Alzheimer's disease brain-derived amyloid-β-mediated inhibition of LTP in vivo is prevented by immunotargeting cellular prion protein

Abstract

Synthetic amyloid-β protein (Aβ) oligomers bind with high affinity to cellular prion protein (PrP(C)), but the role of this interaction in mediating the disruption of synaptic plasticity by such soluble Aβ in vitro is controversial. Here we report that intracerebroventricular injection of Aβ-containing aqueous extracts of Alzheimer's disease (AD) brain robustly inhibits long-term potentiation (LTP) without significantly affecting baseline excitatory synaptic transmission in the rat hippocampus in vivo. Moreover, the disruption of LTP was abrogated by immunodepletion of Aβ. Importantly, intracerebroventricular administration of antigen-binding antibody fragment D13, directed to a putative Aβ-binding site on PrP(C), prevented the inhibition of LTP by AD brain-derived Aβ. In contrast, R1, a Fab directed to the C terminus of PrP(C), a region not implicated in binding of Aβ, did not significantly affect the Aβ-mediated inhibition of LTP. These data support the pathophysiological significance of SDS-stable Aβ dimer and the role of PrP(C) in mediating synaptic plasticity disruption by soluble Aβ.

Figure 1.

SDS-stable Aβ dimer-containing aqueous extract of human Alzheimer's disease brain inhibits LTP of synaptic transmission in the rat hippocampus in vivo. A, Unmanipulated TBS extracts from AD brain (AD), TBS (-ve), extract treated with preimmunine serum (Mock ID) or extract immunodepleted of Aβ (ID) were examined by immunoprecipitation/Western blotting as described in Materials and Methods. The Aβ content of each was estimated by reference to known amounts (2–10 ng) of synthetic Aβ1–42 loaded on the same gel. Molecular weight markers are on the left and the migration of Aβ monomer (M) and SDS-stable dimer (D) are indicated on the right. The blot was trimmed to the 13 kDa molecular weight standard. B, Application of high-frequency stimulation (arrow) induced robust LTP in animals that received vehicle (asterisk, 5 μl, n = 5), whereas acute injection of Aβ-containing AD brain extract (AD, 5 μl) completely inhibited LTP (n = 5). C, AD brain extract immunodepleted of Aβ (ID-AD, 5 μl) did not inhibit LTP (n = 4). In contrast, AD brain extract that had been processed in the same manner but with normal rabbit preimmune antiserum (mock ID-AD, 5 μl, n = 4) strongly inhibited LTP. D, Acute injection of Aβ-containing (asterisk, 5 μl, n = 4) or Aβ-immunodepleted (n = 4) AD brain extract did not affect baseline excitatory synaptic transmission, similar to vehicle-injected controls (5 μl, n = 4). Insets show representative traces at the times indicated. Calibration: 1.5 mV, 10 ms.

Figure 2.

Control human brain extract did not inhibit LTP. Application of HFS induced robust LTP after acute injection (asterisk, 5 μl) of control brain TBS extract (Non-AD) (n = 4) similar to vehicle-injected rats (n = 8). In contrast, injection of Aβ-containing TBS extract from the brain of another AD patient (AD) completely inhibited LTP at 3 h post-HFS (n = 5). Insets show representative traces at the times indicated. Calibration: 1.5 mV, 10 ms.

Figure 3.

PrP^C dependence of human AD brain Aβ-mediated inhibition of LTP. A, Acute injection of Aβ-containing AD brain extract (asterisk, 5 μl) completely inhibited LTP (n = 6) in contrast to animals that received two intracerebroventricular injections of vehicle (n = 5). B, Soluble AD brain Aβ-mediated inhibition of LTP was prevented by the antibody fragment D13 directed to PrP^C_96–104, but not the Fab R1 directed to PrP^C_225–231. Preinjection with D13 (11 μg in 10 μl, n = 5) prevented the inhibition of LTP by the brain extract. In contrast, the brain extract completely inhibited LTP in animals pretreated with R1 (11 μg, n = 5). C, Neither D13 nor R1 significantly affected control LTP. Preinjection with D13 (n = 5) or R1 (n = 4) before injection of Aβ-immunodepleted AD brain TBS extract did not significantly affect LTP compared with animals that received two injections of vehicle (n = 5). Insets show representative traces at the times indicated. Calibration: 1.5 mV, 10 ms.