C-terminal fragment of N-cadherin accelerates synapse destabilization by amyloid-β.

The aetiology of Alzheimer's disease is thought to include functional impairment of synapses and synapse loss as crucial pathological events leading to cognitive dysfunction and memory loss. Oligomeric amyloid-β peptides are well known to induce functional damage, destabilization and loss of brain synapses. However, the complex molecular mechanisms of amyloid-β action resulting ultimately in synapse elimination are incompletely understood, thus limiting knowledge of potential therapeutic targets. Under physiological conditions, long-term synapse stability is mediated by trans-synaptically interacting adhesion molecules such as the homophilically binding N-cadherin/catenin complexes. In this study, we addressed whether inhibition of N-cadherin function affects amyloid-β-induced synapse impairment. We found that blocking N-cadherin function, both by specific peptides interfering with homophilic binding and by expression of a dominant-negative, ectodomain-deleted N-cadherin mutant, resulted in a strong acceleration of the effect of amyloid-β on synapse function in cultured cortical neurons. The frequency of AMPA (alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid) receptor-mediated miniature excitatory postsynaptic currents was reduced upon amyloid-β application much earlier than observed in controls. We further hypothesized that ectodomain-shed, transmembrane C-terminal fragments that are generated during N-cadherin proteolytic processing might similarly enhance amyloid-β-induced synapse damage. Indeed, expression of human N-cadherin C-terminal fragment 1 strongly accelerated amyloid-β-triggered synapse impairment. Ectodomain-shed N-cadherin C-terminal fragment 1 is further proteolytically cleaved by γ-secretase. Therefore, both pharmacological inhibition of γ-secretase and expression of the dominant-negative presenilin 1 mutant L166P were used to increase the presence of endogeneous N-cadherin C-terminal fragment 1. Under these conditions, we again found a strong acceleration of amyloid-β-induced synapse impairment, which could be compensated by over-expression of full-length N-cadherin. Intriguingly, western blot analysis of post-mortem brains from patients with Alzheimer's disease revealed an enhanced presence of N-cadherin C-terminal fragment 1. Thus, an inhibition of N-cadherin function by proteolytically generated N-cadherin C-terminal fragment 1 might play an important role in Alzheimer's disease progression by accelerating amyloid-β-triggered synapse damage.