Carotenoids in the Human Macula and Whole Retina

Garry J. Handelman,t§ Edward A. Drarz,* C. Collin Reay4 and Frederik J. G. M. van Kuijk*

The carotenoid pigments in the whole human retina and in the macular region were measured quantita- tively by high pressure liquid chromatography (HPLC). Approximately a five-fold larger amount of carotenoids was found in the human macula (35-120 ng) than in previously reported work. The dominant carotenoids in the whole retina are lutein and zeaxanthin. Zeaxanthin is concentrated in the macular region, whereas lutein is dispersed throughout the entire retina. Contrary to prior reports, substantial quantities of both carotenoids are present in the infant retina. Increasing variability is observed in carotenoid levels between individuals with advancing age, and some older individuals show very high whole retina carotenoid levels. These quantitative studies were made possible by synthesis of a new, stable carotenoid internal standard. Carotenoids have been proposed to be potent antioxidants, protecting membrane lipids from toxic peroxidation reactions. The method presented in this study will facilitate quantitative investigations of the association between carotenoid levels and health and disease of the retina. Invest Ophthalmol Vis Sci 29:850-855,1988

The first effort at chemical identification of the with the presence of carotenoids. The peak macular human macular pigment was reported by Wald,1 who pigment absorbance measured by this technique var- tentatively identified the pigment as a carotenoid or ied about three-fold between different monkey ret- carotenoids, belonging to the leaf xanthophyll family. inas.7'8 Wald's findings were based on the absorption spec- The biological role of this macular pigment is still trum and organic solvent partition characteristics of uncertain. Various roles have been proposed, includ- the pigment. Recently, Bone et al2 have published ing limiting chromatic aberration at the fovea by fil- evidence that the pigment is a mixture of two carot- tering out blue light,6 quenching of singlet oxygen or enoids, lutein and zeaxanthin. They estimated that free radicals produced in the retina,910 and protecting about 10 ng of total carotenoid could be extracted per the macula from the phototoxicity of blue light.910 macula. The present study is based on a method for separa- Data from psychophysical measurements13"5 indi- tion and quantitative determination of the carot- cate a large variation in the amount of pigment be- enoid pigments that is linear and shows high recov- tween different adults. Pease and Adams4 recently ery. The method was applied to analysis of the whole found a six-fold variation in macular pigment among retina as well as the macular region. Between 20 and different adult subjects and Werner et al5 found about 250 ng of carotenoid/retina was found in different an eight-fold variation. Wald1 and Bone and Spar- individuals. rock3 noted that there was no detectable pigment in some apparently normal adult subjects, and there are Materials and Methods 6 reports that no macular pigment is present in the Reagents and Equipment infant. Pigment was also measured in histological sections of monkey retinas7'8 by microspectropho- HPLC grade acetonitrile, methanol and isopro- tometry, and the absorption spectra were consistent panol (Fisher) were used for HPLC mobile phases. Hexanes were redistilled, reagent grade; ethanol was From the *Department of Chemistry, Montana State University, 200 proof, USP grade, redistilled (Publicker, Linfield, Bozeman, Montana, the fEye Research Institute, Retina Founda- PA); and H2O was HPLC grade (Burdick and Jack- tion, Boston, Massachusetts, and the tLicms Eye Bank, Pacific son, Muskegon, MI). /3-apo-8'-carotenal and O-ethyl- Medical Center, San Francisco, California. hydroxylamine were from Fluka, Ronkonkoma, NY. § Current address: Department of Biochemistry, Tufts Univer- O-methyl-hydroxylamine was from Aldrich. Lutein sity, Health Sciences Campus, Boston, Massachusetts. Supported in part by grants from the National Eye Institute to and zeaxanthin were kindly donated by Hoffman- EAD and partially supported by grants from The Netherlands Or- La Roche (Nutley, NJ) and N. I. Krinsky (Tufts ganization for the Advancement of Pure Research (ZWO) to University School of Medicine, Boston, MA). FJGMvK. /3-carotene and all-trans retinal were from Sigma (St. Submitted for publication: July 7, 1987; accepted December 31, Louis, MO). 1987. Reprint requests: F. J. G. M. van Kuijk, Department of Chemis- Analyses were carried out with an Altex gradient try, Montana State University, Bozeman, MT 59717. HPLC system, with Model 110A pumps, Model 453

850

Downloaded from iovs.arvojournals.org on 09/26/2021 No. 6 CAROTENOIDS IN HUMAN RETINA / Hondelmon er ol. 851

controller, and Rheodyne 725 injector with a 20 ^1 loop. Peak detection at 450 nm was done using a Perkin-Elmer LC-95 HPLC detector equipped with a 1.6 cm path length 18 n\ flow cell. A Spectra-Physics Lutein Minigrator was used to monitor retention times. Peaks were recorded on a Hewlett-Packard 7130 chart recorder. Carotenoid analyses were done with a modification of the method of Krinsky and Welankiwar." Routine analyses were carried out by isocratic elution in 85/15 Zeoxanthin acetonitrile/methanol at 1.0 ml/min. The chromato- graphic column was an Alltech Econosphere C18 with 5 tim particle size and 10% carbon load. Column

dimensions were 25 cm X 0.46 cm, with C18, 5 nm, 1 N-O-C2H5

cm X 0.46 cm integral cartridge pre-column. For de- Carotenal - Ethyl - Oxime terminations of nonpolar carotenoids, such as jS-carotene, a step-gradient to 20% isopropanol in the ace- Fig. 1. Structures of lutein, zeaxanthin, and the internal standard tonitrile/methanol mobile phase was used after the synthesized for this work, called carotenal-ethyl-oxime. more polar carotenoids were eluted. fore analysis. Paired specimens from individual Internal Standard donors were frozen and thawed at the same time. The structure of a new, highly stable internal stan- Retinas were analyzed separately from both eyes dard that was prepared for this work is shown in Fig- from a number of donors. For several donor eye ure 1. The standard is the O-ethylhydroxylamine de- pairs, the macular region from one eye was analyzed rivative of /3-apo-8'-carotenal (called carotenal-ethyl- and compared to the whole retina from the fellow oxime) and was made by a procedure adapted from eye. Only one eye was available from some donors, the method of van Kuijk et al.12 Two milliliters of a 1 and the whole retina was analyzed in these cases. mg/ml solution of apo-carotenal (in methanol) were Precise anatomical collection of the macular region mixed with 200 >A of a solution of 0.1 M ethyl-hy- from thawed frozen eyes was not possible unless the droxylamine (in 0.1 M PIPES buffer, adjusted to pH macular regions were obtained from specimens from 4.7) and incubated 12 hr in the dark at room temper- which the vitreous was removed before freezing. If a ature. The derivative was purified by HPLC, using frozen whole globe was processed, it was used to ob- the isocratic mobile phase conditions described tain the complete retina. The dissection procedure above. The HPLC mobile phase was evaporated, and was directed at obtaining the region of the retina be- the carotenal-ethyl-oxime redissolved in hexanes for tween the superior and inferior temporal vascular ar- storage. The optical density of the stock solution of cades, as this region is easily visualized and the ma- the carotenal-ethyl-oxime was determined at 450 nm cula is fully contained within these boundaries. A and a dilution in hexanes prepared with 0.01 absor- sharp 8 mm trephine was used to punch out the ap- bance from which 1 ml was used in each analysis. If proximately 5 mm diameter macula with some sur- columns with higher carbon loading are used, the rounding tissue, and this region was then lifted off retention time of the internal standard may be exces- with a forceps. All of the macula was removed with sive and other internal standards synthesized from the trephined section, along with about 5% of the shorter precursors (e.g., /3-apo-12'-carotenal from peripheral retina surrounding the macula. Hoffmann-LaRoche) should be substituted. Extraction of Tissues Source and Preparation of Tissues Retina samples were placed in pre-tared glass ho- Specimens were obtained from donor eyes pro- mogenizer sleeves. Buffer (10 mM HEPES, pH 7.4, vided by the Lions Eye Bank, Pacific Presbyterian 0.1 M NaCl, 1 mM Na2EDTA) was added to bring Medical Center, San Francisco, CA. The corneas had the weight of sample plus buffer to 0.5 g, and 0.5 ml been removed from most of the donor eyes. Eyes ethanol containing 50 Mg/ml butylated hydroxy tolu- were either dissected at the Eye Bank to remove the ene (BHT) was added. The sample was homogenized vitreous humor and the posterior poles frozen at for 30 seconds with a Teflon pestle. The homogenate -70°C, or intact globes were frozen. The eyes or pos- was transferred to an 8 ml vial with a Pasteur pipette, terior poles were usually frozen within 48 hr of death. and 1 ml of the internal standard solution in hexanes Storage at -70°C was for a maximum of 1 year be- was added, followed by 3 ml hexanes. The vial was

Downloaded from iovs.arvojournals.org on 09/26/2021 852 INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / June 1988 Vol. 29

sealed with a Teflon-lined cap, mixed on a vortex methanol and aqueous mobile phases should be mixer for 120 seconds, and centrifuged at 800 g for 30 avoided. seconds. The upper phase was transferred to a clean vial and evaporated under nitrogen at 40°C. The dry film was dissolved in 60 fx\ of methanol, which is Results effective in fully solubilizing the extract for HPLC analysis. A 20 n\ portion was analyzed promptly by Carotenoids in Retina and Macula HPLC with detection at 450 nm. All manipulations The chromatogram of a whole-retina extract, were carried out under incandescent bulbs shielded shown in Figure 2A, reveals three dominant peaks, in with orange filters. addition to the internal standard (IS), that absorb at 450 nm. The most prominent two peaks (labeled 2 Calibration and 3) have retention times identical to lutein and The standard solution of zeaxanthin for calibration zeaxanthin, as shown by the standards run in Figure was made in acetonitrile from solid supplied by 2C. Figure 2B shows a chromatogram of the extract Hoffmann-La Roche, and its concentration was de- of the macular region from the fellow eye of the termined spectrophotometrically from the molar ex- donor shown in Figure 2A. Figures 2A and B contain tinction of 134,000 1/mole-cm at 452 nm.13 Lutein the same peaks but in differing amounts. from Hoffmann-La Roche was purified by HPLC, Peak 1 in Figure 2 is much more prominent on and its concentration in acetonitrile was determined analyses of whole retina (Fig. 2 A) than of the macular from the molar extinction at 448 nm of 134,000 region (Fig. 2B) and was attributed to retinal that was 1/mole-cm.13 Standard solutions were divided into extracted from the rhodopsin in the specimen. Evi- small portions, stored in the dark at -20°C, and were dence for the identity of retinal was obtained by pre- shown to be stable for up to 30 days under those paring the O-methyl-hydroxylamine derivative of au- conditions. On the day of analysis, a fresh portion of thentic retinal and of the human retina or macula each standard solution was diluted in acetonitrile to extracts as described by van Kuijk et al,12 and deter- give a concentration of 1.0 jig/ml of each carotenoid mining the effect of derivatization on HPLC reten- in the working stock solution. Fifty microliters of the tion time. Peak 1 in underivatized and derivatized stock solution was added to 0.5 ml EtOH and 0.5 ml extracts had retention times identical with native and H2O and analyzed in parallel with the tissue extracts. identically derivatized all-trans-retina\, respectively. Peak height ratios between the known amounts of The absorption maximum of the native peak 1 col- standards and internal standard were measured to lected from the HPLC was 381 nm with a weak tail calibrate the method for unknown quantitation. The that reached beyond 450 nm, the same as all-tmns- reliability of the use of peak heights was shown with retinal.14 None of the carotenoid peaks changed their generation of linear standard curves from 10-100 ng mobility in the presence of O-methyl hydroxylamine. with r = 0.999 for each carotenoid. Peak heights are After the elution of the internal standard, no peaks more reliable than peak areas when baseline separa- of significant amplitude were detected which could tion is not achieved. Any width differences or extinc- interfere with lutein and zeaxanthin on chromato- tion differences in the peak are accounted for by the graphic analysis of subsequent samples. Only traces calibration procedures. of nonpolar carotenoids, such as /3-carotene were de- To evaluate recovery, bovine retina was homoge- tected (see below). Unidentified satellite peaks are nized as described to provide an amount of lipid present surrounding the main carotenoid peaks comparable to a human retina extraction. Recovery (Fig. 2A). measurements were done by internal addition, as Table 1 shows the comparison of carotenoid con- previously described for retinoids,12 except that stock tent from the left and right retinas from eight sub- solutions of zeaxanthin and lutein were prepared in jects. The agreement between the amounts of lutein acetonitrile instead of hexanes. A small amount of and zeaxanthin in the right and left retina for each lutein present in the bovine retina was accounted for donor is quite good. However, some difficulties in in analysis of recovery. Recovery of lutein was 99%, quantitative dissection of the retina were noted and zeaxanthin was 98%, and /3-carotene was 91%. improvements in quantitative collection of the retina Some HPLC columns produce chromatographi- may show even closer agreement. cally distinct isomers of the carotenoids, which show When intact eyeballs were frozen, the specimens increased absorbance at 340 nm and are thought to could usually be processed to collect the whole retina be cis-isomers. This effect is more serious when after thawing. In about 75% of the cases, most of the smaller samples are analyzed. Suitable columns retina could be collected from a thawed eyeball, with should be stored tightly capped in 85/15 acetonitrile/ little vitreous or choroid contamination. About 25%

Downloaded from iovs.arvojournals.org on 09/26/2021 No. 6 CAROTENOIDS IN HUMAN RETINA / Hondelmon er ol. 853

Table 1. Lutein and zeaxanthin content in whole retinas from donors where both eyes were analyzed (values in nanograms carotenoid per retina)

Lutein Zeaxanthin Donor no. Age Left Right Left Right

1) 1 wk. 24.8 25.4 10.2 10.1 2) 2 mo. 23.4 21.3 10.3 10.0 3) lOyrs. 15.0 16.0 8.3 8.6 4) 15yrs. 24.7 18.0 18.3 13.4 5) 32 yrs. 45.9 50.2 18.8 36.6 6) 56 yrs. 68.6 58.4 35.0 30.0 7) 57 yrs. 62.2 49.3 42.1 30.6 8) 81 yrs. 59.6 43.6 26.0 17.8

of frozen eyebank specimens were suitable for proper collection of the macula. Both procedures, collection of the retina or the macular region, were somewhat approximate techniques from eyebank samples in our hands. Some retinal tissue always adheres to the eyeball. The trephine tends to traumatize the tissue so that it is difficult to collect the complete macula. With more extensive experience, it may be possible to establish conditions to obtain maculas in high yield from eyebank specimens. Figure 3 summarizes the amounts of lutein and zeaxanthin in whole retinas, from 16 subjects over a broad age range. If two retinas were analyzed per subject, the average result is given. There is substantial carotenoid in the infant retinas analyzed. The highest values we observed were found in adulthood and old age, reaching the values of 94 ng and 186 ng of lutein in the whole retina of two elderly subjects. However, for some adult retinas, the lutein value is near the value observed for the very young retinas.

200

• Lutein » Zeoxonihin

'Do ">. , 'l 0 4 8 12 16 20 40 60 80 TIME (MIN) Age (Years) Fig. 3. Carotenoid content of human retinas from donors of Fig. 2. HPLC chromatographs monitored at 450 nm. (A) Whole different ages. The closed circle is the number of nanograms of human retina. (B) Macular region of the fellow eye that provided lutein per retina and the closed triangle is the number of nano- the retina in (A). (C) Reference standard lutein (2), zeaxanthin (3) grams of zeaxanthin per retina. Thirteen of the samples are aver- and carotenal-ethyl-oxime internal standard (IS). The identity of ages from the paired retinas shown in Table 1, and three samples the peak labeled 1 is discussed in the text. are from single retinas.

Downloaded from iovs.arvojournals.org on 09/26/2021 854 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / June 1988 Vol. 29

Table 2. Lutein and zeaxanthin values from donors where the whole retina was analyzed from one eye and the macula analyzed from the other eye (values in nanograms carotenoids per retina or macula)

Whole retina Macular region Macula/retina^ Donor Age no. (yrs.) Lutein Zea.* Lutein Zea. Lutein Zea.

1) 38 78.9 48.2 39.9 41.1 0.505 0.853 2) 65 15.8 17.2 15.7 19.3 0.994 1.122 3) 72 29.0 17.7 18.3 15.9 0.631 0.898 4) 79 93.8 49.4 39.9 41.5 0.425 0.838 5) 81 182.0 83.0 68.3 51.9 0.375 0.625

* Zea. = zeaxanthin. f These columns show the ratio of lutein and zeaxanthin in the macular region.

Figure 3 and Table 1 show that there is less zeax- and we find much larger amounts of carotenoids than anthin than lutein in the whole retina in 15 of the 16 published by Bone et al.2 The method used by Bone samples analyzed (zeaxanthin averaged 44% of the et al2 was designed for identification, not quantita- lutein). tion. Bone et al have previously reported in an ab- Table 2 shows data from subjects where the retina stract15 amounts of macular carotenoid in quantita- was analyzed from one eye and the macular region tive agreement with the finding reported here. from the fellow eye. The amount of lutein in the The data presented indicate that substantial carot- macular region is usually about 30% to 60% of the enoid is found in the retina shortly after birth. The whole retina value. For zeaxanthin, the macular data also suggest that the pigment accumulates in the value is 70-110% of the whole retinal value. Further, whole retina with age to much higher levels in some the two carotenoid pigments seem present in roughly individuals, but not in others (Fig. 3), although there equal amounts in the macular region as was also re- are not enough samples for statistical evaluation. ported by Bone et al.2 This suggestion is consistent with psychophysical measurements1'3"5 that found the density of the mac- Analysis of Retina and Macula For /3-Carotene and ular pigment in different normal subjects ranged Other Nonpolar Carotenoids widely from undetectable to 1.2 absorbance units. While variation within an age group has been re- A step gradient to 20% isopropanol in the acetoni- ported by several investigators, it has also been re- trile/methanol mobile phase after the internal stan- ported that the macular pigment density does not dard has eluted accomplishes the elution of nonpolar change systematically with age, as measured psycho- carotenoids as sharp peaks, including lycopene, a- physically.3"5 Bone et al in a recent abstract also find a carotene, and /3-carotene if they are present. Three lack of systematic accumulation with age.16 It may be macular regions were analyzed for nonpolar carot- that the pigment density remains fairly constant in enoids, and no significant /3-carotene or other non- the macula but that it accumulates, in some individ- polar carotenoid could be detected even though typi- uals, in the peripheral retina. This question should be cal amounts of lutein and zeaxanthin were found. addressed by analysis of a larger number of maculae Two whole retinas were also analyzed for nonpolar and retinas over a broad age range. carotenoids. ^-Carotene equivalent to 1% of the lu- Retinal carotenoids are of interest since their tein was found in one and to 3% in the other whole dietary intake may confer antioxidant protection. retina. Traces of other nonpolar carotenoids could There are indications that human age-related macu- also be seen. lar degeneration may be associated with deficiencies in antioxidant protection.917 Carotenoids are the Discussion most active protective agents known against highly The data presented support the qualitative obser- reactive singlet oxygen,18 and singlet oxygen-induced vation of Bone et al2 that lutein and zeaxanthin are lipid peroxidation has been proposed to be a media- the dominant carotenoids in both retina and macula. tor of light damage in the retina.910 New gas chroma- We hypothesize that the very small amounts of non- tography-mass spectrometry methods19'20 have pro- polar carotenoids found were contributed by blood vided evidence that lipid peroxidation products accu- contamination of the autopsy retinas and that essen- mulate in rat and dog retinas degenerating due to tially no /3-carotene is present in the human retina. vitamin E deficiency.21 Some of the peroxidation The present method is designed to be quantitative, products present are highly cytotoxic.22 Recently, it

Downloaded from iovs.arvojournals.org on 09/26/2021 No. 6 CAROTENOIDS IN HUMAN RETINA / Hondelmon er ol. 855

has been proposed by Krinsky and Deneke23 and by 8. Snodderly DM, Auran JM, and Delori FC: The mascular pig- Burton and Ingold24 that carotenoids may also func- ment: II. Spatial distribution in primate retinas. Invest Oph- thalmol Vis Sci 25:674, 1984. tion as highly effective free radical trap antioxidants 9. Handelman GJ and Dratz EA: The role of antioxidants in the at the low oxygen tension found in tissues. Carot- retina and retinal pigment epithelium and the nature of enoid-deficient monkeys were reported to show pig- prooxidant-induced damage. Advances in Free Radical Biol- mentary changes in the fundus.25 However, controls ogy & Medicine 2:1, 1986. for the specific effects of dietary carotenoid supple- 10. Kirschfeld K: Carotenoid pigments: Their possible role in protecting against photooxidation in eyes and photoreceptor cells. mentation to the deficient diet were not carried out. Proc R Soc Lond B 216:71, 1982. If carotenoid pigments are serving a protective 11. Krinsky NI and Welankiwar S: Assay of carotenoids. Methods function, it would be of interest to compare the con- Enzymol 105:155, 1984. centration of the pigment in humans with degenera- 12. van Kuijk FJGM, Handelman GJ, and Dratz EA: Rapid anal- tive retinal disease, such as age-related macular de- ysis of the major classes of retinoids by step gradient reverse phase high-performance liquid chromatography using retinal generation, with age-matched controls. Studies of (o-ethyl) oxime derivatives. J Chromatogr 348:241, 1985. possible association between carotenoid levels and 13. Strain HH: Leaf Xanthophylls. Washington, DC, Carnegie In- human health are of additional interest because both stitution, 1938. psychophysical and biochemical data suggest sub- 14. Hubbard R: Geometrical isomerization of vitamin A, retinene stantial variation between different individuals. Ex- and retinene oxime. J Am Chem Soc 78:4662, 1956. 15. Bone RA, Landrum JT, Tarsis SC: Distribution of macular periments could be undertaken in primates to alter pigments in human and other retinas. ARVO Abstracts. Invest retinal carotenoid levels by dietary means and to Ophthalmol Vis Sci 27(Suppl):192, 1986. study possible changes in resistance to degeneration 16. Bone RA, Landrum JT, Hime GW, Fernandez L, and Mar- resulting from these manipulations. tinez J: Macular pigments: Mass spectra and an age study. ARVO Abstracts. Invest Ophthalmol Vis Sci 28(Suppl):338, Key words: carotenoids, lutein, zeaxanthin, human retina, 1987. macula, HPLC, lipid peroxides, age related macular degen- 17. Weiter J, Dratz E, Fitch K, and Handelman G: Role of selen- eration ium nutrition in senile macular degeneration. ARVO Ab- stracts. Invest Ophthalmol Vis Sci 26(Suppl):58, 1985. 18. Foote CS, and Denny RW: Chemistry of singlet oxygen: VII. Acknowledgments Quenching by beta-carotene. J Am Chem Soc 90:6233, 1968. 19. van Kuijk FJGM, Thomas DW, Stephens RJ, and Dratz EA: The authors thank Norman Krinsky and Fred Khachik Gas chromatography-mass spectrometry method for determi- for valuable discussions and carotenoid samples. We also nation of phospholipid peroxides: II. Transesterification to thank Carol Buchheit for assistance in preparing the manu- form pentafluorobenzyl esters and detection with picogram script. sensitivity. J Free Radical Biol Med 1:387, 1985. 20. van Kuijk FJGM, Thomas DW, Stephens RJ, and Dratz EA: References Occurrence of 4-hydroxyalkenals in rat tissues determined as pentafluorobenzyl oxime derivatives by gas chromatography- 1. Wald G: The photochemistry of vision. Doc Ophthalmol 3:94, mass spectrometry. Biochem Biophys Res Commun 139:144, 1949. 1986. 2. Bone RA, Landrum JT, and Tarsis SL: Preliminary identifica- 21. van Kuijk FJGM, Dratz EA, Thomas DW, Lowe E, and Ste- tion of the human macular pigment. Vision Res 25:1531, phens RJ: Studies of the role of lipid peroxidation in retinal 1985. degeneration. ARVO abstracts. Invest Ophthalmol Vis Sci 28 3. Bone RA and Sparrock JMB: Comparison of macular pigment (Suppl):141, 1987. densities in human eyes. Vision Res 11:1057, 1971. 22. Esterbauer H: Aldehydic products of lipid peroxidation. In 4. Pease PL, Adams AJ, and Nuccio E: Optical density of human Free Radicals, Lipid Peroxidation and Cancer, McBrien DCH macular pigment. Vision Res 27:705, 1987. and Slater TF, editors. London, Academic Press, 1982, pp. 5. Werner JS, Donnelly SK, and KJiegl R: Aging and human 101-128. macular pigment density. Vision Res 27:257, 1987. 23. Krinsky NI and Deneke SM: Interaction of oxygen and oxy- 6. Nussbaum JJ, Pruett RC, and Delori FC: Historic perspectives. radicals with carotenoids. Journal of the National Cancer Insti- Macular yellow pigment. The first 200 years. Retina 1:296, tute 69:205, 1982. 1981. 24. Burton GW and Ingold KU: j8-carotene: An unusual type of 7. Snodderly DM, Brown PK, Delori FC, and Auran JD: The lipid antioxidant. Science 224:569, 1984. macular pigment: I. Absorbance spectra localization and dis- 25. Malinow MR, Feeney-Bums L, Peterson LH, Klein ML, and crimination from other yellow pigments in primate retinas. Neuringer M: Diet-related macular anomalies in monkeys. In- Invest Ophthalmol Vis Sci 25:660, 1984. vest Ophthalmol Vis Sci 19:857, 1980.

Downloaded from iovs.arvojournals.org on 09/26/2021