Substantial evidence indicates that both amyloid beta (Abeta) oligomerization and cyclooxygenase (COX) activity contribute to the pathogenesis of Alzheimer's disease (AD). This proposal seeks to determine the molecular basis of the interaction between these two processes. The immediate product of the cyclooxygenases is prostaglandin H2(PGH2), which rapidly rearranges in aqueous solution with approximately 20% being converted to levuglandin E2(LGE2) and LGD2. These highly reactive gamma-ketoaldehydes rapidly adduct to amine groups, notably the epsilon-amine of the lysine residues of proteins, and also can crosslink proteins. We have discovered that PGH2, via the levuglandin pathway, accelerates formation of soluble oligomers of Abeta that have the immunochemical, electrophoretic, and ultrastructural characteristics of Abeta-derived diffusible ligands, and these oligomers are neurotoxic. In order to analyze the formation of levuglandinyl adducts of proteins, we have characterized the structure of the levuglandinyl-lysine lactam adduct and developed a method for its analysis with tandem mass spectrometry. Utilizing this sensitive and specific method we have now identified levuglandinyl adducts of protein in hippocampus of brains from patients who had AD, at levels that are increased 12 fold above those found in age-matched control brains. The level of this COX-derived lipid modification of brain proteins is highly correlated with the Braak stage of severity of the Alzheimer's disease and provides tangible new evidence for participation of COX activity in the disease. This proposal addresses the characterization of proteins that are adducted by levuglandins in AD. An initial aim of the proposed research is to determine the presence of levuglandinyl adducts specifically on Abeta in AD brains. This will be addressed by the analysis of adducted amino acid residues and peptides proteolytically derived from immunoprecipitated Abeta, utilizing LC-tandem mass spectrometry for analysis. In addition, the presence of levuglandinyl adducts of proteins involved in biosynthesis of Abeta will be examined, and utilizing an unbiased proteomic analysis, the presence of levuglandinyl adducts of other proteins in AD will be determined. In summary, lipid modification of proteins provides a new paradigm for understanding the consequences of COX activity, and this proposal will characterize proteins that are modified by COX-derived lipid adducts in the brains of AD.
