Transgenic rescue of phenotypic deficits in a mouse model of alternating hemiplegia of childhood

Abstract

Missense mutations in ATP1A3 encoding Na(+),K(+)-ATPase α3 are the primary cause of alternating hemiplegia of childhood (AHC). Most ATP1A3 mutations in AHC lie within a cluster in or near transmembrane α-helix TM6, including I810N that is also found in the Myshkin mouse model of AHC. These mutations all substantially reduce Na(+),K(+)-ATPase α3 activity. Herein, we show that Myshkin mice carrying a wild-type Atp1a3 transgene that confers a 16 % increase in brain-specific total Na(+),K(+)-ATPase activity show significant phenotypic improvements compared with non-transgenic Myshkin mice. Interventions to increase the activity of wild-type Na(+),K(+)-ATPase α3 in AHC patients should be investigated further.

Keywords: Alternating hemiplegia; Atp1a3; Mice; Na+,K+-ATPase α3; Transgenic rescue.

Fig. 1

Body weight (% of +/+; mean ± SEM) of male (n = 6/genotype) and female (n = 6/genotype) Myk/+, Myk/+/Tg, and Tg/+ mice at 4 and 8 weeks of age. For body weight at 4 weeks, main effects of Atp1a3 genotype (F(1, 42) = 8.41, p < 0.01), sex (F(1, 42) = 7.40, p < 0.01), and Atp1a3 genotype by Tg genotype interaction (F(1, 42) = 12.69, p < 0.001) were observed. For body weight at 8 weeks, main effects of Atp1a3 genotype (F(1, 42) = 14.18, p < 0.001), Tg genotype (F(1, 42) = 8.84, p < 0.01), sex (F(1, 42) = 82.72, p < 0.0001), and Atp1a3 genotype by Tg genotype interaction (F(1, 42) = 10.83, p < 0.01) were observed. *p < 0.05; **p < 0.01; ***p < 0.001 compared with +/+ mice

Fig. 2

Balance beam performance of +/+, Myk/+, Myk/+/Tg, and Tg/+ mice (n = 12/genotype). a Number of hind foot slips (mean ± SEM) and b traversal time (s; mean ± SEM) when traversing a narrow beam. For the number of foot slips, main effects of Atp1a3 genotype (F(1, 42) = 14.11, p < 0.001) and Atp1a3 genotype by Tg genotype interaction (F(1, 42) = 5.02, p < 0.05) were observed. For traversal time, main effects of Atp1a3 genotype (F(1, 42) = 22.19, p < 0.0001), Tg genotype (F(1, 42) = 5.52, p < 0.05), and Atp1a3 genotype by Tg genotype interaction (F(1, 42) = 9.54, p < 0.01) were observed. ***p < 0.001; ****p < 0.0001 compared with +/+ mice

Fig. 3

Water maze and fear-conditioning performance of +/+, Myk/+, Myk/+/Tg, and Tg/+ mice (n = 12/genotype). a Escape latency (s; mean ± SEM). Main effects of Atp1a3 genotype (F(1, 42) = 77.09, p < 0.0001) and Tg genotype (F(1, 42) = 7.79, p < 0.01) were observed for latency. b Swim path length (m; mean ± SEM). Main effects of Atp1a3 genotype (F(1, 42) = 67.20, p < 0.0001) and Tg genotype (F(1, 42) = 9.10, p < 0.01) were observed for path length. c Thigmotaxis (% time; mean ± SEM). Main effects of Atp1a3 genotype (F(1, 42) = 53.10, p < 0.0001) and Tg genotype (F(1, 42) = 4.59, p < 0.05) were observed for thigmotaxis. d Freezing levels (% time; mean ± SEM) of mice when test-naïve (baseline) and when returned to the chamber 24 h after training (context). For baseline freezing, a main effect of Tg genotype (F(1, 42) = 11.60, p < 0.01) was observed. For context freezing, main effects of Atp1a3 genotype (F(1, 42) = 5.33, p < 0.05), Tg genotype (F(1, 42) = 7.40, p < 0.01), and Atp1a3 genotype by Tg genotype interaction (F(1, 42) = 10.01, p < 0.01) were observed. **p < 0.01; ***p < 0.001; ****p < 0.0001 compared with +/+ mice. †p < 0.05 compared with Myk/+ mice. ##p < 0.01; ####p < 0.0001 compared with baseline freezing for each genotype. n.s. not significant