Male C57Bl/6 mice (Japan SLC Co., Shizuoka, Japan) aged 7–8 weeks were used in this study. The mice were housed in a controlled environment (25 ± 1°C, 50 ± 10% humidity, and 12-h light/dark cycle) with access to food and tap water ad libitum. All procedures involving experimental mice were approved by the Ethical Review Committee of the University of Tokushima.

Mice received single intraperitoneal injections of nilotinib (5, 10, 25, or 50 mg/kg; Cayman Chemical, Ann Arbor, MI, USA) dissolved in 0.5% aqueous carboxymethyl cellulose 3 days after administration of MPTP or saline. Vehicle-treated (control) mice received an equivalent volume of 0.5% aqueous carboxymethyl cellulose.

In combination with nilotinib (25 mg/kg), A-68930 hydrochloride (2 mg/kg of free base; Sigma-Aldrich, St Louis, MO, USA), SCH-23390 hydrochloride (0.5 mg/kg of free base; Sigma-Aldrich), quinpirole hydrochloride (3 mg/kg of free base; Sigma-Aldrich), or raclopride hydrochloride (1 mg/kg of free base; Sigma-Aldrich) was dissolved in saline and intraperitoneally injected 30 min before sacrifice. The dose of each drug was determined as described previously (Yamamura et al., 2013).

Mice were injected intraperitoneally four times in a day with MPTP hydrochloride (20 mg/kg of free base; Sigma-Aldrich) at 2-h intervals (Aoki et al., 2009; Yamamura et al., 2013). Saline-treated mice received an equivalent volume of 0.9% saline. Our previous study showed that maximal neurodegenerative effects of MPTP on the nigral dopaminergic cells were observed at the 3-day time-point after administration of MPTP (Aoki et al., 2009; Yamamura et al., 2013).

Striatal tissue samples were homogenized in ice-cold lysis buffer containing 50 mM Tris–HCl buffer, pH 7.5, with 0.5 M NaCl, 0.5% Triton X-100, 10 mM EDTA (ethylenediaminetetraacetic acid), 4 mM EGTA (ethylene glycol tetraacetic acid), 1 mM Na₃VO₄, 30 mM Na₂P₂O₇, 50 mM NaF, 0.1 mM leupeptin, 75 μM pepstatin A, 50 μg/ml trypsin inhibitor, 1 mM phenylmethanesulfonyl fluoride, 100 nM calyculin A, and 1 mM dithiothreitol. After centrifugation at 15,000 rpm for 10 min at 4°C, the protein lysates were mixed with Laemmli’s buffer containing 63 mM Tris–HCl, pH 6.8, 2% sodium dodecyl sulfate (SDS), 5% 2-mercaptoethanol, 2.5% glycerol, and 0.01% bromophenol blue, and were then heated at 100°C for 5 min. Each sample, containing the same amount of protein, was applied to a 10% SDS-polyacrylamide gel for electrophoresis (PAGE) followed by blotting onto a PVDF (polyvinylidene fluoride) membrane. The PVDF membranes were then incubated with the desired primary antibodies. Antibodies against tyrosine hydroxylase (TH; 1:1000; Millipore, Billerica, MA, USA), Cdk5-pTyr15 (1:1000; Santa Cruz Biotechnology, Santa Cruz, CA, USA), Cdk5 (1:1000; Santa Cruz Biotechnology), and DARPP-32 (1:1000; Cell Signaling, Danvers, MA, USA) DARPP-32-pThr34 (1:1000; Cell Signaling), and DARPP-32-pThr75 (1:1000; Cell Signaling) were used. Anti-β-actin antibody (1:5000; Sigma-Aldrich) was used to adjust for equal amounts of protein loading into each well. The bound antibodies were detected by the ECL (enhanced chemiluminescence) method with horseradish peroxidase-conjugated secondary antibodies. Gel images were captured using a lumino-imaging analyzer LAS-4000 (Fujifilm, Tokyo, Japan). Optical densities were evaluated using a computerized image analysis system (Dolphin-DOC; Kurabo, Osaka, Japan).

Mice were injected intraperitoneally with a lethal dose of pentobarbital (Sigma-Aldrich) 120 min after drug administration. They were then transcardially perfused with 0.01 M phosphate-buffered saline (PBS) at pH 7.4, followed by cold 4% paraformaldehyde in 0.1 M phosphate buffer (PB) at pH 7.4. The brains were removed, post-fixed overnight in the same fixative at 4°C, and moved through a 10–30% sucrose gradient in 0.1 M PB at 4°C for cryoprotection. Sections were cut on a cryostat at 15-μm thickness, and stored in PBS containing 0.05% NaN₃ until use. Free-floating brain sections were incubated in PBS containing 3% bovine serum albumin (BSA) and rabbit polyclonal antibody to c-Fos (1:50,000; Oncogene Science, Cambridge, MA, USA) at room temperature overnight. The bound primary antibodies were detected by the Histofine Simple Stain Kit (Nichirei, Tokyo, Japan) and the TSA (tyramide signal amplification) system with Cyanine3 (PerkinElmer; Shelton, CT, USA; Goto et al., 2013; Koizumi et al., 2013).

Digital microscopic images were captured using an Olympus BX51 microscope (Olympus, Tokyo, Japan), imported into Adobe Photoshop CS4, and processed digitally for adjustments of contrast, brightness, and color balance. Measurements of the density of c-Fos-labeled nuclei were made on the striatal sections at the level of 0.9–1.1 mm anterior to bregma, according to the atlas of Hof et al. (2000). We counted the number of c-Fos-positive nuclei in five striatal fields of each mouse (n = 5), and calculated the density of c-Fos-positive nuclei/mm² in each animal (Sako et al., 2010).

Prior to pharmacological testing, mice were handled for a week by the same operator to reduce stress, and trained for behavioral tests as described below until their motor performance became reproducible. All behavioral training and tests were performed by the same operator from 10:00 to 16:00 in the order to lower stress. The experimental room environment was kept constant for all tests. Apparatus was cleaned with 70% ethanol on each trial.

All experimental values were expressed as means ± SEM. Statistical significance was evaluated by unpaired two-tailed t-test, or by one-way analysis of variance (ANOVA) followed by Newman–Keuls, Scheffe, or Fisher’s PLSD (protected least significant difference) post hoc test for pairwise comparisons. The significance level was set at P < 0.05. All analyses were conducted in Stat View 5.0 (SAS Institute, Cary, USA).

To test whether nilotinib could affect striatal phosphorylation of Cdk5 and DARPP-32, we first performed western blot analysis on the striatal extracts from mice that received intraperitoneal injections of vehicle or nilotinib at doses of 5, 10, 25, or 50 mg/kg 30 min before they were sacrificed. Striatal levels of Cdk5-pTyr15 (Figure 2A; P < 0.05, ANOVA) and DARPP-32-pThr75 (Figure 2B; P < 0.05, ANOVA) were significantly reduced in mice injected with nilotinib at the doses of 25 and 50 mg/kg, compared to mice treated with vehicle. In contrast, no effects of nilotinib on striatal levels of total Cdk5 (Figure 2A; P > 0.05, ANOVA), DARPP-32-pThr34 (Figure 2B; P > 0.05, ANOVA), and total DARPP-32 (Figure 2B; P > 0.05, ANOVA) were found. Thus, systemic administration of nilotinib could inhibit striatal phosphorylation of both Cdk5 at Tyr15, the substrate site targeted by c-Abl (Zukerberg et al., 2000; Dhavan and Tsai, 2001; Zhang et al., 2007), and DARPP-32 at Thr75, the substrate site targeted by Cdk5 (Greengard, 2001), in naïve mice.

To test if D1R- and D2R-mediated signals differentially affected the decrease of phosphorylation of Tyr15-Cdk5 and Thr75-DARPP-32 by nilotinib, we next performed western blotting on the striatal extracts from naïve mice injected with nilotinib (25 mg/kg) in combination with the D1-agonist A-68930 (2 mg/kg), D1-antagonist SCH-23390 (0.5 mg/kg), D2-agonist quinpirole (3 mg/kg), or D2-antagonist raclopride (1 mg/kg), 30 min before they were sacrificed. Nilotinib-induced decrease of striatal levels of Cdk5-pTyr15 (Figure 3A; P > 0.05, ANOVA) and DARPP-32-pThr75 (Figure 3B; P > 0.05, ANOVA) was unchanged by A-68930 or SCH-23390. In contrast, it was significantly enhanced by quinpirole (Figures 3C,D; P < 0.05, ANOVA), but antagonized by raclopride (Figures 3C,D; P < 0.05, ANOVA). In addition, striatal expression of total Cdk5, total DARPP-32, and DARPP-32-pThr34 was not affected by any dopaminergic compounds (Figures 3A–D). These findings indicate that D2R, but not D1R, activation could positively affect the nilotinib-induced decrease of striatal phosphorylation of Tyr15-Cdk5 and Thr75-DARPP-32.

A striking feature of the neuronal responses to dopamine receptor stimulation in the striatum is a rapid induction of c-Fos, one of the products of the immediate-early genes (LaHoste et al., 1993; Canales and Graybiel, 2000). Strong interactive and synergistic effects of D1Rs and D2Rs have been reported to affect these neuronal responses (LaHoste et al., 1993; Canales and Graybiel, 2000). To assess whether nilotinib could induce striatal c-Fos expression, we performed c-Fos immunostaining on striatal sections from mice that received A-68930 (2 mg/kg) alone, quinpirole (3 mg/kg) alone, nilotinib (25 mg/kg) alone, A-68930 (2 mg/kg) + quinpirole (3 mg/kg), or A-68930 (2 mg/kg) + nilotinib (25 mg/kg) at 120 min before they were sacrificed. Microscopic images showed only a few c-Fos-positive nuclei in the striatum of mice that received A-68930 alone (Figures 4A,F), quinpirole alone (Figures 4B,G), or nilotinib alone (Figures 4C,H). In contrast, a robust induction of c-Fos in striatal neurons was found in mice injected with the combination of A-68930 and quinpirole (Figures 4D,I) and in those injected with the combination of A-68930 and nilotinib (Figures 4E,J). These visual impressions were confirmed by densitometric analyses (Figure 4K). In accordance with the fact that dopamine D1/D2 synergism is required for striatal Fos induction in the intact striatum (LaHoste et al., 1993), a significant increase in the number of c-Fos-labeled nuclei was found in mice injected with the combination of A-68930 and quinpirole (Figure 4K; P < 0.001, unpaired two-tailed t-test), compared to mice injected with A-68930 alone. Interestingly, we also found a significant increase in the number of c-Fos-labeled nuclei in mice injected with the combination of A-68930 and nilotinib (Figure 4K; P < 0.001, unpaired two-tailed t-test), compared to mice injected with A-68930 alone. These findings suggest that, like a D2-agonist, nilotinib can synergize with a D1-agonist for inducing striatal c-Fos expression.

To gain insight into the antiparkinsonian actions of nilotinib, we conducted an experiment in the MPTP mouse model. The experimental design is shown in Figure 5A (see also Materials and Methods). There was a marked (>80%) loss of TH, the rate-limiting enzyme in dopamine synthesis, in MPTP-treated mice compared to saline-treated mice (Figure 5B; P < 0.01, ANOVA). Consistent with a recent landmark report (Viaro et al., 2010), MPTP-treated mice exhibited overt behavioral abnormalities as a parkinsonian-like phenotype (Figures 5C–J). Compared to control mice, MPTP-treated mice showed significant motor deficits, as determined by beam walking (Figure 5C; P < 0.05, ANOVA), bar (Figure 5D; P < 0.05, ANOVA), horizontal wire (Figure 5E; P < 0.05, ANOVA), and rotarod tests (Figure 5F; P < 0.05, ANOVA). A foot printing test showed a significant decrease in stride length of both the hindpaw (Figure 5G; P < 0.001, ANOVA) and forepaw (Figure 5H; P < 0.001, ANOVA) and a significant increase in the overlap length (Figure 5I; P < 0.01, ANOVA) in MPTP-treated mice, compared to saline-treated mice.

Notably, all the behavioral tests (Figures 5C–J) revealed significant recovery of impaired motor performances after intraperitoneal injection of nilotinib in MPTP mice, compared to vehicle-treated MPTP mice. The beam walking (Figure 5C) and bar (Figure 5D) tests showed a significant decrease in traveling time and in catalepsy-like immobilization after administration of nilotinib at a dose of 25 mg/kg (P < 0.05, ANOVA), but not at a dose of 10 mg/kg (P > 0.05, ANOVA). The horizontal wire test (Figure 5E) showed a significant increase in pass rate following injection of nilotinib at doses of 10 mg/kg (P < 0.05, ANOVA) and 25 mg/kg (P < 0.05, ANOVA). The rotarod test (Figure 5F) showed a significant increase in latency to fall after administration of nilotinib at a dose of 25 mg/kg (P < 0.05, ANOVA), but not at a dose of 10 mg/kg (P > 0.05, ANOVA). On the foot printing test, a significant increase in stride length of the hindpaw was found after nilotinib administration at a dose of 25 mg/kg (Figure 5G; P < 0.05, ANOVA), but not at a dose of 10 mg/kg (P > 0.05, ANOVA), while that of the forepaw was found after administration of nilotinib at doses of 10 and 25 mg/kg (Figure 5H; P < 0.05, ANOVA). The foot printing test also showed a significant decrease in overlap length after administration of nilotinib at doses of 10 mg/kg (Figure 5I; P < 0.01, ANOVA) and 25 mg/kg (P < 0.001, ANOVA). These observations indicate that systemic administration of nilotinib could attenuate the motor impairments caused by striatal dopamine depletion in MPTP mice.

To test if nilotinib actually inhibited striatal phosphorylation of Cdk5 and DARPP-32 in MPTP mice, we performed western blot analysis on striatal extracts from MPTP mice that received nilotinib (25 mg/kg) 30 min before they were sacrificed. A significant increase in the striatal levels of Cdk5-pTyr15 (Figure 6A; P < 0.01, ANOVA), but not of total Cdk5 (Figure 6A; P > 0.05, ANOVA), was found in MPTP mice, compared to control mice. Nilotinib treatment reversed the abnormally elevated striatal levels of Cdk5-pTyr15 (Figure 6A; P < 0.01, ANOVA) to the control baseline in MPTP mice. We also found a significant increase in striatal levels of DARPP-32-pThr75 (Figure 6B; P < 0.01, ANOVA), but not of DARPP-32-pThr34 (Figure 6B; P > 0.05, ANOVA) and total DARPP-32 (Figure 6B; P > 0.05, ANOVA), in MPTP mice compared to control mice. Nilotinib treatment reversed the abnormally elevated striatal levels of DARPP-32-pThr75 to the control baseline in MPTP mice (Figure 6B; P < 0.01, ANOVA). Thus, nilotinib could normalize the increased activity of Cdk5/DARPP-32-Thr75 signaling in the striatum of MPTP mice.