Prion-like behaviour and tau-dependent cytotoxicity of pyroglutamylated amyloid-β

Extracellular plaques of amyloid-b and intraneuronal neurofibrillary tangles made from tau are the histopathological signatures of Alzheimer’sdisease.Plaques comprise amyloid-b fibrils that assemble from monomeric and oligomeric intermediates, and are prognostic indicators of Alzheimer’s disease. Despite the importance of plaques to Alzheimer’s disease, oligomers are considered to be the principal toxic forms of amyloid-b. Interestingly, many adverse responses to amyloid-b, suchas cytotoxicity,microtubule loss, impairedmemory and learning, and neuritic degeneration, are greatly amplified by tau expression. Amino-terminally truncated, pyroglutamylated (pE) forms of amyloid-b are strongly associated with Alzheimer’s disease, are more toxic than amyloid-b, residues 1–42 (Ab1–42) and Ab1–40, and have been proposed as initiators of Alzheimer’s disease pathogenesis. Here we report a mechanism by which pE-Ab may trigger Alzheimer’s disease. Ab3(pE)–42 co-oligomerizes with excess Ab1–42 to form metastable low-n oligomers (LNOs) that are structurally distinct and far more cytotoxic to cultured neurons than comparable LNOs made from Ab1–42 alone. Tau is required for cytotoxicity, and LNOs comprising 5% Ab3(pE)–42 plus 95% Ab1–42 (5% pE-Ab) seed new cytotoxic LNOs through multiple serial dilutions into Ab1–42 monomers in the absence of additional Ab3(pE)–42. LNOs isolated from human Alzheimer’s disease brain contained Ab3(pE)–42, and enhanced Ab3(pE)–42 formation in mice triggered neuron loss andgliosis at 3months,butnot in a tau-null background. We conclude that Ab3(pE)–42 confers tau-dependent neuronal death and causes template-induced misfolding of Ab1–42 into structurally distinct LNOs that propagate by a prion-likemechanism.Our results raise the possibility that Ab3(pE)–42 acts similarly at a primary step in Alzheimer’s disease pathogenesis. pE-Ab peptides contain an amino-terminal pyroglutamate, whose modification from glutamate is catalysed by glutaminyl cyclase (QC; also known asQPCT). Themost prominent pE-Ab species in vivo are Ab3(pE)–40, Ab3(pE)–42, Ab11(pE)–40 and Ab11(pE)–42 (ref. 8; Supplementary Fig. 1), with Ab3(pE)–42 being most abundant. pE-Ab is more cytotoxic and aggregates more rapidly than conventional amyloid-b, andQC activity and pE-Ab levels are increased several-fold in Alzheimer’s disease brain. Alzheimer’s disease mouse models also indicate a role for pE-Ab in initiating pathology: oral administration of a QC inhibitor led to improved memory and learning, and reduced levels of pE-Ab and conventional amyloid-b. These data imply that pE-Ab potentiates the neurotoxicity of conventional amyloid-b, but leave open the issue ofmolecularmechanisms.To address that issue,we compared oligomerization of Ab3(pE)–42, Ab1–42, and mixtures of the peptides in vitro, and analysed responses of primary cultured neurons and glial cells (Supplementary Fig. 2) to the oligomers. At5mMpeptide,5%pE-Abaggregated faster thanAb3(pE)–42orAb1–42 alone, based on thioflavin T fluorescence shifts (Supplementary Fig. 3). The ratio of optical densities at 450 nm versus 490nm (OD450nm/OD490nm) forAb3(pE)–42 rose and peakedmore rapidly than for Ab1–42, but peaked at a ,25% lower level. The fastest rise in the