Studies on gamma-secretase activity and products

Abstract: gamma-Secretase is a transmembrane aspartyl protease involved in processing of a multitude of functionally diverse substrates within their transmembrane domains. gamma-Secretase cleavage of the amyloid precursor protein (APP) generates the APP intracellular domain (AICD) and the amyloid beta-peptide (Abeta). Abeta is a key player in the pathogenesis of Alzheimer disease (AD) and the main constituent of senile plaques, one of the hallmarks found in the brain of AD patients. Various lengths of Abeta peptides are produced but the most common forms are 40 or 42 residues long (Abeta40 or Abeta42). gamma-Secretase cleavage of Notch results in the release of the Notch intracellular domain (NICD) which is known to activate transcription. Notch signaling is important for cell differentiation during embryonic development as well as in adulthood. In analogy with APP and Notch, gamma-secretase cleavage of other substrates also generates secreted peptides and intracellular domains (ICDs). gamma-Secretase is a protein complex consisting of presenilin (PS), nicastrin, anterior pharynx defective-1 and presenilin enhancer-2, where PS is suggested to harbor the active site. The work of this thesis has focused on studies of gamma-secretase activity and its products in AD brain and model systems such as cell lines and rat brain. In paper I, we examined the effect of factors such as subcellular localization, pH and detergents, on gamma-secretase activity, i.e. AICD production, in rat brain membranes. A fraction containing Golgi, ER, endosomes and synaptic vesicles demonstrated the highest AICD production and the optimal pH was found to be around 7.0. In addition, the gamma-secretase activity was highly affected by detergents and CHAPSO at a concentration of 0.4% was found to enhance the activity. In paper II, the gamma-secretase dependent production of Abeta40 as well as of ICDs from APP, Notch1, N-cadherin, ephrinB and p75-neurotrophin receptor was found to be down-regulated in adult compared to embryonic rat brain membranes. Processing of all of the substrates was observed in embryonic rat brain membranes while only APP and Notch1 processing was detected in the adult rat brain membranes. In paper III, the large hydrophilic loop of PS1 was examined with respect to its effect on APP and Notch processing. Deletion of the PS1 loop resulted in a striking decrease in production of Abeta38, Abeta39, Abeta40, whereas production of Abeta42 was affected to a lesser extent, and the production of AICD and NICD was not impaired. In particular, the most C-terminal amino acids of the loop were important for the differential effect on APP processing. In paper IV, C-terminal Abeta variants deposited in sporadic and familial AD brains were identified and quantified. Both Abeta deposited in plaque cores and in total amyloid preparations was examined. Apart from Abeta40 and Abeta42, a longer Abeta species, Abeta43, was detected. Abeta43 was detected more frequently than Abeta40, especially in plaque cores. Taken together, we have optimized conditions for studies of gamma-secretase processing of APP in rat brain, demonstrated an age-dependent production of ICDs from several substrates in rat brain and showed that the most C-terminal part of the PS1 loop influences the Abeta profile. Finally, a longer Abeta variant, Abeta43 was frequently detected in amyloid depositions in AD brains.