MicroRNA-128 governs neuronal excitability and motor behavior in mice

Abstract

The control of motor behavior in animals and humans requires constant adaptation of neuronal networks to signals of various types and strengths. We found that microRNA-128 (miR-128), which is expressed in adult neurons, regulates motor behavior by modulating neuronal signaling networks and excitability. miR-128 governs motor activity by suppressing the expression of various ion channels and signaling components of the extracellular signal-regulated kinase ERK2 network that regulate neuronal excitability. In mice, a reduction of miR-128 expression in postnatal neurons causes increased motor activity and fatal epilepsy. Overexpression of miR-128 attenuates neuronal responsiveness, suppresses motor activity, and alleviates motor abnormalities associated with Parkinson's-like disease and seizures in mice. These data suggest a therapeutic potential for miR-128 in the treatment of epilepsy and movement disorders.

Figure 1. miR-128 controls motor behavior in mice

(A) Deficiency in miR-128-2 causes hyperactivity and premature death in mice. (Left panel) Motor activity was determined by measuring total horizontal distance in a 60 min open field assay (n=23 and 12). (Right panel) The lifespans of miR-128-2^−/− mice and littermate controls are shown (n= 20 and 46). (B, C) miR-128 deficiency causes fatal seizures that can be prevented by anti-convulsant treatment (B) Representative display of spontaneous tonic-clonic seizure episodes in miR-128-2^−/− (black) or Camk2a-cre; miR-128-2^fl/fl mice (red) during a 22-day observation period. (C) The lifespans of control miR-128-2^−/− (dotted line, as shown in A) or sodium valproate-treated (red, n=11) miR-128-2^−/− mice are shown (D) Deficiency in miR-128 in postnatal neurons causes hyperactivity and fatal epilepsy. Motor activity and survival rates of Camk2a-cre; miR-128-2^fl/fl mice (n=21 and 25) and littermates (n=8 and 47) are shown. (E) Ectopic expression of miR-128 normalizes hyper-locomotion and prevents death of Camk2a-cre; miR-128-2^fl/fl mice. Motor activity in Camk2a-cre; miR-128-2^fl/fl; Rosa-miR-128 (n=4, blue) and wild-type mice (n=10, gray) are shown. The lifespans of Camk2a-cre; miR-128-2^fl/fl mice in the presence (n=4, blue) or absence (n=9, black) of ectopic miR-128 expression are shown. (F) miR-128 deficiency in D1-neurons causes hyperactivity and fatal epilepsy. Motor activity (n=26 and 42) and lifespans (n= 16 and 28) of mice with a D1-neuron specific miR-128 deficiency or control mice are shown. Error bars show s.e.m., Welch’s t-test, non-significant (ns), * p≤0.05, ** p≤0.01, *** p≤0.001. Kaplan-Meier graph shows survival curves of mutant and littermate control mice, *** p≤0.001, log rank tests.

Figure 2. miR-128 controls signaling protein expression and activation of the ERK signaling network in neurons

(A) Venn diagram shows the RISC-associated mRNA targets of miR-128 (red) and mRNAs that are up-regulated in miR-128 deficient D1-neurons (blue). The overlapping 154 mRNAs are considered as direct miR-128 targets. (B) (Left) Gene ontology annotations of the 154 miR-128 target genes are shown with pathway enrichment presented as −log10 (p-value). The dotted orange line indicates p= 0.05. (Right) The components of the ERK1/2 network (p=10⁻⁴⁶, right-tailed Fisher’s exact test) that are directly targeted by miR-128 are indicated in solid grey. (C) Expression levels of miR-128-targeted ERK regulators in the striatum of Drd1a-cre; miR-128-2^fl/fl and littermate controls were analyzed by Western blotting (n=4 each). (D) Increased ERK2 phosphorylation in the striatum of mice with D1-neuron-specific miR-128 deficiency. Representative Western blot analysis of ERK1/2 phosphorylation in the striatum of control and Drd1a-cre; miR-128-2^fl/fl mice is shown; bar graphs display phospho-ERK/ERK protein ratios (n=4). Error bars show s.e.m., Welch’s t-test, * p≤0.05, ** p≤0.01.

Figure 3. miR-128 controls D1-neuron excitability and responsiveness to dopamine. (A, B) miR-128 regulates D1-neuron dendritic excitability and number of spines

(A) Single action potentials were generated in the soma and action potential invasion was calculated by dividing the distal calcium signal by the maximum proximal calcium signal per cell (n=4 cells, 11–21 shafts per group). Mann-Whitney nonparametric test, *** p≤0.001 (B) Representative maximum intensity projection images of distal dendrites in control and mutant D1-neurons are shown. Boxplots display population spine densities (n=10–11 cells per group). Mann-Whitney nonparametric test, error bars show 90th percentile interval, * p≤0.05. (C–E) miR-128 regulates motor response, ERK2 phosphorylation, and immediate early gene (IEG) induction upon dopamine D1 receptor (Drd1) activation in D1-neurons. (C) Motor activity of Drd1a-cre; miR-128-2^fl/fl and control mice (n=25 and 30) was evaluated in an open-field chamber. Saline and 3mg/kg Drd1 agonist SKF81297 were injected i.p. at 10 and 20 minute, respectively. (D) ERK2 phosphorylation was quantified by Western blotting of striatal lysates derived from Drd1a-cre; miR-128-2^fl/fl and control mice that received saline or D1-agonist SKF81297 injection (n=5 each). Bar graph displays the ratio of phospho-ERK2 to total ERK2 expression. (E) IEG and D1-neuron-expressed Darpp32 gene expression levels were measured by qRT-PCR of D1-neuron specific polyribosome-associated mRNAs purified from saline or SKF81297 treated Drd1a-TRAP; Drd1a-cre; miR-128-2^fl/fl and control mice (n=5 each). Error bars display s.e.m., Welch’s t-test, * p≤0.05, ** p≤0.01, ***p p≤0.001.

Figure 4. Abnormal motor activity caused by miR-128 deficiency is corrected by pharmacological ERK inhibition or ectopic miR-128 expression

(A) Drd1a-cre; miR-128-2^fl/fl and littermate control mice were injected i.p. with either vehicle or 12 mg/kg of the MEK1 inhibitor SL327 (n=5/group). Western blot analysis of ERK2 phosphorylation at 30 min after drug injection (left) and motor activity following vehicle or SL327 injection (right) are shown. 2-way ANOVA followed by Bonferroni post-test. Error bars show s.e.m., * p≤0.05, ** p≤0.01, *** p≤0.001. (B) Overexpression of miR-128 suppresses D1-neuron hyper-responsiveness in the dopamine-depleted striatum. The number of contralateral rotations at baseline and in response to cocaine (10 mg/kg) or D1-agonist SKF81297 (5 mg/kg) in unilateral 6-OHDA lesioned Camk2a-cre; Rosa-miR-128 or control mice (n=11/group) are shown. Error bars show s.e.m., Welch’s t-test, ** p≤0.01. (D) miR-128 reduces the susceptibility to chemically-induced seizures in mice. The numbers of Camk2a-cre; Rosa-miR-128 or littermate control mice (n=12/group) that exhibit tonic-clonic seizures 60 minutes after i.p. injection of pro-convulsive drugs kainic acid (30mg/kg, p-value=0.005) or picrotoxin (3mg/kg, p-value=0.04) are shown. p-values were calculated by Fisher’s exact test.