Deletion of JMJD2B in neurons leads to defective spine maturation, hyperactive behavior and memory deficits in mouse

Abstract

JMJD2B is a histone demethylase enzyme that regulates gene expression through demethylation of H3K9me3. Although mutations of JMJD2B have been suggested to be responsible for neurodevelopmental disorders, the function of JMJD2B in the central nervous system (CNS) remains to be elucidated. Here we show that JMJD2B has a critical role in the development of the CNS. We observed JMJD2B expression, which was especially strong in the hippocampus, throughout the CNS from embryonic periods through adulthood. We generated neuron-specific JMJD2B-deficient mice using the cre-loxP system. We found an increase in total spine number, but a decrease in mature spines, in the CA1 region of the hippocampus. JMJD2B-deficient mice exhibited hyperactive behavior, sustained hyperactivity in a novel environment, deficits in working memory and spontaneous epileptic-like seizures. Together these observations indicate that JMJD2B mutant mice display symptoms reminiscent of neurodevelopmental disorders. Our findings provide evidence for the involvement of histone demethylation in the formation of functional neural networks during development.

Figure 1

Expression of Jmjd2b in mouse. (a) In situ hybridization revealing expression of Jmjd2b mRNA in E17 mouse (left: antisense, right: sense; n=1). (b) In situ hybridization for the detection of Jmjd2b mRNA in the brain at the indicated stages (n=1 per age). (c) In situ hybridization for Jmjd2b mRNA, followed by immunostaining with NeuN (left) and GFAP (right; n=1). Scale bars: (a) 2 mm, (b) 1 mm, (c) 200 μm.

Figure 2

Generation of neuron-specific Jmjd2b mutant mice. (a) Relative expression level of Jmjd2b and Jmjd2c. The expression level was normalized to Gapdh. **P<0.01; NS, not significant. (WT, KO: n=3) Error bars represent s.e.m. (b) Expression of Jmjd2b mRNA in WT and KO mice. P0 mice were used for in situ hybridization. (WT, KO: n=1). (c) Nissl staining showing the gross morphology of the brain in the two genotypes. (WT, KO: n=3). KO, JMJD2B mutant mice; WT, wild-type mice. Scale bars: (b) 1 mm, (c) 50 μm.

Figure 3

Structural analysis of dendrites and spines of hippocampal CA1 of JMJD2B mutant mice.(a) Representative Golgi staining image in the hippocampal CA1 region. (b and c) Representative Golgi staining images of dendritic spines in the hippocampal CA1 region in each genotype. (b′ and c′) Magnified images of b and c. (d) The total spine numbers in the CA1 region of the two genotypes. (e) The percentage of mature mushroom, filopodia, thin and stubby-type spines of CA1 pyramidal neurons. *P<0.05; **P<0.01. WT: n=6, KO: n=5 (a–e). Error bars represent s.e.m. (f; left) Representative images of hippocampal neurons cultured for 7 days, and stained with MAP2. (g) Sholl analysis of hippocampal neurons of each genotype. A concentric circle was drawn every 10 μm from a central focus on the neuronal soma and the number of intersections were counted. NS, not significant. (WT, KO: n=4). (h) The number of dendrites extending from the soma. (WT, KO: n=3). (i) Total dendritic length measured from the soma of cultured hippocampal neurons. Total dendritic length measured from the soma of cultured hippocampal neurons. (WT, KO: n=4). WT, wild-type mice; KO, JMJD2B mutant mice. Scale bars: (a) 10 μm, (b and c) 10 μm, (b'and c') 1 μm, (f) 20 μm. Error bars represent s.e.m. (d, e, h, i).

Figure 4

JMJD2B mutant mice reveal hyperactivity in the open-field test.(a) The graph shows the distance moved during the indicated time period. (b) The graph shows the temporal change in the rate of activity (speed). (c) The graph shows the distance moved in the first 10 min versus the last 10 min (50 to 60 min) of assessment. (d) The graph shows temporal changes in rearing number. (WT: n=12, KO: n=10). WT, wild-type mice; KO, JMJD2B mutant mice. *P<0.05; **P<0.01. Error bars represent s.e.m.

Figure 5

JMJD2B mutant mice show deficits in short-term memory. (a) The result of the social interaction test examining the social affinity to another mouse. The graph shows the relative time of exploration of the novel object. (WT: n=13 KO: n=11). (b) Novel object recognition test to examine the interest in unknown object. (WT: n=13 KO: n=11). (c and d) Y-maze test results examining short-term memory in the two genotypes. (WT: n=10; KO: n=11). (c) The graph shows the percentage of correct entries. (d) The graph shows the number of total entries. KO, JMJD2B mutant mice; NS, not significant; WT, wild-type mice. Error bars represent s.e.m. **P<0.01.