Tau oligomers impair memory and synaptic plasticity through the cellular prion protein

Abstract

Deposition of abnormally phosphorylated tau aggregates is a central event leading to neuronal dysfunction and death in Alzheimer's disease (AD) and other tauopathies. Among tau aggregates, oligomers (TauOs) are considered the most toxic. AD brains show significant increase in TauOs compared to healthy controls, their concentration correlating with the severity of cognitive deficits and disease progression. In vitro and in vivo neuronal TauO exposure leads to synaptic and cognitive dysfunction, but their mechanisms of action are unclear. Evidence suggests that the cellular prion protein (PrP^C) may act as a mediator of TauO neurotoxicity, as previously proposed for β-amyloid and α-synuclein oligomers. To investigate whether PrP^C mediates TauO detrimental activities, we compared their effects on memory and synaptic plasticity in wild type (WT) and PrP^C knockout (Prnp^0/0) mice. Intracerebroventricular injection of TauOs significantly impaired recognition memory in WT but not in Prnp^0/0 mice. Similarly, TauOs inhibited long-term potentiation in acute hippocampal slices from WT but not Prnp^0/0 mice. Surface plasmon resonance indicated a high-affinity binding between TauOs and PrP^C with a K_D of 20-50 nM. Immunofluorescence analysis of naïve and PrP^C-overexpressing HEK293 cells exposed to TauOs showed a PrP^C dose-dependent association of TauOs with cells over time, and their co-localization with PrP^C on the plasma membrane and in intracellular compartments, suggesting PrP^C-may play a role in TauO internalization. These findings support the concept that PrP^C mediates the detrimental activities of TauOs through a direct interaction, suggesting that targeting this interaction might be a promising therapeutic strategy for AD and other tauopathies.

Keywords: Alzheimer’s disease; Memory impairment; Prion protein; Synaptic dysfunction; Tau oligomers.

Fig. 1

Biochemical features of human 2N4R wild-type TauOs. a Scheme of the tau oligomerization procedure. b Representative atomic force microscopy (AFM) image. c Western blot analysis of TauOs. Samples were run on 3–8% tris–acetate gel and incubated with anti-total tau antibody. Both techniques show the oligomeric nature of the preparation

Fig. 2

TauOs, but not the monomer (TauM), impair memory and synaptic plasticity in WT but not in Prnp^0/0 mice. a Scatter plots ± standard error mean (SEM) of the DI of WT mice receiving a single ICV injection of Veh (n = 10), TauM (n = 9) or TauOs (n = 9) and then tested in the NORT; (One-Way ANOVA: F_(2,25) = 16.28, P < 0.001); *** P < 0.001, ****P < 0.0001 Tukey’s post-hoc test. b Scatter plot ± SEM of the DI of WT or Prnp^0/0 mice treated with Veh (WT: n = 6; Prnp^0/0 n = 9) or TauOs (WT: n = 6; Prnp^0/0 n = 10) and tested in the NORT. Two-way ANOVA found an effect of: Tg F_{(1, 27)} = 3,481, P = 0.073, Treatment F_{(1, 27)} = 8.32, P = 0.0076 and a significant Interaction F_{(1, 27)} = 11.01, P = 0.0026; * P < 0.05, **P < 0.01, Tukey’s post-hoc test. c Time course of LTP in the CA1 hippocampal area in slices from WT mice pre-incubated for 60–90 min with either Veh (n = 7), 100 nM monomers (n = 7) or 100 nM TauOs (n = 7). Two-way ANOVA for repeated measures found: VEH vs. TauM F_(1,12) = 0.18, P = 0.68; VEH vs. TauOs F_(1,12) = 14.62, P = 0.002; TauOs versus TauM F_(1,12) = 11.45, P = 0.005; (n = 7/group). d Data are scatter plots ± SEM of residual potentiation calculated by averaging the slopes of the field excitatory postsynaptic potentials (fEPSP) in the last 10 min of LTP recorded in WT slices. One-way ANOVA for treatment factor found: F_(2,18) = 26.39, P < 0.0001; VEH vs. TauOs P < 0.0001; TauM vs VEH P = 0.51; TauOs vs. TauM P < 0.0001. ***P < 0.001, ****P < 0.0001; Tukey’s post-hoc test). e Time course of LTP in the CA1 hippocampal area in slices from Prnp^0/0 mice preincubated with Veh or TauOs. f Scatter plot ± SEM of residual potentiation calculated by averaging the fEPSP slopes in the last 10 min of LTP in Prnp^0/0 slices (t₉ = 2.26, P = 0.985)

Fig. 3

SPR sensorgrams of TauO-PrP^C binding. Sensorgrams were obtained flowing two different concentrations of TauOs (expressed in monomer equivalents) over immobilized 94B4 PrP antibody (a) or TOC1 anti-TauO antibody (b), PrP captured by 94B4 antibody (c) or rhPrP protein (d). TauOs were flowed for 2 min as indicated by the dashed lines. The sensorgrams were obtained after correction for the signal obtained in a reference (empty) surface and are shown in red. For each immobilized protein the sensorgrams with both TauOs concentrations were globally fitted using a two-sites model, and the fitting is shown in black

Fig. 4

Time course of TauO binding and internalization in naïve and PrP^C-overexpressing HEK293 cells. a Representative images of TauOs (green) in naïve (HEK293, upper panels) or PrP-overexpressing (HEK293-PrP, lower panels) cells after 15, 30 and 60 min of incubation (0.5 µM). TauOs immunofluorescence is detectable at early time points only in HEK293-PrP cells, indicating that PrP^C facilitates the association of TauOs to cells. Scale bar 50 µm. b Higher magnification confocal images of TauOs (green) and PrP^C (red) acquired after 60 min of incubation, showing TauOs—PrP^C co-localization in naïve (b¹) or PrP-overexpressing (b²) HEK293 cells. The co-localization channels (yellow) together with TauOs and PrP was reconstructed as 3D rendering and shown in (b³) and (b⁴) for both naïve and HEK293-PrP cells respectively. Major thick 10 µm