Light Exposure and Macular Pigment Optical Density

Abstract

purpose. Biochemical research has demonstrated that lutein and zeaxanthin, the two macular carotenoids, are bleachable pigments. Further, evidence suggests that exposure to UV light can degrade plasma carotenoid levels in vivo. The present study investigated the effects of exposure to normal levels of light on the levels of lutein and zeaxanthin in the retina.

methods. The optical density of macular pigment (MPOD) was measured in two male subjects under four different light-adaptation conditions for 20 days. Heterochromatic flicker photometry was used to measure MPOD at 0.5° eccentricity.

results. The four conditions of light adaptation did not significantly affect MPOD. As in previous studies, however, a significant day-to-day difference was observed for both subjects.

conclusions. The results suggest that lutein and zeaxanthin levels in the eye are unaffected by light and oxidation throughout the day. This justifies current research methods in which MPOD measures are made regardless of the time of day. However, significant between-day variance indicates that multiple MPOD measures may be necessary to evaluate lutein and zeaxanthin levels in the retina accurately.

Lutein and zeaxanthin are naturally occurring yellow pigments called carotenoids. Similar to β-carotene, these pigments are found in various colored fruits and green leafy vegetables. Of the 40 to 50 carotenoids typically consumed in the human diet, ^{1 2} lutein and zeaxanthin, in particular, are deposited in the macular region of the retina. These pigments are collectively referred to as the macular pigment (MP). 

The MP has been postulated to improve acuity through the amelioration of the effects of chromatic aberration ^{3 4} and to preserve ⁵ and protect ^{6 7 8 9} the central retina. The latter hypothesized functions are based on the following two actions. First, MP absorbs light in the harmful short-wave end of the spectrum and dissipates it as heat before it renders damage to the foveal tissue. Second, the macular carotenoids are antioxidants, able to prevent free-radical reactions that can damage cellular components. 

Although several investigators have referred to the macular carotenoids as photostable or unbleachable pigments ^{10 11} (or at least not as rapidly bleachable in the same sense as the retinal photopigments), several biochemical studies have demonstrated that lutein and zeaxanthin can be bleached rapidly in vitro by photolysis ^{12 13} and by reaction with various oxidizing solutions. ^{14 15 16} The bleaching rate, the loss in λ_max absorption as a function of incubation time, is similar for lutein and zeaxanthin in the presence of many radical species (e.g., peroxyl radicals). This result seems to indicate that their antioxidant effectiveness is similar. However, lutein and zeaxanthin have slower bleaching rates than other carotenoids, such as β-carotene and lycopene. ¹⁷ Siems et al. ¹⁶ suggested that the macular carotenoids’ slower bleaching rate may explain their exclusivity in the retina. 

Roe ¹⁸ demonstrated that carotenoids could be photodegraded in vivo. She found that plasma carotenoid levels among her female subjects were significantly lower after 11 days of UVA and UVB light treatment. The daily exposure was equivalent to approximately 8 minutes of sunlight. White et al. ¹⁹ observed similar effects in both male and female subjects who were consuming a controlled diet. In both of these studies, however, it is unclear what carotenoids were measured. 

Previous investigators have observed day-to-day differences in the optical density, or amount, of macular pigment (MPOD) in individual subjects, ^{20 21} where MPOD is defined at 460 nm, the wavelength of peak absorption. Despite the unknown source of this variance, researchers have assumed that MP levels do not fluctuate throughout the day. In fact, numerous investigations have compared the MPOD of subjects without regard to the time of day the measurements were performed. If free radicals and light can bleach or deplete lutein and zeaxanthin in the eye, it may explain some of the difference found day to day. Furthermore, it suggests that macular pigment may decrease throughout the day as a result of light exposure and free radical quenching. Hammond et al. ²² proposed that similar effects may explain why individuals with light-colored irises (blue, gray) tend to have lower MPOD than individuals with dark-colored irises (brown, black). Compared with dark irises, light irises transmit more light, thereby exposing the MP to greater levels of photostress. The present study was designed to investigate the acute effects of normal light exposure on MPOD. 

Methods

Subjects

Two male graduate students (27 and 30 years old), well trained in psychophysics, participated in the experiment. Both subjects were aware of the experimental hypothesis and informed consent was obtained before initiation of the experiment. The experimental procedures adhered to the Declaration of Helsinki. 

Measurement of MPOD

The in vivo assessment of the MPOD takes advantage of the fact that lutein and zeaxanthin are deposited in the inner plexiform and photoreceptor–axonal layers. ²³ Consequently, MP filters light before it reaches the photopigments. Because the absorbance of the macular carotenoids is from 400 to 520 nm, ^{24 25} retinal loci screened by macular pigment have a reduced sensitivity to short-wave stimuli compared with other loci where no MP exists. Several studies have demonstrated that MPOD is highest in the fovea and declines to an optically undetectable point between 6° to 8° eccentricity. ^{26 27} Thus, if it is assumed that the fovea and parafovea have the same spectral sensitivities, then a decrease in foveal short-wave sensitivity would indicate the presence of MP. This assumption has been demonstrated to hold when the testing conditions favor detection by the mid- and long-wave cones. ^{5 10 27 28}  

In the present study, heterochromatic flicker photometry (HFP) was used to measure MPOD (see Werner et al. ¹⁰ ). HFP conducted across the visible spectrum (with varying monochromatic lights) at two retinal locations, one at approximately 7° outside the fovea, where there is no MP, and one in the fovea, where MP is present, yields two spectral sensitivity curves. Any decrease in foveal sensitivity to short-wave light, when these two curves are normalized, is attributable to MP. The MPOD, then, can be calculated for any wavelength by taking the log ratio between the two curves. Absorption spectra obtained in this psychophysical way have been shown to match absorption spectra obtained by other methods such as spectrophotometry used in HPLC ²⁹ and fundus reflectometry. ^{11 30} Given that 460 nm is the absorption peak of MP, it is customary to use only this wavelength to derive MPOD. Thus, the log energy of a 460-nm light necessary to minimize or stop flicker when alternating with a 570-nm light at the foveal location, minus the log energy needed 7° outside the fovea, is a measure of MPOD. 

Apparatus

MPOD in the subjects was assessed with a new free-view device designed specifically to measure MPOD in vivo (for a complete description of this device see Wooten et al. ³¹ ). In short, the new device used two light-emitting diode (LED) arrangements: one to produce the test stimulus, and the other to produce the background. The arrangement used to produce the test stimulus consisted of three LEDs, two with peak energy at 458 nm (20 nm half-bandwidth), and one with peak energy at 570 nm (20 nm half-bandwidth). The three LEDs used to produce the background each had peak energy at 470 nm (20 nm half-bandwidth). Light from both arrangements was collimated by planoconvex lenses (100 mm focal length), defined by apertures (mylar film and Physical Optics Co. [Torrence, CA] holographic type diffusing material), and combined by a 2-in. beam-splitter. The resultant stimulus appeared as a 1°, circular test field that alternated in counterphase between 458 and 570 nm, superimposed on the center of a 6°, 470-nm circular background. 

Procedure

For each measurement, the subject looked directly at the front surface of the beam-splitter (16 in. from the eye) through a viewing hole in the opaque box that housed the optical system. The subject fixated on a tiny black disc (5 minutes of arc) located at the center of the 1° test stimulus. Thus, the MPOD obtained with this stimulus is a measure of the MPOD at 0.5° eccentricity. ¹⁰ The parafoveal value was obtained by the subject’s fixating a small LED through another hole, 7° of visual angle (nasal retina) from the test stimulus. For the foveal measure HFP was performed at 13.0 Hz, and for the parafoveal measure, 6.5 Hz. This lower frequency for the parafovea was used to eliminate Troxler fading. The 458 and 570 nm LEDs were calibrated for light levels before each measurement. 

The experiment consisted of measurements on 20 nonconsecutive days in each subject. The experiment was conducted in 29 calendar days for one subject and 33 calendar days for the other. The night before each of the 20 measurement days, at a specified and constant time, each subject covered his right eye with two eye patches. One eye patch was sealed to the face with medical tape, and a second eye patch was placed over the sealed patch. The two eye patches were worn throughout the night and succeeding morning. The amount of bedtime for each subject was consistent before each experiment day, 6 hours for one subject and 8 hours for the other. Both subjects wore the eye patches for approximately 10 hours. The eye patches were removed immediately before a morning MPOD measurement (condition A). After this measurement, the subject viewed a 17°, white adaptation light (3356 cd/m²) for 5 minutes (condition B). The light source consisted of a 100-W bulb (Softwhite; General Electric, Cleveland, OH) placed immediately behind a spectrally flat milk glass diffusing screen. The lamp, having a color temperature of approximately 2800 K, provided a retinal irradiance of approximately 7 μW/cm² in the wavelength range of 400 to 500 nm. This light source was chosen to provide a luminance level that significantly exceeds average normal viewing conditions, while remaining within an acceptably safe limit of exposure. A second MPOD measure was then conducted after approximately 2 minutes. The time delay was necessary to extinguish the afterimage caused by the adaptation light. After this second measurement, each subject light adapted for 1 hour by reading a book illuminated by a light source located at a constant distance so as to yield a space-averaged luminance of 18 cd/m² emanating from the book (condition C). Each subject then, for a third time, had his MPOD measured. Condition C was intended to approximate a typical indoor viewing situation. A final MPOD measure (condition D) was obtained in the evening, approximately 6.5 hours after the third measure. This final measure was chosen to assess the effects of a full day of typical light exposure on MPOD. Condition D was preceded by 15 minutes of dark adaptation to reproduce the same cone equilibrium before the first measure. 

In sum, the assumption was that if MP were photolabile, condition A would yield relatively high MPODs owing to the period of prolonged darkness preceding the measurement. Conditions B and C were designed to represent two different levels of normal light exposure under experimental control. Condition D was designed to contrast with condition A, where the former included a period of prolonged dark adaptation and the latter a period of extensive light adaptation. Thus, if MP is bleachable under situations of exposure to normal light, then one or more of the light-adaptation conditions should yield values of MPOD lower than those obtained after condition A. 

Each subject thus had four MPOD measures (sessions) per experiment day. For each MPOD measure, the method of adjustment was used to obtain eight foveal and eight parafoveal HFP values. MPOD was calculated by taking the log ratio between the mean foveal and mean parafoveal values. The start times for each session were consistent throughout the experiment. 

Because diet has been shown to affect MPOD, the subjects followed a restricted diet to reduce this source of variability. One week before the experiment started, and throughout its duration, the subjects avoided foods high in lutein and zeaxanthin (e.g., spinach). Also, on experiment days, the first three sessions were conducted while the subjects fasted, and the final session followed an invariant lunch. 

Results

Figure 1 shows that MPOD did not vary significantly as a result of light history. No deviations from normality or homoscedasticity were obvious from visual inspection of the data. A separate one-way analysis of variance (ANOVA) for each subject indicated that the four conditions of light adaptation did not affect MPOD (subject 1 [S₁]: F = 0.195, P = 0.898; subject 2 [S₂] F = 1.476, P = 0.227). 

Given the possibility that MP is replenished in prolonged darkness and then bleaches rapidly after exposure to light (i.e., the stimulus from the first measurement value of the day), an analysis was performed on MPODs calculated from only the first foveal and parafoveal values from each trial. The data are virtually identical with those presented in Figure 1 . A one-way ANOVA for each subject, based only on these data, showed that MPOD was not significantly different after the four conditions of light adaptation (S₁: F = 0.541, P = 0.655; S₂: F = 0.751, P = 0.524), suggesting that MPOD was not affected by the test stimulus immediately after removal of the eye patch. 

In Figure 2 the mean MPOD (n = 4) on each experiment day is presented for both subjects. As in previous studies, day-to-day differences in MPOD were observed. One-way ANOVAs revealed that these differences were statistically significant in both subjects (S₁: F = 2.207, P = 0.01; S₂: F = 2.619, P = 0.002). If these observed day-to-day differences were in fact fluctuations in MPOD, it might be expected that the values obtained with the foveal target would significantly vary across days, whereas values obtained with the parafoveal target would remain invariant. A one-way ANOVA was performed on each subject’s daily foveal (n = 32) and parafoveal (n = 32) energy settings for HFP. According to the four separate ANOVAs, the energy necessary to eliminate flicker for both the foveal and parafoveal targets significantly varied in subject 1 (S_1fovea: F = 4.728, P < 0.0001; S_1parafovea: F = 7.507, P < 0.0001) and subject 2 (S_2fovea: F = 7.795, P < 0.0001; S_2parafovea: F = 7.130, P < 0.0001). Thus, it does not appear that MPOD varies from day to day. 

A Pearson’s r correlation between the two subjects’ daily mean MPOD measures was performed to assess the possibility that an artifact of the apparatus contributed to the variation in MPOD across days. No significant correlation was obtained (r = 0.299, P = 0.242), suggesting that day-to-day differences were not a consequence of an optical system artifact. 

Discussion

In the present study, MPOD remained stable throughout the day, regardless of a variety of light exposures within normal daily experience. This suggests that the macular carotenoids are unbleachable or photostable in the natural retinal environment. It may be that lutein and zeaxanthin in the eye can be degraded by light of greater intensity than the lights used in the current experiment. Both subjects, however, were regularly exposed to substantial, albeit unquantified, amounts of natural daylight before the fourth session of each experiment day. Two studies have shown that ultraviolet light can significantly reduce plasma carotenoid levels. ^{18 19} Although the effects reported by these studies were measured after 11 days of skin exposure, sunlight was avoided by subjects in both studies, and the daily exposure was the equivalent of only 8 minutes of sunlight. Unfortunately, these investigators did not specify which carotenoids were affected by the exposure to ultraviolet light. In vitro, sunlight can photodegrade lutein and zeaxanthin. After 4 hours, Siems et al. ¹⁶ observed significant reductions in the concentration of these carotenoids. Despite these findings, the retina is not commonly exposed to direct sunlight. Further, the eye environment may affect the bleaching rate of the macular pigment. 

Alternatively, the macular pigment may photodegrade throughout the day but be rapidly replaced by lutein and zeaxanthin circulating in the blood or from other body stores. This may be unlikely, given the time (weeks) necessary to increase MPOD by either dietary increases of lutein and zeaxanthin ³² or lutein supplementation. ^{28 30} In contrast, photodegraded lutein and zeaxanthin may be rapidly replenished. The time necessary to increase MPOD after supplementation, then, may involve altering a homeostatic mechanism that regulates the concentrations of these two carotenoids in the retina. 

Similar to several previous studies, day-to-day variation in MPOD was observed in well-trained subjects. Landrum et al. ²⁸ suggested that such variation is the result of measurement error, as opposed to fluctuations in MPOD. In the present study, however, neither subjects’ MPOD significantly varied within experiment days. This suggests that between-day variation may not be due to measurement error, but possibly to day-to-day fluctuations in subjects’ sensitivity to flicker ³³ or relative spectral sensitivity. However, potential measurement error, such as changes in subjects’ criterion of sensation luminance, may vary between days, but remain consistent within a day. Until the source of this variation is realized, multiple MPOD measures may be necessary to evaluate accurately lutein and zeaxanthin levels in the retina. 

 Submitted for publication November 29, 2001; revised May 13, 2002; accepted May 23, 2002.

 Commercial relationships policy: N.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Mean MPOD for each of four conditions of light exposure with associated SEs. MPOD was determined using a 458-nm test.

Figure 2.

 

Mean MPOD at 458 nm and SEs for each experiment day, averaged across conditions.

The authors thank Joanne Curran-Celentano for the use of equipment, Joanne Burke for assistance in the study, and two anonymous referees for helpful comments on an earlier version of the manuscript.