Decreased amyloid-β and increased neuronal hyperactivity by immunotherapy in Alzheimer's models

Abstract

Among the most promising approaches for treating Alzheimer's disease is immunotherapy with amyloid-β (Aβ)-targeting antibodies. Using in vivo two-photon imaging in mouse models, we found that two different antibodies to Aβ used for treatment were ineffective at repairing neuronal dysfunction and caused an increase in cortical hyperactivity. This unexpected finding provides a possible cellular explanation for the lack of cognitive improvement by immunotherapy in human studies.

BACE inhibition-dependent repair of Alzheimer’s pathophysiology

“…Our results further suggest that neuronal hyperactivity within amyloid plaque-bearing local circuits is likely due to prefibrillary Aβ species, which form a halo around the plaque cores. The findings are unexpected in view of previous evidence indicating that at such advanced disease stages, many pathological and cellular processes are chronic and mostly irreversible (7,8,(23)(24)(25). In fact, we found that most dysfunctional neurons in the amyloid-plaque bearing neocortex are still viable, and that their "hyperactive phenotype" can be reversed by reducing Aβ in the brain.…”

“…Experiments were performed on female APP23xPS45 double-transgenic mice aged >6 mo overexpressing the human APP with the Swedish mutation (670/671) and the human G384A-mutated PS1 (7,10,11,17,(35)(36)(37). At this age, mice exhibit widespread cerebral Aβ pathology (10,17,36), neuropil dystrophy (10), and astrocytic (17,38) and microglia (39) impairments, as well as learning and memory deficits (10,11,17).…”

“…After the behavioral testing, in vivo two-photon Ca 2+ imaging was performed as described previously (7,10,17). In brief, mice were anesthetized with isoflurane 1-1.5% (vol/vol) for induction and during surgery.…”

“…Image analysis was performed offline as described previously (7,10,17,36). In brief, regions of interest (ROIs) were drawn around individual somata, and then relative fluorescence change (ΔF/F) vs. time traces were generated for each ROI.…”

“…The amyloid hypothesis posits that the abnormal accumulation of amyloid-β (Aβ) peptides in the brain, and their aggregation, is an essential feature of AD (2,3); however, results from clinical studies using several Aβ-targeting compounds have called into question the existence of a direct link between a reduction in Aβ and improvement of brain function, particularly in more advanced disease stages (4)(5)(6). In addition, recent evidence obtained in mouse models carrying genetic mutations that cause AD in humans revealed that immunotherapy with antibodies against Aβ worsened rather than reversed neuronal dysfunction (7). Despite reducing plaque burden, the anti-Aβ antibodies caused a massive increase in cortical hyperactivity and promoted abnormal synchrony of neurons in a subset of the treated mice.…”