Enhanced striatal dopamine transmission and motor performance with LRRK2 overexpression in mice is eliminated by familial Parkinson's disease mutation G2019S

Abstract

PARK8/LRRK2 (leucine-rich repeat kinase 2) was recently identified as a causative gene for autosomal dominant Parkinson's disease (PD), with LRRK2 mutation G2019S linked to the most frequent familial form of PD. Emerging in vitro evidence indicates that aberrant enzymatic activity of LRRK2 protein carrying this mutation can cause neurotoxicity. However, the physiological and pathophysiological functions of LRRK2 in vivo remain elusive. Here we characterize two bacterial artificial chromosome (BAC) transgenic mouse strains overexpressing LRRK2 wild-type (Wt) or mutant G2019S. Transgenic LRRK2-Wt mice had elevated striatal dopamine (DA) release with unaltered DA uptake or tissue content. Consistent with this result, LRRK2-Wt mice were hyperactive and showed enhanced performance in motor function tests. These results suggest a role for LRRK2 in striatal DA transmission and the consequent motor function. In contrast, LRRK2-G2019S mice showed an age-dependent decrease in striatal DA content, as well as decreased striatal DA release and uptake. Despite increased brain kinase activity, LRRK2-G2019S overexpression was not associated with loss of DAergic neurons in substantia nigra or degeneration of nigrostriatal terminals at 12 months. Our results thus reveal a pivotal role for LRRK2 in regulating striatal DA transmission and consequent control of motor function. The PD-associated mutation G2019S may exert pathogenic effects by impairing these functions of LRRK2. Our LRRK2 BAC transgenic mice, therefore, could provide a useful model for understanding early PD pathological events.

Figure 1.

BAC transgenic LRRK2-Wt and LRRK2–G2019S mouse lines: transgene levels and striatal DA content. A, Schematic showing the original BAC with the mouse LRRK2 gene, modified LRRK2-Wt BAC in which the FLAG tag was inserted in-frame after the ATG translation start codon, and LRRK2–G2019S BAC carrying G-to-S mutation at amino acid 2019. B, Western blot using anti-LRRK2 antibody shows that transgenes LRRK2-Wt and LRRK2–G2019S produce similar protein levels, approximately sixfold higher than endogenous brain LRRK2. C, D, HPLC analysis indicated that striatal DA and HVA contents were ∼25% lower in LRRK2–G2019S than in littermate control mice at 12 months (*p < 0.05 vs control 2, n = 7 mice per group). Striatal DA or HVA in LRRK2-Wt mice did not differ significantly from littermate controls (p > 0.05 vs control 1, n = 7 mice per group). Quantitative data are expressed as mean ± SEM and were analyzed by one-way ANOVA with Bonferroni's post hoc analysis.

Figure 2.

Examination of TH activity and protein levels for VMAT2, DAT, and D₂R in LRRK2-Wt and LRRK2–G2019S mice. A, Western blot analysis (left) and quantification (right) of striatal TH, VMAT2, DAT, and D₂R protein levels. No significant differences were detected among LRRK2-Wt, LRRK2–G2019S, and littermate control mice (n = 3 mice per group). B, Striatal TH activity, determined by release of tritium from [3,5-³H]-l-tyrosine, was not changed in LRRK2-Wt or LRRK2–G2019S compared to littermate control mice at 9–10 months (n = 3 mice per group). Data are expressed as percentage of control. C, Western blot analysis of striatal tissue from mice at 9–10 months using anti-TH and anti-phospho-TH (P-Ser31 and P-Ser40) antibodies (left), with quantification (right). No significant differences in total TH or phospho-TH (P-Ser31 and P-Ser40) protein levels were detected among control, LRRK2-Wt, and LRRK2–G2019S mice (n = 3 mice per group). Quantitative data are expressed as mean ± SEM and were analyzed by one-way ANOVA with Bonferroni's post hoc analysis.

Figure 3.

Opposing regulation of striatal DA release and uptake in LRRK2-Wt versus LRRK2–G2019S mice. A, Comparison of evoked DA release in 12 month transgenic LRRK2 mice (3 mice per group). Left, Averaged single-pulse evoked [DA]_o versus time obtained in striatal slices from LRRK2-Wt and littermates (control 1) (n = 43–50 sites) and in LRRK2–G2019S and littermates (control 2) (n = 43–53 sites) (error bars omitted for clarity). Right, Peak [DA]_o was ∼25% higher in LRRK2-Wt mice than in control 1 (*p < 0.05), whereas evoked [DA]_o was ∼35% lower in LRRK2–G2019S mice than in control 2 (***p < 0.001); control groups did not differ (p > 0.05, control 1 vs 2). B, DA uptake in 12 month transgenic LRRK2 mice. Left, Representative evoked [DA]_o from control striatum showing the data points used for curve fitting to extract V_max. Middle, Fit of representative data points to Michaelis–Menten uptake kinetics with a fixed K_m of 0.9 μm gave peak [DA]_o = 1.09 μm, V_max = 4.03 μm/s, and r² = 0.996. Right, Averaged V_max values did not differ between LRRK2-Wt and control 1 mice (p > 0.05; n = 41–50 sites), whereas V_max was lower in LRRK2–G2019S than in control 2 mice (*p < 0.05; n = 45–51 sites), indicating less efficient DA uptake after G2019S mutation. Data are expressed as mean ± SEM and were analyzed by one-way ANOVA with Bonferroni's post hoc analysis.

Figure 4.

Impaired sustainability of striatal DA release in LRRK2–G2019S mice. Sustainability of single-pulse evoked DA release was assessed in 12 month mice (3 mice per group) by comparing the decline in peak [DA]_o evoked at 2 min intervals over a 20 min period. A, Representative [DA]_o records evoked at a given site in a control 1 and LRRK2-Wt slice (top), as well as in a control 2 and LRRK2–G2019S slice (bottom). The examples shown for each mouse group were selected to have an initial starting peak [DA]_o of ∼1 μm to aid visual comparison of the decline in peak [DA]_o. B, Average peak [DA]_o with time, normalized to peak [DA]_o evoked by the first stimulus (P1) of a series at a given recording site taken as 100%. Left, Control 1 (n = 9 sites) and control 2 (n = 11 sites) mice showed a similar decline in peak evoked [DA]_o of ∼30% over a 20 min period of stimulation (p > 0.05; control 1 vs control 2). Middle, The average decline in peak [DA]_o over a 20 min stimulation period in LRRK2-Wt (n = 10 sites) was similar to control 1 mice (p > 0.05). Right, In contrast, DA release was less well sustained in LRRK2–G2019S (n = 11 sites) versus control 2 mice (***p < 0.001). Quantitative data are expressed as mean ± SEM and were analyzed by two-way ANOVA with Bonferroni's post hoc analysis.

Figure 5.

LRRK2-Wt, but not LRRK2–G2019S, mice show hyperactivity and enhanced motor performance. A–C, In the open-field test, LRRK2-Wt mice displayed increased rearing compared to LRRK2–G2019S or control mice at 6 and 12 months (A), traveled longer distances than LRRK2–G2019S or control mice at 12 months (B), and spent more time moving at 12 months (C). C, All 12 month mice were less active than 6 month mice regardless of genotype. D–F, In the challenge beam test, LRRK2-Wt and LRRK2–G2019S mice used a similar number of steps when crossing a beam covered with metal mesh (D); however, LRRK2-Wt mice had fewer slips (E) and fewer slips per step (F) than LRRK2–G2019S or control mice. Littermate controls for each transgenic line did not differ in any behavioral test; for each age, control data from these animals were therefore pooled (*p < 0.05; **p < 0.01; control, n = 30 mice; LRRK2-Wt, n = 15 mice; LRRK2–G2019S, n = 15 mice). Quantitative data are expressed as mean ± SEM and were analyzed by one-way ANOVA with Bonferroni's post hoc analysis.

Figure 6.

Enhanced brain kinase activity in LRRK2–G2019S does not cause loss of DAergic neurons, but correlates with increased of striatal phospho-tau staining versus LRRK2-Wt. A, Higher kinase activity of transgenic brain LRRK2–G2019S protein versus LRRK2-Wt. Left, Kinase activity was assayed using MBP phosphorylation (2.4-fold increase) and autophosphorylation (2.7-fold increase) with equal protein levels. Right, Quantification of MBP phosphorylation and LRRK2 autophosphorylation from three independent experiments. B, No loss of SNc TH+ neurons with stereological cell counting or change in cell morphology was seen even at 60× magnification in either mouse line at 12 months (n = 6 per group). Scale bar, 100 μm. C, TH staining in the striatum of LRRK2-Wt and LRRK2–G2019S mice at 12 months. Optical density analysis indicates no difference in TH staining among the different genotypes (n = 6 per group). Scale bar, 100 μm. D, Immunohistochemical staining with anti-phospho-tau antibodies (PHF-1 and CP13) in dorsal striatum. LRRK2-Wt mice had fewer PHF-1- and CP13-positive cells than control or LRRK2–G2019S mice. Scale bar, 50 μm. Quantitative data show attenuation of PHF-1-positive cell number in the dorsal striatum of LRRK2-Wt mice (**p < 0.01; n = 3 per group). Quantitative data are expressed as mean ± SEM and were analyzed by one-way ANOVA with Bonferroni's post hoc analysis.