Phenotype Distinctions in Mice Deficient in the Neuron-Specific α3 Subunit of Na,K-ATPase: Atp1a 3tm1Ling/+ and Atp1a3 +/D801Y

Abstract

ATP1A3 is a Na,K-ATPase gene expressed specifically in neurons in the brain. Human mutations are dominant and produce an unusually wide spectrum of neurological phenotypes, most notably rapid-onset dystonia parkinsonism (RDP) and alternating hemiplegia of childhood (AHC). Here we compared heterozygotes of two mouse lines, a line with little or no expression (Atp1a3^tm1Ling/+) and a knock-in expressing p.Asp801Tyr (D801Y, Atp1a3 ^+/D801Y). Both mouse lines had normal lifespans, but Atp1a3 ^+/D801Y had mild perinatal mortality contrasting with D801N mice (Atp1a3 ^+/D801N), which had high mortality. The phenotypes of Atp1a3^tm1Ling/+ and Atp1a3 ^+/D801Y were different, and testing of each strain was tailored to its symptom range. Atp1a3^tm1Ling/+ mice displayed little at baseline, but repeated ethanol intoxication produced hyperkinetic motor abnormalities not seen in littermate controls. Atp1a3 ^+/D801Y mice displayed robust phenotypes: hyperactivity, diminished posture consistent with hypotonia, and deficiencies in beam walk and wire hang tests. Symptoms also included qualitative motor abnormalities that are not well quantified by conventional tests. Paradoxically, Atp1a3 ^+/D801Y showed sustained better performance than wild type on the accelerating rotarod. Atp1a3 ^+/D801Y mice were overactive in forced swimming and afterward had intense shivering, transient dystonic postures, and delayed recovery. Remarkably, Atp1a3 ^+/D801Y mice were refractory to ketamine anesthesia, which elicited hyperactivity and dyskinesia even at higher dose. Neither mouse line exhibited fixed dystonia (typical of RDP patients), spontaneous paroxysmal weakness (typical of AHC patients), or seizures but had consistent, measurable neurological abnormalities. A gradient of variation supports the importance of studying multiple Atp1a3 mutations in animal models to understand the roles of this gene in human disease.

Keywords: ATP1A3; Na,K-ATPase; disease mutation; motor testing; mouse model.

Visual Abstract

Figure 1.

Body weights of Atp1a3^tm1Ling/+ and Atp1a3^+/D801Y. A, Weights of Atp1a3^tm1Ling/+ mice were recorded at the time of motor testing, from 4 to 6 months of age. No differences were seen in males between WT (n = 8) and heterozygotes (het; n = 13) or females (n = 10 for both WT and het). Solid lines represent linear regressions for wild type. Dashed lines represent linear regressions for Atp1a3^tm1Ling/+, not significant by analysis of covariance (ANCOVA). B, Weights of Atp1a3^+/D801Y mice from 5 weeks to 6 months of age. Green symbols indicate wild type, ochre symbols indicate Atp1a3^+/D801Y. Solid lines represent linear regressions for wild type. Dashed lines represent linear regressions for Atp1a3^+/D801Y. Analysis of covariance (ANCOVA) was used to compare regression lines between WT and Atp1a3^+/D801Y. While slopes were not different in females (WT = 0.004605; Atp1a3^+/D801Y= 0.009857; F = 0.1559; DFn = 1, DFd = 50; p = 0.06946), a statistically significant difference (p = 0.0356) was observed between WT and Atp1a3^+/D801Y slopes in males (0.06053 and 0.04080, respectively; F = 4.585; DFn = 1, DFd = 72).

Figure 2.

Expression levels of α3 and ATPase activity in Atp1a3^tm1Ling/+ and Atp1a3^+/D801Y mice. A, Relative expression level of the α3 subunit was determined by Western blots of crude homogenates. Each datapoint is a within-experiment ratio of observed antibody stain for mutant relative to WT for Atp1a3^tm1Ling/+ (tm1Ling) and Atp1a3^+/D801Y (D801Y) mice. Statistical analysis was performed with Student's t test. p values are as follows. For the ratio of each mutant to WT, p < 0.0001 (gray asterisks); for comparison between mutant strains: protein expression p = 0.18, coincidentally the same. B, The maximal ouabain-sensitive ATP hydrolysis was similarly measured in preparations from Atp1a3^tm1Ling/+ and Atp1a3^+/D801Y mice. For reduction of ATPase activity relative to WT, p ≤ 0.0001 for both strains. In comparing the strains, the greater reduction in Atp1a3^+/D801Y was also p ≤ 0.0001. C, α3 subunit Western blots were also carried out loading equal amounts of protein (left) or equal amounts of activity (right). The protein concentrations and ATPase activities for each preparation were determined together. We then prepared a single 80 µl of complete LDS (SDS) sample with equal amounts of protein for each preparation. On the equal protein side of the gel, 20 µl of these were loaded into gel wells. On the equal ATPase activity side, a calculated volume of each master sample was added corresponding to equal amounts of measured ATPase activity. The samples on both sides of the gel thus came from the same tubes.

Figure 3.

Activity and motor function of Atp1a3^tm1Ling/+ mice. Baseline open-field chamber activity was evaluated in WT (n = 6) and Atp1a3^tm1Ling/+ (n = 6), by quantifying optical beam breaks during a 1 h test (A). No difference in activity was observed between WT and Atp1a3^tm1Ling/+. Data are shown as mean ± SD (shaded areas). Statistical analysis was performed by two-way ANOVA, followed by uncorrected Fisher's LSD test. The same activity protocol was repeated after administration of 100 mg/kg ketamine (B). No significant difference was recorded either in the duration of full immobility at ∼5 min or in the rate of recovery. Motor coordination and balance was quantified by measuring the time necessary to cross the beam (C) and the number of paw slips (D) that occur in the process. The accelerating rod also did not show difference between Atp1a3^tm1Ling/+ and WT under baseline conditions (E) or 1 month after stress elicited by forced swimming for 15 min every day for 5 d (F).

Figure 4.

Atp1a3^tm1Ling/+ mice exhibit normal running parameters. Running on metered wheels was used to look for deficits in the Atp1a3^tm1Ling/+. Data were collected for 9 consecutive days (n = 16 mice per group). There was no effect of genotype on two parameters, kilometer run per day (A), and speed in km/hour (C). Time (min) spent running per day (B) showed a statistically significant difference between the two groups (p = 0.0472). Data are shown as mean ± SEM.

Figure 5.

Ethanol stress reveals neurological vulnerability in Atp1a3^tm1Ling/+ mice. WT and Atp1a3^tm1Ling/+ mice were held in a vapor chamber (layout in A) and exposed for 6 h every weekday for 2 weeks. B, C, The mutants are either unrecovered (B) or experiencing repetitive dystonic or myoclonic movements (C). The wild type in B is labeled with an asterisk. See also Movie 1.

Figure 6.

Atp1a3^+/D801Y hyperactivity and paradoxical response to ketamine. Open-field activity was assessed for 60 min in WT and Atp1a3^+/D801Y mice at baseline (A; n = 6 per group) and upon administration of 60 mg/kg (B) and 120 mg/kg (C) ketamine (n = 5 per group for both doses). Atp1a3^+/D801Y mice displayed a hyperactive behavior relative to WT at most of the time points tested (p < 0.0001). A subanesthetic dose (B) produced robust rebound-associated hyperactivity. At high anesthetic dose (C), Atp1a3^+/D801Y mice never lost consciousness and exhibited dyskinesias. See also Movie 3. Data are shown as mean ± SD (shaded areas). Statistical analysis was performed by two-way ANOVA (or RELM mixed-model for data relative to 120 mg/kg dose), followed by uncorrected Fisher's LSD post hoc test. Asterisks denote statistically significant differences between groups at different time points. Individual time point t tests that were significant had p values as follows: baseline: at 5’ = 0.0341; at 15 min = 0.0192; at 20 min = 0.0182; at 30 min = 0.0395; at 40 min = 0.0108; at 45 min = 0.0155; at 50 min = 0.0322. 60 mg/kg ketamine: at 5’ = 0.0004; at 10’ = 0.0018; at 50’ = 0.0385. 120 mg/kg ketamine: at 5’ = 0.0036; at 10’ = 0.0426; at 15’ = 0.0050; at 20’ = 0.0009; at 25’ = 0.0002; at 30’ = 0.0499.

Figure 7.

Beam cross and performance in Atp1a3^+/D801Y. Motor coordination and balance were assessed by measuring performance on the balance beam. This test, with females only, was conducted a total of three times for each mouse and triplicates were averaged. Time to cross the beam (A) and number of paw slips (B) were quantified. Statistical analysis was performed by unpaired Student's t test. p values are as follows: p = 0.0195 and 0.0136 for time to cross beam and foot slip, respectively.

Figure 8.

Weakness of the Atp1a3^+/D801Y heterozygotes on the narrow rod. A 3 mm rod suspended between two posts was a more stringent challenge than the beam. Each mouse was placed on the rod and given the opportunity to grasp it. Wild types clung to the rod and inched their way toward a post. Atp1a3^+/D801Y (D801Y) mice had great difficulty with their hindlimbs and rarely stayed on for the full 60 s. A, Sequential photographs; WT is on the left. B, Quantification for a cohort with n = 14 per group. Data were analyzed with unpaired t test with Welch's correction for unequal variance (p ≤ 0.0001; <0.0002; and <0.0001, respectively, for trials 1, 2, and 3).

Figure 9.

Atp1a3^+/D801Y mice exhibit exceptional performance on the rotarod. A, The rotarod (average of two trials carried out 3 d apart) showed increased baseline activity in Atp1a3^+/D801Y (D801Y) mice compared with WT. The test was terminated at 180 s. B, In a separate cohort, mice underwent testing again at baseline and then were stressed by forced swimming for 15 min. Then they were tested on the rotarod again at 7, 30, and 90 d later. Statistical analysis was performed by two-way ANOVA followed by Fisher's LSD test. Asterisks indicate statistically significant differences between WT and het (p < 0.0001 for all comparisons).

Figure 10.

Hyperactivity of Atp1a3^+/D801Y in forced swimming; impairments during recovery. A, C57BL/6 mice reach ∼70% immobility at 4–5 min, and so the time spent resting between 5 and 15 min was scored from video recordings. (The Atp1a3^+/D801Y mice that swam continuously were in unfilmed replicate experiments.) B, C, The tremor test detected an increase in amplitude over baseline in both males and females, but no significant difference in frequency.