Glycogen synthase kinase-3beta haploinsufficiency mimics the behavioral and molecular effects of lithium

Abstract

Lithium is widely used to treat bipolar disorder, but its mechanism of action in this disorder is unknown. Several molecular targets of lithium have been identified, but these putative targets have not been shown to be responsible for the behavioral effects of lithium in vivo. A robust model for the effects of chronic lithium on behavior in mice would greatly facilitate the characterization of lithium action. We describe behaviors in mice that are robustly affected by chronic lithium. Remarkably, these lithium-sensitive behaviors are also observed in mice lacking one copy of the gene encoding glycogen synthase kinase-3beta (Gsk-3beta), a well established direct target of lithium. In addition, chronic lithium induces molecular changes consistent with inhibition of GSK-3 within regions of the brain that are paralleled in Gsk-3beta+/- heterozygous mice. We also show that lithium therapy activates Wnt signaling in vivo, as measured by increased Wnt-dependent gene expression in the amygdala, hippocampus, and hypothalamus. These observations support a central role for GSK-3beta in mediating behavioral responses to lithium.

Figure 1.

Dosing regimen and serum LiCl concentrations. A, Experimental design for the dose-response experiments. Animals were placed on a diet of 0, 0.2, or 0.2 followed by 0.4% LiCl in standard mouse chow for the times indicated and then were tested sequentially in the FST, elevated zero maze, and exploratory behavior (hole board) as indicated. B, Experimental design to test the effect of LiCl on behaviors after a 2 d drug washout. In contrast to the sequence in A, each behavior was tested with an independent cohort of animals to rule out an order effect. C, Serum lithium concentrations of mice from dose-response in A. ^*p < 0.05 compared with control and washout; ^#p < 0.05 compared with 0.2% LiCl.

Figure 4.

A, Reduction in GSK-3β protein in Gsk-3β heterozygous mice. Hypothalamus, hippocampus, frontal cortex, and cerebellum were isolated from adult brains of wild-type and heterozygous knock-out mice (n = 3) and were subjected to Western blotting for both isoforms of GSK-3.β-Tubulin serves as a loading control. B, Hypothalamus, hippocampus, frontal cortex, and cerebellum were isolated from adult brains of wild-type and heterozygous knock-out mice; GSK-3 and β-tubulin band intensities were measured independently for each animal (n = 5 for each group) by chemifluorescent immunoblotting and quantitation on a Storm/Phosphor-Imager. ^*p ≤ 0.05 compared with untreated wild-type mice (see Materials and Methods).

Figure 2.

The effect of lithium treatment on the FST. A, Mice (n = 10 per group) were placed on a diet with increasing concentrations of LiCl as described in Figure 1 A. Time immobile in the FST was measured as described in Materials and Methods. B, Animals were placed on the higher dose LiCl diet (n = 8) or control (CON) diet (n = 7) for 10 d; one group (n = 8) was placed on the LiCl diet 2 d earlier, switched to control diet after 10 d, and then tested after a 2 d wash out, as shown in Figure 1 B. The intermediate response of the group removed from LiCl for 2 d suggests that the behavior effect persists after the drug is removed. ^*p < 0.05 compared with control group; ^#p < 0.05 compared with 0.2% LiCl group. PER, Persistent effect of lithium after withdrawal of LiCl diet.

Figure 3.

The effect of lithium on exploratory behavior and overall activity. Animals (n = 10 per group) were placed on the higher dose LiCl diet, as described in Figure 1 B, and were tested for exploratory behavior in the hole board apparatus. A, Frequency of hole pokes. B, Overall activity (including movement in center and periphery of cage as well as rearing). PER, Persistent effect of lithium after withdrawal of LiCl diet; CON, control. ^*p < 0.05.

Figure 5.

Heterozygous loss of Gsk-3β mimics the effect of lithium on the FST and hole board-exploratory behavior. A-C, Gsk-3β^+/- mice and wild-type littermates were tested in FST (A) (n = 32, 31, and 23 for Wt, Wt/Li, and KO, respectively), exploratory behavior (B) (n = 19, 22, and 18, as noted above), and overall activity (C). Wild-type littermates received 0.2%/0.4% LiCl or control diet as described in Figure 1 A. ^*p < 0.05 compared with wild-type group. wt, Wild-type littermates; wt/Li, wild-type littermates on LiCl diet; KO, Gsk-3β^+/- heterozygotes.

Figure 6.

A,β-Catenin protein levels in hypothalamus of wild-type, wild-type plus lithium, heterozygous Gsk-3β (Gsk-3β^+/-), and lithium-treated Gsk-3β^+/- animals were assessed by immunoblotting. β-Tubulin serves as a loading control. B, Hypothalamus of six each of untreated controls, lithium-treated mice, and Gsk-3β^+/- mice was harvested and analyzed as in A, except that samples from each animal were analyzed individually.β-Catenin band intensity was measured by chemifluorescent immunoblotting and quantitation on a Storm/PhosphorImager. ^*p ≤ 0.05 compared with untreated wild-type mice (see Materials and Methods).

Figure 7.

Increased Wnt-Tcf-dependent transcription in lithium-treated mice. Coronal vibratome sections of brains from control and lithium-treated BAT-gal mice after incubation with X-gal. β-Galactosidase activity is increased in the dentate gyrus, amygdala, and to a lesser extent the hypothalamus. β-Galactosidase activity was exclusively in neurons (see Materials and Methods).