Evidence that increased Kcnj6 gene dose is necessary for deficits in behavior and dentate gyrus synaptic plasticity in the Ts65Dn mouse model of Down syndrome

Abstract

Down syndrome (DS), trisomy 21, is caused by increased dose of genes present on human chromosome 21 (HSA21). The gene-dose hypothesis argues that a change in the dose of individual genes or regulatory sequences on HSA21 is necessary for creating DS-related phenotypes, including cognitive impairment. We focused on a possible role for Kcnj6, the gene encoding Kir3.2 (Girk2) subunits of a G-protein-coupled inwardly-rectifying potassium channel. This gene resides on a segment of mouse Chromosome 16 that is present in one extra copy in the genome of the Ts65Dn mouse, a well-studied genetic model of DS. Kir3.2 subunit-containing potassium channels serve as effectors for a number of postsynaptic metabotropic receptors including GABAB receptors. Several studies raise the possibility that increased Kcnj6 dose contributes to synaptic and cognitive abnormalities in DS. To assess directly a role for Kcnj6 gene dose in cognitive deficits in DS, we produced Ts65Dn mice that harbor only 2 copies of Kcnj6 (Ts65Dn:Kcnj6++- mice). The reduction in Kcnj6 gene dose restored to normal the hippocampal level of Kir3.2. Long-term memory, examined in the novel object recognition test with the retention period of 24h, was improved to the level observed in the normosomic littermate control mice (2N:Kcnj6++). Significantly, both short-term and long-term potentiation (STP and LTP) was improved to control levels in the dentate gyrus (DG) of the Ts65Dn:Kcnj6++- mouse. In view of the ability of fluoxetine to suppress Kir3.2 channels, we asked if fluoxetine-treated DG slices of Ts65Dn:Kcnj6+++ mice would rescue synaptic plasticity. Fluoxetine increased STP and LTP to control levels. These results are evidence that increased Kcnj6 gene dose is necessary for synaptic and cognitive dysfunction in the Ts65Dn mouse model of DS. Strategies aimed at pharmacologically reducing channel function should be explored for enhancing cognition in DS.

Keywords: Cognition; Down syndrome critical region; Fluoxetine; Genotype-phenotype relationship; Kcnj6; Kir3.2; Learning; Locomotor activity; Long-term potentiation; Mouse models; Novel object recognition; Synaptic plasticity; Ts65Dn; Y-maze.

Fig. 1

Changes in body weight. The body weight was reduced in both Ts65Dn:Kcnj6+++ and Ts65Dn:Kcnj6++− vs. 2N:Kcnj6++ mice at all ages. There was no difference between the body weight of Ts65Dn:Kcnj6+++ vs. Ts65Dn:Kcnj6++− in young animals (1–4 months). However, the body weight was increased in older (5–6 mo and 9–12 mo) Ts65Dn:Kcnj6++− vs. Ts65Dn:Kcnj6+++ mice. The animal numbers and other statistical parameters are given in Table 1.

Fig. 2

Hippocampal levels of proteins. A. Level of Kir3.1 was not altered, while level of Kir3.2 was increased in Ts65Dn:Kcnj6+++ (n = 6) vs. 2N:Kcnj6++ (n = 5) samples. B. There was no difference in the levels of either Kir3.1 or Kir3.2 in Ts65Dn:Kcnj6++− (n = 5) vs. 2N:Kcnj6++ (n = 6) samples. C. Kir3.1 level was not different, but Kir3.2 was increased in Ts65Dn:Kcnj6+++ (n = 5) vs. Ts65Dn:Kcnj6++− (n = 6) mice.

Fig. 3

Spontaneous locomotion. A. Ambulatory distance was significantly increased in both Ts65Dn:Kcnj6+++ (n = 12) and Ts65Dn:Kcnj6++− (n = 14) vs. 2N:Kcnj6++ (n = 19) mice. No difference between Ts65Dn:Kcnj6+++ vs. Ts65Dn:Kcnj6++− mice was observed. B. Ambulatory time was also increased in both Ts65Dn:Kcnj6+++ and Ts65Dn:Kcnj6++− mice. C, D. Ambulatory distance and ambulatory time spent on periphery were greater in Ts65Dn:Kcnj6+++ vs. 2N:Kcnj6++ mice, but not different in Ts65Dn:Kcnj6++− vs. 2N:Kcnj6++ animals.

Fig. 4

Working memory and exploratory activity in the Y-maze. A. The rate of spontaneous alternations was reduced in both Ts65Dn:Kcnj6+++ (n = 12) and Ts65Dn:Kcnj6++− (n = 14) vs. 2N:Kcnj6++ (n = 19) mice. Reduction of Kcnj6 gene dose had no effect on performance of Ts65Dn mice (p = 0.84 for Ts65Dn:Kcnj6+++ vs. Ts65Dn:Kcnj6++−). B. Number of ‘arm entries’ during the Y-maze test was greater in Ts65Dn:Kcnj6+++ vs. 2N:Kcnj6++ mice. The number of arm entries was reduced in Ts65Dn:Kcnj6++− mice to the level of 2N:Kcnj6++ controls.

Fig. 5

Long-term memory. Performance in the novel object recognition test with a retention period of 24 h. A: Time for object exploration was not different between genotypes. B: The discrimination index was reduced in Ts65Dn:Kcnj6+++ (n = 12) vs. 2N:Kcnj6++ (n = 19) mice, but it was increased in Ts65Dn:Kcnj6++− (n = 14) mice to the level observed in 2N:Kcnj6++ control mice.

Fig. 6

Synaptic plasticity: effect of the Kcnj6 gene dose on STP and LTP in DG. A. Averaged changes in the initial slope of field EPSP during the experiment. B. Averaged values for STP (1–15 min) and LTP (30–60 min). STP and LTP were reduced in Ts65Dn:Kcnj6+++ (n = 8) vs. 2N:Kcnj6++ (n = 12) slices. Both parameters were normalized in slices from Ts65Dn:Kcnj6++− (n = 11) mice.

Fig. 7

Synaptic plasticity: Effect of fluoxetine on STP and LTP in DG. A. Averaged changes in the initial slope of field EPSP during the experiment. Application of fluoxetine or vehicle marked by the black line under the graph. B. Averaged values for STP (1–15 min) and LTP (30–60 min). Both STP and LTP were reduced in Ts65Dn (n = 14) vs. 2N (n = 13) slices. Fluoxetine increased both parameters in Ts65Dn slices (n = 10) to control levels, but did not affect these parameters in 2N slices (n = 7).