Dopamine Transporter, PhosphoSerine129 α-Synuclein and α-Synuclein Levels in Aged LRRK2 G2019S Knock-In and Knock-Out Mice

The G2019S mutation in leucine rich-repeat kinase 2 (LRRK2) is a major cause of familial Parkinson’s disease. We previously reported that G2019S knock-in mice manifest dopamine transporter dysfunction and phosphoSerine129 α-synuclein (pSer129 α-syn) immunoreactivity elevation at 12 months of age, which might represent pathological events leading to neuronal degeneration. Here, the time-dependence of these changes was monitored in the striatum of 6, 9, 12, 18 and 23-month-old G2019S KI mice and wild-type controls using DA uptake assay, Western analysis and immunohistochemistry. Western analysis showed elevation of membrane dopamine transporter (DAT) levels at 9 and 12 months of age, along with a reduction of vesicular monoamine transporter 2 (VMAT2) levels at 12 months. DAT uptake was abnormally elevated from 9 to up to 18 months. DAT and VMAT2 level changes were specific to the G2019S mutation since they were not observed in LRRK2 kinase-dead or knock-out mice. Nonetheless, dysfunctional DAT uptake was not normalized by acute pharmacological inhibition of LRRK2 kinase activity with MLi-2. Immunoblot analysis showed elevation of pSer129 α-syn levels in the striatum of 12-month-old G2019S KI mice, which, however, was not confirmed by immunohistochemical analysis. Instead, total α-syn immunoreactivity was found elevated in the striatum of 23-month-old LRRK2 knock-out mice. These data indicate mild changes in DA transporters and α-syn metabolism in the striatum of 12-month-old G2019S KI mice whose pathological relevance remains to be established.


Introduction
The c.6055G>A transition in the leucine-rich repeat kinase 2 (LRRK2) gene results in the G2019S substitution in the LRRK2 protein, which is associated with familial, autosomal dominant forms of Parkinson's Disease (PD) and represents a risk factor for idiopathic PD [1,2]. LRRK2 is a multidomain protein encompassing a kinase and a GTPase domain surrounded by protein-protein interaction domains [3,4]. The G2019S mutation causes a 2-3-fold increase in kinase activity which is instrumental to LRRK2 neurotoxicity in vitro [5][6][7] and in vivo [8][9][10]. LRRK2 G2019S-associated PD and idiopathic PD share clinical and neuropathological features. Indeed, they are characterized by hypokinetic symptoms, nigrostriatal dopamine (DA) neuron degeneration and Lewy body pathology [1,11],

VMAT2 Activity Assay
Whole-brain synaptic vesicles were obtained, and VMAT2 activity assay was carried out as previously described [14]. Synaptic vesicles were incubated for 5 min at 37 • C with 20 nM [ 3 H]-DA isotopically diluted with varying concentrations of unlabeled DA. Radioactivity was counted with a PerkinElmer Tri Carb 2810 TR scintillation counter.

Data Presentation and Statistical Analysis
Data are expressed as mean ± SEM (standard error of the mean) of n mice. Statistical analysis was implemented in Graphpad Prism 8.4.3 (GraphPad; LaJolla, CA, USA). The Student t-test, two-tailed for unpaired data, was used to compare two groups; in the remaining cases, one-way or two-way ANOVA followed by the Tukey test for multiple comparisons was used. Outliers were identified using the ROUT method implemented

Figure 2.
Dysfunctional DAT activity in G2019S KI mice appears at 9 months of age. Kinetic analysis of [ 3 H]-DA uptake in synaptosomes were performed in the striata of 6-month-old (A), 9-month-old (B), 12-month-old (C) and 18-month-old (D) G2019S KI mice versus age-matched WT controls. Data are expressed as mean ± SEM of n = 4 mice per group. Statistical analysis was performed using the Student t-test, two-tailed for unpaired data. * p < 0.05, different from WT mice.

DAT Abnormalities Are Associated with the G2019S Mutation but Independent of Ongoing LRRK2 Kinase Activity
To verify whether the alterations of DAT levels observed in G2019S KI mice were specifically related to the G2019S mutation, striatal DAT levels were measured in a different cohort of 12-month-old G2019S KI mice in comparison with age-matched LRRK2 KO, KD and WT mice ( Figure 3A). Immunoblot analysis revealed changes across genotypes (F3,20 = 4.02, p = 0.0216), and confirmed the elevation of DAT levels in G2019S KI mice (+46%). Conversely, no changes were observed in LRRK2 KO and KD mice compared to WT controls.
Since immunoblot data suggested that the increase of DAT levels relied on the G2019S mutation, the possibility was tested that the abnormal increase of DAT activity observed in G2019S KI mice was sustained by ongoing LRRK2 kinase activity. [ 3 H]-DA uptake kinetics were then analyzed in striatal synaptosomes obtained from 12-month-old G2019S KI and WT mice acutely treated with MLi-2 (10 mg/kg, i.p.) or vehicle ( Figure 3B). MLi-2 failed to normalize the elevation of Vmax in G2019S KI mice (genotype effect F1,12 = 31.44, p = 0.0001; MLi-2 effect F1,12 = 0.0008, p = 0.99). To confirm effective LRRK2 targeting by MLi-2, striatal pSer1292 LRRK2 levels were monitored in G2019S KI mice as a readout of kinase activity [43]. MLi-2 caused a 75% reduction of pSer1292 LRRK2 levels (t = 8.838 df = 10, p < 0.0001; Figure 3C) without affecting total protein levels ( Figure 3D), indicating that MLi-2 impaired striatal LRRK2 kinase activity. Dysfunctional DAT activity in G2019S KI mice appears at 9 months of age. Kinetic analysis of [ 3 H]-DA uptake in synaptosomes were performed in the striata of 6-month-old (A), 9-month-old (B), 12-month-old (C) and 18-month-old (D) G2019S KI mice versus age-matched WT controls. Data are expressed as mean ± SEM of n = 4 mice per group. Statistical analysis was performed using the Student t-test, two-tailed for unpaired data. * p < 0.05, different from WT mice.

DAT Abnormalities Are Associated with the G2019S Mutation but Independent of Ongoing LRRK2 Kinase Activity
To verify whether the alterations of DAT levels observed in G2019S KI mice were specifically related to the G2019S mutation, striatal DAT levels were measured in a different cohort of 12-month-old G2019S KI mice in comparison with age-matched LRRK2 KO, KD and WT mice ( Figure 3A). Immunoblot analysis revealed changes across genotypes (F 3,20 = 4.02, p = 0.0216), and confirmed the elevation of DAT levels in G2019S KI mice (+46%). Conversely, no changes were observed in LRRK2 KO and KD mice compared to WT controls.
Since immunoblot data suggested that the increase of DAT levels relied on the G2019S mutation, the possibility was tested that the abnormal increase of DAT activity observed in G2019S KI mice was sustained by ongoing LRRK2 kinase activity. [ 3 H]-DA uptake kinetics were then analyzed in striatal synaptosomes obtained from 12-month-old G2019S KI and WT mice acutely treated with MLi-2 (10 mg/kg, i.p.) or vehicle ( Figure 3B). MLi-2 failed to normalize the elevation of Vmax in G2019S KI mice (genotype effect F 1,12 = 31.44, p = 0.0001; MLi-2 effect F 1,12 = 0.0008, p = 0.99). To confirm effective LRRK2 targeting by MLi-2, striatal pSer1292 LRRK2 levels were monitored in G2019S KI mice as a readout of kinase activity [43]. MLi-2 caused a 75% reduction of pSer1292 LRRK2 levels (t = 8.838 df = 10, p < 0.0001; Figure 3C) without affecting total protein levels ( Figure 3D), indicating that MLi-2 impaired striatal LRRK2 kinase activity.  To confirm the elevation of DAT signal observed in immunoblot experiments, we performed DAT immunohistochemistry in the striatum of 12-month-old G2019S KI mice. Although DAT levels were slightly higher in G2019S KI and LRRK2 KO mice with respect to controls, ANOVA did not reveal significant changes in DAT density across genotypes (F2,15 = 1.12, p = 0.35; Figure S1A). Immunohistochemistry was repeated in striatal slices of 23-month-old mice ( Figure S1B). DAT immunoreactivity was unchanged across genotypes (F2,21 = 2.707, p = 0.089) although the DAT signal trended to a 7% reduction in G2019S KI mice with respect to controls (p = 0.09).

VMAT2 Dysfunction in G2019S KI Mice
VMAT2 protein levels were analysed in the striatum of 6, 9, 12 and 18-month-old WT and G2019S KI mice ( Figure 4). VMAT2 levels were similar between genotypes in 6month-old and 9-month-old mice ( Figure 4A,B), whereas a 53% decrease was found in 12month-old G2019S KI mice (t = 9.806, df = 10, p < 0.0001; Figure 4C) [14]. VMAT2 protein levels were similar between genotypes at 18 months ( Figure 4D). To investigate the LRRK2 To confirm the elevation of DAT signal observed in immunoblot experiments, we performed DAT immunohistochemistry in the striatum of 12-month-old G2019S KI mice. Although DAT levels were slightly higher in G2019S KI and LRRK2 KO mice with respect to controls, ANOVA did not reveal significant changes in DAT density across genotypes (F 2,15 = 1.12, p = 0.35; Figure S1A). Immunohistochemistry was repeated in striatal slices of 23-month-old mice ( Figure S1B). DAT immunoreactivity was unchanged across genotypes (F 2,21 = 2.707, p = 0.089) although the DAT signal trended to a 7% reduction in G2019S KI mice with respect to controls (p = 0.09).

VMAT2 Dysfunction in G2019S KI Mice
VMAT2 protein levels were analysed in the striatum of 6, 9, 12 and 18-month-old WT and G2019S KI mice ( Figure 4). VMAT2 levels were similar between genotypes in 6-month-old and 9-month-old mice ( Figure 4A,B), whereas a 53% decrease was found in 12-month-old G2019S KI mice (t = 9.806, df = 10, p < 0.0001; Figure 4C) [14]. VMAT2 protein levels were similar between genotypes at 18 months ( Figure 4D). To investigate the LRRK2 kinase dependence of the effect observed, VMAT2 levels were measured in G2019S KI mice in comparison with age-matched LRRK2 KD, KO and WT mice. Marked (−55%) reduction of VMAT2 levels was confirmed in the new cohort of G2019S mice, whereas no changes were observed in KD and KO mice (F 3,20 = 6.487, p = 0.0030; Figure 4E). VMAT2 uptake activity was then analysed in a whole-brain synaptic vesicle preparation of 9-month-old and 18month-old mice ( Figure 5)   Figure 4E). VMAT2 uptake activity was then analysed in a whole-brain synaptic vesicle preparation of 9month-old and 18-month-old mice ( Figure 5).  Figure 5B).

pSer129 α-Syn and Total α-Syn in G2019S KI and LRRK2 KO Mice
Immunoblot analysis was performed in the striatum of 12-month-old WT, G2019S KI and LRRK2 KO mice, to investigate whether changes in DAT levels were associated with changes in total α-syn and pSer129 α-syn levels ( Figure 6). Total α-syn levels (normalized to the houskeeping gene GAPDH) were unchanged across genotypes (F 2,23 = 0.06, p = 0.91; Figure 6A), whereas pSer129 levels ( Figure 6B) were almost doubled in G2019S KI mice when compared to both WT and LRRK2 KO mice (F 2,23 = 4.69, p = 0.019; Figure 6B). However, only a trend towards an increase was observed when pSer129 α-syn levels were normalized to total α-syn levels since ANOVA yielded a p-value just above the limit of significance (F 2,23 = 2.93, p = 0.07, Figure 6C). Western blot analysis of (A) total α-syn and (B) pSer129 α-syn levels in the striatum of 12-monthold G2019S KI and LRRK2 KO mice in comparison with age-matched WT controls. (C) pSer129 αsyn/total α-syn ratio is also shown. Data are expressed as mean ± SEM of n = 8 WT, n = 10 G2019S KI and n = 8 LRRK2 KO mice. Statistical analysis was performed using one-way ANOVA followed by the Tukey test for multiple comparisons. * p < 0.05, different from WT mice. α-syn/total α-syn ratio is also shown. Data are expressed as mean ± SEM of n = 8 WT, n = 10 G2019S KI and n = 8 LRRK2 KO mice. Statistical analysis was performed using one-way ANOVA followed by the Tukey test for multiple comparisons. * p < 0.05, different from WT mice.

Discussion
A temporal analysis of DA transporter levels and activity in relation to α-syn and pSer129 α-syn immunoreactivity was performed in the striatum of G2019S KI mice. DAT dysfunction emerged earlier than VMAT2 dysfunction (as early as age 9 months) and outlasted it. The increase of DAT and the reduction of VMAT2 levels were observed in G2019S KI mice but not mice where LRRK2 was deleted or LRRK2 kinase activity silenced, suggesting that they are related to the mutation-associated increase of LRRK2 kinase activity. Nonetheless, the abnormal DAT activity was independent of ongoing LRRK2 kinase activity since MLi-2 failed to normalize it at a dose reducing pSer1292 levels by 75% [36,44]. The most parsimonious explanation is that abnormal DAT activity ensues from permanent changes associated with chronic enhancement of LRRK2 kinase activity at the dopaminergic synapse. Nonetheless, the possibility that repeated MLi-2 administrations are necessary to normalize DAT activity needs to be tested.
Previous studies in LRRK2 mice revealed that LRRK2 modulates DAT levels and activity [14,20,45] (for a review see [46]). A dampened response of striatal DA release to DAT blockers was detected in R1441R KI mice [47], G2019S KI mice [14] and 18-month-old hG2019S rats overexpressing the transgene in the adulthood [20]. Paradoxically, here we confirm that the response in G2019S KI mice is associated with an increase of DAT protein levels and DAT velocity (Vmax) without changes in affinity for DA [14]. As it has been shown that the density of DAT expressed in plasma membranes inversely correlates with sensitivity to DAT blockers [48,49], we have speculated that the reduced sensitivity to DAT blockers might indeed reflect a greater accumulation of DAT, and possibly DAT uptake activity, at the plasma membrane. Immunohistochemistry failed to replicate the results of Western analysis since DAT immunoreactivity was non-significantly elevated in striatal slices of 12-month-old G2019S KI mice. However, it should be recalled that, despite very sensitive, indirect immunohistochemistry is not quantitative and the signal cannot be unequivocally referred to a specific protein as in Western analysis. Anyway, the lack of changes of DAT immunoreactivity would suggest that DAT changes are mild. Elevated DAT activity has been associated with cell death in vitro [30,[50][51][52] and experimental Parkinsonism in vivo [30] due to facilitation of toxin (e.g., MPTP) entry and/or buildup of cytosolic DA resulting in the production of toxic species and oxidative stress [30,50,51,53]. This process might be amplified by the impairment of VMAT2 activity [28] since, consistent with the VMAT2 role in maintaining cytosolic DA concentrations within a physiological subtoxic range, deletion of VMAT2 results in DA neuron loss [31,32,54], whereas VMAT2 overexpression is neuroprotective [31,33]. The divergent nature of the time-courses of DAT and VMAT2 levels in G2019S KI mice indicates that they are not related to changes of striatal DA terminal density. Indeed, striatal TH density does not change in G2019S KI mice compared to WT controls over aging [13,14]. It is also difficult to infer whether changes in DA transporters are linked to aging or to strain differences since across-genotype and not within-genotype longitudinal design of immunoblot analysis was performed. However, DA homeostasis undergoes age-related changes in G2019S KI mice and, specifically, DA transmission appears to be elevated at 3 but not at 12 months of age [45]. Since vesicular DA release appears similar in 12-month-old WT and G2019S KI mice [14,45], the elevation of DAT activity we observed in G2019S KI mice at 12 months might explain the normalization of DA transmission observed at this age [45]. The fact the functional changes in DA uptake outlast changes in protein levels might be due to the higher sensitivity of the functional assay, since the immunoblot analysis cannot discriminate between active and inactive DAT pools. It is unlikely that dysregulation of DA homeostasis ensues in nigral DA cell loss since G2019S KI mice do not show overt nigrostriatal DA neuron or terminal degeneration during their lifespan [13,14]. Nonetheless, they show signs of synaptic dysregulation [14,35,55] and increased susceptibility to Parkinsonian toxins along with aging [40], suggesting that they can model the presymptomatic/premotor phases of the disease. In support of this view, changes in DAT and VMAT2 levels/activity in the striatum of 12-month-old G2019S KI mice were associated with an increased phosphorylation of α-syn at Ser129, as revealed by immunoblot analysis. This can be considered an early sign of neuron demise. In fact, although the role of this posttranslational α-syn modification in neurodegeneration remains to be established [34], presynaptic pools of pSer129 α-syn unrelated to neurodegeneration have been identified [56]. Previous dual immunofluorescence analysis revealed that the increase of striatal pSer129 levels in G2019S KI mice occurs in neurons [37], consistent with the view that G2019S LRRK2 might orchestrate DAT, VMAT2, and α-syn changes within the dopaminergic synapse. The present study cannot prove a causal link between DAT and pSer129 α-syn changes or tell whether dysregulation of DA homeostasis precedes the elevation of Ser129 α-syn phosphorylation or viceversa. Both hypotheses might be proven true. In fact, partial genetic knockdown (~50%) of VMAT2 in rats has been reported to bring about impaired striatal DA transmission, nigro-striatal DA cell loss and increased pSer129 α-syn levels [32]. On the contrary, it is well known that α-syn binds to DAT [52,[57][58][59] and that α-syn regulates DAT trafficking [58,60] and surface localization, which is key to the modulation of DA uptake and DA homeostasis. Both positive [52,61,62] and negative [57,59,63] α-syn modulations of DAT function have been reported. The positive association of DAT levels, DAT uptake and pSer129 α-syn levels in aged G2019S KI mice is consistent with a study in SH-SY5Y cells showing that pSer129 α-syn increased Vmax without changing Km of [ 3 H]-DA uptake, suggesting that pSer129 α-syn increases DA uptake independent of DAT trafficking [62]. Therefore, although the present data cannot prove a causal link between DAT, VMAT2 and pSer129 α-syn changes, they suggest that these changes might be part of an overall (synaptic) adaptive response associated with the G2019S mutation. pSer129 α-syn immunoreactivity analysis did not confirm immunoblot data, again revealing different sensitivities of the two techniques. The discrepancy with previous immunohistochemical finding in G29019S KI mice [26] might be due to the different method of pSer129 α-syn quantification (area-of-threshold vs. optical density) and the different model used (AAV-GFP-injected vs. naïve G2019S KI mice). Immunohistochemistry, however, revealed an increase in α-syn immunoreactivity in the striatum of 23-month-old LRRK2 KO mice. This is consistent with the elevation of α-syn levels in the kidneys of 15-month-old LRRK2 KO mice [64], although no such increase was observed in the brain of these mice. Conversely, α-syn immunoreactivity was elevated in the striatum of 15-month-old LRRK1 and LRRK2 double KO mice compared to LRRK1 and LRRK2 single KO mice and WT controls [65]. In this respect, the present data would suggest that LRRK1 compensation of LRRK2 functions [65] is not lifelong and might fade at very late ages (i.e., 23 months) leaving LRRK2 function uncompensated.

Conclusions
We previously reported aberrant DAT and VMAT2 levels and dysfunctional DAT in 12-month-old G2019S KI mice [14]. Here, a careful time-course of DAT and VMAT2 expression/activity and α-syn immunoreactivity in G2019S KI mice up to 23 months of age was carried out. In confirming aberrant DAT and VMAT2 levels and dysfunctional DAT in 12-month-old G2019S KI mice, we show for the first time that DAT changes preceded and outlasted VMAT2 changes, and that DAT dysfunction is not sustained by ongoing LRRK2 kinase activity. Various clinical studies failed in showing changes of DAT imaging in asymptomatic LRRK2 mutant carriers [66][67][68]. Nonetheless, a more recent analysis performed in a significantly larger and homogeneous patient population showed a better PET ligand binding to DAT in LRRK2 G2019S non-symptomatic carriers with respect to sporadic PD patients, which has been interpreted as being due to a slower DAT decline or a reduced DA release [69]. Our data lend support to the hypothesis that better DAT signal is due to an increase in DAT availability, although the possibility that G20919S KI mice also have reduced DA levels cannot be ruled out [13,14]. Further studies are needed to explore the mechanisms by which the G2019S mutation regulates DAT function. Immunoblot analysis in the striatum of 12-month-old G2019S KI mice also confirms the elevation of pSer129 α-syn levels previously reported with immunohistochemistry [14], although a different quantification method of immunostaining can influence the outcome of the analysis. Finally, immunohistochemical analysis extended from 12-month-old G2019S KI and WT mice [14] to 12, 18 and 23-month-old G2019S KI, LRRK2 KO and WT mice reveals for the first time a very late elevation of α-syn levels in LRRK2 KO mice, which confirms the role of LRRK2 in α-syn clearance and, more in general, proteostasis [37,65,70,71].

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.