Silencing Parkinson’s risk allele Rit2 sex-specifically compromises motor function and dopamine neuron viability

Parkinson’s disease (PD) is the second most prevalent neurodegenerative disease and arises from dopamine (DA) neuron death selectively in the substantia nigra pars compacta (SNc). Rit2 is a reported PD risk allele, and recent single cell transcriptomic studies identified a major RIT2 cluster in PD DA neurons, potentially linking Rit2 expression loss to a PD patient cohort. However, it is still unknown whether Rit2 loss itself impacts DA neuron function and/or viability. Here we report that conditional Rit2 silencing in mouse DA neurons drove motor dysfunction that occurred earlier in males than females and was rescued at early stages by either inhibiting the DA transporter (DAT) or with L-DOPA treatment. Motor dysfunction was accompanied by decreased DA release, striatal DA content, phenotypic DAergic markers, DA neurons, and DAergic terminals, with increased pSer129-alpha synuclein and pSer935-LRRK2 expression. These results provide clear evidence that Rit2 loss is causal for SNc cell death and motor dysfunction, and reveal key sex-specific differences in the response to Rit2 loss.

in both VTA and SNc for 4-6 weeks had no significant effect on either male or female baseline locomotion in the open field 34 .Given that PD is typically late onset, we aimed to comprehensively investigate the impact of conditional Rit2 knockdown (KD) in DANs on more complex motor behaviors at either short-term (4-6 weeks; ST) or long-term (5-6 months; LT) timepoints in both male and female mice.Pitx3 IRES-tTA mouse VTA were bilaterally injected with AAV9-TRE-shRit2, which we previously reported drives shRit2 expression selectively in DANs in both VTA and SNc, due to AAV9 spread 9,34 .Consistent with our previous reports, AAV9-TRE-shRit2 significantly decreased Rit2 mRNA in both ST and LT male and female Pitx3 IRES-tTA mouse midbrains, as compared to AAV9-TRE-eGFP injected controls (Supplemental Fig. 1a-d).We previously reported that ST Rit2 KD had no effect on baseline locomotion.Similar to our previous finding in ST shRit2 mice, LT Rit2 KD likewise did not significantly affect horizontal locomotion in either male or female mice (Supplemental Fig. 1e,f), nor was their fine movement significantly affected (Supplemental Fig. 1g,i).There was additionally no change in male vertical motion (Supplemental Fig. 1h), whereas female mice exhibited significantly increased vertical locomotion (Supplemental Fig. 1j).
We next asked whether either ST or LT DAergic Rit2 KD impacted more complex motor behaviors, such as motor learning and coordination, assessed on rotarod and challenge balance beam assays.In male mice, both ST and LT DAergic Rit2 KD significantly decreased performance on the accelerating rotarod compared to controls (Fig. 1a,b), whereas female mouse rotarod performance was not significantly affected by either ST or LT DAergic Rit2 KD (Fig. 1c,d).Accelerating rotarod deficits could be due to either learning or coordination deficits.To discriminate between these possibilities, we assessed performance on the fixed-speed rotarod and challenge balance beam.Despite their accelerating rotarod deficits, male ST shRit2 mice did not exhibit any significant deficits on either the fixed-speed rotarod (Fig. 1e), or challenge balance beam (Fig. 1i,k), as compared to controls.Similar to males, female ST shRit2 mice did not exhibit any significant deficits on either the fixed-speed rotarod (Fig. 1g) or challenge balance beam (Fig. 1j,l), as compared to controls.We further assessed both mouse gait and grip strength following ST Rit2 KD, and found no differences between shRit2 mice and controls, in either males or females (Supplemental Fig. 2).
Given that ST Rit2 KD only affected male accelerating rotarod performance, we next asked whether longer Rit2 silencing would lead to motor dysfunction in females, and/or the development of motor dysfunction on other motor tasks.Following LT Rit2 KD, males continued to exhibit significantly poorer performance than controls on the accelerating rotarod (Fig. 1b), and females still exhibited no significant effect on accelerating rotarod performance (Fig. 1d).However, LT Rit2 KD drove a significant deficit in fixed-rotarod performance in both males (Fig. 1f) and females (Fig. 1h), and both foot faults and beam traversal times were significantly increased in both male (Fig. 1i, k) and female (Fig. 1j, l) LT shRit2 mice on the balance beam, as compared to both control and ST Rit2 KD mice.Gait analysis revealed that multiple gait parameters were unaffected by ST Rit2 KD in both males (Supplemental Fig. 2a, e, I, m) and females (Supplemental Fig. 2c, g, k, o).However, there was a differential impact on gait in males and females following LT Rit2 KD.LT shRit2 males completed significantly fewer gait analysis trials (Supplemental Fig. 2b) and had significantly narrower forelimb stride widths (Supplemental Fig. 2j), whereas LT shRit2 females had significantly wider hindlimb stride widths (Supplemental Fig. 2l).Despite the observed coordination and gait deficits, all LT shRit2 mice also had significantly increased four-limb grip strength (Supplemental Fig. 2r, t).Taken together, the behavioral data suggests that DAergic Rit2 is specifically required for male motor learning and that prolonged Rit2 suppression leads to motor coordination and gait deficits in both males and females.

Short-term DAergic Rit2 silencing blunts DA release in males
Motor learning deficits in males in response to ST DAergic Rit2 silencing could be due to altered DA release.To test this possibility, we leveraged fast-scan cyclic voltammetry (FSCV) to measure both DA release and clearance in ex vivo dorsal striatal slices (Fig. 2a, b).Given the viability issues inherent to acute brain slices prepared from older animals, we limited our FSCV studies to ST shRit2-and control-injected males, in which we observed a motor learning deficit.We 9 and others previously reported that DRD2 autoreceptors significantly blunt DA transient amplitudes and accelerate DA clearance.Indeed, in control mice DA transient amplitudes were significantly smaller when evoked in ACSF as compared to those evoked in the presence of L-741,626 (25 nM, Fig. 2c), a DRD2-specific antagonist, as we previously reported 9 .In shRit2 mice, DA transient amplitudes recorded in ACSF were not significantly different transients from control mice (Fig. 2c).However, unlike control DA transients, DA amplitudes recorded in the presence of L-741,626 were not significantly greater than those recorded in ACSF, and were significantly smaller than amplitudes recorded in L-741,626 from control mice.Moreover, we previously reported that ST Rit2 silencing in males decreased DAT surface levels by ~50% in dorsal striatum (DS).Despite this reduction in DAT, Rit2 silencing did not significantly affect DA clearance times (Fig. 2d), and we still observed DRD2-mediated enhancement of DA clearance in slices from both control and shRit2 mice, suggesting that DRD2 signaling was intact (Fig. 2d).Taken together, these results suggest that males locomotor deficits following ST Rit2 silencing may be, in part, due aberrant DA signaling.
Rit2 silencing suppresses the DAergic phenotype with earlier manifestation in males than females Given our FSCV results, we hypothesized that motor deficits observed in ST males, and in both males and females following LT Rit2 silencing, could potentially be due to a loss in DAergic tone.To test this possibility, we first measured striatal DA content using mass spectroscopy in male and female dorsal (DS) and ventral (VS) striata following ST and LT Rit2 silencing.In ST shRit2 mice, total DA content was not significantly affected in either DS or VS from either male or female mice as compared to their respective controls (Fig. 3a,c).However, DA content was significantly reduced in LT shRit2 male DS (Fig. 3b) and LT shRit2 female VS and DS (Fig. 3d) as compared to controls.Importantly, total striatal GABA content was not altered in male or female VS or DS at any timepoint (Fig. 3e-h), demonstrating specific changes in DANs and not global changes in striatal neurotransmitter content.Given Rit2's association with PD and the profound changes in motor function and DAergic tone observed with LT Rit2 silencing, we hypothesized that Rit2 silencing may decrease DAN viability.To test this possibility, we first measured DAergic gene and protein expression in isolated ventral midbrain (vMB) and striatum, respectively, following ST and LT Rit2 silencing.In males, RT-qPCR studies revealed that ST Rit2 KD significantly decreased tyrosine hydroxylase (TH) and DAT mRNA in vMB (Fig. 4a, b), and quantitative immunoblotting revealed that striatal TH and DAT protein were also significantly reduced (Fig. 4i, j).Unsurprisingly, following LT Rit2 KD, male TH and DAT vMB mRNA (Fig. 4c, d) and striatal TH and DAT protein (Fig. 4k, l) remained significantly diminished compared to controls.In females, TH and DAT mRNA (Fig. 4e, f) and protein (Fig. 4m, n) were unaffected following ST Rit2 KD.However, following LT Rit2 silencing females exhibited robust and significant loss in vMB TH and DAT mRNA (Fig. 4g, h), as well as striatal TH and DAT protein (Fig. 4o, p).We also found that while TH and DAT were proportionally decreased in both ST (TH: 72.4 ± 8 vs DAT: 73.5 ± 7.4%control levels, p = 0.95) and LT (TH: 51.9 ± 7.8 vs. DAT: 50.9 ± 5.6%control levels, p = 0.92) males, TH was disproportionately diminished more than DAT in LT females (TH: 29.5 ± 3.8 vs DAT: 44.6 ± 3.8%control levels, p = 0.02, unpaired, two-tailed, Student's t test).
We further tested whether Rit2 silencing impacted TH activation, by measuring pSer40-TH via immunoblot.When normalized to actin, pSer40-TH was significantly reduced in ST shRit2 male striatum (Supplemental Fig. 3a) and trended to decrease in LT shRit2 males (Supplemental Fig. 3b).However, the proportion of pSer40-TH to total TH was not significantly different in ST or LT shRit2 male mice as compared to controls (Supplemental Fig. 3e, f), suggesting that functional TH regulation is intact, despite diminished total TH levels.In females, ST Rit2 silencing had no effect on pSer40-TH or the fraction of pSer40-TH (Supplemental Fig. 3c, g).However, there was a drastic pSer40-TH loss following LT Rit2 silencing in females (Supplemental Fig. 3d), as well as the fraction of pSer40-TH (Supplemental Fig. 3h), suggesting that TH is dysregulated following LT Rit2 KD in females.
Given the profound losses in TH and DAT expression, as well as DA content, we further asked whether other characteristic ventral midbrain DAergic mRNAs were affected by shRit2.In both ST and LT shRit2 males, DRD2 and Pitx3 mRNA were significantly decreased (Supplemental Fig. 4a, b, e, f), and Nurr1 was significantly diminished following LT, but not ST, Rit2 silencing (Supplemental Fig. 4i, j).In females, ST Rit2 silencing did not significantly affect DRD2, Pitx3, or Nurr1 mRNA levels (Supplemental Fig. 4c, g, k).However, by the LT timepoint all three DAergic markers were significantly diminished (Supplemental Fig. 4d, h, l).Taken together these data demonstrate that DAergic Rit2 silencing results in downregulation of all DAergic genes, consistent with the loss in DAergic tone.
Prolonged DAergic Rit2 silencing increases PD markers in the striatum Given Rit2's association with PD, and the striking motor dysfunction we observed following Rit2 silencing, we next asked whether the loss in DAergic markers following Rit2 silencing was accompanied by an appearance of PD markers.To test this, we measured striatal αSyn and pSer129-αSyn levels, which markedly increase in idiopathic PD 35 .Total αSyn levels were not significantly affected in either ST or LT shRit2 males (Fig. 5a,b) nor in LT females (Fig. 5d), but were significantly increased in ST shRit2 females (Fig. 5c).Importantly, shRit2 drove a significant increase in pSer129-αSyn in ST and LT shRit2 males (Fig. 5e,f) and in LT females (Fig. 5h), and strongly trended for an increase in ST females (Fig. 5f).We additionally measured whether there were any changes in leucine-rich repeat kinase 2 (LRRK2), whose variants have been linked to PD and cause LRRK2 phosphorylation anomalies 36 .Total LRRK2 levels did not significantly differ from controls in all ST and LT shRit2 mice (Supplemental Fig. 6).However, LT Rit2 KD significantly increased pSer935-LRRK2 in both LT males and females, while ST Rit2 KD did not significantly affect pSer935-LRRK2 in males or females.
Together, these data indicate that in addition to profound motor deficits and decreased DAergic gene expression, LT DAergic Rit2 silencing drives an increase in both phosphorylated α-synuclein and LRRK2.

Long-term Rit2 silencing results in DAN degeneration
DAergic marker loss and pSyn increases may be due to cell death, or quiescence of the DAergic phenotype without cell death.To discriminate between these possibilities, we performed stereological counting to measure the number of total (Nissl stain) and dopaminergic (TH + ) SNc neurons at the LT Rit2 KD timepoint, where both males and females exhibited significant motor dysfunction.In LT Rit2 KD males, there was no significant loss of total SNc neurons (Fig. 6a), and a strong trend towards decreased TH + cells in the SNc (p = 0.07, Fig. 6b).Moreover, when normalized to the total SNc cells, there was a significant decrease in the %TH+ cells in male SNc (Fig. 6c).In LT Rit2 KD females, there was a significant decrease in both total SNc neurons (Fig. 6d), as well as TH+ cells (Fig. 6e).There was, however, no difference in the % neurons that were TH+ in LT Rit2 KD females as compared to controls (Fig. 6f), suggesting that the SNc neuronal losses occurred primarily in the DAergic population.Together, these results are consistent with DAN cell death following LT Rit2 KD.We additionally performed immunohistochemistry to ask whether LT Rit2 KD either 1) deleteriously affected DAergic terminals in the DS, or 2) impacted pSer129-Syn aggregation, which is a hallmark of PD.Both males and females exhibited a gross loss of TH+ methylphenidate (Ritalin), may have therapeutic potential in PD 37,38 .We asked whether such pharmacological intervention could rescue the motor deficits observed on the rotarod due to Rit2 silencing.We first tested whether increasing extracellular DA levels by inhibiting DAT with methylphenidate would rescue motor learning.ST shRit2 males were assessed on the accelerating rotarod, injected with either saline or methylphenidate (MPH, 5 mg/kg, I.P.), and were reassessed 15 min post-injection (see schematic, Fig. 7a).MPH treatment significantly improved rotarod performance as compared to saline-injected mice (Fig. 7b).MPH is equipotent at DAT and the norepinephrine transporter (NET) 39 .Therefore, to rule out any adrenergic contributions to the observed motor learning rescue, we tested whether rotarod performance was improved with desipramine (DMI), a NET-specific inhibitor.DMI treatment had no significant effect on shRit2 mouse performance (Fig. 7b), suggesting that DAT inhibition was specifically responsible for rescued rotarod performance in ST shRit2 males.L-DOPA is the immediate chemical precursor to DA and is the prevailing PD treatment 40 .Therefore, we next asked whether L-DOPA treatment could rescue motor learning deficits in ST and LT shRit2 mice.Mice were scored on the accelerating rotarod, injected with either saline or L-DOPA (20 mg/ kg, I.P.), and rescored 1-hour postinjection.In ST shRit2 mice, L-DOPA significantly improved rotarod performance (Fig. 7c).However, L-DOPA treatment had no effect on rotarod performance in LT shRit2 mice, (Fig. 7d).Taken together, these data demonstrate that while pharmacological intervention can rescue motor deficits exhibited by ST shRit2 mice, the loss of DAergic tone and DANs caused by LT Rit2 KD drives deficits that were not rescuable by pharmacological means.

Discussion
Multiple GWAS studies report Rit2 as a PD risk allele 16,17,[19][20][21][22]24,26 . Additonally, Rit2 mRNA is among the more highly downregulated genes in PD patient substantia nigra 3 , and defines a specific transcriptomic cluster in single-cell RNAseq studies from PD patients 32 .Despite these findings, it was unknown whether diminished Rit2 levels are causal or consequential for PD symptoms.Our results demonstrate that prolonged, conditional Rit2 silencing in DANs leads to significant DAergic deficits, both at the molecular and behavioral level.Specifically, DAergic Rit2 KD resulted in motor dysfunction (Fig. 1), accompanied by decreases in DA release (Fig. 2), striatal DA content (Fig. 3), DAergic gene and protein expression (Fig. 4), as well as decreased DAN numbers and striatal arborization (Fig. 5).These were accompanied by increased pSer129-αSyn (Fig. 6) and pSER935-LRRK2 (Supplemental Fig. 6).While these findings are consistent with a PD-like phenotype, further studies testing whether DAN viability would progressively deteriorate are necessary to definitively determine if Rit2 loss leads to a bona fide PD state.In most instances, males were affected at an earlier timepoint than females, consistent with sex-specific differences in PD prevalence and onset in patients.Interestingly, female performance on the accelerating rotarod was completely resistant to Rit2 loss, even at the LT timepoint, despite significant deficits in gait, fix-speed rotarod, and challenge balance beam, and diminished dopaminergic tone. Moor learning on the accelerating rotarod is DA-dependent 41,42 , however, the observed sexual dimorphism raises the possibility that there may be a strong DA-independent component to motor learning in females.Moreover, females, but not males, exhibited a disproportionately greater loss in TH than DAT at the LT timepoint, raising the possibility that differential TH/DAT protein levels potentially contributed to the observed behavior.Alternatively, it is possible that distinct, sex-specific SNc DAN subpopulations are vulnerable following Rit2 silencing.Future single-cell transcriptomic studies will be necessary to test this possibility directly.
While motor learning deficits are specific to shRit2 male mice, shRit2 coordination deficits are sex-independent and are only apparent with prolonged Rit2 silencing (Fig. 1).Coordination deficits were accompanied by modest alterations in shRit2 mouse gait (Supplemental Fig. 2).Furthermore, LT male and female mice both exhibited increased four-limb grip strength (Supplemental Fig. 2).While grip strength typically decreases with PD progression in patients 5,43 , increased mouse grip strength may also reflect rigidity or bradykinesia present in PD.Indeed, increased grip strength is observed in a rat 6-OHDA lesion PD model 44 .
Surprisingly, despite coordination deficits, conditional Rit2 silencing did not perturb horizontal locomotion, even with prolonged silencing (Supplemental Fig. 1).PD is a late-onset neurodegenerative disorder and motor symptoms are often not overtly apparent until >75% of SNc DANs have died 3,4 .Indeed, our stereological data suggest that at the LT timepoint (~25 weeks post-injection), there is ~32% loss of TH+ neurons in the SNc, and a gross loss of TH+ terminals in DS (Supplemental Fig. 6).Thus, it is possible that longer Rit2 silencing would lead to even further losses in the TH+ population, and more pronounced baseline motor deficits.Of note, the loss of TH+ striatal terminals appears grossly disproportionate to the cell number loss in SNc (Fig. 6).Recent studies indicate that DAergic axons likely deteriorate at a faster pace than their respective soma 45 , which may explain our observations.While Rit2 silencing diminished DAN numbers, the mechanism(s) downstream of Rit2 loss that led to decreased DAN viability are unknown.To date, the function of Rit2 in neurons remains poorly defined.Rit2 is required for EGF-and NGF-mediated neurite outgrowth in cell culture models 46 , and is required for NGF-mediated ERK phosphorylation and cell viability 13,14 .Our lab previously reported that Rit2 binds directly to the DAT 10 and is required for both PKC-and mGluR5-stimulated DAT endocytosis in ex vivo striatal slices 8,9 .Indeed, ST DAergic mGluR5 silencing in male mice blocked DAT internalization, increased DAT plasma membrane presentation, and likewise drove motor learning dysfunction on the accelerating rotarod that was rescuable with a DAT-selective inhibitor 9 .Thus, ST shRit2 effects may be due, in part, to DAT dysregulation, while LT shRit2 effects are more likely due to diminished DAN viability.We also previously reported that ST DAergic Rit2 KD differentially modulates acute cocaine locomotor responses in males and females, wherein male shRit2 mice exhibit increased cocaine sensitivity, and females exhibit decreased cocaine sensitivity 34 .Thus, the requirement for Rit2 in DANs likely extends beyond motor behaviors.
Despite ST Rit2 KD significantly decreasing DAT protein levels (Fig. 4j), there was no significant impact on DA clearance in parallel FSCV studies (Fig. 2d).Interestingly, we instead observed that Rit2 silencing dampened DA release as compared to controls (Fig. 2c).Decreased DA release was accompanied by decreases in both pSer40-TH (Supplemental Fig. 3a) and DRD2 mRNA (Supplemental Fig. 4a), raising the possibility that DA synthesis may be altered due to Rit2 silencing.However, DRD2-mediated regulation of DA clearance remained intact following Rit2 KD (Fig. 2d), suggesting that DRD2 dysregulation is not likely responsible for the observed decrease in DA release.Rit2 may play some previously undefined role in DA synthesis and/or release that is independent of DRD2 regulation.Interestingly, DA release is adversely affected prior to DAN degeneration in several experimental PD models 45,47 , raising the possibility that the altered DA release we observed following ST Rit2 KD may, likewise, forbode the imminent DAN degeneration.
We also found that LT Rit2 KD significantly diminished the expression of several hallmark DAergic genes, including TH, DAT, DRD2, and Pitx3, that are critical for DA signaling and the DA neuronal phenotype.In contrast, the DAN-specific transcription factor, Nurr1, was only decreased in LT shRit2 mice (Supplemental Fig. 4).Nurr1 is associated with PD progression and DAergic Nurr1 ablation results in decreased, DAT and TH expression, reduced striatal DA content and locomotor deficits 48,49 .Surprisingly, the ubiquitously expressed genes, Rit1 and Vps35 were upregulated (Supplemental Fig. 5).Rit1 is the closest homolog to Rit2 and may be upregulated or stabilized to compensate for Rit2 loss, however, whether they functionally overlap is unknown.Vps35 is a core retromer protein required for the endocytic delivery of DAT, and several neuronal receptors, to the plasma membrane 9,50-52 , and Vps35 mutations have been identified in PD patients.Thus, increased Vps35 expression may also be a compensatory mechanism to deliver additional DAT and/or receptors to the membrane in response to Rit2 loss.DAergic Rit2 KD also drove an increase in Ser129-αSyn, a PD biomarker (Fig. 5).pSer129-αSyn accumulates in DAergic nuclei and negatively regulates Nurr1 expression 53 .Whether Rit2 directly regulates DAergic gene expression, or whether the observed changes are consequences of viability and cell death will need to be determined in future studies.
We used two pharmacological approaches to test whether boosting DA availability could rescue motor dysfunction at ST and LT timepoints: DAT inhibition with MPH, and L-DOPA treatment.Both MPH and L-DOPA treatments rescued male accelerating rotarod performance following ST Rit2 KD (Fig. 7b,c).We previously reported that a DAT inhibitor (CE-158) can rescue motor learning dysfunction in response to DAergic mGluR5 KD, but has no effect on wildtype mice 9 .Thus, MPH-mediated rotarod rescue is likewise not likely to generally stimulate locomotion.While L-DOPA rescued ST motor learning deficits, it was unable to rescue motor performance following LT Rit2 KD (Fig. 7d).PD patients typically lose responsiveness to L-DOPA during disease progression, and L-DOPA efficacy requires DA uptake and release mechanisms to be in place 54 .Thus, the striking loss in TH + terminals in DS (Fig. 6) may explain the failure to rescue with L-DOPA at the LT timepoint.Alternatively, since peripheral amino acid decarboxylases (AADCs) can convert L-DOPA to DA and thereby limit its CNS bioavailability, AADC inhibitors are frequently administered with L-DOPA to increase clinical efficacy 6,55 .We did not find that AADC inhibitors were necessary in order to achieve L-DOPA rescue in ST shRit2 mice and therefore did not include them in any of our treatments.However, since the DA tone is poorer in LT shRit2 mice, the failure to achieve L-DOPA rescue might be due to peripheral AADC activity limiting CNS L-DOPA bioavailability, rather than a non-rescuable condition.
In summary, our study demonstrates that DAergic Rit2 is required for DA neuron viability and several DA-dependent motor behaviors.We identified a sex-dependent role for Rit2 in motor learning and demonstrated severe deficits in the DAergic phenotype following conditional Rti2 silencing, that included gene expression, DA content, and biomarker changes.Future studies will illuminate the mechanisms that lead from Rit2 silencing to decreased viability.

Methods
Materials L-DOPA (3788) and L-741,626 (1003) were from Tocris.Methylphenidate and desipramine were from Sigma.All other reagents were from either Sigma-Aldrich or Fisher Scientific and were of the highest possible grade.

Mice
Pitx3 IRES-iTA/+ mice were continuously backcrossed onto the C57Bl/6 J background and were the generous gift of Dr. Huaibin Cai (National Institute on Aging).Mice were maintained on 12 h light/dark cycle (lights on at 0700) at constant temperature and humidity.Food and water were available ad libitum and mice were maintained on standard chow.All studies were conducted in accordance with University of Massachusetts Medical School-approved IACUC Protocol 202100046 (formerly A-1506 H.E.M.).
Survival surgeries.Mice aged 3-4 weeks were anesthetized with I.P. 100 mg/kg ketamine (Vedco Inc.) and 10 mg/kg xylazine (Akorn Inc).To increase viral spread, mice were administered 20% mannitol (Neogen-Vet) at least 15 min prior to viral delivery 56 .Anesthetized mice were prepared and placed in a stereotaxic frame (Stoelting Inc.). 1 µL of the indicated viruses were administered at a rate of 0.2 µL/min bilaterally to the VTA using coordinates from bregma: anterior/posterior: -3.08 mm, medial/lateral: ±0.5 mm, dorsal/ventral: -4.5 mm.Syringes were maintained in position for a minimum of 5 minutes post-infusion prior to removal.Viral incubation was for either 4-5 weeks or 5-6months.Viral expression was confirmed by visualizing midbrain GFP reporters encoded in the viral constructs and/or by RT-qPCR.
RNA extraction and RT-qPCR Bilateral 1.0mm 2 tissue punches were obtained from 300 µm coronal ventral midbrain slices of experimental mice.Punches were collected while visualizing GFP on an inverted fluorescence microscope and RNA was extracted immediately, or following tissue storage at -70°C, using RNAqueous®-Micro Kit RNA isolation (Thermo Fisher Scientific).Extracted RNA was reverse transcribed using RETROscript® reverse transcription kit (Thermo Fisher Scientific).Quantitative PCR was performed using the Applied Biosystems® 7500 Real-Time PCR System Machine and software or using the Bio-Rad C1000 Touch Thermal Cycler with CFX96 Real-Time system and software using Taqman® gene expression assays for mouse Rit2

Mouse Behavior
Locomotion.Mouse activity was monitored individually in photobeam activity chambers (San Diego Instruments) as previously described 34 .
Horizontal, vertical, and fine movements were measured in 5-minute binds for 90 minutes total.
Accelerating and fixed-speed rotarod.Mice were habituated to the testing room in home cage for >30 min with ambient lighting and the rotarod (UgoBasile 47600) running at 4 RPM.Accelerating rotarod: Mice were placed on the rod moving at 4 RPM and rod speed was increased linearly from 4 to 40 RPM over 5 minutes.Trials were terminated and latency determined by either triggering the strike plate during a fall or if the mouse made >1 consecutive passive rotation.For drug administration studies, performance indices were calculated as the average latency time to fall (sec) across three trials (pre-or postdrug administration) normalized to the maximal trial time (300 sec).Fixed-speed: Mice were placed on the rod moving at the indicated speeds (20, 25, 30, 35, 40, 45  RPM) for 60 second trials.Latency to fall was measured or trial was stopped following >1 passive rotation.Two consecutive trials were performed per speed and latencies were averaged per animal.
Challenge balance beam.Mice were habituated to the testing room for >30 min with overhead lights off and only a single light source placed approximately 1.5 feet over the beam origin illuminated.On day one, mice were trained over 5 trials to traverse a 1.0 m, step-wise tapered (widths: 35 mm, 25 mm, 15 mm, 5 mm) elevated beam (#80306, Lafayette Neuroscience) at an incline of 15°.A dark box with home-cage bedding was placed at the far end of the beam.On day two, a challenge grid with 1 cm x 1 cm openings (custom 3D-printed, Thingiverse #4869650) was placed over the beam and mice traversed the beam in 3 independent trials.Traversal initiation and completion were determined by breaking an IR beam at each end of the beam.Traversals were video captured and scored for foot faults and traversal time, averaged over the first two completed trials.Both the experimenter and an independent scorer were double-blinded to the mouse ID.
Gait analysis.Gait analysis protocol was adapted from Wertman, et al. 57 .Gait testing apparatus consisted of a 10 cm x 36 cm runway with 14 cm high foamboard walls and a dark box at the opposing end.Fresh, legalsize paper was placed on the benchtop under the runway for each trial.Mouse forepaws and hind-paws were dipped in non-toxic orange and blue tempera paint, respectively, and mice were placed on the paper at the open end of the runway and allowed to traverse to the closed box at the opposing end.Three trials were performed per mouse and stride length, stride width and tow spread were measured for both fore-and hindlimbs.Number of completed trials was also quantified.Experimenters and data analysts were double-blinded to mouse IDs.
Grip strength.Four-limb grip strength was measured using the Bioseb Grip Strength Test (BIO-GS3) equipped with mesh grip grid for mice.Mice were suspended by tail over the mesh and lowered onto it until all 4 paws grasped the mesh.The mouse was then pulled backwards gently on the horizontal plane until it released from the mesh.The maximal force applied was recorded for 3 consecutive trials and averaged.

Tissue harvesting and immunoblotting
Striata were collected by preparing 300 µm coronal sections on a Vibratome as previously described 8,9 .Sections were collected through the entire striatum, dorsal and ventral striata were sub-dissected, and slices encompassing each region were pooled for each independent mouse.Hemislices were moved to the recording chamber and were perfused with oxygenated ASCF supplemented with 500 µM Na-Ascorbate.Glass pipettes containing a 7 µm carbon-fiber microelectrode were prepared and preconditioned in ASCF by applying triangular voltage ramps ( − 0.4 to +1.2 and back to −0.4 V at 400 V/s), delivered at 60 Hz for 1 hour.Recordings were performed at 10 Hz.Electrodes were calibrated to a 1 µM DA standard prior to recording.Electrodes were positioned in DS and DA transients were electrically evoked with a 250 µA rectangular pulse every 2 min, using a concentric bipolar electrode placed ~100 µm from the carbon fiber electrode.Data were collected with a 3-electrode headstage, using an EPC10 amplifier (Heka) after low-pass filter at 10 kHz and digitized at 100 kHz, using Patchmaster software (Heka).A stable baseline was achieved after evoking six consecutive DA transients, after which experimental data were collected.Each biological replicate is the average of three evoked DA transients/slice, and a minimum of 3 independent mice were used to gather data from the indicated number of slices in each experiment.Data were analyzed in Igor Pro, using the Wavemetrics FSCV plugin (gift of Veronica Alvarez, NIAAA).Peak amplitudes were measured for each individual DA transient, and tau was calculated as 1/e according to the equation: y = y 0 + A ((x-x 0 )/tau)) .
Mass spectrometry Sample preparation.Brains were harvested, 1.0 mm coronal sections were prepared and bilateral 1.0mm 2 punches were each taken from dorsal and ventral striata.Each bilateral pair was solubilized in 10 µL internal standard solution (200 µM 13C 4 -GABA and 1 µM 2H 3 -DA in water with 500 µM ascorbic acid and 0.1% formic acid) and 50 µl ice-cold acetonitrile with 1% formic acid.Samples were vortexed twice for 0.5 min with a 1 min incubation on ice between vortexing and were sonicated in an icewater bath until tissue was completely disrupted.Samples were centrifuged (10 min, 16,000 x g) and supernatants were collected for LC/MS/ MS analysis.A standard (STD) solution containing 200 µM GABA, 1 µM dopamine, 500 µM ascorbic acid and 0.1% formic acid was also prepared.Stereological analysis and Immunohistochemistry/confocal microscopy.Briefly, mice were perfused and fixed with freshly made 4% paraformaldehyde (PFA) in PBS.Brains were removed immediately and fixed again in 4% PFA followed by equilibration in 30% sucrose in PBS.Midbrains were removed for stereological analysis, and forebrains were used for immunohistochemistry/confocal microscopy.
Stereological analysis.SNc total and TH+ neurons were quantified as previously described 58 .Fixed brains were imbedded in the OCTcompound media (Sakura) and frozen in liquid nitrogen.40 µm cryosections were prepared through the midbrain a Leica CM3050s cryostat, and were stored in an antifreeze media containing 30% ethylene glycol, 25% glycerol, and 5% phosphate buffer.For stereology counting, 1 in every 5 sections was selected with a random start and a total of 6 brain slices on average were used for each mouse for IHC labeling for TH, including DAB enhancement, followed by Cresyl violet staining to reveal all neurons.Substantia nigra pars compacta was imaged using a Zeiss Axioplan 2 microscope equipped with a 20X objective, and Stereo Investigator was used to estimate the total number of neurons in the region of interest using the following parameters: frame sizes, 150 ×150 µm; grid sizes, 250 ×250 µm; top guard zone height, 2 µm; and optical dissector height, 8 µm.These parameters yielded a coefficient of error <10% throughout the analysis.Total cell numbers measured were weighted to section thickness for each mouse and were averaged across each cohort.Investigators performing stereological counting were blinded to mouse identity.
Immunohistochemistry/confocal microscopy.25 µm coronal sections through the striatum were prepared using a microtome and were co-stained with rabbit anti-pSer129-Syn (Cell Signaling #23706; 1:500) and chicken anti-TH (Millipore #AB9702, 1:500), followed by staining with secondary antibodies preabsorbed against mouse (Alexa568-goat anti-rabbit, Alexa647-donkey anti-chicken, Jackson ImmunoResearch).Z-stacks in the dorsal striatum were acquired with a Zeiss 700 LSM scanning confocal microscope using 555 nm and 647 nm lasers, and were pseudocolored to red and green, respectively.Acquisition settings within each channel (pinhole size, digital gain, and laser strength) were identical across all samples.Z-stacks were imported into ImageJ software where a representative plane was chosen, channels were separated (for individual red and green images), and images exported as tiff files.Files were subsequently imported into Adobe Photoshop, and levels were adjusted identically across all images.

Statistics
Data analysis was performed with GraphPad Prism software.All data were assessed for normality and nonparametric tests were applied if data distribution was non-Gaussian.Outliers in each data set were identified using either Grubb's or Rout's outlier tests, with a or Q values set at 0.05 or 5%, respectively, and were removed from further analysis.Significant differences between two values were determined using either a one-tailed, two-tailed, or paired Student's t test, as indicated.Differences amongst more than two conditions were determined using one-way or two-way ANOVA, as appropriate, and significant differences among individual values within the group were determined by post-hoc multiple comparison tests, as described for each experiment.

Fig. 6 |
Fig. 6 | Prolonged DAergic Rit2 silencing decreases the substantia nigra DA neurons and striatal DAergic terminals.Immunohistochemistry Pitx3 IRES-tTA males and female mouse VTA were bilaterally injected with either AAV9-TRE-eGFP or AAV9-TRE-shRit2 and brains were fixed, sectioned, stained with the indicated dyes/ antibodies and analyzed at the LT timepoint as described in Methods.a-f Stereological analysis of SNc neurons.Midbrain sections were stained with cresyl violet and TH-specific antibodies.Total and TH+ cell numbers were counted as described in Methods.Average values ± S.E.M. are presented and significance was tested with a two-tailed, unpaired, Student's t test.N values indicate the number of independent mouse SNc analyzed.a-c.Males: Rit2 KD did not significantly decrease total Nissl+ neurons a. p = 0.46), trended towards a decrease in TH+ neurons b. p = 0.07) cells, and significantly decreased the %TH+ neurons c. *p = 0.02) in SNc, n = 5. d-f Females: Rit2 KD significantly decreased both total Nissl+ neurons (d.*p = 0.02), and TH+ neurons e. *p = 0.01) cells, but did not significantly decrease the %TH+ neurons f. p = 0.48) in SNc, n = 5-6.

Fig. 7 |
Fig.7| Male motor learning deficits are rescuable with Parkinson's therapeutics at early, but not late, Rit2 silencing timepoints.Accelerating rotarod rescue studies.a Experimental schematic.Pitx3 IRES-tTA male mouse VTA were bilaterally injected with AAV9-TRE-shRit2 and were assessed on the accelerating rotarod at the indicated timepoints as described in Methods.Mice were initially tested for three trials, injected ±the indicated treatment drugs (I.P), and were retested for an additional three trials.Performance indices for each mouse were calculated pre-and post-test, and performance were assessed with a two-tailed, paired Student's t test.N values indicate the number of independent mice.b Methylphenidate (MPH) treatment: ST shRit2 mice received either saline, 5 mg/kg MPH, or 0.5 mg/kg DMI, and were retested 15 min postinjection.Left: Raw rotarod results presented as average latency to fall during the trial, ±S.E.M. Right: Paired pre-and post-test rotarod performance indices for each independent mouse assessed.shRit2 mice treated with MPH performed significantly better than pre-injection (*p = 0.03), whereas performance was not enhanced by either saline (p = 0.28) or DMI (p = 0.60), n = 5-6.c L-DOPA treatment on ST shRit2 males: ST shRit2 mice were treated ±20 mg/kg L-DOPA and were retested 1 hour postinjection.Left: Raw rotarod results presented as average latencies ±S.E.M. Right: Paired pre-and post-test rotarod performance indices for each independent mouse.shRit2 mice treated with L-DOPA (**p = 0.005), but not saline (p = 0.98) performed significantly better than pre-injection, n = 6-7.d L-DOPA treatment on LT shRit2 males: LT shRit2 mice were treated ±20 mg/kg L-DOPA and were retested 1 hour postinjection.Left: Raw rotarod results presented as average latencies ±S.E.M. Right: Paired pre-and post-test rotarod performance indices for each independent mouse.Neither L-DOPA (p = 0.48) nor saline (p = 0.82) treatment significantly improved performance as compared to preinjection performance, n = 6.
Tissue was lysed in RIPA buffer (10 mM Tris, pH 7.4; 150 mM NaCl; 1.0 mM EDTA; 0.1% SDS, 1% Triton X-100, 1% Na deoxycholate) supplemented with protease inhibitors (1.0 mM phenylmethylsulfonyl fluoride and 1.0 g/mL each leupeptin, aprotinin, and pepstatin) and Phosphatase inhibitor cocktail V (EMD Millipore).Mechanical tissue disruption was also performed by triturating sequentially through a 200 µL pipette tip, 22-, and 26-gauge tech tips and solubilized by rotating (30 min 4°C).Rad) and were quantified using Quantity One software (Bio-Rad).Representative blots shown for a given condition were cropped from the same exposure of the same immunoblot and spliced together for presentation purposes only.Splice margins are indicated with a line.All compared blots were processed in parallel and derive from the same experiments.