencies is not compensated for by an in- 12. Famsworth, C. C, and Dratz, E. A.: Oxidative damage of retinal rod outer segment membranes creased rate of ROS renewal. Antioxidant and the role of vitamin E, Biochim. Biophys. Acta mechanisms may therefore be necessary for 443:556, 1976. maintaining the functional integrity of photo- 13. Young, R. W.: The renewal of rod and cone outer receptors in the vertebrate eye. segments in the Rhesus monkey, J. Cell Biol. 49:303, 1971. We acknowledge helpful discussions with Drs. Dean 14. Tappel, A. L.: Selenium-glutathione peroxidase and Bok and James O'Donnell, the late Dr. Michael Hogan, vitamin E, Am. J. Clin Nutr. 27:960, 1974. and Mr. Garry Handelman. We thank Ms. Pat Murphy 15. Mertz, W.: Chromium as a dietary essential for and Dr. Matt LaVail for demonstrating the perfusion man. In Hockstra, W. G., Suttie, J. W., Ganther, fixation and tissue handling procedures used in their lab- H. E., and Mertz, W., editors: Trace Elements Me- oratory. Our thanks also to Mr. Chris Farnsworth for the tabolism in Animals—2, Baltimore, 1974, Univer- vitamin E assays, Mr. Larry deGhetaldi for some of the sity Park Press. glutathione peroxidase assays, and Mr. Marco Martinez, 16. Schwarz, K., and Foltz, C. M.: Factor 3 activity of Jr., for his assistance with the photographic work. selenium compounds, J. Biol. Chem. 233:245, 1958. 17. Nutrient Requirements of Laboratory Animals, Na- tional Research Council Publication No. 10, Wash- REFERENCES ington, D.C., 1972, National Academy of Sciences, 1. Porta, E. A., and Hartroft, W. S.: Lipid pigments in p. 56. relation to aging and dietary factors (lipofuscin). In 18. Dugan, D. E.: Spectrofluorometric determination of Wolman, M., editor: Pigments in Pathology, New tocopherols, Arch. Biochem. Biophys. 84:116, 1959. York, 1969, Academic Press, Inc., pp. 192-236. 19. Little, C, and O'Brien, P. J.: An intracellular 2. Feeney, L., Grieshaber, J. A., and Hogan, M. J.: GSH-peroxidase with a lipid peroxide substrate, Studies on human ocular pigment. In Rowen, J. W., Biochem. Biophys. Res. Commun. 31:145, 1968. editor: Eye Structure. II. Symposium, Stuttgart, 20. Machlin, L. J., Filipski, R., Nelson, J., Horn, L. R., 1963, Schattauer-Verlag, pp. 535-548. and Brin, M.: Effects of prolonged vitamin E 3. Kolb, H., and Gouras, P.: Electron microscopic ob- deficiency in the rat, J. Nutr. 107:1200, 1977. servations of human retinitis pigmentosa, domi- 21. Nishiyama, J., Ellison, E. C, Mizuno, G. R., and nantly inherited, INVEST. OPIITHALMOL. 13:487, Chipault, J. R.: Micro-determination of a-tocoph- 1974. erol in tissue lipids, J. Nutr. Sci. Vitaminol. 21:355, 4. Streeter, B. W.: The sudanophilic granules of the 1975. human retinal pigment epithelium, Arch. Ophthal- 22. Chow, C. K., and Tappel, A. L.: Response of mol. 66:391, 1961. glutathione peroxidase to dietary selenium in rats, J. 5. Wiggert, B. N., Bergsma, D. R.: Funahashi, M., Nutr. 104:444, 1974. Kuwabara, T., and Chader, G. J.: Vitamin A recep- 23. Hafeman, D. G., Sunde, R. A., and Hoekstra, tors in normal and dystrophic human retina, Nature W. G.: Effect of dietary selenium on erythrocyte 265:66, 1977. and liver glutathione peroxidase in the rat, J. Nutr. 6. Reddy, K., Fletcher, B., Tappel, A. L., and Tappel, 104:580, 1974.

Downloaded from iovs.arvojournals.org on 09/28/2021 1058 Katz, Stone, and Dratz

24. Wolfe, L. S., NgYingKin, N. M. K.,, Baker, R. R., duced by deficiencies of vitamin E or A, INVEST. Carpenter, S., and Andermann, F.: Identification of OPHTHALMOL. 13:499, 1974. retinoyl complexes as the autofluorescent compo- 27. Thomson, C. D., Rea, H. M., Robinson, M. F., and nent of neuronal storage material in Batten disease, Chapman, O. W.: Low blood selenium concen- Science 195:1360, 1977. trations and glutathione peroxidase activities in el- 25. Tappel, A. L.: Lipid peroxidation and fluorescent derly people, Proc. U. Otago Med. School 55:18, molecular damage to membranes. In Trump, B. F., 1977. and Arstila, A. V., editors: New York, 1975, 28. Zeman, W.: The neuronal ceroid-lipofuscinoses- Academic Press, Inc., p. 145. Batten-Vogt syndrome: a model for human aging? 26. Hayes, K. C.: Retinal degeneration in monkeys in- Adv. Gerontol. Res. 3:147, 1971.

Information for authors Most of the provisions of the Copyright Act of 1976 became effective on January 1, 1978. Therefore all manuscripts must be accompanied by the following statement, signed by each author: "The undersigned author(s) transfers all copyright ownership of the manuscript entitled (title of article) to The Association for Research in Vision and Ophthalmology, Inc., in the event the work is published. The author(s) warrant that the article is original, is not under consideration by another journal, and has not been previously published." Authors will be consulted, when possible, regarding republication of their material.

Downloaded from iovs.arvojournals.org on 09/28/2021