Riluzole attenuates glutamatergic tone and cognitive decline in AβPP/PS1 mice

Abstract

We have previously demonstrated hippocampal hyperglutamatergic signaling occurs prior to plaque accumulation in AβPP/PS1 mice. Here, we evaluate 2-Amino-6-(trifluoromethoxy) benzothiazole (riluzole) as an early intervention strategy for Alzheimer's disease (AD), aimed at restoring glutamate neurotransmission prior to substantial Beta amyloid (Aβ) plaque accumulation and cognitive decline. Male AβPP/PS1 mice, a model of progressive cerebral amyloidosis, were treated with riluzole from 2-6 months of age. Morris water maze, in vivo electrochemistry, and immunofluorescence were performed to assess cognition, glutamatergic neurotransmission, and pathology, respectively, at 12 months. Four months of prodromal riluzole treatment in AβPP/PS1 mice resulted in long-lasting procognitive effects and attenuated glutamatergic tone that was observed six months after discontinuing riluzole treatment. Riluzole-treated AβPP/PS1 mice had significant improvement in long-term memory compared to vehicle-treated AβPP/PS1 mice that was similar to normal aging C57BL/6J control mice. Furthermore, basal glutamate concentration and evoked-glutamate release levels, which were elevated in vehicle-treated AβPP/PS1 mice, were restored to levels observed in age-matched C57BL/6J mice in AβPP/PS1 mice receiving prodromal riluzole treatment. Aβ plaque accumulation was not altered with riluzole treatment. This study supports that interventions targeting the glutamatergic system during the early stages of AD progression have long-term effects on disease outcome, and importantly may prevent cognitive decline. Our observations provide preclinical support for targeting glutamate neurotransmission in patients at risk for developing AD. Read the Editorial Highlight for this article on page 399.

Keywords: Alzheimer's disease (AD); alpha-7 nicotinic acetylcholine receptor (α7nAChR); amyloid-beta (Aβ); biosensor; learning and memory; prodromal intervention.

Figure 1:. Experimental Design

An outline of the experimental design. Riluzole treatment was conducted when mice were 2-6 months of age. At 6 months of age, treatment was discontinued and mice were returned to untreated water for the remainder of the study. Animal numbers are located underneath each analysis. *One C57BL/6J vehicle treated mouse died due to complications from anesthetized surgical procedures. ^#One C57BL/6J vehicle treated and two AβPP/PS1 mice were excluded from immunofluorescent analysis because of inadequate slice preparation and staining procedures. Abbreviations: Morris water maze (MWM).

Figure 2:. MWM Training and Probe Challenge

Mice were trained to locate a hidden escape platform over five consecutive days. Each training day consisted of three trials that were averaged into a single data point for each treatment group to analyze swimming speed (A) CIPL (B) and cumulative distance (C). CIPL adjusts for average swim speed and corrects the difference in distance between different start positions and distance to the platform. Representative track plots of the 60 s probe challenge are shown for each treatment group (D). No differences in swimming speed were observed during the probe challenge (E) indicating motor performance does not account for differences in cognitive measurements. C57BL/6J vehicle and AβPP/PS1 riluzole mice, but not AβPP/PS1 vehicle mice, selectively searched the target quadrant compared to the other quadrants for the former location of the hidden escape platform (F). AβPP/PS1 vehicle mice entered the annulus 40 (immediate area surrounding the platform) fewer times compared to C57BL/6J vehicle and AβPP/PS1 riluzole mice (G). N= number of mice. Figures A-C: two-way ANOVA, Dunnett’s post-hoc; Figures E-G: one-way ANOVA, Dunnett’s post-hoc; *p<0.05, **p<0.01, AβPP/PS1 vehicle (n=12) vs C57BL/6J vehicle mice (n=14); §§p<0.01, §§§p<0.001 AβPP/PS1 vehicle vs AβPP/PS1 riluzole mice (n=11).

Figure 3:. Representative Glutamate Traces and Glutamate Measures

A) Representative traces of basal glutamate and stimulus-evoked glutamate release from 70 mM KCl stimulation. Treatment groups are placed into columns and hippocampal subfields are shown in rows. The inset trace at the top of each panel displays the reproducibility of glutamate release and clearance, while the single response is a magnified view of the first inset signal to give a clearer presentation of dynamics of glutamatergic signaling. Concentration and time axes are consistent in all panels for comparisons. B) Basal glutamate was determined prior to local application of 70 mM KCl in each hippocampal subfield. C) The volume of stimulus ejected was consistent across treatment groups for direct comparisons of evoked glutamate release. D) The average glutamate release was determined by subtracting the peak amplitude from the basal measure obtained prior to each ejection of the stimulus. E) Glutamate uptake rate was calculated by multiplying the peak response (μM) by logarithmic slope (k₋₁) of glutamate decay versus time (s⁻¹) as estimated by the use of regression analysis (R²≥0.9). N= number of mice. A one-way ANOVA with Dunnett’s post-hoc was used in each hippocampal subfield. *p<0.05; **p<0.01, ***p<0.001, ****p<0.0001 AβPP/PS1 vehicle (11-12) versus C57BL/6J vehicle (n=12-13) mice; §p<0.05, §§p<0.01, §§§p<0.001, §§§§p<0.0001 AβPP/PS1 vehicle versus AβPP/PS1 riluzole (n=11) mice.

Figure 4:. Hippocampal α7nAChR Expression

A) Representative 40x images (scale bar = 20 μM) of DAPI (blue) and α7nAChR (red) expression in the DG (A-C), CA3 (D-F), and CA1 (G-I) for the three treatment groups separated into columns. The average mean α7nAChR staining density is shown in J. A one-way ANOVA with Dunnett’s post-hoc was used in each hippocampal subfield. N= number of mice; C57BL/6J vehicle (n=13), AβPP/PS1 vehicle (n=12); AβPP/PS1 riluzole (n=9) mice.

Figure 5:. Hippocampal GFAP/Aβ Expression

A) Representative 10x images (Scale bar = 100 μM) of GFAP (red) and Aβ plaque (blue) expression, with 40X insets (scale bar = 20 μm), in the DG (A-C) and CA1/CA3 (D-F) hippocampal subregions for the three treatment groups (columns). Subregion specific changes in GFAP expression (G) was compared using a one-way ANOVA with Dunnett’s post-hoc, while a t-test was used for comparing plaque accumulation between riluzole and vehicle treated AβPP/PS1 groups (H). N= number of mice.