Commentary: Aβ N- Terminal Isoforms: Critical contributors in the course of AD pathophysiology

Article type: Research Article

Authors: Tekirian, Tina L.

Affiliations: Genetics & Aging Unit & Department of Neurology, Massachusetts General Hospital & Harvard Medical School, Building 149, 13th Street, Room 6133, Charlestown, MA 02129, USA. E-mail: [email protected]

Abstract: The assessment of protein or amino acid variations across evolution allows one to glean divergent features of disease-specific pathology. Within the Alzheimer's disease (AD) literature, extensive differences in Aβ processing across cell lines and evolution have clearly been observed. In the recent past, increased levels of amyloid β Aβ1-42 have been heralded to be what distinguishes whether one is prone to the development of AD [59]. However, observations in naturally occurring, non-transgenic animals which display a great deal of parenchymal Aβ1-42 (Aβ found within extracellular plaque deposits) and a complete lack of β1-40 within these same Aβ1-42 plaques raise the issue of whether Aβx-42 (Aβ that is truncated or modified at the N- terminus), rather than Aβ1-42, is instead the critical mediator of Aβ production and pathogenesis [47,49]. Distinct ratios of Aβ N-terminal variants (i.e. Aβ1-x, Aβ3-x, Aβ11-x, β17-x) have been assessed in human amyloid plaques [18,21,40,41,42,47,48,49,52]. Moreover, ratios of specific Aβ N-terminal variants separate naturally occurring, non-transgenic animals which develop abundant levels of Aβx-42 and not Aβx-40 from human AD participants who harbor plaques that contain both the Aβx-42 and Aβx-40 variants [49]. Next, Teller and colleagues have demonstrated the presence of N-terminal truncated soluble 3kD (likely Aβ17-x) and 3.7kD peptides (in addition to 4kD Aβ) well before the appearance of amyloid plaques in Down Syndrome brain [51], indicating an early contribution of the β N-terminus to the formation of amyloid pathology. Additional critical facts concerning the major contribution of the Aβ N-terminus in AD pathogenesis include observations which support thatβ generated by rodent neurons is predominantly truncated at Aβ11-x [13], the major form of APP C-terminal fragments in mice lacking functional PS1 is AβPP11-98 [9], β11-x expression is increased as a function of BACE expression [55], and an interrelationship between presenilin-1 mutations and increased levels of N-terminally truncatedβ [40]. This commentary highlights current understanding and potential biochemical, pathological, and cell biological contributions of Aβ N-terminal variants implicated during the course of AD pathogenesis. Although the amyloid β protein precursor (AβPP) gene and Aβ are highly conserved across mammalian species, there are species-specific differences. For instance, the primate, guinea pig, canine, and polar bear share an identical Aβ sequence to that observed in human brain while the rat displays a distinct amino acid sequence with substitutions at residues 5 (Arg), 10 (Tyr), and 13 (His) [24,37]. All of these mammals generate Aβ1-42 via cleavage by at least two enzymes, beta (β-) secretase and gamma (γ-) secretase (Fig. 1). The enzyme that liberates the N- terminus of the Aβ peptide (`β-secretase') is also termed BACE (beta-site AβPP cleaving enzyme) [55]. Cathepsin D, which accumulates within AD neurons [15], also cleaves at the N-terminal side of the first aspartate residue of amyloid β [2]. β-secretase activity is necessary in order to initiate 4kD β1-x formation by cleaving AβPP at the N-terminus and results in the release of a soluble 100kD AβPP N- terminal fragment and a 12kD membrane bound C-terminal fragment (C99/C100) [55]. The carboxyl-terminus of the Aβpeptide is liberated through cleavage by the enzyme termed γ-secretase. In the past, potential AD therapeutic strategies have mainly been geared towards gamma-secretase inhibition. However, such strategies alone no longer appear sound as it is clear that the AβPP C99/C100 fragment itself, which requires β-, but not γ-, secretase cleavage for generation and includes the entire Aβ peptide, is neurotoxic when evaluated in cultured cells [12,30,34]. Thus, γ-secretase inhibition alone would not preclude the generation of the neurotoxic C99/C100 fragment.

DOI: 10.3233/JAD-2001-3209

Journal: Journal of Alzheimer's Disease, vol. 3, no. 2, pp. 241-248, 2001