| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1From the Australian Cataract Research Foundation, University of Wollongong, Wollongong, New South Wales, Australia; and the 2Department of Chemistry, Macquarie University, Sydney, New South Wales, Australia.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
METHODS. Indian cataract lenses (n = 39) and normal lenses (n = 6) were examined by HPLC to quantify levels of UV filter compounds, the UV filter precursor amino acid tryptophan (Trp), as well as tyrosine (Tyr) and uric acid.
RESULTS. The levels of the two major primate UV filters, 3-hydroxykynurenine glucoside (3OHKG) and 4-(2-amino-3-hydroxyphenyl)-4-oxobutanoic acid glucoside (AHBG), in cataract lenses were markedly decreased compared with levels in normal lenses. By contrast, the levels of Trp were greatly increased. Mean Trp concentrations were an order of magnitude higher than in normal lenses, with 86% of dark-colored cataract lens nuclei having Trp concentrations greater than the mean level in the normal lenses. The concentrations of Tyr were also higher in cataract lenses. The levels of Kyn, however, were unchanged, and the uric acid levels were substantially lower.
CONCLUSIONS. The levels of the free UV filter compounds 3OHKG and AHBG, and also of Trp, Tyr, and uric acid were different in cataract lenses compared to normal lenses. These data suggest that the metabolism of a large proportion of patients with cataract may be substantially different than in persons with normal lenses. Although the mechanism of such metabolic defects are unknown, the authors speculate that an amino acid transporter system may be upregulated in patients with cataract. Because kynurenine levels in cataract were not significantly different from those of normal lenses, there may be a defect in the lenticular UV filter pathway at one, or both, of the steps that convert kynurenine to 3OHKG.
Age-related nuclear (ARN) cataract is characterized by opacification, tanning, and hardening of the lens. It has been proposed that tryptophan (Trp) metabolites that are present in the lens as UV filters may be involved.2 3 4 5 6 7 8 9 10 11 Some of these metabolites are indicated in Scheme 1 . Of these, two of the major UV filters, kynurenine (Kyn) and 3-hydroxykynurenine O-ß-D-glucoside (3OHKG), are known to bind in vivo to lens proteins, leading to tanned and fluorescent proteins.4 5 6 This process increases with age, and age is by far the greatest risk factor for cataract.
|
The purpose of this investigation was to determine the levels of the free UV filters 3OHKG, Kyn, and AHBG; their biosynthetic precursor Trp, in addition to tyrosine (Tyr) and uric acid, in cataract lenses from India compared with normal lenses, to determine whether there were differences and whether there was a correlation with the brunescence of the lens. This approach of profiling the low-molecular-weight compounds present in tissues or fluids of diseased states, particularly in patients with suspected inborn errors of metabolism, has traditionally been used to determine the site of the principal metabolic defect.16
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Instrumental Conditions
Reversed-phase HPLC was performed on a system (Shimadzu, Kyoto, Japan) with a photodiode array detector and controlled by computer (Class-VP software). Standard curves and separations were performed on a 250 x 4.6 mm, 10.0 µm C18 column (Microsorb; Varian, Sunnyvale, CA), fitted with a C18 guard column (Phenomenex, Ltd., Macclesfield, UK). Solvents used were water with 0.05% (vol/vol) TFA (solvent A) and 80% (vol/vol) acetonitrile and water with 0.05% (vol/vol) TFA (solvent B). The percentage of solvent B in the gradient was 0% (5 minutes), 0% to 50% (50 minutes), 50% to 0% (5 minutes), and 0% (15 minutes). The flow rate was 0.5 mL/min. Standard curves for 3OHKyn, DL-kynurenine sulfate, L-tryptophan, L-tyrosine, and uric acid were determined and analysis of the lens extracts were performed in the above conditions. Detection was at 360 nm for the UV filters and 280 nm for the other compounds.
Mass spectra were obtained on a commercial system (Q-TOF2; Micromass, Manchester, UK) equipped with a nanospray source, in positive ion mode.
UV Filter Extraction
Each lens was cored (5 mm bore) and the nucleus and cortex separated. Approximately 0.5 mm was cut from each end of the nucleus and added to the cortex fraction. All dissections were performed at 0°C. The average weights of the nucleus and cortex for each lens type were as follows: light-colored cortex, 45 ± 34 mg; light-colored nucleus, 22 ± 4 mg; dark-colored cortex, 50 ± 32 mg; dark-colored nucleus, 24 ± 7 mg; normal cortex, 154 ± 25 mg; and normal nucleus, 49 ± 10 mg.
Immediately after dissection, the lens fractions were weighed and homogenized in 100% (vol/vol) ethanol (300 µL). The homogenate was stored at –20°C for 1 hour and then centrifuged (14,000g, 10°C, 15 minutes). The supernatant was removed and stored at –20°C, and the pellet was re-extracted with 80% (vol/vol) ethanol (300 µL). The homogenate was stored at –20°C for 1 hour and then centrifuged as just described, and the supernatants were combined and dried.17 Each sample was analyzed by HPLC, and the relevant peaks were collected and analyzed by tandem mass spectrometry. The average recovery of compounds from the lenses was measured by adding 250 picomoles Kyn in 100 µL ethanol to one half of both a normal and a cataract nucleus and then extracting and analyzing the lens as just described. The recoveries obtained were 97% and 73%, respectively.
Amino Acid Analysis
Amino acid analysis for Trp was performed by the Australian Proteome Analysis Facility (APAF; Sydney), using the a precolumn derivatization method (AccQ-Tag; Waters, Milford, MA).
Statistical Analysis
Statistical analysis was performed with a one-way ANOVA t-test, to compare the data points obtained from cataract and normal lenses for significance. P < 0.05 was considered significant.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Normal lenses (n = 6) ranged in age from 51 to 64 years with an average age of 56 years. Cataract lenses (n = 39) ranged in age from 26 to 90 years, with an average age of 62 years.
UV Filters
Each lens extract was analyzed by HPLC. Representative elution profiles of light- and dark-colored cataract lens ethanol extracts are shown in Figures 1A and 1B . Peak 1 (23 to 24 minutes) corresponded to 3OHKG, peak 2 (27 to 28 minutes) to Kyn, and peak 3 (32 to 34 minutes) to AHBG. The standard curve for 3OHKyn was used to determine the concentration of 3OHKG and AHBG, and the standard curve for Kyn was used to determine the concentration of Kyn, from the HPLC peak area at 360 nm.
|
3-Hydroxykynurenine Glucoside.
The concentration of 3OHKG was markedly lower in cataract lenses than in normal lenses (Table 1) . This was true of both light- and dark-colored lenses. As has been found in a previous study,12 there was little difference between the average values in the cortical and nuclear extracts of normal lenses (cortex, 396 pmol/mg; nucleus, 423 pmol/mg). Light- and dark-colored cataract lenses also had similar average levels of 3OHKG, both being approximately five times lower than normal lenses, with only a small variance between the cortex and nucleus (light-colored cortex, 71 pmol/mg; light-colored nucleus, 86 pmol/mg; dark-colored cortex, 75 pmol/mg; dark-colored nucleus, 78 pmol/mg). Cataract lenses had lower levels of 3OHKG than did normal lenses, and the difference was statistically significant (dark-colored nucleus, P < 0.0001; dark-colored cortex, P = 0.0005; light-colored nucleus, P < 0.0001; light-colored cortex, P < 0.0001).
|
Kynurenine.
As shown in Table 1 , the average concentration of Kyn in cataract lenses was slightly higher than that in normal lenses, but the magnitude of the difference was small compared with 3OHKG. In normal lenses, there was a slight increase noted in the Kyn levels in the nucleus (15 pmol/mg average) compared with the cortex (10 pmol/mg average). There was less difference between the average levels in the cortical and nuclear fractions of cataract lenses, with light-colored (cortex, 13 pmol/mg; nucleus, 17 pmol/mg) and dark-colored (cortex, 15 pmol/mg; nucleus, 17 pmol/mg) lenses having comparable quantities of this UV filter. Statistically, the levels in cataract compared with normal lenses were not significant (dark-colored nucleus, P = 0.7749; dark-colored cortex, P = 0.5148; light-colored nucleus, P = 0.4240; light-colored cortex, P = 0.3834).
There appeared to be no correlation of Kyn concentration with age in cataract lenses (data not shown).
AHBG.
Table 1 demonstrates that the average levels of AHBG in cataract lenses were approximately four to five times lower than those in normal lenses. This difference in levels was similar to its metabolic precursor 3OHKG. As expected in the normal lenses,12 the concentration of AHBG in the nucleus was higher than in the cortex (cortex, 105 pmol/mg average; nucleus, 203 pmol/mg average). In the case of cataract lenses, the average values for cortical and nuclear regions were similar for both light-colored (cortex, 37 pmol/mg; nucleus, 39 pmol/mg) and dark-colored (cortex, 35 pmol/mg; nucleus, 34 pmol/mg) lenses, with the dark-colored lenses having a slightly lower amount than the light-colored lenses. The levels in cataract compared with normal lenses were all lower, and the difference was statistically significant (dark-colored nucleus, P < 0.0001; dark-colored cortex, P = 0.0505; light-colored nucleus, P < 0.0001; light-colored cortex, P = 0.0004).
There appeared to be no relationship between the age of the lens and AHBG levels in cataract lenses (data not shown).
Tryptophan, Tyrosine, and Uric Acid
The lenses were also analyzed for the low-molecular-weight compounds Trp, Tyr, and uric acid. These elute as shown in Figure 2 , in which peak 1 (12 minutes) is uric acid, peak 2 (23 to 24 minutes) is Tyr, and peak 3 (36 minutes) corresponds to Trp. The corresponding standard curve for each compound was used to determine the concentration of each from the peak area based on absorbance at 280 nm.
|
|
Tyrosine.
As depicted in Table 2 , normal lenses had, on average, a lower concentration of Tyr than cataract lenses, although there was considerable scatter in the data. The average levels in the cortical and nuclear regions of the normal lenses were similar (cortex, 367 pmol/mg; nucleus, 381 pmol/mg). Both light- and dark-colored lenses appeared to have comparable average concentrations, although there was a difference in the light-colored lenses between the cortex and nucleus (cortex, 806 pmol/mg; nucleus, 507 pmol/mg). There was little difference between the two regions in dark-colored lenses (cortex, 704 pmol/mg average; nucleus, 711 pmol/mg average). A comparison of cataract lens levels with normal lens levels, showed that the higher Tyr levels in cataract lenses were not statistically significant (dark-colored nucleus, P = 0.3616; dark-colored cortex, P = 0.5151; light-colored nucleus, P = 0.5935; light-colored cortex, P = 0.4136).
Again, no correlation was observed between the age of the lens and the quantities of Tyr in cataract lenses (data not shown).
Tryptophan.
The levels of Trp detected in normal lenses were much lower than those in most cataract lenses (Table 2) . The mean levels in cataract lenses were an order of magnitude higher than those in normal lenses. There was remarkably little variation in the concentrations of Trp in normal lenses with the levels detected corresponding to those described previously by Bessems and Hoenders.19 By contrast, the concentrations of Trp in cataract lenses varied markedly. Seventy percent of the dark-colored and 58% of the light-colored cataract lenses contained levels of Trp greater than the mean levels in the normal lenses.
To confirm that in fact only Trp was in the HPLC peak and that other components had not interfered with the assay, the Trp peak from HPLC of one lens extract was collected and sent for amino acid analysis (AAA). This confirmed that our value for Trp was correct (HPLC, 6.9 nanomoles; AAA, 8.3 nanomoles). Both dark-colored (cortex, 882 pmol/mg; nucleus, 942 pmol/mg) and light-colored (cortex, 1203 pmol/mg; nucleus, 896 pmol/mg) lenses had similar average Trp concentrations. As can be seen in these data, the cortices and nuclei of dark-colored lenses had similar levels of Trp; however, in the light-colored lenses the cortices had slightly increased levels in comparison to the corresponding nuclei. Although the majority of cataract lenses had higher Trp concentrations than normal lenses (cortex, 63 pmol/mg; nucleus, 65 pmol/mg), the differences were not statistically significant (dark-colored nucleus, P = 0.4492; dark-colored cortex, P = 0.4721; light-colored nucleus, P = 0.3629; light-colored cortex, P = 0.3023). These probabilities, however, reflect the wide variation of Trp levels throughout cataract lenses, as shown by the SEM. The SEM was very low in normal lenses, indicating that the level of Trp in normal lenses is tightly regulated.
There appeared to be no link between the ages of the lenses and the levels of Trp in the lens (Fig. 3) .
|
GSH-3OHKG concentrations were also examined in all the cataract lenses studied, as this compound is known to be in normal lenses.20 Levels were below the detection limit, however.
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This investigation into the low-molecular-weight compounds present in cataract lenses revealed some remarkable findings. Two major primate UV filters, 3OHKG and AHBG, were found to be present in markedly reduced levels compared with levels in normal lenses. For example, 3OHKG was detected in cataract lenses on average at approximately one sixth of the concentration in normal lenses. This factor alone would render the eyes of such patients more susceptible to UV damage. Although UV filter bound proteins also contribute to the absorption of UV light,21 recent results in studies by our group show that these bound proteins, when subjected to UV light, are susceptible to photosensitization. This appears to accelerate photooxidative damage of the bound protein, thus reducing protection from UV light.
In normal lenses, 3OHKG is known to be the major free UV filter,2 22 23 24 25 26 followed by AHBG and Kyn.17 24 These were also the major free UV filter compounds present in cataract lenses, although the relative levels of the individual constituents were altered significantly. Our results with cataract lenses show that whereas the Kyn levels were comparable to those in normal lenses, 3OHKG and AHBG concentrations were markedly decreased. These low levels in the cataract lenses were also lower than those in normal lenses of similar ages in the comprehensive study by Bova et al.12 Because Trp levels in the cataract lenses are elevated in both nuclear and cortical regions, it is clear that the reason for the markedly lower concentrations of 3OHKG and AHBG is not due to a general loss of small metabolites from the lens.
An increase in protein binding could also be indicated by these findings; however, conjugation to protein represents only a small proportion of the total amount of UV filters that are synthesized by the lens each day.4 6 The total amount bound presumably reflects a slow accumulation of UV filters and, on this basis, would not be expected to make a large difference in the measured free UV filter concentrations in the lens. In addition, studies within our group27 have shown that the levels of protein-bound Kyn in cataract lenses, particularly those that are at an advanced stage, are lower than in normal lenses. Therefore, our finding of a marked disparity between the free levels of Kyn and 3OHKG, which are known to bind to proteins, and also the levels of AHBG, which does not deaminate and cannot therefore bind to lens proteins, were greatly reduced, could indicate that the low levels of UV filters in the cataract lenses in this study are not solely the result of increased protein binding.
Given that Trp is the precursor for UV filter synthesis and indoleamine 2,3-dioxygenase (IDO), the first enzyme in the Trp metabolic pathway, is rate limiting,28 in a normal lens one may expect that high Trp levels would be associated with correspondingly high concentrations of the downstream UV filters. However, in the cataract lenses there was no direct relationship between the concentrations of Trp and 3OHKG in individual lenses (data not shown). To illustrate this, the eight cataract lenses with Trp concentrations >1000 pmol/mg (approximately 10 times greater than in normal lenses) contained 3OHKG concentrations <50 pmol/mg (approximately eight times less than in normal lenses). The finding of significantly increased Trp, together with slightly elevated Kyn, coupled to markedly reduced 3OHKG and AHBG, points to an impairment in the normal UV filter metabolic pathway in these cataract lenses.
Takikawa et al.28 showed no noticeable decrease in IDO activity with age in normal lenses. Studies have shown that the IDO enzyme is inhibited by H2O2 29 30 and, as it is thought that H2O2 is elevated in cataract lenses,31 this inhibition could cause a decrease in the activity of IDO. IDO activity in the cataract lenses, however, was not determined, since the lenses were collected in India and stored frozen for some time. Therefore, it is feasible that IDO activity could be different in these lenses.
Two further enzymatic steps in the pathway that could also be involved are the hydroxylation of Kyn, catalyzed by kynurenine 3-monooxygenase (EC 1.14.13.9), and the glucosylation of 3OHKyn by uridine diphosphate (UDP) glucosyl transferase (EC 2.4.1.15). Because 3OHKyn is present at low levels in lenses and is also highly susceptible to oxidation, we did not think that measurement of 3OHKyn in cataract lenses that had been stored for some time would be meaningful. Both the hydroxylation and the glucosylation steps require cellular organelles.11 It should be noted that aromatic hydroxylation is a mitochondrial reaction,32 that the glucosylation is thought to involve microsomes, and that these can only occur in the outer regions of the lens where these organelles are present. Therefore, at this stage, we cannot properly differentiate between these two enzymatic steps. As noted, the hydroxylation is a mitochondrial reaction and so impairment at this point in the pathway could be indicative of a more general metabolic dysfunction.
In contrast, if the glucosylation reaction does not operate properly in the presence of cataract, one would expect that the levels of 3OHKyn would be elevated in cataract lenses. This would have major untoward consequences for the lens and would promote oxidation.33 Oxidation of proteins is the characteristic feature of ARN cataract. Assays of enzyme activities in fresh lenses extracted from patients are the only way to confirm the site(s) of the supposed metabolic problem.
It has been reported that cataract lenses contain a novel UV filter, xanthurenic acid glucoside (XA8OG).13 14 We therefore looked for the presence of this compound in all the lenses we analyzed. In all but one of the cataract lenses, we did not detect XA8OG. It was found in very low levels in one dark-colored lens, and its identity was confirmed by tandem mass spectrometry. It should be recognized that this novel UV filter can form artifactually quite readily. For example, if a lens extract containing 3OHKG is left to stand, deamination followed by cyclization and oxidation can occur, because of the inherent instability of 3OHKG, thus leading to XA8OG.9 34
The other striking finding from this study was that the levels of Trp, and to a lesser extent Tyr, were considerably elevated in most of the cataract lenses. The reason for this is not clear, but it was true of both light- and dark-colored lenses. When we consulted the literature for precedents, we discovered a comprehensive study performed in 1968 by G. Winston Barber.15 In his investigation of free amino acid concentrations in cataract and normal lenses, he reported a dramatic increase in the lens levels of proteogenic amino acids, although he did not measure Trp. In his study, approximately 60% of patients with cataract had lens levels of the proteogenic amino acids greater than normal lenses. Barber proposed that proteolysis of lens proteins could be responsible for the increased amino acid levels. However, the relative amounts of the free proteogenic amino acids do not correspond to those present as lens proteins.
Our results demonstrate (Fig. 4) that those lenses with higher Trp concentrations also had increased Tyr concentrations and those with less Trp also contained a lower concentration of Tyr. These findings could suggest that the same phenomenon occurs with some of the other proteogenic amino acids that were measured by Barber. The increased levels that we found of Trp and Tyr and those found by Barber of other proteogenic amino acids could be caused by an increase in proteolysis in the cataract lens. However, we did not measure levels of the other amino acids. It would be difficult to confirm whether increased proteolysis is indeed the reason for these higher levels; however, we do not believe that increased proteolysis of lens proteins alone can be responsible for our observations. Free Trp levels were one fifth those of Tyr in normal lenses, but in the cataract lenses Trp levels were higher on average than levels of Tyr.
|
In our study, we found that although Tyr levels were above normal, uric acid concentrations were markedly reduced in cataract lenses. Uric acid is an antioxidant present in the lens that is more stable than ascorbate or glutathione; thus, the lower levels found in this study may give an indication of the antioxidant status of the cataract lenses. The study by Bessems and Hoenders19 used HPLC to quantify the levels of Tyr and uric acid, as well as Trp, in cataract and normal lenses. Concentrations of Trp, Tyr, and uric acid in our normal lenses were comparable to those that they found. The level of Tyr found by Barber15 was also similar. In the case of uric acid, the quantities in the cataract lenses studied by Bessems and Hoenders19 were higher than those in our study, although the levels of Trp they reported in cataract lenses were lower. The reasons for this are unclear, but may be a reflection of the site of collection or the classification of cataracts used by Bessems and Hoenders.
Our cataract lenses were obtained from India, and the normal lenses from Australia. Although it would have been preferable to look at lenses from the same population, it was not possible because of the surgical procedures used for cataract surgery in Western countries. Because Barber15 obtained results similar to ours on amino acid contents and he used lenses derived from patients in the United States, it seems unlikely that our elevated amino acid levels were due to genetic or dietary factors related to their Indian origin.
The data reported from our study are an indication that there may be metabolic defects associated with most human cataracts at the level of amino acid and UV filter metabolism. It is tempting to suggest that these defects could be, in part, causative of human ARN cataract.
|
Acknowledgements |
|---|
|
Footnotes |
|---|
Submitted for publication February 19, 2004; revised May 25 and July 14, 2004; accepted July 21, 2004.
Disclosure: I.M. Streete, None; J.F. Jamie, None; R.J.W. Truscott, None
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.
Corresponding author: Roger J. W. Truscott, Australian Cataract Research Foundation, Department of Chemistry, University of Wollongong, Wollongong, NSW 2522, Australia; rjwt{at}uow.edu.au.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. J. Ortwerth, J. Bhattacharyya, and E. Shipova Tryptophan Metabolites from Young Human Lenses and the Photooxidation of Ascorbic Acid by UVA Light Invest. Ophthalmol. Vis. Sci., July 1, 2009; 50(7): 3311 - 3319. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. S. Lim, E-S. Tai, T. Aung, W. T. Tay, S. M. Saw, M. Seielstad, and T. Y. Wong Relation of Age-related Cataract With Obesity and Obesity Genes in an Asian Population Am. J. Epidemiol., May 15, 2009; 169(10): 1267 - 1274. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Mailankot, D. Smith, S. Howell, B. Wang, J. W. Jacobberger, T. Stefan, and R. H. Nagaraj Cell Cycle Arrest by Kynurenine in Lens Epithelial Cells Invest. Ophthalmol. Vis. Sci., December 1, 2008; 49(12): 5466 - 5475. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Korlimbinis, J. A. Aquilina, and R. J. W. Truscott Protein-Bound UV Filters in Normal Human Lenses: The Concentration of Bound UV Filters Equals That of Free UV Filters in the Center of Older Lenses Invest. Ophthalmol. Vis. Sci., April 1, 2007; 48(4): 1718 - 1723. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. M. Staniszewska and R. H. Nagaraj 3-Hydroxykynurenine-mediated Modification of Human Lens Proteins: STRUCTURE DETERMINATION OF A MAJOR MODIFICATION USING A MONOCLONAL ANTIBODY J. Biol. Chem., June 10, 2005; 280(23): 22154 - 22164. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
