Lithium reduces tau phosphorylation but not A beta or working memory deficits in a transgenic model with both plaques and tangles

Abstract

Glycogen synthase kinase 3 (GSK-3) is a major kinase implicated in the pathogenesis of Alzheimer's disease (AD), and reducing its activity may have therapeutic efficacy. Two variants exist, referred to as GSK-3 alpha and GSK-3beta. In addition to the latter's well-described role in the phosphorylation of tau, reports also suggest that GSK-3 alpha may regulate amyloid precursor protein processing and Abeta formation. The activities of both GSK-3 alpha and GSK-3beta are reduced by lithium, a well-tolerated drug used in humans to combat bipolar disorder. Here, we investigate the therapeutic efficacy of chronic lithium administration in aged 3xTg-AD mice that harbor both plaques and tangles. We found that lithium reduced tau phosphorylation but did not significantly alter the A beta load. Despite the reduction in phosphotau, lithium treatment did not improve deficits in working memory. Although other studies have investigated the effects of lithium on tau biochemistry, this study represents the first to address comprehensively its therapeutic potential on other critical aspects of AD including its effect on A beta and learning and memory. It remains to be determined from human clinical trials whether lithium treatment alone will improve the clinical outcome in AD patients. These results, however, suggest that the most efficacious treatment will be combining lithium with other anti-A beta interventions.

Figure 1

The Aβ load is not altered by lithium treatment in 3xTg-AD mice with established plaque pathology. Low (A, B)- and high (C, D)-magnification photomicrographs of hippocampal brain sections from lithium-treated and untreated mice stained with the anti-Aβ antibody 6E10. No changes in the intraneuronal Aβ staining or plaque number were observed between treated and untreated mice (n = 10 per group). E and F: The steady-state levels of soluble and insoluble Aβ40 and Aβ42 were not significantly different between treated and untreated mice as determined by sandwich enzyme-linked immunosorbent assay (n = 10 per group). T-PER, tissue protein extraction reagent (Pierce); FA, formic acid.

Figure 2

Reduced selective phosphotau epitopes in brains of lithium-treated mice. Representative microphotographs of sections from treated and untreated mice stained with different anti-tau antibodies. A–D: Low- and high-magnification views of the CA1 region of the hippocampus stained with antibody HT7 showing that lithium does not affect total tau levels. E–H: Lithium-treated mice showed a significant decrease in the somatodendritic accumulation of tau phosphorylated at Thr181 and Ser202/Thr205 compared with untreated mice in the CA1 region of the hippocampus. I–L: In contrast, lithium did not affect tau phosphorylation at the PHF-1 and 12E8 sites (n = 10 per group).

Figure 3

Lithium selectively lowers steady-state levels of certain phosphotau epitopes. A: Protein extract from brains of treated and untreated mice were separated in a Western blot probed with different anti-tau antibodies (n = 10 per group). Representative Western blots are shown. B–H: Densitometric analysis of the Western blots showed that the steady-state levels of tau, as detected by antibody HT7, were not significantly different between treated and untreated mice (B), whereas there was a significant decrease in the steady-state levels of AT270 (P = 0.037, C), AT8 (P = 0.011, D), AT180 (P = 0.031, E), and AT100 (P = 0.033, F). Lithium, however, did not significantly change the steady-state levels of PHF-1 and 12E8 (G, H). Veh, vehicle; Li, lithium; A.U., arbitrary units.

Figure 4

Lithium lowers GSK-3α and GSK-3β activity in the brains of the 3xTg-AD mice. A: Protein extract from brains of treated and untreated mice were separated in Western blots probed with different antibodies to determine the activity of GSK-3α, GSK-3β, CDK5, cdc2, JNK, and MAP kinase (n = 10 per group). B: Densitometric analysis of the Western blots showed that the steady-state levels of total GSK-3α and GSK-3β were similar between treated and untreated mice (P > 0.05). C and D: In contrast, the levels of the inactive forms of GSK-3α and GSK-3β were significantly increased in the lithium-treated mice compared with vehicle-treated mice (P < 0.05). E and F: To assess the changes in GSK-3 activity, we directly measured the enzymatic activity of these two kinases and found that lithium administration significantly decreased the activity of GSK-3α and GSK-3β (P = 0.002 and P < 0.0001, respectively). G–K: The effects of lithium were selective for GSK-3 because we found that the levels of CDK5 or its activators p35 and p25, cdc2, JNK, and p44/42 MAP kinase were not significantly different between treated and untreated mice. Veh, vehicle; Li, lithium; A.U., arbitrary units.

Figure 5

Lithium failed to rescue working memory deficits in the 3xTg-AD mice. Lithium treatment did not rescue the working memory deficits present in aged 3xTg-AD mice because no statistically significant difference was detected between the alternation rate of treated and untreated mice (n = 10 per group).