Alzheimer's disease mutations and cellular signalling

Abstract

Alzheimer's disease is the most frequent cause of dementia in elderly and is characterised neuropathologically by the extracellular deposition of amyloid plaques (containing the 39-43 amino acid Abeta peptide), as well as intracellular neurofibrillary tangles (NFTs) composed primarily of an abnormally hyperphosphorylated form of the microtubule-associated protein tau. A proportion (10- 15%) of Alzheimer's disease cases is familial, the remainder being sporadic. Familial Alzheimer's disease is genetically heterogeneous and can be caused by defects in at least three early- onset genes located on chromosomes 21 (amyloid precursor protein, APP), 14 (presenilin 1, PS 1) and 1 (presenilin 2, PS2).

Increased production of Abeta is a common pathogenic phenotype in familial Alzheimer's disease due to APP and PS mutations. The major aim of this thesis was to determine whether the phenotypes of Alzheimer's disease causing APP and PSI mutations also include altered signalling. For this we used peripheral skin fibroblasts from sporadic as well as from familial Alzheimer's disease cases with the Swedish APP KM670/67 1 NI, mutation or different PSI mutations.

Experiments were also performed on SH-SY5Y neuroblastoma cells transfected with wild-type and mutated forms of PSI. We studied signalling pathways known to be disrupted in sporadic Alzheimer's disease and thought to be implicated in either Abeta production, tau hyperphosphorylation, intracellular calcium homeostasis or apoptosis.

The results obtained provide supportive evidence that an altered signalling occurs in Alzheimer's disease, but those groups with different disease origins (sporadic or familial) are affected in different ways. Mutations in different genes (APP vs PSI) differentially affect cellular signalling. The clearest finding was that the Swedish APP KM670/67INL mutation does not change any of the signal transduction parameters studied (beta-adrenoceptor-stimulated adenylyl cyclase, PKC activity, bradykinin-stimulated calcium release), the exception being a reduced bombesin-induced calcium release. The only abnormality previously reported in fibroblasts with this mutation is that of an increased secretion of Abeta. This effect appears to be relatively clean and is the most likely explanation for disease pathology in this family.

The PSI mutations studied in primary skin fibroblasts gave alterations in a number of signalling parameters. These included an increased beta-adrenoceptor-stimulated adenylyl cyclase activity (as compared to a decrease seen in sporadic Alzheimer's disease fibroblasts), as well as a decreased PKC-mediated APPs secretion.

We also conclude that the reported increased susceptibility of PSI mutation transfected SH-SY5Y neuroblastoma cells to undergo apoptosis after serum deprivation and high glucose stress, is unlikely due to a downregulation of Akt. Finally, the regulation of GSK-3 beta in response to the pro-apoptotic stimuli of either serum deprivation, P13 kinase inhibition or staurosporine treatment did not differ in PSI wild-type and mutation transfected SH-SY5Y cells.