Mice with deficient BK channel function show impaired prepulse inhibition and spatial learning, but normal working and spatial reference memory

Abstract

Genetic variations in the large-conductance, voltage- and calcium activated potassium channels (BK channels) have been recently implicated in mental retardation, autism and schizophrenia which all come along with severe cognitive impairments. In the present study we investigate the effects of functional BK channel deletion on cognition using a genetic mouse model with a knock-out of the gene for the pore forming α-subunit of the channel. We tested the F1 generation of a hybrid SV129/C57BL6 mouse line in which the slo1 gene was deleted in both parent strains. We first evaluated hearing and motor function to establish the suitability of this model for cognitive testing. Auditory brain stem responses to click stimuli showed no threshold differences between knockout mice and their wild-type littermates. Despite of muscular tremor, reduced grip force, and impaired gait, knockout mice exhibited normal locomotion. These findings allowed for testing of sensorimotor gating using the acoustic startle reflex, as well as of working memory, spatial learning and memory in the Y-maze and the Morris water maze, respectively. Prepulse inhibition on the first day of testing was normal, but the knockout mice did not improve over the days of testing as their wild-type littermates did. Spontaneous alternation in the y-maze was normal as well, suggesting that the BK channel knock-out does not impair working memory. In the Morris water maze knock-out mice showed significantly slower acquisition of the task, but normal memory once the task was learned. Thus, we propose a crucial role of the BK channels in learning, but not in memory storage or recollection.

Figure 1. Hearing thresholds measured with auditory brainstem response.

(A) Click thresholds of WT and BKα^−/− mice. Mean and S.D. (filled symbols). Single animal thresholds are shown as smaller, open symbols beside either group mean. The difference in threshold (3 dB) was statistically not significant. (B) Pure tone thresholds (mean and S.D.) as function of stimulus frequency for WT (black circles) and BKα^−/− mice (grey squares). Threshold curves were just significantly different, based on significant elevated thresholds at 32 kHz (*:p’ = 0.048, 1-sided t-test without alpha correction) and a drop out of threshold for BKα-/- mice at 45 kHz (star: p = 0.0262, Fisher Exact Probability Test).

Figure 2. Muscle functions.

(A) Muscular tremor was measured employing data recorded during the startle testing. During periods of no acoustic stimulation the noise of the recorded signal was significantly higher in the BKα^−/− mice. (B) In addition to the tremor the BKα^−/− mice also showed significant reduced muscle strength judging by their front limb grip force.

Figure 3. Gait strait analysis.

(A) Examples of the paw print and their timing for WT (left) and BKα^−/− mice (right). The bars indicating the timing of each single paw on the floor (from left to right: left hind limb, left front limb, right front limb, right hind limb). (B) Stride length for all paws is reduced in the BKα^−/− mice, but (C) no significant differences were found for the paw print areas. (D) The inter-limb coordination as measured with the regularity index in reduced in BKα^−/− mice. The dashed line shows a regularity index of 100% indicating perfect inter-limb coordination.

Figure 4. Locomotion.

(A) BKα^−/− mice show normal locomotion in the open field box as well as (B) normal rearing activity.

Figure 5. Prepulse inhibition (PPI) of the acoustic startle response.

(A) On the first day of testing no significant differences in prepulse inhibition were found between genotypes for all tested inter-stimulus intervals. (B) Across days the WT and BKα^+/− mice showed improvement in PPI whereas the BKα^−/− mice did not. The average PPI for all tested ISI is shown, since no effect of ISI was reported by the repeated measurements ANOVA. The right panel shows the average difference in PPI between the fifth and first day of testing. Differences were calculated for each single animal and then averaged for genotypes.

Figure 6. Spontaneous alternation in the Y-maze.

( A) Spontaneous alternations are presented as percent of total arm entries. There was no difference between the genotypes. (B) The number of total arm entries is significantly increased in the BKα^−/− mice.

Figure 7. Spatial learning and memory in the Morris water maze.

(A) The average escape latency improved across the training trails in all genotypes, but especially on days one to three, the BKα^−/− mice improve less within the training sessions. Note that on the last training trail on day 4 they caught up and performed as well as the WT mice. (B) Escape latency as a function of days of experience with the task. The average escape latency for each day was fitted with a one component exponential function for each genotype and time constants were derived from the resulting functions. The BKα^−/− mice have a longer time constant then the WT and the BKα^+/− mice. (C) Performance on the probe day. All mice spend a significant higher amount of time in the quadrant where previously the platform was hidden (target, solid dark grey) than in any other quadrant. The dashed line indicates 25% chance criterion.