Major Changes in Human Ocular UV Protection with Age

Abstract

purpose. Age-dependent human lens coloration may be explained by the binding of UV filters to crystallins. It has been proposed that glutathione may compete for reaction with UV filter degradation products and therefore protect crystallins from modification. To understand this process, UV filters were quantified together with oxidized and reduced glutathione in human lenses of varying age.

methods. Lens tissues were homogenized in ethanol to extract the UV filters. Metabolites were quantified by HPLC and correlations between them in the nuclear and cortical regions of the lens were examined.

results. The concentrations of the UV filters 3-hydroxykynurenine, kynurenine, and 3-hydroxykynurenine glucoside decreased linearly with age, with slightly lower levels in the nucleus than the cortex. 4-(2-Amino-3-hydroxyphenyl)-4-oxobutanoic acid glucoside was found in higher levels in the nucleus than the cortex and decreased slowly in both regions with age. Glutathionyl-3-hydroxykynurenine glucoside was present in higher concentrations in the nucleus, barely detectable in young lenses, but increased significantly after age 50. Reduced glutathione levels were lower in the nucleus and decreased in both regions with age, yet oxidized glutathione increased in the nucleus but remained constant in the cortex.

conclusions. Results are consistent with a predominantly nuclear origin for both 4-(2-amino-3-hydroxyphenyl)-4-oxobutanoic acid glucoside and glutathionyl-3-hydroxykynurenine glucoside. This is in accord with their proposed mechanism of formation, which involves an initial deamination of 3-hydroxykynurenine glucoside. This process is more pronounced in older lenses, possibly because of the barrier to diffusion. The barrier may also explain the increase in nuclear oxidized glutathione that is observed with age.

One of the main functions of the lens is to absorb UV radiation transmitted by the cornea in the 295 nm to 400 nm band. This is performed by a group of low molecular weight, fluorescent compounds, termed UV filters. ^{1 2 3} Absorption of this band protects both the lens and retina from UV-induced photodamage and aids visual acuity by decreasing chromatic aberration. ²  

The UV filters are formed in the lens from l-tryptophan via the kynurenine pathway. ^{1 2 3 4} Indoleamine 2,3-dioxygenase (IDO) ^{5 6} catalyzes the formation of N-formylkynurenine, which is then hydrolyzed to yield the UV filter kynurenine (Kyn). ⁵ The major pathway for kynurenine metabolism is hydroxylation to 3-hydroxykynurenine (3OHKyn), followed by glycosylation to the major UV filter l-3-hydroxykynurenine O-β-d-glucoside (3OHKG). ^{3 7} These compounds are not good photosensitizers, befitting their role as UV filters. ⁸ It has been postulated that the interaction of UV filters with crystallins arises via an initial side-chain deamination to yield a highly reactiveα ,β-ketoalkene. ⁹ Such a process is also responsible for the formation of the second most abundant UV filter 4-(2-amino-3-hydroxyphenyl)-4-oxobutanoic acid O-β-d-glucoside (AHBG) ¹⁰ and the recently identified UV filter, a glutathione (GSH) adduct of 3OHKG (GSH-3OHKG). ¹¹  

With age, the human lens undergoes numerous biochemical changes including a yellowing of the nucleus and an overall increase in fluorescence. ¹² A number of agents, for example, malondialdehyde and glucose, have been proposed to play a role in these age-dependent changes (reviewed in Harding). ¹³ The importance of UV filters in normal human lens coloration and possibly cataract formation has been documented recently. ⁹ As well as 3OHKG being covalently attached to the crystallins, ⁹ under oxidative conditions, 3OHKyn can react with the crystallins to produce cross-linked products with features characteristic of those observed in age-related cataract lenses. ^{14 15}  

The role of reduced glutathione (GSH) as the essential and primary lenticular antioxidant is well established. ¹⁶ A lowered concentration of GSH is thought to increase the rate of posttranslational modification of crystallins. ¹⁶ It has been proposed recently that a permeability barrier at the nucleus/cortex interface causes GSH levels to drop in the nucleus with age. ¹⁷ It is possible that an insufficient concentration of GSH in the nucleus allows the UV filters to covalently link to crystallins, thereby coloring them and disrupting their conformation. Hydroxyl radical damage to crystallins ¹⁸ may also be observed if GSH levels drop below 1 mM, ¹⁹ resulting in the onset of nuclear cataract. 

In this study, the nuclear and cortical concentrations of the major UV filters and GSH in human lenses of varying ages are reported. 

Materials and Methods

Materials

l-3-Hydroxykynurenine, l-kynurenine, trifluoroacetic acid (TFA), meta-phosphoric acid (mPA), potassium hydrogenphosphate (K₂HPO₄), sodium borohydride (NaBH₄), 5,5′-dithiobis-(2-nitrobenzoic acid) (DTNB), tris(hydroxymethyl)aminomethane (Tris-HCl), cysteine, and oxidized glutathione (GSSG) were purchased from the Sigma–Aldrich Chemical Co. (St. Louis, MO). d,l-3-Hydroxykynurenine O-β-d-glucoside and 4-(2-amino-3-hydroxyphenyl)-4-oxobutanoic acid O-β-d-glucoside were synthesized by the method of Manthey and coworkers. ²⁰ Glutathionyl-3-hydroxykynurenine glucoside was synthesized by the method of Garner and coworkers. ¹¹ Ethanol, methanol, and acetonitrile were purchased from Ajax (Auburn, NSW, Australia). Human lenses were obtained postmortem from donor eyes at the National Disease Research Interchange (Philadelphia, PA) and The Sydney Eye Hospital Lions Eye Bank (Sydney, Australia). After removal, lenses were placed immediately into sterile plastic screw-capped vials and kept at −20°C until analyzed. 

Dissection of Lenses

The nucleus (∼50 mg) was separated from the cortex (100 to 200 mg) by coring through the visual axis with a 3-mm cork borer. Approximately 0.5 mm was cut from each end of the core and added to the doughnut-shaped remainder, referred to as the cortex. All dissections were performed at −18°C. Immediately after dissection, lens tissues were weighed and extracted. 

Protein-Free Lens Extracts

Two types of protein-free lens extracts were prepared. Extraction of UV filters was performed by homogenizing the lens tissue in 100% ethanol (0.15 ml/nucleus, 0.35 ml/cortex). The homogenate was left at −20°C for 60 minutes and then centrifuged (13,000g, 20 minutes, 4°C). The supernatant was removed and stored at −20°C while the pellet was re-extracted in 80% ethanol. The combined supernatants were lyophilized. Extraction of GSH and GSSG was performed by sonication of the lens tissue in 3% mPA on ice. The precipitated proteins were pelleted by centrifugation (13,000g, 20 minutes, 4°C) and the supernatant was removed. 

HPLC Methods

HPLC system consisted of: two ICI LC 1150 pumps; a Rheodyne 7125 sample injector; an ICI SD 2100 Variable Wavelength UV-VIS Detector set at 365 nm; and an ICI LC 1250 Fluorescence Detector. Chromatograms were recorded and peak areas integrated using the WinChrom Chromatography Data System (GBC Scientific Equipment, Castlehill, Australia). Standard curves and separations were performed on a 250 mm × 4.6 mm Varian Microsorb C18 column using an acetonitrile/H₂O gradient in 0.05% (v/v) TFA. The percentage acetonitrile in the gradient was 0% (5 minutes), 0% to 40% (50 minutes), and 40% to 0% (5 minutes). The flow rate was 0.6 ml/min. Standard curves for 3OHKyn, Kyn, d,l-3OHKG, GSH-3OHKG, and AHBG were prepared using these standard HPLC conditions. Lyophilized protein-free lens extracts were redissolved in water and injected on to the column. 

HPLC system consisted of: a Varian 2010 HPLC Pump; a Rheodyne 7125 sample injector; and a Varian 2050 Variable Wavelength UV-VIS Detector set at 200 nm. Standard curves for GSH and GSSG were prepared and separations were performed on a 250 mm × 4.6 mm Activon Goldpak Spherisorb S50DS2 C18 column. An isocratic gradient of 1% methanol, 10 mM K₂HPO₄ adjusted to pH 2.0 with mPA, at 25°C, 1.0 ml/min, was used. These conditions were modified from a method by Liu and coworkers. ²¹ The supernatant was injected directly on to the HPLC column. To confirm that the GSSG peak detected represented only GSSG and no other coeluting species, GSSG peaks were collected and reduced using NaBH₄ and reinjected onto the HPLC system. The GSSG peak disappeared and a GSH peak was observed with a stoichiometrically equivalent area. 

Non-protein Sulfhydryl Determination

Next, 10 μl of the mPA extraction supernatant was mixed with 300 μl of Tris-HCl buffer (1.0 M, pH 9.0) and the absorbance at 412 nm noted. Then, 10 μl of DTNB (10 mM in methanol) was added and the absorbance read after 30 seconds. Quantification was determined by reference to a standard curve of cysteine (0 mM to 1 mM). 

Statistical Analysis

Linear regression analysis was used to evaluate the relationships between age and UV filters or GSH. Paired comparisons such as nucleus/cortex were evaluated using the two-sample paired t-test to determine whether the data points were distinct. A P value < 0.01 was considered significant. 

Results

UV Filters

The HPLC chromatogram of a 54-year-old human lens ethanol extract, with UV detection at 365 nm, is shown in Figure 1 . The two major peaks correspond to 3OHKG (peak 2) and AHBG (peak 5). The smaller peaks are 3OHKyn (peak 1), Kyn (peak 3), GSH-3OHKG (peak 4), and a compound of unknown identity (peak 6) that elutes after AHBG. Peak 4 is a double peak as it represents the two diastereoisomers of GSH-3OHKG. ¹¹ Peak areas of the UV filters were compared to the appropriate standard curves to determine the concentrations present in each sample. 

Kynurenine

As depicted in Figure 2 , the concentration of Kyn was found to decrease linearly in both the nucleus (r ² = 0.58, P < 0.0005, n = 50) and cortex (r ² = 0.55, P < 0.0005, n = 50) as a function of age. Highest levels were detected in lenses below the age of 20 years and lowest levels were detected in lenses of 80 years of age or older. Although the data points are scattered, it is clear that from early adulthood (second decade) through to old age (eighth decade) the average concentration of this UV filter decreased from ∼30 nmol/g to 5 nmol/g, which is equivalent to a decrease of ∼12% per decade, in both the nucleus and cortex. No statistically significant difference (P > 0.01) was found between the levels detected in the nucleus and cortex. 

3-Hydroxykynurenine

3OHKyn showed a marked linear decrease in the nuclear (r ² = 0.52, P < 0.0005, n = 50) and cortical (r ² = 0.51, P < 0.0005, n = 50) concentrations with age (Fig. 3) . From the second decade to eighth decade, the 3OHKyn levels decreased from ∼7 nmol/g to 2 nmol/g in the nucleus and from ∼15 nmol/g to 5 nmol/g in the cortex. This represents a decrease of ∼12% per decade in both regions of the lens. All lenses contained statistically significant (P < 0.0005) lower levels in the nucleus than the cortex, with lenses above the age of 80 years displaying nuclear concentrations at near undetectable levels. 3OHKyn is readily oxidized at physiological pH; however, this oxidation is inhibited by the lenticular antioxidant GSH. ^{14 15} To establish whether the determined nuclear levels of 3OHKyn were artifactually decreased by oxidation during extraction, the nucleus of a 53-year-old lens was halved and extracted in either the presence or absence of 10 mM GSH. No significant difference in 3OHKyn concentration was detected. Therefore, it is concluded that the methodology for extraction of 3OHKyn is valid. 

3-Hydroxykynurenine Glucoside

Previous studies using small sample populations have reported that the concentration of the most abundant UV filter, 3OHKG, decreases in an exponential manner with age. ^{1 22} In this study, however, a much larger sample population was investigated (n = 50), covering a more comprehensive age-range and specifically analyzing the nuclear and cortical regions of the lens. The level of 3OHKG was found to decrease linearly in both the nucleus (r ² = 0.67, P < 0.0005, n = 50) and cortex (r ² = 0.69, P < 0.0005, n = 50) with age (Fig. 4) . Young lenses (second decade), contained the highest levels (∼400 nmol/g) in the nucleus and cortex, whereas older lenses (eighth decade), contained lower levels (∼100 nmol/g). This represents a substantial 50 nmol per decade loss of 3OHKG, which is equivalent to∼ 12% loss per decade. Individual lenses generally displayed slightly lower levels in the nucleus than the cortex (P = 0.001). 

Glutathionyl-3-Hydroxykynurenine Glucoside

In contrast to Kyn, 3OHKyn, and 3OHKG, the GSH-3OHKG adduct was always detected in higher concentrations in the nuclear region of the lens (P = 0.003), as seen in Figure 5 . Although some scatter of data are evident, after ∼50 years of age, the level of the GSH-3OHKG adduct displayed a significant increase, most pronounced in the nucleus. In particular, a 93-year-old lens contained the highest nuclear level of 684 nmol/g (omitted from Fig. 5 ). This lens appeared visibly brown and was difficult to homogenize, which is consistent with lenses that have nuclear cataract. The medical history of the donors and the appearance of the majority of the lenses before dissection did not suggest that they were cataractous. It should also be noted that the curves fitted to the nuclear and cortical data points in Figure 5 are for illustrative purposes and are examples of several curves that could be fitted to the data. 

4-(2-Amino-3-Hydroxyphenyl)-4-Oxobutanoic Acid Glucoside

The data obtained for AHBG contains a relatively large amount of variation. In a previous study using whole human lenses, no correlation was observed between the levels of AHBG and age. ¹⁰ Although it is difficult to justify the use of linear lines for the nucleus (r ² = 0.27, P = 0.0001, n = 50) and cortex (r ² = 0.28, P < 0.0005, n = 50), this study does show a slight decrease in AHBG concentration associated with age (Fig. 6) . Lenses below the age of ∼50 years all displayed concentrations above 20 nmol/g, whereas one-third of the lenses above ∼50 years of age displayed concentrations less than 20 nmol/g. Further, individual lenses consistently contained higher levels in the nucleus than the cortex (P < 0.0005). 

Oxidized and Reduced Glutathione

As shown in Figure 7 , the concentration of GSSG increased significantly in the nucleus as a function of age (r ² = 0.26, P = 0.001, n = 37). This concentration was lowest in lenses below 20 years of age, which displayed levels between ∼0.1 mM and 0.3 mM. With increasing age, this nuclear concentration increased, but with a high degree of variability, to∼ 0.5 mM to 0.6 mM by the eighth decade. In contrast, the concentration of GSSG in the cortex remained close to ∼0.2 mM. Linear regression analysis shows no significant trend (P > 0.01) with age. This observation may reflect the fact that there is little change in lenticular GSH reductase activity as the lens ages ²³ and that same is localized in the lens cortex. ²⁴  

The concentration of GSH decreased linearly in both the nucleus (r ² = 0.22, P = 0.002, n = 38) and cortex (r ²= 0.46, P < 0.0005, n = 38) with age. As seen in Figure 8 , in the cortex, young lenses below the age of 20 years displayed an average concentration of ∼6 mM. This cortical level decreased by ∼7% per decade, such that by the ninth decade the average concentration had halved to ∼3 mM. The average nuclear concentration of GSH in lenses less than 20 years of age was ∼4.5 mM. This nuclear level decreased by ∼10% per decade to an average value of ∼1 mM by the ninth decade. None of the lenses analyzed that were below 40 years of age had a nuclear GSH concentration less than 2 mM, whereas the great majority above 40 years of age had concentrations less than 2 mM (∼70%) and a number of these (∼35%) contained GSH concentrations less than 1 mM. When nuclear and cortical data are combined to give a picture of the whole lens, the data are very similar to those reported by Lou and Dickerson. ²⁵ Individual lenses consistently displayed lower levels in the nucleus than the cortex (P < 0.0005). These overall trends are consistent with other studies. ^{26 27 28 29 30}  

The redox environment of the lens can be inferred from the GSH:GSSG ratio. From early adulthood through to old age, the cortical ratio decreased from 30:1 to 11:1, indicating that the cortex had become a less reducing environment. The nuclear region exhibited an even greater (sixfold) decrease from 29:1 to 5:1, indicating that the nucleus had become a relatively more oxidizing environment. 

Discussion

This comprehensive study of human lenses, across the whole age range, has revealed a marked linear decline in the concentrations of the UV filters 3OHKG, Kyn, and 3OHKyn. This finding may have important consequences, in terms of our susceptibility to lenticular and retinal photodamage, particularly in later years. This decrease in UV protection is only partly compensated for by an increase in the concentration of GSH-3OHKG. The role of protein bound chromophores, which increase with age, ⁹ clearly warrants investigation in terms of impact on UV protection. The lens pathways linking the UV filters and glutathione species are depicted diagrammatically in Figure 9 . This summarizes the biochemistry of UV filter synthesis and glutathione distribution and their possible roles in coloration of the lens nucleus. 

Primate UV filters are synthesized from l-tryptophan. ^{1 2 3 4} The first metabolite in this pathway is N′-formylkynurenine, which cannot be detected by HPLC because it is rapidly hydrolyzed by a nonspecific formamidase enzyme to yield Kyn. ¹ In the present study, the concentration of Kyn in both the nucleus and cortex was found to decline linearly as a function of age. The reason for this is not clear because both tryptophan concentrations ¹ and in vitro IDO activity (Takikawa O, Littlejohn TK, Truscott RJW, manuscript submitted 2000) remain relatively constant with age. 

Kyn is the biosynthetic precursor of 3OHKyn; thus, it is logical that as the concentration of Kyn decreases toward 0 with age, a similar decrease in 3OHKyn is also observed. In addition, the concentration of 3OHKyn was significantly lower in the nucleus than the cortex. Previous studies in our laboratory have shown that the oxidation products of 3OHKyn formed in the presence of O₂ readily react with crystallins. ^{14 15} This tanning process occurs when the GSH concentration declines below 1.0 mM. ^{14 15} Therefore, the extent of tanning would be expected to increase as the concentration of GSH decreases in the lens with age. The consistently low concentration of 3OHKyn in the nucleus of older lenses could be explained by the oxidation products covalently linking to crystallins, or possibly by 3OHKyn being scavenged by GSH to form a new UV filter adduct analogous to GSH-3OHKG. 

With age, the major UV filter 3OHKG decreases markedly in both the nucleus and cortex. This is consistent with the decrease in concentration of the 3OHKG metabolic precursors (Kyn and 3OHKyn). Interestingly, these UV filters all declined at the same percentage rate (∼12% per decade). There are three identified pathways leading to the loss of 3OHKG. First, the formation of GSH-3OHKG. ¹¹ Based on age-dependent loss of 3OHKG and the increase of GSH-3OHKG in the nucleus, it is estimated that conjugation with GSH accounts for∼ 25% of the overall loss of 3OHKG at 85 years of age (Figs. 4 and 5) . As there is considerable variation in lens-to-lens GSH-3OHKG concentration, this figure can only be viewed as a guide. GSH-3OHKG appears to be formed mainly in the nucleus, as the nuclear concentration in any given lens was consistently higher than in the cortex. Further, within individual lenses a negative correlation was observed between the concentration of nuclear GSH-3OHKG and GSH. High levels of GSH-3OHKG were associated with low levels of free GSH and vice versa. This suggests an important role for GSH in the nucleus is scavenging deaminated 3OHKG to prevent this reactive α,β-ketoalkene covalently binding to the crystallins. Second, the formation of AHBG. ¹⁰ AHBG is also more concentrated in the nucleus than the cortex, which suggests that it is also formed in the nuclear region of the lens. The overall concentration of AHBG decreases slightly with age, which is in accord with the decline in its metabolic precursor 3OHKG. Third, the covalent linkage of 3OHKG to lens crystallins. This process has been reported to account for at least 50% of the increase in age-related coloration of the lens. ⁹ Unlike GSH-3OHKG and AHBG, 3OHKG bound to protein cannot diffuse out of the lens. Therefore, it gradually accumulates in the lens and permanently increases the yellow color of the lens. 

The formation of GSH-3OHKG, AHBG and 3OHKG-protein adducts all appear to occur primarily in the nucleus. Deamination of the 3OHKG amino acid side chain must occur before these reactions can proceed. ^{9 10 11} The observation that nuclear GSH-3OHKG concentrations, and 3OHKG-protein adducts ⁹ both begin to rise exponentially after the fifth decade suggests that some change to the biochemical conditions in the nucleus has occurred, thus promoting the deamination process. Accumulation of these UV filter deamination-derived products in the nuclear region may be related to the development of a barrier to diffusion within the lens. It has been proposed previously that with age, a barrier to diffusion develops at the nuclear/cortex interface that slows GSH in its movement from the cortex into the nucleus. ¹⁷ The results obtained in this study are consistent with this hypothesis. Both UV filters and GSH would have a substantially increased half-life in the nuclear region, as has been demonstrated for older normal lenses using D₂O diffusion. ³¹ This would result in more deamination and an increase in GSSG and a lowered GSH:GSSG ratio as the GSH scavenges potentially damaging oxidants. 

In summary, this work shows that there is a marked age-dependent linear decline in both the nuclear and cortical levels of all the major UV filters, with the exception of GSH-3OHKG. The distribution of this UV filter adduct indicates that it is formed in the nucleus and its rate of formation is markedly increased after 50 years of age. Because the mechanism of its formation involves nucleophilic attack by the sulfhydryl group of GSH on deaminated 3OHKG, ¹¹ this finding implies that deamination of UV filters is more prominent in the center of the lens, particularly after 50 years of age. Thus reaction with lens crystallins is also more likely in the nucleus, and this is supported by data on covalent interaction with 3OHKG. ⁹ Further, it emphasizes the important role of GSH in protecting the lens from such protein modification ³² since GSH levels also decrease linearly with age. One can speculate that once levels of GSH in the nucleus fall below a certain level posttranslational modification of crystallins may become much more extensive. 

²  Present affiliation: Department of Chemistry, Division of Environmental and Life Sciences, Macquarie University, NSW, Australia.

 Supported in part by grants from the National Health and Medical Research Council (NHMRC).

 Submitted for publication August 15, 2000; accepted September 20, 2000.

 Commercial relationships policy: N.

 

Figure 1.

 

Human lens ethanol extracts were analyzed for UV filter content by gradient HPLC. UV detection at 365 nm. Chromatogram is for a 54-year-old lens. 1, 3OHKyn; 2, 3OHKG; 3, Kyn; 4, GSH-3OHKG; 5, AHBG; 6, unknown.

Figure 2.

 

Concentration of Kyn in human lenses as a function of age. No statistical difference was evident between nuclear and cortical concentrations. Linear regression lines shown, nucleus (r = −0.76, P < 0.0005, n = 50) and cortex (r = −0.74, P < 0.0005, n = 50).—•— , nucleus; - -○- -, cortex.

Figure 3.

 

Concentration of 3OHKyn in human lenses as a function of age. The nuclear concentration was lower than the cortex in all lenses. Linear regression lines shown, nucleus (r = −0.72, P < 0.0005, n = 50) and cortex (r = −0.71, P < 0.0005, n = 50). Paired comparison t-test, nucleus/cortex (P < 0.0005). —•—, nucleus; - -○- -, cortex.

Figure 4.

 

Concentration of 3OHKG in human lenses as a function of age. The nuclear concentration was slightly lower than the cortex. Linear regression lines shown, nucleus (r = −0.82, P < 0.0005, n = 50) and cortex (r = −0.83, P < 0.0005, n = 50). Paired comparison t-test, nucleus/cortex (P = 0.001). —•—, nucleus; - -○- -, cortex.

Figure 5.

 

Concentration of the GSH-3OHKG adduct in human lenses as a function of age. The nuclear concentration was higher than the cortex in all lenses and increased significantly more than the cortex with age. Paired comparison t-test, nucleus/cortex (P = 0.003). —•—, nucleus; - -○- -mdash;- -○- -cir;- -○- -mdash;, cortex.

Figure 6.

 

Concentration of AHBG in human lenses as a function of age. Higher levels were detected in the nucleus than the cortex. Linear regression lines shown, nucleus (r = −0.52, P = 0.0001, n = 50) and cortex (r = −0.53, P < 0.0005, n = 50). Paired comparison t-test, nucleus/cortex (P < 0.0005). —•—, nucleus; - -○- -, cortex.

Figure 7.

 

Concentration of GSSG in human lenses as a function of age. Levels remained constant in the cortex, whereas in the nucleus the concentration increased with age. Only the linear regression line for the nucleus is shown (r = 0.51, P = 0.001, n = 37). Paired comparison t-test, nucleus/cortex (P < 0.0005). —•—, nucleus; ○, cortex.

Figure 8.

 

Concentration of GSH in human lenses as a function of age. Both the nuclear and cortical concentrations decreased with age. All lenses contained higher levels in the cortex than the nucleus. Linear regression lines shown, nucleus (r = −0.47, P = 0.002, n = 38) and cortex (r = −0.68, P < 0.0005, n = 38). Paired comparison t-test, nucleus/cortex (P < 0.0005). —•—, nucleus; - -○- -, cortex.

Figure 9.

 

Proposed interaction of UV filters and glutathione and their role in the modification of crystallins in the human lens. Kyn, 3OHKyn, and 3OHKG are all formed in the cortex from l-tryptophan. 3OHKG diffuses into the nucleus and undergoes deamination to form a reactive intermediate (α,β-ketoalkene). This intermediate can either reduce to form AHBG, form an adduct with GSH (GSH-3OHKG), or covalently bind to the crystallins. Similar reactions may occur with Kyn and 3OHKyn with the additional feature that 3OHKyn may also oxidize and bind to crystallins. The lens is permeable to all tryptophan metabolites and GSH-containing species, thus necessitating continuous synthesis of the UV filters and GSH as they diffuse from the lens.