Calpain-regulated p35/cdk5 plays a central role in dopaminergic neuron death through modulation of the transcription factor myocyte enhancer factor 2

Abstract

The mechanisms underlying dopamine neuron loss in Parkinson's disease (PD) are not clearly defined. Here, we delineate a pathway by which dopaminergic loss induced by 1-methyl-4-phenyl 1,2,3,6 tetrahydropyridine (MPTP) is controlled in vivo. We reported previously that calpains play a central required role in dopamine loss after MPTP treatment. Here, we provide evidence that the downstream effector pathway of calpains is through cyclin-dependent kinase 5 (cdk5)-mediated modulation of the transcription factor myocyte enhancer factor 2 (MEF2). We show that MPTP-induced conversion of the cdk5 activator p35 to a pathogenic p25 form is dependent on calpain activity in vivo. In addition, p35 deficiency attenuates MPTP-induced dopamine neuron loss and behavioral outcome. Moreover, MEF2 is phosphorylated on Ser444, an inactivating site, after MPTP treatment. This phosphorylation is dependent on both calpain and p35 activity, consistent with the model that calpain-mediated activation of cdk5 results in phosphorylation of MEF2 in vivo. Finally, we provide evidence that MEF2 is critical for dopaminergic loss because "cdk5 phosphorylation site mutant" of MEF2D provides neuroprotection in an MPTP mouse model of PD. Together, these data indicate that calpain-p35-p25/cdk5-mediated inactivation of MEF2 plays a critical role in dopaminergic loss in vivo.

Figure 1.

MPTP induces calpain-dependent formation of p25 and increase in cdk5 activity. A, Representative Western blot showing MPTP-induced increase in p25 expression. Protein samples were extracted from the nigral region of mice treated with either calpastatin adenovirus (Ad. CalSt) or LacZ control. Calpastatin-treated animals showed significant reduction in MPTP-induced p25 levels compared with LacZ controls. β-Actin was used as a protein loading control. B, Quantification of p25 band intensity (∼25 kDa molecular weight) measured against β-actin protein loading control at 24 h, 3 d, and 7 d after MPTP treatment (n = 3 per treatment group) with calpain inhibition (Ad. CalSt) or LacZ control (Ad. LacZ). C, Representative kinase assay of nigral protein extracts subject to in vitro cdk5 kinase assay, using Histone H1 as the substrate. Cdk5 activity is significantly increased 7 d after MPTP treatment. MPTP-induced increase in cdk5 activity is attenuated by calpain inhibition (Ad. CalSt.) and p35 deficiency. Loading control (LC) is Coumassie blue-stained band in destained kinase assay gel. Similar results were obtained in two independent experiments. Error bars represent ±SEM.

Figure 2.

p35 deficiency attenuates MPTP-induced dopamine neuron loss. A, Representative photomicrographs showing significant reduction in total number of TH-positive cells after MPTP treatment in wild-type mice. MPTP-induced loss of TH was attenuated in p35 mutant mice 14 d after MPTP treatment. B, Quantification of total TH-positive cell bodies in the SNc, after saline or chronic MPTP treatment, in p35 mutants and littermate controls reveal significant attenuation of MPTP-induced dopamine neuron loss in the substantia nigra of p35 null mice. C, Western blot analysis reveals no significant differences in TH expression (∼58 kDa molecular weight) in nigral samples extracted from p35 null mice when compared with littermate control. Seven days after MPTP treatment, both p35 null mice and adenoviral calpastatin (Ad. CalSt.)-treated mice showed attenuation of MPTP-induced TH loss in the nigral region. The membrane was blotted simultaneously with β-actin (as a loading control). D, Quantification of CV-positive cells in the SNc (MTN region) of saline controls (p35 wild type and p35 null; n = 7) and MPTP-treated p35 wild-type (n = 5) and MPTP-treated p35 null (n = 6) mice. Asterisks represent p < 0.01 (ANOVA; Newman–Keuls post hoc; n = 6 per treatment group). KO, Knock-out mice; WT, wild-type mice. Error bars represent ±SEM.

Figure 3.

p35 null mice show improvement in MPTP-induced loss of striatal function. A, C, Representative photomicrographs show that p35 null mice display normal expression of DAT. B, Significant reduction in DAT expression was observed after MPTP treatment in wild-type littermates. D, This MPTP-induced loss of DAT was partially attenuated by p35 deficiency. E, Quantification (optical density measurement) of striatal DAT staining at 14 d after saline or MPTP treatment in wild-type and p35 mutant mice (n = 6 per treatment group). F, Dopamine levels, measured by HPLC, in striatal extracts 14 d after MPTP or saline treatment in wild-type and p35 mutant mice show that p35 deficiency reduces MPTP-induced loss of striatal dopamine (n = 6–8 per treatment group). G, Striatal MPP⁺ measurement 90 min after a single 30 mg/kg MPTP injection shows no significant differences in metabolism of MPTP in the different genotypes (n = 4 per treatment group). *p < 0.05, ANOVA; **p < 0.01, ANOVA; Newman–Keuls post hoc test. Error bars represent ±SEM.

Figure 4.

p35-deficient mice show significant attenuation of MPTP-induced increase in markers of denervation and behavioral deficits. A, Quantification of δ FosB expression in the striatal region after saline or MPTP treatment in wild-type and p35 null mice (n = 4–5 per treatment group). B, Home cage “beam break” locomotor activity of p35 wild-type and mutant mice after saline or MPTP treatment show that p35 deficiency alone did not influence locomotor activity; p35 deficiency attenuates MPTP-induced behavioral deficits. The bar graph represents total locomotor activity before initiation of MPTP treatment (basal) and 7 and 14 d after MPTP treatment. Activity was monitored over a 24 h period (n = 6–11 per group). *p < 0.05, ANOVA, **p < 0.01, ANOVA; Newman–Keuls post hoc test. Error bars represent ±SEM.

Figure 5.

MEF2D is a critical downstream target of cdk5 that is induced in response to MPTP. A, Representative Western blot analysis showing an increase in phosphorylation of MEF2D at Serine 444 (pMEF2D) in nigral samples extracted 24 h, 3 d, and 7 d after the first MPTP injection. MPTP also induced cleavage of pMEF2D, which is dependent on p35, because p35 null mice show reduced pMEF2D and reduced formation of the cleavage product; similar results were obtained with calpain inhibition (Ad. CalSt.) at 7 d after MPTP. Similar results were obtained in three independent experiments. Immunofluorescence analysis reveals that pMEF2D is expressed in the nigral region after MPTP treatment. B, Quantification of TH-positive neurons in the SNc (MTN anatomical region) after MEF2D (S444A) mutant or GFP adenoviral protein expression with saline or MPTP treatment (n = 6 per group). C, Inset, Western blot analysis showing increased MEF2 protein expression in the nigral region of GFP-treated and MEF2D (S444A) adenovirus-treated mice. **p < 0.01, ANOVA; Newman–Keuls posthoc test. Error bars represent ±SEM.