Familial Alzheimer’s disease mutations in amyloid protein precursor alter proteolysis by γ-secretase to increase amyloid β-peptides of ≥45 residues

Amyloid β-protein (Aβ) is produced by the membrane-embedded γ-secretase complex. Early onset familial Alzheimer’s disease (FAD) is associated with mutations in the transmembrane domain of amyloid β-protein precursor (APP) as it alters the ratio of aggregation-prone 42-residue Aβ (Aβ42) to 40-residue Aβ (Aβ40). γ-secretase progressively proteolyzes the APP substrate along two pathways. However, due to difficulties in detecting and quantifying longer Aβ peptides, the effects of FAD mutations on each proteolytic step are unknown. The pathogenic Aβ variant in Alzheimer’s disease has been accepted as Aβ42. This specific particular Aβ variant is the main component of the cerebral extraneuronal amyloid plaques that are a defining pathological feature of Alzheimer’s disease. In addition, the ratio of Aβ42 to Aβ40 can be elevated by the FAD mutations in APP and the presenilins. However, it has been difficult to find irrefutable identification of pathogenic assemblies of aggregation-prone Aβ42 and explanation of neurotoxic mechanisms. Recent studies have demonstrated that many presenilin-1 FAD mutations do not elevate Aβ42/Aβ40 and this has raised difficulty on understanding Aβ42 as a required initiator of pathogenesis. A comprehensive and quantitative analysis of all proteolytic steps is required to understand how FAD mutations alter APP processing by γ-secretase but to date this has not been performed for any FAD mutations.

To address this, University of Kansas researchers Dr. Sujan Devkota, Dr. Todd Williams and Professor Michael Wolfe examined the levels of each Aβ peptide produced by purified γ-secretase from wild-type and 14 different FAD-mutant APP substrates. Their results showed that all the 14 disease-causing-mutations demonstrated an elevation of Aβ peptides of 45- to 49-residues in length.  The original research work is now published in the Journal of Biological Chemistry.

The research team noticed that approximately tenfold more Aβ40 and Aβ42 peptides were produced by the detergent-solubilized condition compared to when reconstituted in proteoliposomes. The 14 FAD-mutant and wild-type substrates had similar relative effects on both Aβ40 and Aβ42 under the two prepared conditions (detergent-solubilized and reconstitution in proteoliposomes). In addition, for all the substrate variants between the two different conditions, the ratios of Aβ42 to Aβ40 were also similar.

FAD mutations behave in a similar fashion as ELISA on Aβ40 and Aβ42 and Aβ42/Aβ40. Of the 14 mutations examined in the study, five had the highest elevation of Aβ42, while two did not increase Aβ42/Aβ40. For the wild-type substrate, the authors found that in all except two mutant substrates, the sum of Aβ peptides was nearly equal to the total AICD level, and the sum of Aβ peptides along the Aβ49→Aβ40 or Aβ48→Aβ38 pathway was comparable to levels of the respective AICD isoforms. A less efficient trimming of Aβ peptides ranging from 45 to 49 residues in length was noticed to be caused by all the 14 different FAD mutations in the APP transmembrane domain.

In summary, through this study the authors have identified that long Aβ peptides are potentially pathogenic, especially in FAD. They speculate that these peptides affect FAD progression through their formation of oligomers and membrane pore or lateral diffusion and interaction with other membrane proteins. However, the authors believes that the pathogenesis of FAD is closely linked to the inability of γ-secretase to effectively process Aβ peptides of 45 residues.