Asp 58 modulates lens αA-crystallin oligomer formation and chaperone function.

Senile cataract onset is caused by insolubilization of lens proteins. The lens crystallin protein family correctly orders the formation of homo- or hetero-oligomers in lens fiber cells. Because lens fiber cells do not divide, covalent post-translational modifications, such as isomerization of aspartate residues, accumulate with aging. Although many isomerization sites of αA-crystallin have been reported, their structural and functional contributions have never been identified. In this study, αA-crystallin was extracted from aged human lens and separated into each oligomeric state by size exclusion chromatography and electrophoresis. The novel combination methodology of in-solution/gel tryptic digestion with liquid chromatography equipped with mass spectrometry (LC-MS/MS) was used to evaluate the isomerization of Asp 58. The contributions of isomerization to assembly, solubility, and chaperone functions of αA-crystallin were estimated using a series of mutations of Asp 58 in αA-crystallin. The results indicated that the isomerization of Asp 58 depended on the oligomer size and age of the lens. The substitution of Asp 58 for hydrophobic residues increased αA-crystallin oligomer size and decreased solubility. All substitutions decreased the chaperone function of αA-crystallin for aggregates of bovine ßL-crystallin and alcohol dehydrogenase. The data indicated that Asp 58 in αA-crystallin was critical for intermolecular interactions in the lens. Our results also suggested that LC-MS/MS-based isomerization analyses of in-gel-digested products could be useful for investigating the isomerization of Asp residues in oligomeric states. This method could also be used to analyze d/l ratios of amino acid residues in soluble protein aggregates.

Evidence of Highly Conserved ß-Crystallin Disulfidome that Can be Mimicked by In Vitro Oxidation in Age-related Human Cataract and Glutathione Depleted Mouse Lens.

Low glutathione levels are associated with crystallin oxidation in age-related nuclear cataract. To understand the role of cysteine residue oxidation, we used the novel approach of comparing human cataracts with glutathione-depleted LEGSKO mouse lenses for intra- versus intermolecular disulfide crosslinks using 2D-PAGE and proteomics, and then systematically identified in vivo and in vitro all disulfide forming sites using ICAT labeling method coupled with proteomics. Crystallins rich in intramolecular disulfides were abundant at young age in human and WT mouse lens but shifted to multimeric intermolecular disulfides at older age. The shift was ∼4x accelerated in LEGSKO lens. Most cysteine disulfides in ß-crystallins (except ßA4 in human) were highly conserved in mouse and human and could be generated by oxidation with H(2)O(2), whereas γ-crystallin oxidation selectively affected γC23/42/79/80/154, γD42/33, and γS83/115/130 in human cataracts, and γB79/80/110, γD19/109, γF19/79, γE19, γS83/130, and γN26/128 in mouse. Analysis based on available crystal structure suggests that conformational changes are needed to expose Cys42, Cys79/80, Cys154 in γC; Cys42, Cys33 in γD, and Cys83, Cys115, and Cys130 in γS. In conclusion, the ß-crystallin disulfidome is highly conserved in age-related nuclear cataract and LEGSKO mouse, and reproducible by in vitro oxidation, whereas some of the disulfide formation sites in γ-crystallins necessitate prior conformational changes. Overall, the LEGSKO mouse model is closely reminiscent of age-related nuclear cataract.

Disability for function: loss of Ca(2+)-binding is obligatory for fitness of mammalian ßγ-crystallins.

Vertebrate ßγ-crystallins belonging to the ßγ-crystallin superfamily lack functional Ca(2+)-binding sites, while their microbial homologues do not; for example, three out of four sites in lens γ-crystallins are disabled. Such loss of Ca(2+)-binding function in non-lens ßγ-crystallins from mammals (e.g., AIM1 and Crybg3) raises the possibility of a trade-off in the evolutionary extinction of Ca(2+)-binding. We test this hypothesis by reconstructing ancestral Ca(2+)-binding motifs (transforming disabled motifs into the canonical ones) in the lens γB-crystallin by introducing minimal sets of mutations. Upon incorporation of serine at the fifth position in the N/D-N/D-X-X-S/T(5)-S motif, which endowed a domain with microbial characteristics, a decreased domain stability was observed. Ca(2+) further destabilized the N-terminal domain (NTD) and its serine mutants profoundly, while the incorporation of a C-terminal domain (CTD) nullified this destabilization. On the other hand, Ca(2+)-induced destabilization of the CTD was not rescued by the introduction of an NTD. Of note, only one out of four sites is functional in the NTD of γB-crystallins responsible for weak Ca(2+) binding, but the deleterious effects of Ca(2+) are overcome by introduction of a CTD. The rationale for the onset of cataracts by certain mutations, such as R77S, which have not been clarified by structural means, could be explained by this work. The findings presented here shed light on the evolutionary innovations in terms of the functional loss of Ca(2+)-binding and acquisition of a bilobed domain, besides imparting additional advantages (e.g., protection from light) required for specialized functions.

Comparative analysis of crystallins and lipids from the lens of Antarctic toothfish and cow.

Animal model systems of senile cataract and lens crystallin stability are essential to understand the complex nature of lens transparency. Our aim in this study was to assess the long-lived Antarctic toothfish Dissostichus mawsoni (Norman) as a model system to understand long-term lens clarity in terms of solubility changes that occur to crystallins. We compared the toothfish with the mammalian model cow lens, dissecting each species' lens into a cortex and nuclear region. In addition to crystallin distribution, we also assayed fatty acid (FA) composition by negative ion electrospray ionization mass spectrometry (ESI-MS). The majority of toothfish lens crystallins from cortex (90.4%) were soluble, whereas only a third (31.8%) from the nucleus was soluble. Crystallin solubility analysis by SDS-PAGE and immunoblots revealed that relative proportions of crystallins in both soluble and urea-soluble fractions were similar within each species examined and in agreement with previous reports for bovine lens. From our data, we found that both toothfish and cow crystallins follow patterns of insolubility that mirror each animals lens composition with more γ crystallin aggregation seen in the toothfish lens nucleus than in cow. Toothfish lens lipids had a large amount of polyunsaturated fatty acids that were absent in cow resulting in an unsaturation index (I(U)) four-fold higher than that of cow. We identified a novel FA with a molecular mass of 267 mass units in the lens epithelial layer of the toothfish that accounted for well over 50% of the FA abundance. The unidentified lipid in the toothfish lens epithelia corresponds to either an odd-chain (17 carbons) FA or a furanoid. We conclude that long-lived fishes are likely good animal models of lens crystallin solubility and may model post-translational modifications and solubility changes better than short-lived animal models.