Singlet oxygen production correlates with his and Trp destruction in brunescent cataract water-insoluble proteins

Small molecules, both dietary and endogenous, influence the onset of lens cataracts

How the lens ages successfully is a lesson in biological adaption and the emergent properties of its complement of cells and proteins. This living tissue contains some of the oldest proteins in our bodies and yet they remain functional for decades, despite exposure to UV light, to reactive oxygen species and all the other hazards to protein function. This remarkable feat is achieved by a shrewd investment in very stable proteins as lens crystallins, by providing a reservoir of ATP-independent protein chaperones unequalled by any other tissue and by an oxidation-resistant environment. In addition, glutathione, a free radical scavenger, is present in mM concentrations and the plasma membranes contain oxidation-resistant sphingolipids what compromises lens function as it ages? In this review, we examine the role of small molecules in the prevention or causation of cataracts, including those associated with diet, metabolic pathways and drug therapy (steroids).

Glycation by ascorbic acid oxidation products leads to the aggregation of lens proteins

Previous studies from this laboratory have shown that there are striking similarities between the yellow chromophores, fluorophores and modified amino acids released by proteolytic digestion from calf lens proteins ascorbylated in vitro and their counterparts isolated from aged and cataractous lens proteins. The studies reported in this communication were conducted to further investigate whether ascorbic acid-mediated modification of lens proteins could lead to the formation of lens protein aggregates capable of scattering visible light, similar to the high molecular aggregates found in aged human lenses. Ascorbic acid, but not glucose, fructose, ribose or erythrulose, caused the aggregation of calf lens proteins to proteins ranging from 2.2 x 10(6) up to 3.0 x 10(8 )Da. This compared to proteins ranging from 1.8 x 10(6) up to 3.6 x 10(8 )Da for the water-soluble (WS) proteins isolated from aged human lenses. This aggregation was likely due to the glycation of lens crystallins because [U-(14)C] ascorbate was incorporated into the aggregate fraction and because NaCNBH(3), which reduces the initial Schiff base, prevented any protein aggregation. Reactions of ascorbate with purified crystallin fractions showed little or no aggregation of alpha-crystallin, significant aggregation of beta(H)-crystallin, but rapid precipitation of purified beta(L)- and gamma-crystallin. The aggregation of lens proteins can be prevented by the binding of damaged crystallins to alpha-crystallin due to its chaperone activity. Depending upon the ratios between the components of the incubation mixtures, alpha-crystallin prevented the precipitation of the purified beta(L)- and gamma-crystallin fractions during ascorbylation. The addition of at least 20% of alpha-crystallin by weight into glycation mixtures with beta(L)-, or gamma-crystallins completely inhibited protein precipitation, and increased the amount of the high molecular weight aggregates in solution. Static and dynamic light scattering measurements of the supernatants from the ascorbic acid-modified mixtures of alpha- and beta(L)-, or gamma-crystallins showed similar molar masses (up to 10(8 )Da) and hydrodynamic diameter (up to 80( )nm). These data support the hypothesis, that if the lens reducing environment is compromised, the ascorbylation of lens crystallins can significantly change the short range interactions between different classes of crystallins leading to protein aggregation, light scattering and eventually to senile cataract formation.

Identification of the new UV filter compound cysteine-l-3-hydroxykynurenineO-β-d-glucoside in human lenses

UV filters protect the human lens and retina from UV light-induced damage. Here, we report the identification of a new UV filter, cysteine-l-3-hydroxykynurenine O-beta-d-glucoside, which is present in older normal human lenses. Its structure was confirmed by independent synthesis. It is likely this novel UV filter is formed in the lens by nucleophilic attack of cysteine on the unsaturated ketone derived from deamination of 3-hydroxykynurenine O-beta-d-glucoside. Quantitation studies revealed considerable variation in normal lens levels that may be traced to the marked instability of the cysteine adduct. The novel UV filter was not detected in advanced nuclear cataract lenses.

Age-related nuclear cataract—oxidation is the key

Age is by far the biggest risk factor for cataract, and it is sometimes assumed that cataract is simply an amplification of this aging process. This appears not to be the case, since the lens changes associated with aging and cataract are distinct. Oxidation is the hallmark of age-related nuclear (ARN) cataract. Loss of protein sulfhydryl groups, and the oxidation of methionine residues, are progressive and increase as the cataract worsens until >90% of cysteine and half the methionine residues are oxidised in the most advanced form. By contrast, there may be no significant oxidation of proteins in the centre of the lens with advancing age, even past age 80. The key factor in preventing oxidation seems to be the concentration of nuclear glutathione (GSH). Provided that nuclear GSH levels can be maintained above 2 mm, it appears that significant protein oxidation and posttranslational modification by reactive small molecules, such as ascorbate or UV filter degradation products, is not observed. Adequate coupling of the metabolically-active cortex, the source of antioxidants such as GSH, to the quiescent nucleus, is crucial especially since it would appear that the cortex remains viable in old lenses, and even possibly in ARN cataract lenses. Therefore it is vital to understand the reason for the onset of the lens barrier. This barrier, which becomes apparent in middle age, acts to impede the flow of small molecules between the cortex and the nucleus. The barrier, rather than nuclear compaction (which is not observed in human lenses), may contribute to the lowered concentration of GSH in the lens nucleus after middle age. By extending the residence time within the lens centre, the barrier also facilitates the decomposition of intrinsically unstable metabolites and may exacerbate the formation of H(2)O(2) in the nucleus. This hypothesis, which is based on the generation of reactive oxygen species and reactive molecules within the nucleus itself, shifts the focus away from theories for cataract that postulated a primary role for oxidants generated outside of the lens. Unfortunately, due to marked variability in the lenses of different species, there appears at present to be no ideal animal model system for studying human ARN cataract.

Protective mechanisms against peptide and protein peroxides generated by singlet oxygen

Reaction of certain amino acids, peptides, and proteins with singlet oxygen yields substrate-derived peroxides. Recent studies have shown that these species are formed within intact cells and can inactivate key cellular enzymes. This study examines potential mechanisms by which cells might remove or detoxify such peroxides. It is shown that catalase, horseradish peroxidase, and Cu/Zn superoxide dismutase do not react rapidly with these peroxides. Oxymyoglobin and oxyhemoglobin, but not the met (Fe3+) forms of these proteins, react with peptide but not protein, peroxides with oxidation of the heme iron. Glutathione peroxidase, in the presence of reduced glutathione (GSH) rapidly removes peptide, but not protein, peroxides, consistent with substrate size being a key factor. Protein thiols, GSH, other low-molecular-weight thiols, and the seleno-compound ebselen react, in a nonstoichiometric manner, with both peptide and protein peroxides. Cell lysate studies show that thiol consumption and peroxide removal occur in parallel; the stoichiometry of these reactions suggests that thiol groups are the major direct, or indirect, reductants for these species. Ascorbic acid and some derivatives can remove both the parent peroxides and radicals derived from them, whereas methionine and the synthetic phenolic antioxidants Probucol and BHT show little activity. These studies show that cells do not have efficient enzymatic defenses against protein peroxides, with only thiols and ascorbic acid able to remove these materials; the slow removal of these species is consistent with protein peroxides playing a role in cellular dysfunction resulting from oxidative stress.

Advanced glycation endproducts as UVA photosensitizers of tryptophan and ascorbic acid: consequences for the lens

Upon aging, the lens accumulates brown fluorophores, mainly derived from the Maillard reaction between vitamin C oxidation products and crystallins lysine residues. At the same time, the concentration of UVA filters decreases, allowing some radiation to be absorbed by lenticular advanced glycation endproducts (AGEs). This paper quantifies the photosensitizing activity of AGEs at various oxygen pressures, and compares it to that of lenticular riboflavin (RF). Solutions containing the sensitizer and the substrates tryptophan (Trp) and ascorbate (AH(-)) were irradiated at 365 nm. We show that the AGEs-photosensitized Trp oxidation rate increases with AGEs concentration and is optimal at 5% oxygen, the pressure in the lens. By contrast, for AH(-), the photooxidation rate increases with oxygen concentration. Despite the higher quantum yield of RF-depending reactions, its low concentration as compared to that of AGEs in aging lenses induces significantly higher Trp and AH(-) photodegradation rates with AGEs than with RF. As ascorbate is more rapidly photodegraded than Trp, the antioxidant competitively protects Trp from oxidation up to 1 mM, although not absolutely. We conclude that in the aging lens, AH(-) exerts a strong UVA protecting activity, but does not impede some Trp residue to be photodegraded proportionally to the AGEs concentration.

"Mitochondrial" photochemical drugs do not release toxic amounts of 1O(2) within the mitochondrial matrix space

Previously, we demonstrated that mitochondrial NAD(P)H is the primary target of singlet oxygen (1O(2)) generated by photoactivation of mitochondria-selective rhodamine derivatives. Hence, local NAD(P)H oxidation/fluorescence decrease may be used to reveal the site of intracellular 1O(2) generation. Therefore, in addition to the previously used tetramethylrhodamine methylester (TMRM), 2('),4('),5('),7(')-tetrabromorhodamine 123 bromide (TBRB) and rhodamine 123 (Rho 123), we tested here whether mitochondrial NAD(P)H of cultured hepatocytes is directly oxidized upon irradiation of different "mitochondrial" photosensitizers (Photofrin; protoporphyrin IX; Al(III) phthalocyanine chloride tetrasulfonic acid; meso-tetra(4-sulfonatophenyl)porphine dihydrochloride; Visudyne). In contrast to TMRM and Rho 123, which directly oxidized NAD(P)H upon irradiation, irradiation of intracellular TBRB and the photochemical drugs only indirectly affected mitochondrial NAD(P)H due to loss of mitochondrial integrity. In line with this result only TMRM and Rho 123 exclusively localized within the mitochondrial matrix. Due to these results it is doubtful whether real mitochondrial photosensitizers actually exist among the photochemical drugs applicable/used for photodynamic therapy.

Low-temperature photosensitized oxidation of a guanosine derivative and formation of an imidazole ring-opened product

An organic-soluble guanosine derivative, 2',3',5'-O-(tert-butyldimethylsilyl)guanosine (1), was prepared and its photosensitized oxidation was carried out in several solvents at various temperatures. Singlet oxygen is the reactive oxidizing agent responsible for this reaction. Neither an endoperoxide nor a dioxetane intermediate was detected by low-temperature NMR even at -78 degrees C. A product (A) with an oxidized imidazole ring was the only major product detected at room temperature; this compound could be isolated by low-temperature column chromatography and was characterized by (1)H and (13)C and mass spectroscopy. CO(2) was the other major product. A small amount of the corresponding 8-oxo-7,8-dihydroguanosine derivative B was detected during the initial stage of the photooxidation and was shown to be intermediate in the formation of two products of extensive degradation, C and D. Reaction of 1 with the singlet oxygen analogues 4-methyl-1,2,4-triazoline-3,5-dione (MTAD) and 4-phenyl-1,2,4-triazoline-3,5-dione (PTAD) gave products consistent with a proposed mechanism involving the rearrangement of an initially formed endoperoxide to give A and B from reaction of 1 with singlet oxygen.

Inhibition of glyceraldehyde-3-phosphate dehydrogenase by peptide and protein peroxides generated by singlet oxygen attack

Reaction of certain peptides and proteins with singlet oxygen (generated by visible light in the presence of rose bengal dye) yields long-lived peptide and protein peroxides. Incubation of these peroxides with glyceraldehyde-3-phosphate dehydrogenase, in the absence of added metal ions, results in loss of enzymatic activity. Comparative studies with a range of peroxides have shown that this inhibition is concentration, peroxide, and time dependent, with H2O2 less efficient than some peptide peroxides. Enzyme inhibition correlates with loss of both the peroxide and enzyme thiol residues, with a stoichiometry of two thiols lost per peroxide consumed. Blocking the thiol residues prevents reaction with the peroxide. This stoichiometry, the lack of metal-ion dependence, and the absence of electron paramagnetic resonance (EPR)-detectable species, is consistent with a molecular (nonradical) reaction between the active-site thiol of the enzyme and the peroxide. A number of low-molecular-mass compounds including thiols and ascorbate, but not Trolox C, can prevent inhibition by removing the initial peroxide, or species derived from it. In contrast, glutathione reductase and lactate dehydrogenase are poorly inhibited by these peroxides in the absence of added Fe2+-EDTA. The presence of this metal-ion complex enhanced the inhibition observed with these enzymes consistent with the occurrence of radical-mediated reactions. Overall, these studies demonstrate that singlet oxygen-mediated damage to an initial target protein can result in selective subsequent damage to other proteins, as evidenced by loss of enzymatic activity, via the formation and subsequent reactions of protein peroxides. These reactions may be important in the development of cellular dysfunction as a result of photo-oxidation.

Studies on singlet oxygen formation and UVA light-mediated photobleaching of the yellow chromophores in human lenses

The protein-bound chromophores, which increase with aging in the human lens, act as UVA sensitizers, producing almost exclusively singlet oxygen in vitro. Direct irradiation of whole, aged human lenses with high intensity UVA light (200 mW cm(-2) for 24 hr), however, failed to produce singlet oxygen damage, as evidenced by the lack of either His or Trp photodestruction. Total homogenates of human lenses prepared in a cuvette under air did show destruction of His and Trp residues by UVA light, but no destruction was seen when equivalent homogenates were prepared under argon. These data are consistent with the idea that the low oxygen levels in the lens prevent singlet oxygen damage in vivo.UVA irradiation of aged human lenses in culture caused an extensive photobleaching of the yellow chromophores. A time course indicated that the photobleaching increased with time, with significant color loss apparent after 6 hr. Homogenization of the irradiated and dark control lenses in 6 M guanidine-HCl, followed by determination of the difference spectrum, showed approximately 50% bleaching of compounds with a lambda(max) at 355 nm. Similarly, fluorophores with a lambda(max) for excitation of 355 nm and for emission of 420 nm were 50% destroyed by the UVA light. Similar results were obtained in vitro by the anaerobic irradiation of a sonication-solubilized WI fraction from type II brunescent cataracts and from aged human lenses. In this system, there was an initial bleaching of 15% after 30 min of irradiation, followed by a slow increase over the next 6 hr to a final bleaching of 30%. The addition of 1.0 m M ascorbic acid, but not 1.0 m M glutathione (GSH), increased the photobleaching to 60% under argon, and the loss of ascorbate could be detected under these anaerobic conditions. In the presence of air, UVA light produced no photobleaching, but rather caused a three-fold increase in absorbance at 345 nm, which was prevented by the inclusion of 1.0 m M ascorbic acid and almost 50% inhibited by 1.0 m M GSH. The data are consistent with the conversion of the triplet state of the sensitizers to anion and cation radicals in the absence of oxygen. Photobleaching may occur either by dismutation of the anion radical or by reduction of the anion radical by ascorbate via type I chemistry. UVA irradiation of an enriched fraction of sensitizers from a proteolytic digest from type II cataract lenses produced a 63% bleaching at 330 nm in the absence of oxygen, and the almost complete loss of the A(330) absorbing and 350/450 nm fluorescent peaks upon HPLC separation. This loss correlated with the loss of the ability of the irradiated fraction to produce singlet oxygen in vitro upon subsequent UVA irradiation.

Chromatographic comparison of the UVA sensitizers present in brunescent cataracts and in calf lens proteins ascorbylated in vitro

The water-insoluble (WI) fraction from aged human lenses contains yellow chromophoric sensitizers, which generate reactive oxygen species (ROS) when irradiated with UVA light. The WI proteins from type I to V brunescent cataract lenses were assayed for UVA-dependent superoxide anion synthesis. Rates varied from 8.4-15 nMol h(-1)mg protein(-1), but there was no significant difference in specific activity between cataract types. When calf lens soluble proteins were incubated with ascorbic acid for 4 weeks and dialyzed, they were capable of generating 30-40 nMol h(-1)mg protein(-1)superoxide anion when irradiated with UVA light. Two preparations each of brunescent cataract WI proteins and bovine lens proteins ascorbylated in vitro were extensively digested with proteolytic enzymes and the released amino acids separated by normal phase HPLC. The elution profiles of the digests were very similar based upon the absorbance at 330 nm and fluorescence at 350 nm excitation/450 nm emission. Each peak was pooled and analyzed for the UVA-dependent generation of both superoxide anion and singlet oxygen. Every peak exhibited sensitizer activity, and the UVA-dependent ROS generation was roughly proportional to the absorbance at 330 nm. In addition, the ratio of superoxide anion to singlet oxygen generated was similar with both preparations. These data argue that it is the brown, fluorescent compounds which accumulate during aging and cataract formation that are responsible for the UVA-dependent ROS formation, and that these browning products may be similar to the advanced glycation endproducts produced by ascorbylation of lens proteins under oxidative conditions. This work also presents an initial report of a chromatographic method to separate the UVA-sensitizers present in each of these protein preparations without the use of acid or base hydrolysis.