Anatomy and Neuroscience - Theses
Permanent URI for this collection
Search Results
1 - 1 of 1
-
ItemPyroglutamate-Aß in the pathogenesis of Alzheimer's disease( 2013)Background: Amyloid-beta (Aβ) peptides are central contributors to Alzheimer’s disease (AD) pathogenesis. Although Aβ peptides are present in all human brains, the AD brain is particularly enriched for oligomeric Aβ species and Aβ peptides containing post-translational modifications such as oxidation, amino-truncation and pyroglutamate (pE). Metal interactions are a critical aspect of Aβ-induced neurotoxicity, however the effects of pE formation on Aβ-metal reactions such as nucleated oligomerisation, redox cycling and the production of reactive oxygen species (ROS) have not been investigated. Recent reports have indicated that pE-Aβ peptides are more neurotoxic than full-length Aβ, although a mechanistic difference in the toxic properties of these peptides has yet to be established. Increased levels of glutaminyl cyclase (QC) are thought to be responsible for the abundance of pE-Aβ in the AD brain, via the cyclisation of exposed N-terminal glutamate to pyroglutamate on amino-truncated Aβ. The relative levels of QC protein and mRNA are reported to be elevated in the temporal cortex and peripheral blood of individuals with AD compared to healthy controls. However, there are no published values of QC enzymatic activity in human central nervous system tissues. The involvement of QC in pE-Aβ formation and AD pathogenesis has led to the recent generation of QC inhibitors as a potential therapeutic intervention for AD. For this reason, there is a critical need to establish standardised levels of QC protein and activity in populations of healthy individuals and people with AD. Furthermore, there is a lack of animal models of pE-Aβ expression, thus the generation of simple pE-Aβ expression models may facilitate the study of potential QC inhibitors as an AD therapeutic. Objectives: I aimed to assess potential differences between synthetic pE-Aβ and full-length Aβ peptides in terms of their oligomerisation rate, fibril ultrastructure, cellular life-span and neurotoxicity. I also sought to compare the Aβ variants for their capacity to undergo nucleated polymerisation in the presence of Cu2+ or Zn2+, in addition to the generation of ROS and oxidative modifications such as dityrosine via redox cycling reactions with Cu2+ and ascorbate. To determine whether changes in soluble QC (sQC) expression and activity are a feature of AD pathogenesis, I aimed to establish standardised ranges of sQC protein and activity in the human brain through analysis of post-mortem cortical tissue samples from a cohort of AD and control brains. Finally, I sought to generate a Caenorhabditis elegans nematode model of pE-Aβ expression for in vivo comparisons of Aβ variant cytotoxicity and cellular interactions. Results: Vastly different rates of fibrilisation and fibril ultrastructures were observed for amino-truncated and pE-Aβ peptides compared with full-length peptides. Amino-truncated Aβ showed accelerated fibril seeding compared to full-length Aβ, while further addition of Cu2+ inhibited fibrilisation and produced aggregates of different ultrastructures between the seeded mixtures. In contrast, Zn2+ promoted fibrilisation but was also found to rapidly and reversibly aggregate Aβ peptides in short incubation periods. Redox-cycling reactions of Aβ, Cu2+ and ascorbate demonstrated significant differences between full-length Aβ and pE-Aβ peptides in the profiles of oligomers produced as well as the rate of hydroxyl radical production and dityrosine formation. The reaction of Aβ1-40 with Cu2+ and ascorbate was further found to cause amide-bond hydrolysis and the formation of amino-truncated Aβ peptides. Both the Aβ1-42 and Aβ3pE-42 peptides were toxic to cortical neurons and inhibited hippocampal long-term potentiation, however methodological differences in the preparation of peptides were found to significantly alter the relative Aβ neurotoxicity. Aβ1-42 was the only peptide to significantly increase neuronal ROS levels, suggesting that the toxicity observed for Aβ3pE-42 was ROS-independent. The levels of Aβ3pE-42 were much higher than Aβ1-42 following 48 h treatment of the peptides on cortical neurons, indicating that Aβ3pE-42 is highly resistant to proteolysis in neurons. Mean levels of sQC protein were modestly, though significantly, elevated in the frontal cortex of individuals with AD compared with healthy controls. No significant difference in the mean levels of total sQC activity or specific activity were observed between AD and control subjects. Gel electrophoresis and mass-spectrometry analyses of a C. elegans strain designed to express Aβ1-42 (CL2120) unexpectedly revealed that the predominant peptide expressed was actually Aβ3-42 – a precursor substrate for pE-Aβ formation. Genetic manipulation of this strain to co-express human sQC resulted in production of an additional Aβ species in these worms with hydrophobic properties consistent with Aβ3pE-42. Conclusions: Previous studies have established that amino-truncation and pE formation greatly enhance the oligomerisation and fibrilisation of Aβ peptides. The data presented here demonstrate that these modifications also affect the capacity of Aβ to undergo facile redox cycling with Cu2+, thus altering the relative production of cytotoxic ROS and oxidative protein modifications such as dityrosine. Aβ3pE-42 showed either comparable or enhanced toxicity to cortical neurons compared with Aβ1-42, although different peptide dissolution methods were seen to skew the relative toxicity of each peptide. Aβ1-42 greatly increased cytosolic ROS in neurons, whereas pE-Aβ peptides did not, suggesting that pE-Aβ induced neurotoxicity is ROS-independent. Furthermore, pyroglutamate formation renders Aβ resistant to proteolysis in neurons, indicating that pE-Aβ peptides are both neurotoxic and biologically persistent. The reported abundance of pE-Aβ in AD brains is not however a function of increased QC activity in the frontal cortex, suggesting that the rate of pE-Aβ formation is either due to regional-specific changes in QC activity or increased production of amino-truncated Aβ precursors, or both. The process of Aβ amino-truncation in vivo may also be due to multiple processes such as aminopeptidase activity and the interactions of Aβ with Cu2+. C. elegans models of pE-Aβ expression may facilitate further studies into the biological properties of these amyloidogenic peptides and the screening of potential therapeutics to inhibit their formation.