Chronic Sodium Selenate Treatment Restores Deficits in Cognition and Synaptic Plasticity in a Murine Model of Tauopathy

Abstract

A major goal in diseases is identifying a potential therapeutic agent that is cost-effective and can remedy some, if not all, disease symptoms. In Alzheimer's disease (AD), aggregation of hyperphosphorylated tau protein is one of the neuropathological hallmarks, and Tau pathology correlates better with cognitive impairments in AD patients than amyloid-β load, supporting a key role of tau-related mechanisms. Selenium is a non-metallic trace element that is incorporated in the brain into selenoproteins. Chronic treatment with sodium selenate, a non-toxic selenium compound, was recently reported to rescue behavioral phenotypes in tau mouse models. Here, we focused on the effects of chronic selenate application on synaptic transmission and synaptic plasticity in THY-Tau22 mice, a transgenic animal model of tauopathies. Three months with a supplement of sodium selenate in the drinking water (12 μg/ml) restored not only impaired neurocognitive functions but also rescued long-term depression (LTD), a major form of synaptic plasticity. Furthermore, selenate reduced the inactive demethylated catalytic subunit of protein phosphatase 2A (PP2A) in THY-Tau22 without affecting total PP2A.Our study provides evidence that chronic dietary selenate rescues functional synaptic deficits of tauopathy and identifies activation of PP2A as the putative mechanism.

Keywords: Alzheimer’s disease; chronic oral treatment; long-term depression; neurocognitive functions; protein phosphatase 2A (PP2A); synaptic plasticity; synaptic transmission; tau hyperphosphorylation.

Figure 1

Selenate treatment restores behavioral and neurocognitive deficits in THY-Tau22 (henceforth referred to as Tg) mice to levels comparable with wildtype (WT) controls. (A) Timeline of the experiments. The behavioral experiments started at the age of 12 months, the electrophysiological examination at 14 months. At the same time, tissue samples were collected for immunohistochemistry and Western blotting. (B) Analysis of cage activity revealed higher activity of Tg mice at the beginning of the recording period and increased nocturnal activity as compared with WT controls. These differences were abolished by chronic selenate treatment. Note that selenate treatment did not affect WT animals. (C–E) Open-field behavior. Tg animals traveled less in the arena (C) showed higher latencies to enter the center of the arena (D) and displayed less exploratory rearings in the center (E) as compared with WT mice. Selenate treatment remediated these behavior changes. Mean ± SEM is given; group sizes: cage activity n = 10 per group; open field test Tg-veh n = 9, Tg-sel n = 10, WT-veh n = 8, WT-sel n = 10. Bar with ^# indicates the significance level p ≤ 0.05 in RM-ANOVA with a Dunnett post hoc test using Tg vehicle as comparison; bars with a * and ** indicate p ≤ 0.01 and p ≤ 0.001, respectively, in the Two-way ANOVA; ^# and ^## represent p ≤ 0.01 and p ≤ 0.001, respectively, in the Dunnett post hoc test.

Figure 2

Impaired spatial learning and memory of Tg mice in the Morris water maze (MWM) was restored by treatment with selenate. (A) Tg mice provided with dietary selenate displayed escape latencies for locating the hidden platform comparable with the selenate-treated WT and non-treated WT groups. In contrast, the vehicle-treated Tg group were poor learners. (B) In the probe test, both WT groups and selenate-treated Tg groups confirmed memory for the platform location by spending significantly more time in the platform quadrant compared to vehicle-treated Tg mice. The vehicle-treated Tg group, in contrast, spent significantly less time in the target quadrant and showed a preference for the opposite quadrant instead (adj = adjacent 1/2, opp = opposite and target; hashes denote statistically significant differences between vehicle-treated Tg mice and the other three groups). (C–F) Examples of heat maps of the swim patterns during probe tests. The location of the platform is marked in the heat map as a black circle in the bottom right quadrant. (C) WT vehicle, (D) WT selenate-treated, (E) Tg vehicle, (F) Tg selenate-treated. Note the clear target preference of Tg after selenate treatment in panel (F) as compared to vehicle-treated Tg in (E). The heat map scale bar indicates the time in seconds. Mean ± SEM is given; group sizes: Tg-veh n = 9, Tg-sel n = 10, WT-veh n = 8, WT-sel n = 9. Bar with ^### indicates significance level p ≤ 0.001 of RM-ANOVA with a Dunnett post hoc test using Tg vehicle as comparison; ^# and ^## represent p ≤ 0.01 and p ≤ 0.001, respectively, in the Dunnett post hoc test.

Figure 3

Selenate restored impaired retention memory of Tg mice in the Passive Avoidance test. Mean ± SEM is given. Group sizes: Tg-veh n = 9, Tg-sel n = 10, WT-veh n = 8, WT-sel n = 10. A bar with a * indicates p ≤ 0.01 in the Two-way ANOVA; ^### represents p ≤ 0.0001 in the Dunnett post hoc test.

Figure 4

Enhanced basal synaptic transmission by selenate treatment in Tg mice. (A) Chronic selenate application increased basal synaptic transmission in Tg mice. Two-way ANOVA confirmed a significant between-group difference (F_(3,35) = 3.389, p < 0.029) and post hoc comparisons revealed a significant enhancement of synaptic transmission in selenate-treated Tg mice (n = 10) compared to selenate-treated WT mice (n = 8; p ≤ 0.020, Dunnett test with WT-sel as the control group). (B) Selenate-treated Tg mice (n = 12) had higher paired-pulse values than selenate-treated WT mice (n = 12) at short interpulse intervals of 10 and 20 ms. Interestingly, selenate caused a significant enhancement of paired-pulse values in WT mice at 200 ms. Two-Way ANOVA plus Dunnett post hoc test compared with WT-selenate. Mean ± SEM is given. Bar with ^#indicates significance level ≤0.05 of Dunnett post hoc test, likewise ^#p ≤ 0.05; ^##p ≤ 0.01.

Figure 5

Chronic selenate treatment restored synaptic plasticity [long-term depression (LTD)] in Tg mice. RM-ANOVA of all four groups confirmed a significant main effect of group F_(2,28) = 6.843, p = 0.001. (A) Vehicle-treated Tg mice (n = 10) failed to express robust LTD which was intact in vehicle-treated WT siblings (n = 8). (B) Selenate rescued LTD in Tg (n = 6) compared with vehicle-treated siblings (n = 10). (C) The rescued LTD in selenate-treated Tg mice matched exactly the time-course of LTD in selenate-treated WT animals (n = 8). Mean ± SEM is given. Bar with cross(es) indicates a significant difference in Tukey’s post hoc multiple comparisons, ⁺p ≤ 0.05, ⁺⁺p ≤ 0.01.

Figure 6

Selenate reduces AT8 immunoreactivity and antagonizes PP2A-demethylation in Tg mice. (A) Representative immunohistochemical sections from a vehicle-treated Tg mouse (left) and an animal that received selenate (right). Note the reduced AT8 immunoreactivity in the mouse treated with selenate. (B) Quantification of immunohistochemical AT8-staining from both groups. Selenate treatment reduced AT8-immunoreactivity to about 50% (p = 0.0027). (C) Immunoblots of PP2Ac and demethylated PP2Ac in vehicle-treated and selenate-treated Tg. (D,E) Quantification of the levels of PP2Ac and demethylated PP2Ac, respectively. The amount of demethylated PP2Ac is significantly lower after treatment with selenate, supporting higher PP2A activity under treatment. (F) The amount of demethylated PP2Ac (deMe-PP2Ac) normalized to the amount of PP2AC. Mean ± SEM is given. *p < 0.05, n = 3 per group.