Reactions of long-lived lens proteins

Abstract

The human lens contains the highest protein concentration of any tissue in the body, yet there is no protein turnover. As a result, proteins found in the centre of the lens (the nucleus) are present for a lifetime. This tissue can therefore be used to examine major posttranslational events that take place in long-lived proteins. Age-dependent deterioration of long-lived proteins in humans may have wide-ranging effects on health, fitness and diseases of the elderly [1]. To a large extent, denaturation of old proteins appears to result from the intrinsic instability of certain amino acids, however these reactions are incompletely understood. In this thesis, to understand more about these reactions, the breakdown of peptides was studied under controlled conditions, typically at physiologically relevant pHs and with elevated temperatures used to promote the reaction. Significant truncation of long-lived proteins has been shown to occur in the aged human lens. In the case of α-crystallin, one notable feature of the sequences of two of the most abundant truncations (αA 67-80 and αB 1-18) was that sites of cleavage were adjacent to Ser residues. While the truncation of proteins at Asp/Asn residues via the formation of a succinimide ring has been well characterised, our understanding of the processes that enable truncation at Ser is incomplete. The first part of this thesis aimed at understanding the mechanisms behind this truncation. A secondary aim was to understand the mechanism behind the age-related racemisation of Ser residues seen in the lens, and determine if it occurs via a mechanism analogous to that seen with Asp residues. Model peptides based on the sequence of known Ser truncation sites in human α-crystallin were exposed to elevated temperatures at physiological pH. Non-enzymatic truncation at the N-terminus of Ser, similar to that seen in the aged-lens was demonstrated. A range of additional factors were also examined for their ability to promote truncation. The role of the Ser hydroxyl group was investigated and found to play an important role in truncation at Ser. Interestingly Ser racemisation was also observed under these conditions, and it occurred regardless of the presence of a free or blocked Ser hydroxyl group. This was at odds with our initial hypothesis that both racemisation of Ser and truncation the N-terminal side of Ser might be a linked process, occurring via formation of a cyclic tetrahedral intermediate. Another possible source of Ser racemisation was investigated by examining the potential of phosphoser (pSer) residues to form dehydroalanine (DHA) via beta elimination. While DHA was generated from a model peptide at physiological pH, the results in this thesis did not find any evidence of water adding to the double bond, thus ruling it out as a possible explanation for Ser racemisation in the lens. However the observed formation of DHA at physiological pH does provide a potential explanation for the extensive non-covalent cross linking seen in aged lens proteins. The second part of this thesis examined a range of modifications that can occur at a peptide N-terminus. Approximately 70% of soluble proteins in eukaryotic cells have an acetylated α amino group. One proposed role of this is that it protects the protein from a range of N-terminal modifications. However following age-related internal truncations, such as those observed at the N-terminus of Ser and other known truncations, the resulting protein fragments (with free amino groups) could then be subject to N-terminal degradation. Using model peptides based on human lens crystallin sequences, facile racemisation of N-terminal residues incubated under physiological conditions was demonstrated. It was shown to occur across a range of N-terminal residues, buffers and temperatures. Unexpectedly the racemisation rate of L-residues was almost twice that of D-residues but the reasons for this are as yet unclear. A novel mechanism to explain these findings, involving the formation of a Schiff base has been proposed. Racemisation of the N-terminal residue was also shown to render the peptides resistant to amino peptidase degradation, suggesting a protective role could be provided by this modification. The prevalence of this modification in the humans lens was then demonstrated using the integral membrane protein Aquaporin 0 (AQP0). By the age of 68, 13% of the N-terminal Met residue of AQP0 had racemised in the cortex, increasing to 28% in the nucleus which is the oldest part of the lens. Other N-terminal modifications studied included the degradation of model peptides via sequential loss of their N-terminal residues. Again it was shown to occur at physiological conditions using model peptides based on crystallin sequences, however further investigation revealed that it was more prominent in phosphate buffer. One potential mechanism involves phosphate buffer acting as a nucleophile. It is worth noting that phosphate buffer is present in the human lens and that this may still be a biologically relevant degradation process. Aged proteins demonstrating sequential loss of amino acid residues or “laddering” have been described in the literature [2]. The potential for proteins to degrade via loss of two amino acids at a time through a cyclic diketopiperazine (dkp) from the N-terminus was also investigated. A range of factors were considered and it was shown to be another potential degradation pathway for long-lived proteins which have a free amino termini. Peptides that have a penultimate Pro residue were particularly prone, with significant dkp formation occurring even at physiological temperatures. For peptides without a penultimate Pro, dkp formation was still observed for a range of peptides, but elevated temperatures were required. The final part of this thesis detailed the isolation and characterisation of a novel UV filter found in the lens of the thirteen-lined ground squirrel. The structure of this UV filter is of interest due to the lens of the thirteen-ground squirrel having a similar UV filter profile to that of the human lens. Characterisation by mass spectrometry and NMR spectroscopy revealed the likely structure to be an N-acetylated 3OH Kynurenine adduct with the incorporation of a proline. A potential structure has been proposed involving the formation of an imine bond between the proline and the 3OH Kynurenine.