Transgenic mice overexpressing glycogen synthase kinase 3beta: a putative model of hyperactivity and mania

Abstract

Lithium is used as treatment for bipolar disorder with particular efficacy in the treatment of mania. Lithium inhibits glycogen synthase kinase 3beta (GSK-3beta) directly or indirectly via stimulation of the kinase Akt-1. We therefore investigated the possibility that transgenic mice overexpressing GSK-3beta could be of relevance to model bipolar disorder. Transgenic mice showed hypophagia, an increased general locomotor activity, and decreased habituation as assessed in an open field, an increased acoustic startle response, and again decreased habituation. The forced swim test revealed a reduced immobility in transgenic mice, but this is probably related to the hyperactivity of the animals. There were no differences in baseline and stress-induced increases of plasma adrenocorticotrophic hormone and corticosterone levels. Molecular analysis suggests compensatory mechanisms in the striatum of these transgenic mice for the overload of active GSK-3beta by dimming the endogenous GSK-3beta signaling pathway via upregulation of Akt-1 expression. Brain-derived neurotrophic factor protein levels were increased in the hippocampus of the transgenic mice. This suggests some kind of compensatory mechanism to the observed reduction in brain weight, which has been related previously to a reduced size of the somatodendritic compartment. Together, in mice overexpressing GSK-3beta, specific intracellular signaling pathways are affected, which is accompanied by altered plasticity processes and increased activity and reactivity, whereas habituation processes seem to be decreased. The behavioral observations led to the suggestion that the model at hand recapitulates hyperactivity as observed in the manic phase of bipolar disorder.

Figure 1.

Averaged food intake (means ± SEM) (A), food intake relative to body weight (B), water intake (C), and water intake relative to body weight (D) over 5 consecutive days in female WT mice (n = 15) and female GSK mice (n = 16) during the light and dark phase. ^#p < 0.05 (genotype × phase effect); *p < 0.01 and **p < 0.001 (overall genotype effect).

Figure 2.

Brain region wet weights (means ± SEM; **p < 0.01, t tests) (A) and DNA synthetic rate as an index of proliferation rate expressed as a percentage [³H]thymidine incorporation in the hippocampus and SVZ (no differences between groups) (B) of male WT mice (n = 8) and male GSK mice (n = 7). OB, Olfactory bulb.

Figure 3.

Locomotor activity (means ± SEM) in an open field during 1 h (in 5 min bins). A, Male GSK mice (n = 15) traveled a longer distance than male WT mice (n = 17). B, Percentage habituation of the distance traveled. **p < 0.01 (overall genotype effect); ^‡p < 0.05 (genotype × time bin effect).

Figure 4.

Startle responses (means ± SEM) in the noise burst intensity function test with 20 ms startle stimuli of 100, 105, and 110 dB (A), immobility time in the FST for 6 min (B), and BDNF protein concentrations in homogenates of the hippocampus (C) of male WT mice (n = 4), female WT mice (n = 13), male GSK mice (n = 10), and female GSK mice (n = 11). ^†p < 0.05 and ^††p < 0.01 (overall sex effect); **p < 0.01 (overall genotype effect).