Conditional Knockout of Ankyrin-G In The Mouse Forebrain Results In Bipolar-Like Behavior

Abstract

Background To date, few animal models of bipolar disorder have been developed, and fewer still are based on genetic loci despite the fact that genome wide association studies (GWAS) have now identified a number of candidate genes. Across studies, one of the most consistent findings is in ANK3, which encodes Ankyrin-G, a protein that organizes sodium and potassium channels at the axon initial segments (AIS). To explore the potential pathophysiological role of Ankyrin-G in bipolar disorder, we generated a mouse model with conditional deletion of all major isoforms of Ank3 in the adult forebrain. Methods Ank3 flox mice (Ank3 flox/flox) were crossed with Camk2a-Cre mice, which begin to express Cre in the forebrain at post-natal day 19. Beginning at 3 months of age, the behaviors of conditional knockout mice and control littermates were assessed in a 24 hour locomotor activity test, open field, elevated plus maze, forced swim test, and prepulse inhibition test. In a second cohort, behavioral effects of two mood-stabilizers, lithium and valproic acid, as well as the stimulant methylphenidate were assessed. In separate cohorts, mice were exposed to 14 day cycles of social defeat stress, a commonly used paradigm that results in a depression-like phenotype among wild-type mice. Finally, the behavior of heterozygous mice was characterized at baseline and then after exposure to social defeat stress and treatment with the antidepressant fluoxetine. Results Conditional knockout of Ank3 (cKO) resulted in striking hyperactivity, sleep-cycle disruption, and decreased anxiety-like behaviors. Ank3 cKO mice were indistinguishable from controls in prepulse inhibition, where deficits are often found in models of schizophrenia. Consistent with a mania-like phenotype, the hyperactivity of Ank3 cKO mice was normalized by treatment with lithium and valproic acid. When treated with methylphenidate, both Ank3 cKO and control mice displayed increased locomotor activity. After 14 days of social defeat, Ank3 and control mice displayed a depression-like phenotype. Consistent with a bipolar depression-like phenotype, Ank3 cKO mice expose to social defeat were indistinguishable from socially defeated controls. When allowed to recover for 14 days and then re-exposed to social defeat, Ank3 cKO mice, but not controls, displayed evidence of ‘cycling’ between mania and depression-like behaviors. Finally, we found that Ank3 heterozygous mice are indistinguishable from controls at baseline. However, Ank3 heterozygous mice appear to be more sensitive to both social defeat stress and the antidepressant fluoxetine. Discussion Overall cKO of Ank3, a genetic loci that has repeatedly been implicated in human bipolar disorder and is involved in organization of the axon initial segment, recapitulates major aspects of both bipolar mania and bipolar depression. We believe that this novel genetic model may provide a better understanding of the role of of Ankyrin-G in the brain, further insights into the molecular biology of bipolar disorder, and allow for the development of novel therapeutic approaches.