Adenosine A2A receptor inactivation alleviates early-onset cognitive dysfunction after traumatic brain injury involving an inhibition of tau hyperphosphorylation

Abstract

Tau is a microtubule-associated protein, and the oligomeric and hyperphosphorylated forms of tau are increased significantly after neurotrauma and considered important factors in mediating cognitive dysfunction. Blockade of adenosine A_2A receptors, either by caffeine or gene knockout (KO), alleviates cognitive dysfunction after traumatic brain injury (TBI). We postulated that A_2AR activation exacerbates cognitive impairment via promoting tau hyperphosphorylation. Using a mouse model of moderate controlled cortical impact, we showed that TBI induced hyperphosphorylated tau (p-tau) in the hippocampal dentate gyrus and spatial memory deficiency in the Morris water maze test at 7 days and 4 weeks after TBI. Importantly, pharmacological blockade (A_2AR antagonist ZM241385 or non-selective adenosine receptor antagonist caffeine) or genetic inactivation of A_2ARs reduced the level of tau phosphorylation at Ser404 and alleviated spatial memory dysfunction. The A_2AR control of p-tau is further supported by the observations that a KO of A_2AR decreased the activity of the tau phosphorylation kinases, glycogen synthase kinase-3β (GSK-3β) and protein kinase A (PKA) after TBI, and by that CGS21680 (A_2AR agonist) exacerbated okadaic acid-induced tau hyperphosphorylation in cultured primary hippocampal neurons. Lastly, CGS21680-induced neuronal tau hyperphosphorylation and axonal injury were effectively alleviated by individual treatments with ZM241385 (A_2AR antagonist), H89 (PKA antagonist) and SB216763 (GSK-3β antagonist), or by the combined treatment with H89 and SB216763. Our findings suggest a novel mechanism whereby A_2AR activation triggers cognitive dysfunction by increasing the phosphorylation level of tau protein after TBI and suggest a promising therapeutic and prophylactic strategy by targeting aberrant A_2AR signaling via tau phosphorylation.

Figure 1

Astrogliosis and edema indicate the moderate degree of our traumatic brain injury (TBI) mouse model. (a) Astrogliosis was obvious in the contralateral dentate gyrus (DG) (red box) and the ipsilateral impacted cortex (blue box) at 24 h and 7 days after TBI. Scale bar, 50 μm. (b) Integrated optical density of positive GFAP staining in the contralateral DG. n=3 mice at each time point. Data represent mean±s.e.m., *P<0.01 compared to the SHAM group, ^nsP>0.05 compared to the SHAM group, one-way ANOVA. (c) Brain water content after TBI; n=6 per group. Data represent mean±s.e.m., *P<0.01 compared to the SHAM group, one-way ANOVA.

Figure 2

Hyperphosphorylation of tau in wild-type mice after traumatic brain injury (TBI). (a) The contralateral dentate gyrus (DG) showed a significant increase in the level of tau phosphorylated at Ser404 at 24 h, 7 days and 4 weeks post-injury. The levels of tau phosphorylation at Thr205 and Ser262 did not change obviously. Scale bar, 50 μm. (b) The integrated optical density of Ser404 phospho-tau staining in the contralateral DG of SHAM mice and mice 24 h, 7 days and 4 weeks after TBI. n=3 per group. Data represent mean±s.e.m., *P<0.05 compared to the SHAM group, ^#P<0.05 compared to the 24 h group, one-way ANOVA. (c) Western blot analysis indicated that only phosphorylation at Ser404 increased at 24 h after TBI. (d) Relative levels of phospho-tau/total tau based on western blot results for the contralateral hippocampus at 24 h after TBI; n=3 per group. Data represent mean±s.e.m., *P<0.05 between the two groups, two-tailed t-test.

Figure 3

Inactivation of A_2A receptors attenuated spatial memory impairment and tau hyperphosphorylation after traumatic brain injury (TBI). (a) Neurological deficit scores at 1 day, 3 days and 6 days after TBI. The differences between genotypes were not significant at 24 h (P=0.662), 3 days (P=0.852) and 6 days (P=0.950) after TBI. n=6 per time point for knockout (KO) group, n=8 per time point for WT group, Mann–Whitney U-test. (b) All groups exhibited comparable total swimming distances on the first training day for the Morris water maze (MWM) (P>0.05, one-way ANOVA); n=25 in the WT group, n=12 in the KO group and n=12 in the WT+Caffeine group for each time point. Data represent mean±s.e.m. (c) Spatial reference memory was ameliorated in A_2AR KO mice. There was a significant difference in the escape latency between the WT+TBI and KO TBI+TBI groups at 7 days after TBI (P<0.05). This tendency was also present but not significant at 4 weeks after TBI (P>0.05). Furthermore, there were no significant differences in the escape latency between the SHAM groups of WT and KO mice (P>0.05). Data represent mean±s.e.m., *P<0.05 between WT and KO groups, two-way ANOVA. (d) Spatial working memory impairment was also alleviated in A_2AR KO mice compared with that of the WT group at 4 weeks after TBI. Data represent mean±s.e.m., *P<0.05, one-way ANOVA. (e) The recovery amplitude of spatial working memory. Data represent mean±s.e.m., *P<0.05, one-way ANOVA. (f) Representative swimming traces of mice at 4 weeks after TBI in the spatial working memory paradigm. (g) The schematic representation of the method and process for MWM, edema evaluation and neurologic deficit scoring. (h) Hyperphosphorylation of tau at Ser404 (red) was mainly observed in the granule neurons (green) in the contralateral DG. The level of tau phosphorylation at Ser404 increased significantly at 7 days (B) and 4 weeks (C) after TBI compared with that of the SHAM group (A). A_2AR knockout attenuated this effect (E, F), scale bar, 50 μm. (i, j) A significant increase in tau phosphorylation at Ser404 in the contralateral hippocampus was also detected, and A_2AR knockout could also alleviate the hyperphosphorylation of tau at 7 days and 4 weeks after TBI. n=3 per group, data represent mean±s.e.m., *P<0.05, one-way ANOVA.

Figure 4

Chronic caffeine pretreatment attenuated the decrease in dendritic spine density in the dentate gyrus. Representative images of Golgi-impregnated dentate gyrus (DG) (a1, b1, c1), granule cells (a2, b2, c2) and dendritic spines (a3, b3, c3). Quantitative analysis of the densities of mushroom-shaped spines (d), thin spines (e) and stubby spines (f). (g) Quantitative analysis of the dendritic density of granule cells. (h) Changes in the proportions of spine morphological subtypes in the DG granule neurons. Data are presented as the mean±s.e.m. n=3 mice per group, *P<0.05, **P<0.01, one-way ANOVA. Scale bars represent 200 μm in a1–c1, 20 μm in a2–c2 and 5 μm in a3–c3.

Figure 5

PKA and GSK-3β were involved in the neurotoxic effects of okadaic acid (OA) and CGS21680. (a) Treatment with OA for 12 h produced tau hyperphosphorylation at Ser404 (red) and significant decreases in axons and dendrites of primary hippocampal neurons (green). Increased p-tau signals were observed in the cell bodies (arrows) and the end of disconnected neurites (circles). CGS21680 treatment exacerbated OA-induced neuronal damage and tau hyperphosphorylation. H89 and/or SB216763 could alleviate the neurite loss and the hyperphosphorylation of tau at Ser404. ZM241385 by itself ameliorated OA-induced or OA+CGS-induced neuronal damage and tau hyperphosphorylation. Scale bar, 50 μm. CGS, CGS21680; SB, SB216763. (b) Levels of tau phosphorylation were measured using the average optical density of p-tau Ser404 staining. *P<0.05 compared with the DMSO group, ^#P<0.05 compared with the OA group, ^△P<0.05 compared with the OA+CGS group and ^&P<0.05 between two groups, one-way ANOVA. (c) Western blot analysis indicating the level of tau phosphorylation after treatments with OA, CGS21680 and ZM241385. (d) Phosphorylation levels of PKA and GSK-3β were analyzed by western blot. (e) In vivo testing demonstrated significant increases in the p-PKA levels in the contralateral hippocampus at 7 days and 4 weeks after traumatic brain injury (TBI). However, PKA activity increased only slightly in A_2AR KO mice. (f) Phosphorylation of GSK-3β at Y216 indicated that its activity did not change significantly at 4 weeks after TBI. (g) Inhibition of GSK-3β activity was observed in A_2AR KO mice at 7 days and 4 weeks after TBI based on S9 phosphorylation. n=4 mice per group, data represent mean±s.e.m., *P<0.05, one-way ANOVA. (h) Co-localization of A_2AR with p-tau (Ser404) (yellow box) in brain tissue samples from TBI patients. With no positive A_2AR signal, no p-tau could be detected (blue box). (i) Co-localization of A_2AR with p-GSK-3β (Y216) (yellow box). Scale bars, 50 μm.