Exp. Eye Res. (1997) 64, 211–218
`
`Distribution of Lutein and Zeaxanthin Stereoisomers in the
`Human Retina
`
`R I C H A R D A. B O N Ea*, J O H N T. L A N D R U Mb, L A R R Y M. F R I E D E Sb, C H R I S T I N A M. G O M E Zb,
`M A R K D. K I L B U R Nb, E U G E N I O M E N E N D E Za, I V O N N E V I D A La    W E I L I W A N Gb
`
`Departments of a Physics and b Chemistry, Florida International University, Miami, FL 33199, U.S.A.
`
`(Received Cleveland 14 June and accepted in revised form 16 July 1996 )
`
`The distribution of macular pigment stereoisomers in the human retina has been mapped and a pathway
`to account for the presence of the non-dietary carotenoid, meso-zeaxanthin, is proposed. Adult neural
`retinas were cut into three concentric areas centered on the fovea, and the extracted carotenoids were
`analysed and purified by high-performance liquid chromatography. The dicarbamate or dibenzoate
`derivatives of the collected zeaxanthin fractions for each tissue sample were further analysed by HPLC to
`determine their stereoisomer composition. Whole retinas from infant eyes were similarly analysed. The
`results show that, relative to zeaxanthin, the concentration of lutein in the adult neural retina increases
`with radial distance from the fovea while that of meso-zeaxanthin decreases. Infant retinas were found
`to have more lutein and less meso-zeaxanthin, relative to zeaxanthin, than adult retinas. Small quantities
`of (3S, 3￿S)-zeaxanthin were also found in the adult retina, particularly in the macula. It is proposed that
`lutein and zeaxanthin are transported into an individual’s retina in the same proportions found in his or
`her blood serum. Some of the lutein is then converted into meso-zeaxanthin, primarily in the macula, by
`a mechanism which is less developed in infants than adults.
`￿ 1997 Academic Press Limited
`Key words : macular pigment ; carotenoids ; zeaxanthin ; lutein ; stereoisomers.
`
`1. Introduction
`
`The macular pigment, consisting principally of the
`carotenoids zeaxanthin and lutein (Bone, Landrum
`and Tarsis, 1985), has an interesting and unresolved
`distribution in the human retina. In the inner macula,
`the concentration of zeaxanthin is about twice that of
`lutein. As eccentricity from the fovea increases, the
`ratio of concentrations continuously changes with
`lutein becoming the dominant component in the
`peripheral retina. At distances exceeding 6 mm from
`the fovea, the lutein : zeaxanthin ratio is between 2 : 1
`and 3 : 1 (Bone et al., 1988 ; Handelman et al., 1988).
`Noting that
`this ratio increased from the fovea
`outwards in rough proportion to the increasing
`rod : cone ratio, Bone et al. (1988) suggested that this
`might reflect an association of lutein and zeaxanthin
`primarily with rods and cones respectively. Snodderly,
`Handelman and Adler (1991) on the other hand
`proposed that particular
`lutein : zeaxanthin ratios
`might be associated with specific cone types whose
`relative abundance varied across the retina.
`After the age of about 2 years, no age-related
`changes appear to occur in either the concentration or
`composition (lutein : zeaxanthin) of the pigment in the
`maculas of normal individuals, though wide variation
`in the former has been observed (Bone and Sparrock,
`1971 ; Bone et al., 1988 ; Hammond et al., 1996 ;
`Pease, Adams and Nuccio 1987 ; Werner, Donnelly
`
`* Correspondence to : Dr Richard A. Bone, Department of Physics,
`Florida International University, Miami, FL 33199, U.S.A.
`
`0014–4835￿97￿020211￿08 $25.00￿0￿ey960210
`
`and Kliegl, 1987). Also prenatal whole retinas and
`postnatal retinas up to the age of about 2 years differ
`from older age groups, having a higher lutein
`: zeaxanthin ratio (Bone et al., 1988).
`Zeaxanthin extracted from the macula has been
`shown to consist of similar amounts of the (3R, 3￿R)
`and (3R, 3￿S) stereoisomers, and possibly a much
`smaller amount of the (3S, 3￿S) stereoisomer (Bone et
`al., 1993). [Lutein is present as the single stereoisomer,
`(3R, 3￿R, 6￿R). The
`structure of
`this and the
`zeaxanthin stereoisomers may be found in Straub
`(1987).] Of
`the zeaxanthin stereoisomers, only
`(3R, 3￿R)-zeaxanthin has been detected in human
`blood serum (Bone et al., 1993). The other two
`stereoisomers may be present in the serum, but if so,
`they are below current
`levels of detection. This
`suggests that their presence in the eye is the result of
`carotenoid transformations occurring in the eye. In
`particular, it has been postulated that the presence of
`(3R, 3￿S)-zeaxanthin might be due to isomerization of
`lutein within the retina (Bone et al., 1993).
`Current interest in the macular pigment has been
`sparked by the intriguing possibility that it may play a
`protective role against some forms of age-related
`macular degeneration (AMD). Protection could be
`provided by the pigment in two ways : through its
`ability to quench free radicals and singlet oxygen, and
`by absorbing blue light before it reaches the sensitizers
`which initiate photochemical damage. AMD is the
`leading cause of vision loss among the elderly (Hyman,
`1992) and is without cure. Assessing risk factors for its
`development is therefore of considerable importance.
`
`￿ 1997 Academic Press Limited
`ZeaVision Exh. 2023
`
`

`

`212
`
`R. A. B O N E E T A L.
`
`Preliminary data have shown that AMD donor eyes
`have, on average, a somewhat lower concentration of
`zeaxanthin and lutein throughout the retina than do
`non-AMD controls (Landrum et al., 1995 ; Landrum,
`Bone and Kilburn, 1996a). In the same vein, Seddon
`et al., 1994 found that among dietary carotenoids, a
`higher intake of lutein and zeaxanthin was associated
`with a reduced risk of neovascular AMD. The Eye
`Disease Case-Control Study Group (1993) found the
`same to be true when examining serum levels of
`carotenoids, including lutein and zeaxanthin. How-
`ever, this result is inconclusive at present. Another
`study (Mares-Perlman et al., 1995) has revealed no
`such association.
`The ability to increase the amount of macular
`pigment by dietary supplementation with lutein has
`been demonstrated (Landrum et al., 1996a ; Landrum
`et al., 1996b). Such a strategy may become recognized
`as an effective means of reducing the risk, and￿or
`progression, of AMD in some individuals. Under-
`standing the mechanism of pigment accumulation
`and possible transformation in the macula is vitally
`important to the development of such a therapy. If, for
`example,
`lutein is partially isomerized into meso-
`zeaxanthin, the retina may gain an advantage owing
`to the slightly greater quenching efficiency of
`zeaxanthin over lutein (Foote, Chang and Denny,
`1970). In this study we have measured the distri-
`butions of lutein and the zeaxanthin stereoisomers
`across the retina using pooled samples from many
`eyes, as well as individual eyes. The results support the
`isomerization hypothesis and provide a straightfor-
`ward explanation of the carotenoid distributions.
`
`2. Methods
`
`Sample Preparation
`
`Human donor eyes were obtained fresh from the
`Florida Lions Eye Bank. These were stored in the dark
`at ￿20￿C prior to analysis for periods not exceeding
`two weeks. Additional eyes, fixed in formaldehyde
`within approximately 6 hr of death, were provided by
`the National Disease Research Interchange (NDRI).
`These eyes were stored in the dark at 4￿C for one to
`four weeks prior to analysis. All eyes were either from
`adults over 55, or from infants 0 to 7 months old.
`Some of
`the eyes in the older group, obtained
`exclusively from the NDRI, were from AMD patients.
`(They were also used in a parallel study on AMD
`whose results are not reported here.) These were
`diagnosed prior to death but the extent of the disease
`was unknown. Normal eyes were those for which no
`diagnosis of AMD or other eye disease had been made
`and which showed no visible abnormalities of the
`retina upon dissection. Procurement methods for
`tissues used in this study were humane, including
`proper consent and approval, and complied with the
`tenets of the Declaration of Helsinki. In order to obtain
`
`whole, untorn, neural retinas, each eye was immersed
`in 0￿9 % saline solution during dissection. Care was
`taken to minimize exposure of the tissues to bright
`light. In the case of adults, the intact retina was
`allowed to settle on a 1￿ Lucite sphere which was
`raised from the solution and placed in a device,
`previously described (Bone et al., 1988), which
`permitted cutting the retina into pieces of
`tissue
`concentric with the fovea. For the present study, the
`device included trephines of 3, 11, and 21 mm
`diameter, resulting in a central disk of tissue of area
`7￿1 mm￿ containing the yellow spot, and two con-
`centric annuli of areas 93 and 343 mm￿. For the 0 to
`7 month old age group, the entire retina was used.
`In the earlier phase of the study, corresponding
`disks and annuli of tissue from 10 eyes were pooled for
`carotenoid analysis. A total of 60 eyes were analysed
`in this way. An alternative procedure was developed
`in the later phase to permit the analysis of individual
`tissue samples. Thirty seven eyes from 24 donors were
`thus analysed. To extract the carotenoids, the pooled
`or individual tissue samples were homogenized in a
`glass tissue grinder with 2 ml of ethanol￿water (1 : 1).
`For the individually analysed eyes, 0￿5 ml of an
`ethanol solution of lutein monohexyl ether were added
`as an internal standard. The solution was routinely
`calibrated spectrophotometrically, 0￿5 ml containing
`￿ 12 ng of
`the standard. The homogenate was
`transferred to a large culture tube and the tissue
`grinder rinsed with three 2 ml aliquots of ethanol￿
`water and two 5 ml aliquots of hexane, the rinses
`being added to the culture tube. After vortexing and
`centrifuging, the hexane layer was transferred to a
`pear-shaped flask and dried under a stream of N￿. The
`final preparation step was to concentrate the samples
`in 30 µl of the HPLC mobile phase (see below).
`
`Reversed-phase HPLC
`
`In order to separate and quantify the zeaxanthin
`and lutein components in each sample, a reversed-
`phase HPLC system was employed. This included a
`250￿2 mm C-18
`column
`packed with
`3 µm
`Ultracarb ODS (Phenomenex, Torrance, CA, U.S.A.).
`The mobile phase was 90 % acetonitrile and 10 %
`methanol, with 0￿1 % (v￿v) of triethyl amine added to
`inhibit degradation of carotenoids during elution. The
`flow rate was 0￿2 ml min−￿ and detection was at
`452 nm. (The advantage of a 2 mm column over the
`standard 4￿6 mm column used previously (Bone et al.,
`1988) cannot be over-emphasized. Peak height was
`increased by a factor of ￿ 5 and solvent consumption
`reduced by the same factor.)
`
`Stereoisomer Analysis
`
`Zeaxanthin stereoisomers were readily separable
`from each other by either of two methods. For the
`pooled samples,
`the zeaxanthin collected during
`
`

`

`L U T E I N A N D Z E A X A N T H I N I N T H E R E T I N A
`
`213
`
`L
`
`ZT
`
`(A) In n er
`
`(B) Media l
`
`(C) Ou t er
`
`ZT
`
`L
`
`L
`
`ZT
`
`2 Min s
`
`F. 1. HPLC chromatograms, obtained with a reversed-phase column, of macular pigment extracts from three different
`regions of a single human retina. (A) ‘ Inner ’ – disk centered on fovea obtained with 3 mm trephine, area 7￿1 mm￿. (B) ‘ Medial ’
`– annulus obtained with 3 and 11 mm trephines, area 93 mm￿. (C) ‘ Outer ’ – annulus obtained with 11 and 21 mm trephines,
`area 343 mm￿. L ￿ lutein, Z T
`￿ combined zeaxanthin stereoisomers. The chromatograms have been truncated and do not
`show the internal standard.
`
`reversed-phase HPLC was converted to the dibenzoate
`derivative and analysed on a chiral column as
`described elsewhere (Bone et al., 1993). The reaction,
`however, is highly inefficient and did not permit the
`analysis of carotenoids in single retinas or portions
`thereof. For these the following procedure, described
`by Ru￿ ttimann, Schiedt and Vecci (1983), was adopted.
`Zeaxanthin collected during sample elution on the
`reversed-phase column was thoroughly dried under a
`stream of N￿ in a siliconized microcentrifuge tube. The
`tube was transferred to a glove box containing a dry
`N￿ atmosphere in order to carry out the derivatization
`procedure. The zeaxanthin, which had been concen-
`trated into the bottom of the tube, was dissolved in
`20 µl of anhydrous pyridine￿benzene (50 : 50 v￿v). To
`this was added 1 µl of (S)-(￿)-1-(1-Naphthyl) ethyl
`isocyanate, and the reaction allowed to proceed at
`room temperature for ￿ 48 hr.
`The dicarbamate derivatives thus produced were
`analysed by HPLC, using a 250￿2 mm normal-phase
`column
`packed with
`silica
`5 µm Prodigy
`(Phenomenex, Torrance, CA, U.S.A.). The mobile
`phase was 88 % hexane and 12 % isopropyl acetate at
`a flow rate of 0￿2 ml min−￿. Detection was at 451 nm.
`No internal standard was necessary ; the total quantity
`of zeaxanthin stereoisomers was obtainable from the
`reversed-phase chromatography and the normal-
`phase separation permitted measurement of
`their
`relative proportions. The same was true for the
`dibenzoate derivatives analysed on the chiral column.
`It should be stressed that in a significant number of
`cases the normal-phase system was operating at, or
`close to, its signal-to-noise threshold (peak height ￿
`twice noise amplitude). As a result, a number of
`samples did not yield useful data and were therefore
`excluded from further analysis.
`
`For comparison purposes, standards of the three
`zeaxanthin stereoisomers were prepared from rhodo-
`xanthin as previously described (Maoka et al., 1986),
`and converted to the dibenzoate and dicarbamate
`derivatives.
`
`3. Results
`
`Retinal Distribution
`
`Reversed-phase chromatograms of extracts from the
`central disk (‘ inner ’) and two concentric annuli of
`tissue (‘ medial ’ and ‘ outer ’) consistently displayed the
`trend, previously observed (Bone et al., 1988 ;
`Handelman et al., 1988), of an increasing lutein
`(L) : zeaxanthin ratio with increasing distance from the
`fovea. This is apparent from the sample chromato-
`grams, obtained from a single eye, shown in Fig. 1.
`The
`zeaxanthin collected during reversed-phase
`chromatography from the inner, medial and outer
`regions for the pooled samples subsequently yielded
`chiral column chromatograms such as those shown in
`Fig. 2. Figure 3 depicts a similar set obtained for a
`single eye,
`the collected zeaxanthin having been
`converted to the dicarbamate derivative and analysed
`on a normal-phase column. Note that the order of
`elution of the stereoisomers is different for the two
`chromatographic systems.
`The chromatograms of Figures 2 and 3 were
`generally representative of all adult donor eyes
`analysed, exhibiting a decreasing ratio of meso-
`zeaxanthin (MZ) to zeaxanthin (Z) with increasing
`eccentricity from the fovea. For the inner, medial and
`outer regions, the average MZ : Z ratios￿.. obtained
`using individuals eyes were 0￿83￿0￿15, 0￿39￿0￿12,
`and 0￿24￿0￿16 respectively.
`In computing such
`
`

`

`214
`
`R. A. B O N E E T A L.
`
`Z
`
`MZ
`
`Z
`
`MZ
`
`Z
`
`MZ
`
`(A) In n er
`
`(B) Media l
`
`(C) Ou t er
`
`4 Min s
`
`SZ
`
`F. 2. HPLC chromatograms, obtained with a chiral column, of dibenzoate esters of zeaxanthin stereoisomers. These were
`obtained from the three different regions defined in Fig. 1 and represent the pooled extracts from 10 retinas. In order of elution,
`the stereoisomers are meso-zeaxanthin (MZ), zeaxanthin (Z) and (3S, 3￿S)-zeaxanthin (SZ).
`
`Z
`
`MZ
`
`Z
`
`MZ
`
`(A) In n er
`
`(B) Media l
`
`(C) Ou t er
`
`Z
`
`MZ
`
`SZ
`
`4 Min s
`
`SZ
`
`F. 3. HPLC chromatograms, obtained with a normal-phase column, of dicarbamate esters of zeaxanthin stereoisomers.
`These were obtained from the three different regions, defined in Fig. 1, of a single human retina. In order of elution, the
`stereoisomers are (3S, 3￿S)-zeaxanthin (SZ), meso-zeaxanthin (MZ) and zeaxanthin (Z).
`
`left￿right differences in macular pigment density have
`been found (Bone et al., 1988 ; Landrum, Bone and
`Kilburn, 1996a ; Landrum et al., 1996b). A repeated
`measures, one way analysis of variance indicated a
`significant difference among the three regions [F(2,
`46) ￿ 150￿07, P ￿ 0￿001]. A Fisher’s least significant
`difference post hoc test performed at a 5 % significance
`level indicated that all pairs of means differed from one
`another. These MZ : Z ratios, obtained from individually
`analysed eyes, were generally consistent with those
`obtained from the pooled samples where the average
`ratios were 0￿79￿0￿06, 0￿51￿0￿06, and 0￿30￿0￿08
`respectively. (An independent samples, two-tailed t-
`test returned P values of ￿ 0￿5, ￿ 0￿02, and ￿ 0￿2 for
`the differences between the pooled and individual
`inner, medial and outer regions respectively.) From
`these results, the combined averages for all donor eyes
`were 0￿82￿0￿12, 0￿41￿0￿10, and 0￿25￿0￿13 re-
`spectively.
`The small
`
`leading peak in Fig. 3(A) has been
`
`250
`
`300
`
`350
`
`400
`
`450
`
`500
`
`550
`
`Wa velen gt h (n m )
`
`F. 4. UV-visible absorption spectrum of the SZ peak in
`Figs 2 and 3. Solvent : hexane￿isopropyl acetate (88 : 12).
`The spectrum is identical to that of zeaxanthin and meso-
`zeaxanthin.
`
`averages, here and elsewhere, left and right eye data
`for a donor, where available, were first averaged. This
`conservative approach was adopted even though
`
`

`

`L U T E I N A N D Z E A X A N T H I N I N T H E R E T I N A
`
`215
`
`T I
`
`Concentrations* of lutein (L), zeaxanthin (Z ), meso-zeaxanthin (MZ ) and (3S, 3￿S )-zeaxanthin (SZ ) in the inner,
`medial and outer regions of 37 individual retinas from 24 donors
`
`Inner (7￿1 sq. mm)
`
`Medial (93 sq. mm)
`
`Outer (343 sq. mm)
`
`Donor
`number†
`
`L
`
`Z
`
`MZ
`
`SZ
`
`L
`
`Z
`
`MZ
`
`SZ
`
`L
`
`Z
`
`MZ
`
`SZ
`
`1A
`2A
`3A
`4A
`5A
`5B
`6A
`6B
`7A
`7B
`8A
`9A
`9B
`10A
`10B
`11A
`11B
`12A
`12B
`13A
`14A
`15A
`15B
`16A
`17A
`17B
`18A
`18B
`19A
`19B
`20A
`20B
`21A
`21B
`22A
`23A
`24A
`
`Average
`Stdev
`
`4￿610
`4￿590
`4￿390
`4￿200
`4￿160
`0￿343
`3￿540
`3￿080
`2￿730
`2￿620
`2￿510
`2￿030
`1￿390
`2￿010
`1￿600
`1￿910
`1￿730
`1￿810
`1￿630
`1￿800
`1￿710
`1￿540
`1￿400
`1￿480
`1￿400
`1￿110
`0￿737
`0￿636
`0￿677
`0￿609
`0￿455
`0￿159
`0￿423
`0￿374
`0￿266
`0￿206
`0￿189
`
`1￿92
`1￿42
`
`2￿520
`3￿030
`3￿090
`2￿190
`4￿470
`0￿275
`2￿630
`2￿790
`3￿750
`3￿740
`2￿030
`2￿250
`1￿920
`1￿870
`1￿450
`1￿890
`1￿190
`1￿330
`1￿020
`0￿804
`1￿510
`1￿600
`1￿290
`1￿500
`1￿120
`0￿779
`0￿720
`0￿543
`0￿817
`0￿700
`0￿413
`0￿146
`0￿264
`0￿302
`0￿166
`0￿064
`0￿320
`
`1￿54
`1￿04
`
`2￿010
`2￿470
`2￿470
`1￿670
`3￿430
`0￿095
`1￿350
`2￿130
`2￿590
`2￿590
`1￿840
`1￿320
`1￿210
`0￿831
`0￿832
`1￿550
`1￿200
`1￿170
`0￿825
`0￿737
`1￿520
`1￿230
`1￿290
`1￿230
`0￿754
`0￿439
`0￿742
`0￿597
`0￿866
`0￿768
`0￿325
`0￿131
`0￿289
`0￿283
`0￿133
`0￿061
`0￿285
`
`1￿20
`0￿76
`
`0￿107 0￿950 0￿281 0￿103 0￿0120
`0￿412 0￿295 0￿146 0￿055 0￿0097
`0￿430 0￿567 0￿211 0￿086 0￿0220
`0￿315 0￿180 0￿133 0￿042 0￿0260
`0￿744 0￿442 0￿306 0￿116 0￿0160
`**
`0￿260 0￿160 0￿044 0￿0085
`0￿215 0￿287 0￿182 0￿071 0￿0173
`0￿394 0￿160 0￿089 0￿039 0￿0097
`0￿381 0￿236 0￿133 0￿088 0￿0130
`0￿464 0￿385 0￿289 0￿091 0￿0186
`0￿312 0￿091 0￿052 0￿027 0￿0047
`0￿196 0￿094 0￿072 0￿022 0￿0042
`0￿183 0￿067 0￿064 0￿019 0￿0021
`0￿115 0￿259 0￿165 0￿046 0￿0042
`0￿085 0￿146 0￿078 0￿013
`**
`0￿293 0￿134 0￿079 0￿024
`**
`0￿387 0￿109 0￿056 0￿029 0￿0061
`0￿224 0￿082 0￿035 0￿011 0￿0149
`0￿297 0￿032 0￿009 0￿004 0￿0020
`0￿102 0￿137 0￿044 0￿020 0￿0017
`0￿340 0￿254 0￿155 0￿075
`**
`0￿220 0￿105 0￿069 0￿028 0￿0051
`0￿339 0￿088 0￿056 0￿026 0￿0079
`0￿298 0￿104 0￿211 0￿029
`**
`0￿119 0￿250 0￿142 0￿055 0￿0066
`0￿097 0￿232 0￿169 0￿060
`**
`0￿105 0￿081 0￿028 0￿018 0￿0029
`0￿080 0￿072 0￿026 0￿016
`**
`0￿101 0￿040 0￿006 0￿003
`**
`0￿144 0￿041 0￿021 0￿012 0￿0022
`0￿050 0￿015 0￿009 0￿002 0￿0014
`0￿025 0￿019 0￿010 0￿002 0￿0007
`0￿117 0￿037 0￿028 0￿015 0￿0050
`0￿046 0￿049 0￿024 0￿016 0￿0021
`0￿042 0￿041 0￿018 0￿007 0￿0013
`**
`0￿043 0￿018 0￿008
`**
`**
`0￿017 0￿016 0￿005
`**
`
`0￿21
`0￿14
`
`0￿19
`0￿21
`
`0￿10
`0￿08
`
`0￿037 0￿0064
`0￿031 0￿0072
`
`0￿259
`0￿079
`0￿123
`0￿044
`0￿012
`0￿116
`0￿086
`0￿079
`0￿109
`0￿133
`0￿041
`0￿033
`0￿029
`0￿092
`0￿068
`0￿054
`0￿049
`0￿034
`0￿053
`0￿070
`0￿132
`0￿038
`0￿033
`0￿048
`0￿114
`0￿093
`0￿036
`0￿042
`0￿020
`0￿022
`0￿015
`0￿014
`0￿015
`0￿014
`0￿020
`0￿015
`0￿014
`
`0￿064
`0￿055
`
`0￿0735
`0￿0140
`0￿0430
`0￿0117
`0￿0529
`0￿0562
`0￿0389
`0￿0327
`0￿0650
`0￿0749
`0￿0489
`0￿0175
`0￿0207
`0￿0557
`0￿0359
`0￿0211
`0￿0085
`0￿0067
`0￿0234
`0￿0136
`0￿0570
`0￿0164
`0￿0150
`0￿0256
`0￿0424
`0￿0304
`0￿0099
`0￿0105
`0￿0019
`0￿0126
`0￿0072
`0￿0061
`0￿0108
`0￿0041
`0￿0073
`0￿0054
`0￿0138
`
`0￿027
`0￿021
`
`0￿0228
`0￿0037
`0￿0082
`0￿0093
`0￿0194
`0￿0133
`0￿0060
`0￿0052
`0￿0160
`0￿0208
`**
`0￿0042
`0￿0044
`0￿0034
`0￿0048
`0￿0048
`0￿0018
`0￿0014
`0￿0076
`0￿0030
`0￿0176
`0￿0039
`0￿0023
`**
`0￿0057
`0￿0067
`0￿0036
`0￿0037
`0￿0009
`**
`0￿0013
`0￿0012
`0￿0043
`0￿0024
`0￿0014
`0￿0019
`0￿0013
`
`0￿0060
`0￿0064
`
`0￿0036
`0￿0008
`0￿0015
`0￿0056
`0￿0016
`0￿0003
`**
`**
`0￿0045
`0￿0025
`**
`0￿0013
`0￿0007
`**
`**
`**
`0￿0004
`0￿0037
`0￿0031
`**
`**
`0￿0029
`0￿0013
`**
`0￿0008
`**
`**
`**
`**
`**
`0￿0007
`**
`0￿0032
`0￿0007
`**
`**
`**
`
`0￿0011
`0￿0015
`
`* Units are pmole of carotenoid per sq. mm of tissue.
`** Undetectable.
`† A and B refer to the 2 eyes of a donor. Where only one eye was fully analysed, only A is used.
`
`tentatively identified as (3S, 3￿S)-zeaxanthin (SZ). The
`same component appears as the small trailing shoulder
`in Fig. 2(A). Accumulated material
`from many
`experiments yielded the UV-visible absorbance spec-
`trum shown in Fig. 4. The spectrum was indis-
`tinguishable from those of the other two zeaxanthin
`stereoisomers. (The different spatial orientations of the
`hydroxyl groups of the three stereoisomers do not
`affect their spectra.) Furthermore, coinjection with the
`mixture of derivatized stereoisomer standards (Z, MZ
`and SZ) on the normal-phase HPLC column produced
`enhancement of the SZ peak.
`For each eye that was analysed individually, the
`
`mass of total zeaxanthin stereoisomers, ZT, in an inner,
`medial or outer tissue sample was determined from the
`reversed-phase chromatogram as the product of the
`ratio of the ZT peak area to that of the internal
`standard, the mass of the standard, and a weighting
`factor which accounted for the different extinction
`coefficients of ZT and the standard at the detection
`wavelength. A similar calculation gave the mass of L.
`[A linear relationship between peak area and the mass
`of analyte (L, ZT and standard) was confirmed by
`independent measurements in which a series of
`samples with masses throughout the range of interest
`was analysed.] From the relative peak areas on the
`
`

`

`216
`
`R. A. B O N E E T A L.
`
`OH
`
`H
`
`5'
`
`6'
`
`7'
`
`4'
`
`1'
`
`3'
`2'
`
`OH
`
`7
`
`11
`
`15
`
`13'
`
`9'
`
`9
`
`13
`
`15'
`
`11'
`
`Lu t ein
`(3R,3'R,6'R)-β,ε-Ca r ot en e-3,3'-diol
`
`6
`
`5
`
`1
`
`4
`
`2 3
`
`H O
`
`H O
`
`m eso-Zea xa n t h in
`(3R,3'S)-β,β-Ca r ot en e-3,3'-diol
`
`F. 5. Structures of lutein and meso-zeaxanthin. Con-
`version of lutein into meso-zeaxanthin would require only
`the migration of the 4￿,5￿ double bond in lutein to the 5￿,6￿
`position to form meso-zeaxanthin.
`
`Media l
`
`0.6
`
`MZ:Z
`
`In n er
`
`0.8
`
`1.0
`
`Ou t er
`
`4
`
`3
`
`2
`
`L:Z
`
`0.4
`
`.2
`
`10
`
`F. 6. For the pooled retinal extracts, meso-zeaxanthin
`decreases, while lutein increases, relative to zeaxanthin,
`with increasing distance from the fovea.
`
`considerable. From Table I, MZ represents on average
`approximately 25 % of the total amount of carotenoids
`in the central disk of tissue. We previously put forward
`the hypothesis that this stereoisomer might be derived
`in the eye from L (Bone et al., 1993). The process
`would require only the migration of a double bond in
`the lutein molecule, leaving the spatial configuration
`of the hydroxyl groups unaltered. This may be seen in
`Fig. 5. Evidence in support of this possible origin of MZ
`may be found in the variation in carotenoid com-
`position with retinal eccentricity given in Table I.
`Normalizing with respect to Z, it is seen that in the
`central disk, the MZ : Z ratio is high and the L : Z ratio
`is low ; in the outer annulus the situation is reversed,
`and an intermediate situation exists in the inner
`annulus. This trend is also illustrated graphically in
`Fig. 6, which is based on the data obtained from the
`pooled samples. We might
`imagine a conversion
`mechanism concentrated
`in the macula
`and
`decreasing in density with distance from the fovea. If
`the mechanism were to operate everywhere in the
`retina with 100 % efficiency, the (L￿MZ) : Z ratio
`
`normal-phase chromatograms, it was then possible to
`apportion the mass of ZT among the zeaxanthin
`stereoisomers
`represented.
`The
`numbers
`of
`pmoles mm−￿ of the different carotenoids present in
`the inner, medial, and outer regions of 37 retinas (24
`donors) are shown in Table I, together with the
`averages. Data were included in this study only when
`both reversed- and normal-phase HPLC analyses were
`successfully accomplished for all three regions.
`
`Age Effects
`
`The average L : Z mass ratio in the whole adult
`retina was calculated to be 1￿86￿0￿63. This was
`based on the amounts of L and Z in the three combined
`regions of the individually analysed retinas. [The effect
`on this L : Z ratio of
`the very small amounts of
`carotenoids which we have measured in the retina
`beyond the outer annulus (data not shown)
`is
`negligible.] By comparison, the same ratio determined
`for five donors comprising the 0 to 7 month old age
`group was 3￿3￿1￿8. The difference is significant at P
`￿ 0￿005 in a two-tailed, independent samples t test.
`Also from Table I, the average MZ : Z ratio in the whole
`adult retina was 0￿50￿0￿13, whereas for the 0 to 7
`month old group, it was 0￿29￿0￿28. This difference is
`significant at P ￿ 0￿02.
`
`4. Discussion
`
`the main
`Lutein and zeaxanthin are two of
`xanthophylls found in higher plants (Goodwin, 1992),
`with lutein generally predominating, and this is the
`principal source of these carotenoids in the human
`diet. Higher animals are unable to synthesize caro-
`tenoids de novo. Analysis of zeaxanthin in a number of
`higher plants including corn and pumpkin, and also in
`commercial chicken egg yolks, indicates that it has the
`(3R, 3￿R) configuration,
`i.e.
`it is Z (Maoka et al.,
`1986). It is perhaps not surprising therefore, con-
`sidering the dietary origin of carotenoids in animals,
`that, among the zeaxanthin stereoisomers, only Z has
`been detected in human blood plasma (Bone et al.,
`1993).
`Carotenoids ingested by animals may undergo
`metabolic transformations into other carotenoids, and
`this is presumably what is occurring in the human eye
`where MZ and SZ are present. The lack of detectable
`levels of these two stereoisomers in human blood
`argues against their formation elsewhere and sub-
`sequent transportation to the retina. Such a mech-
`anism might exist but would require an efficiency of
`uptake of MZ by the macula greatly exceeding that of
`L and Z. Apart from minor components identified
`isomers
`(Bone and
`tentatively as carotenoid cis
`Landrum, 1992), zeaxanthin and lutein are the only
`known carotenoids deposited in the retinal tissues. As
`such, they are obvious candidates for precursors of MZ
`and SZ. The amount of MZ in the macula is
`
`

`

`L U T E I N A N D Z E A X A N T H I N I N T H E R E T I N A
`
`217
`
`would not vary with eccentricity and, furthermore,
`would match the L : Z ratio in the blood. In fact the
`average values of (L￿MZ) : Z for the inner, medial and
`outer regions of the individually analysed adult retinas
`are 2￿16￿0￿60, 2￿32￿0￿77, and 2￿91￿1￿21 re-
`spectively. (For the pooled samples, the corresponding
`values, which differ insignificantly (P ￿ 0￿5) from the
`above, are 2￿22￿0￿33, 2￿28￿0￿36, and 3￿13￿0￿60.)
`These figures are consistent with various analyses of
`blood serum carotenoids which indicate L : Z ratios in
`the range of about 2 : 1 to 4 : 1 (Krinsky et al., 1990 ;
`Craft, 1992 ; Handelman, Shen and Krinsky, 1992).
`Analysis of variance of the (L￿MZ) : Z ratios indicated
`a significant difference among the three regions [F(2,
`46) ￿ 7￿57, P ￿ 0￿003]. A Fisher’s LSD post hoc test
`at a 5 % significance level
`indicated no difference
`between the inner and medial regions but a difference
`between these and the outer region. The higher
`(L￿MZ) : Z ratio here is compatible with an imperfect
`conversion mechanism which degrades a fraction of L
`during the postulated conversion process.
`In the
`central region, where the conversion rate is high, the
`loss of L would be correspondingly high and the
`(L￿MZ) : Z ratio would be lowered. The low rate of
`conversion in the outer region would mean a smaller
`loss of L and a relatively higher (L￿MZ) : Z ratio.
`A logical candidate for the conversion mechanism
`would be an enzyme, specifically an isomerase, capable
`of catalysing the reaction of L into MZ. If this were
`found only in the cone cell axons, the density of which
`decreases with eccentricity, the corresponding de-
`crease in conversion rate could be rationalized. We
`have also considered the possibility of a photochemical
`process. A [1,3]-sigmatropic shift of hydrogen from
`the 6￿ to 4￿ carbon of L’s ε-ionene ring would convert
`L into MZ. This process is thermally forbidden but
`photochemically allowed. There are, however, a
`number of objections to this mechanism. (1) In vitro
`experiments with L show no similar conversion into
`MZ. (2) There is no reason to believe that the average
`retinal illuminance decreases rapidly with eccentricity
`from the fovea in a way that would account for the
`changing conversion rate of L into MZ. (3) A similar
`process occurring in the β-ionene ring of Z would be
`equally likely. This would produce [3R, 3￿S, 6￿R]-
`lutein, an isomer not found in the retina.
`Further support for our hypothesis regarding the
`conversion of L into MZ comes from our 0 to 7 month
`old data. Here, a somewhat higher L : Z ratio in the
`whole retina, compared with adults, is once again
`associated with a low MZ : Z ratio, reminiscent of the
`outer region of the adult retina. To reconcile this
`observation with our model, the purported enzyme
`would be presumed to be absent or less effective in the
`newborn.
`The origin in the retina of the small quantity of SZ,
`which is not found in the serum, could be due to a
`process involving loss of configuration at the 3 and 3￿
`carbons of Z. This could occur through the biological
`
`oxidation of the two hydroxyl groups and subsequent
`non-stereospecific reduction to produce the three
`zeaxanthin stereoisomers.
`
`Acknowledgements
`
`The authors thank the National Disease Research In-
`terchange and the Florida Lions Eye Bank for providing
`human donor eyes, and Paulette Johnson for providing
`statistical assistance. Support was provided by an NIH grant
`GM08205.
`
`References
`
`Bone, R. A. and Landrum, J. T. (1992). Distribution of
`macular pigment components, zeaxanthin and lutein,
`in human retina. In : Methods in enzymology 213.
`(Packer, L., Ed.) Pp. 360–6. Academic Press, Inc. : San
`Diego, U.S.A.
`Bone, R. A., Landrum, J. T., Fernandez, L. and Tarsis, S. L.
`(1988). Analysis of the macular pigment by HPLC :
`Retinal distribution and age study. Invest. Ophthalmol.
`Vis. Sci. 29, 843–9.
`Bone, R. A., Landrum, J. T., Hime, G. W., Cains, A. and
`Zamor,
`J.
`(1993). Stereochemistry of
`the human
`macular carotenoids. Invest. Ophthalmol. Vis. Sci. 34,
`2033–40.
`(1985).
`J. T. and Tarsis, S. L.
`Bone, R. A., Landrum,
`Preliminary identification of the human macular pig-
`ment. Vis. Res. 25, 1531–5.
`Bone, R. A. and Sparrock, J. M. B. (1971). Comparison of
`macular pigment densities in human eyes. Vis. Res. 11,
`1057–64.
`Craft, N. E. (1992). Carotenoid reversed-phase high-per-
`formance liquid chromatography methods : reference
`compendium. In : Methods in enzymology 213. (Packer,
`L., ed.). Pp. 185–205. Academic Press, Inc. : San Diego,
`U.S.A.
`(1970).
`Foote, C. S., Chang, Y. C. and Denny, R. W.
`Chemistry of singlet oxygen. X. Carotenoid quenching
`parallels biological protection. J. Am. Chem. Soc. 92,
`5216–8.
`Goodwin, T. W. (1992). Distribution of carotenoids. In :
`(Packer, L., Ed). Pp.
`Methods in enzymology 213.
`167–72. Academic Press, Inc. : San Diego, U.S.A.
`Hammond, B. R., Curran-Celentano, J., Judd, S., Fuld, K.,
`Krinsky, N. I., Wooten, B. R. and Snodderly, D. M.
`(1996). Sex differences in macular pigment optical
`density : relation to plasma carotenoid concentrations
`and dietary patterns. Vis. Res. 36, 2001–12.
`Handelman, G. J., Dratz, E. A., Reay, C. C. and van Kuijk, F.
`J. G. M. (1988). Carotenoids in the human macula and
`whole retina. Invest. Ophthalmol. Vis. Sci. 29, 850–5.
`Handelman, G. J., Shen, B. and Krinsky, N. I. (1992). High
`resolution analysis of carotenoids in human plasma by
`high-performance liquid chromatography. In : Methods
`(Packer, L. ed.). Pp. 336–46.
`in enzymology 213.
`Academic Press, Inc. : San Diego, U.S.A.
`Hyman, L. (1992). Epidemiology of AMD. In Age-related
`macular degeneration : principles and practice. (Hampton,
`G. R. and Nelson, P. T., Eds). Pp. 1–35. Raven Press :
`New York, U.S.A.
`Krinsky, N. I., Russett, M. D., Handelman, G. J. and
`Snodderly, D. M. (1990). Structural and geometrical
`isomers of carotenoids in human plasma. J. Nutr. 120,
`1654–62.
`Landrum, J. T., Bone, R. A. and Kilburn, M. D. (1996a). The
`macular pigment : a possible role in protection from
`age-related macular degeneration. Adv. Pharmacol. 38,
`537–56.
`
`

`

`218
`
`R. A. B O N E E T A L.
`
`Landrum, J. T., Bone, R. A., Kilburn, M. D., Joa, H. and
`Gomez, C. (1996b). Dietary lutein supplementation
`increases macular pigment (MP). FASEB J. 10, A242.
`Landrum, J. T., Bone, R. A., Vidal, I., Menendez, E. and
`Kilburn, M. (1995). Macular pigment stereomers in
`individual eyes : a comparison between nor