Midbrain dopaminergic degeneration differentially modulates primary motor cortex activity and motor behavior in hemi-parkinsonian rats

Parkinson’s disease (PD) is marked by degeneration in the nigrostriatal dopaminergic pathway, affecting motor control via complex changes in the cortico-basal ganglia-thalamic motor network, including the primary motor cortex (M1). The modulation of M1 neuronal activity by dopaminergic inputs, particularly from the ventral tegmental area (VTA) and substantia nigra pars compacta (SNc), plays a crucial role in PD pathophysiology. This study investigates how nigrostriatal dopaminergic degeneration influences M1 neuronal activity in rats using in vivo calcium imaging. Histological analysis confirmed dopaminergic lesion severity, with high lesion level rats showing significant motor deficits. Levodopa treatment improved fine motor abilities, particularly in high lesion level rats. Analysis of M1 calcium signals based on dopaminergic lesion severity revealed distinct M1 activity patterns. Animals with low dopaminergic lesion showed increased calcium events, while high lesion level rats exhibited decreased activity, partially restored by levodopa. These findings suggest that M1 activity is more sensitive to transient fluctuations in dopaminergic transmission, rather than to chronic high or low dopaminergic signaling. This study underscores the complex interplay between dopaminergic signaling and M1 neuronal activity in PD symptoms development. Further research integrating behavioral and calcium imaging data can elucidate mechanisms underlying motor deficits and therapeutic responses in PD.


Introduction
8][9] Dopaminergic inputs from the ventral tegmental area (VTA) and, to a lesser extent, from the substantia nigra pars compacta (SNc) directly modulate M1 neuronal activity via D1-like (D1R) and D2-like (D2R) dopaminergic receptors 10,11 by modulating the ring rate and synchronization of M1 neurons.8][29] In contrast, studies in hemiparkinsonian rats have shown increased synchronicity and beta-frequency oscillations between M1 and striatum. 30Therefore, a cohesive interpretation of the pathological changes in M1 activity in the context of pathological behaviors and therapeutic responses remains elusive.
The objective of this study is to characterize how midbrain dopaminergic degeneration can induce changes in M1 activity that can be visualized and quanti ed using in vivo calcium imaging in awake rats.To achieve this goal, we used single-photon uorescent calcium imaging, a technique that combines the bene ts of imaging and electrophysiological recording techniques to provide an assessment of both the systems and cellular responses in the disease and treatment states in a single analysis platform.We imaged calcium activity in the M1 of GCaMP6f-expressing rats as they transitioned from a naïve baseline to a 6-hydroxydopamine (6-OHDA)-lesioned hemi-parkinsonian state and then to a levodopa-treated state over the course of three months.Motor function and M1 calcium activity were chronically evaluated in the single pellet-reaching test (SPRT) (Fig. 1a).Here, we demonstrate that levodopa treatment of 6-OHDA-lesioned rats not only improves ne motor abilities but modulates M1 neuronal calcium activity in a manner dependent of the levels of dopaminergic lesion.

Con rmation of GCaMP6f expression and PRISM lens placement
Calcium activity data from seven of 13 rats were included in the analysis following both veri cation of PRISM lens placement in M1 and GCaMP6f expression (Fig. 2a).Of the six rats excluded, one rat (R5) was removed from the analysis because the baseplate detached from the headcap precluding microscope connection.An additional ve rats (R9 -R13) were excluded because calcium signals were not detected in one or more imaging sessions.Histological analysis of GCaMP6f expression and PRISM lens localization showed that four of these rats had the PRISM lens misplaced relative to the GCaMP6fexpressing area, and one rat had extensive in ammatory reaction with cortical damage that hindered proper calcium imaging (see Table 1 for details).
Levodopa improves ne motor abilities only in high lesion level hemi-parkinsonian rats Following SPRT training, a set of 3-5 test sessions with 25 trials each was performed.For each session, M1 activity was recorded while rats extended either their left or right forelimb through the grid fence and grasped a sucrose pellet placed in the receptacle located just outside the behavioral setup.Their motor performance was recorded on video and synchronized to the calcium recordings.Categorization and quanti cation of ne motor movements was conducted by a trained investigator who was blinded to the treatment (Fig. 1b, Supplementary Table 1).
Levodopa treatment improved overall performance in the SPRT compared to both naive (p = 0.03) and 6-OHDA-lesioned (p = 0.02) states by reducing the number of attempts executed to successfully reach and grasp the pellet in each trial (Fig. 3a).Interestingly, movements executed when a sucrose pellet was placed in the receptacle (full grasp and reach without grasp) were not affected (Figs.3b and 3c).
However, levodopa treatment resulted in a signi cant decrease in grasp movements when a sucrose pellet was not in the receptacle relative to naive and 6-OHDA-lesion, respectively (p = 0.028, p = 0.005) (Fig. 3d).This suggests that levodopa improves voluntary movements contributing to improved performance in the SPRT.
Additionally, the levodopa treatment state had an increased total attempt duration compared to the 6-OHDA lesion (Fig. 3e).We divided the total attempt duration into reaching duration (i.e., time to reach and grasp the sucrose pellet) and grasping duration (i.e., time during which grasp is maintained to bring the pellet to the animal's mouth).Levodopa treatment increased the reaching duration relative to naïve and 6-OHDA-lesion (p = 0.029, p = 0.015), respectively (Fig. 3f).However, levodopa treatment did not affect grasping duration (Fig. 3g).This suggests an improved motor coordination and movement control of forelimbs and paws during the levodopa treatment state.Indeed, during the 6-OHDA-lesioned state, rats exhibited frequent shorter and incomplete ballistic paw/forelimb movements that did not t the criteria for a full extension to reach or grasp a pellet beyond the grid fence.
2][33] Thus, unilateral 6-OHDA lesioning of the nigrostriatal pathway should result in signi cant motor de cits evident in the SPRT.However, our results did not show signi cant motor de cits in ne motor function of rats in the 6-OHDA-lesioned state (Fig. 3).The success rate, de ned as the number of successful attempts in which the rat grasped and ate a sucrose pellet divided by the total number of attempts, was used to evaluate the effect of dopaminergic lesion on motor behavior.Rats with high loss of TH + neurons (R8, R9, R13) showed a decreased success rate following 6-OHDA lesion followed by improvement after levodopa treatment.This pattern was not observed in rats with mild lesions (R6, R7, R10, R11) (Fig. 4).
Importantly, rats 1 through 5 received two days of training during the naïve state and a total of 15 trials per session (Table 1).These rats did not receive su cient SPRT training during naïve state, which could bias comparisons in subsequent states, 7,34 and were thus analyzed separately.For these rats (R1 -R5), we observed reduced performance in the SPRT following 6-OHDA lesion as shown by an increased number of attempts per trial and full grasps, and no improvement during levodopa treatment (Supplementary Figure S1a, Supplementary Table S2).Additionally, the success rate for rats 1-5 does not follow a similar pattern to rats 6-13 (Supplementary Figure S1b).Speci cally, rat 2 does not show a clear pattern of reduced success rate during the 6-OHDA lesioned state, as expected from rats with high levels of dopaminergic neuronal loss.
Changes in M1 calcium activity in response to dopaminergic lesion and levodopa treatment We recorded calcium signals from M1 neurons using a miniature head-mounted microscope while rats executed the SPRT.We assessed global changes in calcium activity by determining the total number of calcium events, the frequency of calcium events (Hz), and the magnitude of calcium in ux using the average of the area under the curve (AUC) of each event.We obtained consistent calcium signals in all three states for seven rats (Table 1).We successfully maintained focal planes and tracked neuronal ensembles for 3-4 days within each state.However, we were unable to track the same neuronal ensembles across the three states due to intrinsic technical limitations of the device and surgical approach that creates slight shifts in the eld of view over a prolonged period of time (> 2 weeks).
Analysis of calcium activity showed that the 6-OHDA lesion was associated with an increase in the number of calcium events in M1 (p = 0.07) but had no effects on the frequency (p = 0.992) or magnitude (p = 0.948) of calcium events relative to naïve state.In contrast, levodopa treatment was not associated with changes in the total number of calcium events relative to naïve or 6-OHDA-lesion (p = 0.185, p = 0.563), respectively.Similarly, levodopa treatment was not associated with calcium frequency relative to naïve or 6-OHDA-lesion (p = 0.222, p = 0.225), respectively.However, it was associated with an increase in calcium in ux magnitude relative to naïve and 6-OHDA-lesion (p = 0.090, p = 0.101), respectively (Fig. 5).
We further assessed calcium activity in rats with high (R2, R8), mild (R3, R6, R7), and low (R1, R4) lesion levels to determine if SNc dopaminergic transmission in uences neuronal activity in M1.The frequency of calcium events was signi cantly reduced for rats with low dopaminergic lesion levels during the 6-OHDA-lesioned state (p = 0.046) and the levodopa treated state (p = 0.056), the magnitude of calcium in ux increased in the 6-OHDA-lesioned state (p = 0.038) with no changes in the levodopa treated state relative to naïve and 6-OHDA-lesion (p = 0.164, p = 0.367), respectively (Fig. 6a).The calcium events in R4 showed a signi cant reduction in frequency (p < 0.0001) and an increase in the magnitude of calcium in ux in the 6-OHDA-lesioned state (p < 0.0001) followed by improvement of calcium event frequency (p = 0.0002) and calcium in ux (p < 0.0001) with levodopa treatment.In contrast, no effects were observed in the 6-OHDA-lesioned state in R1 (calcium frequency p = 0.9675, calcium in ux p = 0.7716) (Fig. 6b).No signi cant effects in the frequency and magnitude of calcium events were observed in rats with mild lesions during the 6-OHDA-lesion and levodopa treatment states (Fig. 6d).Rats with a mild lesion level (R3 and R7) showed increased calcium event frequency (p < 0.0001).However, levodopa treatment only improved calcium event frequency in R3 (p = 0.0004).These rats showed no changes in calcium in ux magnitude (Fig. 6e).Levodopa treatment showed a reduction in frequency of calcium events in rats with high lesion levels compared to 6-OHDA-lesioned state (p = 0.07) and signi cantly increases in calcium in ux magnitude compared to naïve (p = 0.014) and 6-OHDA-lesioned (p = 0.003) states (Fig. 6g).Rats with high lesion levels exhibited different responses: R2 showed increased calcium event frequency in the 6-OHDA-lesioned state (p < 0.0001) that was recovered by levodopa treatment (p < 0.0001) and increased calcium in ux after levodopa treatment (p < 0.0001).In contrast, Rat 8 did not show signi cant changes in calcium event frequency (p = 0.7544) or calcium in ux magnitude (p = 0.1428) in the 6-OHDAlesioned state but showed decreased calcium event frequency (p < 0.0001) and increased calcium in ux magnitude (p < 0.0001) in the levodopa treatment state (Fig. 6h).
We con rmed the in uence of dopaminergic transmission on M1 neurons by tting the frequency and magnitude of calcium events and percentage of remaining SNc dopaminergic neurons using a simple linear regression model.The model showed that the frequency of calcium events in M1 was not affected by the gradual and steady loss of dopaminergic neurons in SNc during the 6-OHDA-lesioned state (F(1,571) = 2.40), p = 0.122).Instead, the model showed only a trend to decrease calcium in ux magnitude (F(1,571) = 3.77), p = 0.053).However, when dopaminergic transmission is replenished with levodopa treatment, the frequency of calcium events in M1 decreases (F(1,633) = 24.54,p < 0.0001) while the in ux magnitude of such events increases (F(1,633) = 63.24,p < 0.0001) with the loss of dopaminergic neurons (Fig. 7).

Discussion
M1 excitability is modulated by dopaminergic projections from the VTA and by the net effect of the direct and indirect pathways of the cortico-basal ganglia-thalamic circuitry. 10,116][17] However, multiple animal model and clinical studies report con icting results, showing increased, 5,18,19,25,26,[35][36][37][38] decreased, 29,[39][40][41][42][43][44] or unchanged 27,28,34,38 activation of M1.Here, we propose that such inconsistencies may result from studying M1 excitability at different levels of dopaminergic loss.An initial analysis of M1 activity shows subtle trends towards M1 overexcitation following 6-OHDA lesion and levodopa treatment (Fig. 5).However, when we consider the level of midbrain dopaminergic depletion achieved with 6-OHDA lesion, we identify distinct patterns of M1 activation.Speci cally, rats with less than 70% of dopaminergic depletion (low lesion level, Fig. 6a-c) showed decreased frequency and increased magnitude of calcium events with no signi cant effects in response to levodopa treatment.In contrast, rats with more than 30% of dopaminergic depletion (high lesion level, Fig. 6g-i) showed reduced frequency and signi cant increase in the magnitude of calcium events during levodopa treatment, with no signi cant changes during 6-OHDA-lesioned state.Additionally, rats with mild (30-69%) dopaminergic depletion (Fig. 6d-f) did not show changes in M1 activity.Importantly, it is worth noting that linear regression analysis reveals that the frequency of calcium events in the M1 tends to decrease while calcium in ux magnitude tends to increase as the dopaminergic lesion level increases when rats are treated with levodopa but remain unaffected during the 6-OHDA lesion state, i.e., in the absence of treatment (Fig. 7).
D1R and D2R may have similar effects in M1 depending on the cortical layers and neuronal types in which they are expressed, as well as the anatomical origin of dopaminergic depletion. 7,9,45However, their overall signaling effect through cortico-basal ganglia-thalamic circuitry may have opposing effects.In our study, the meso-cortical dopaminergic pathway and cortico-basal ganglia-thalamic circuitry were not severely damaged in rats with low dopaminergic depletion, but the latter might still have minor dysfunction given that direct 6-OHDA administration into the MFB would affect dopaminergic projections primarily in the nigrostriatal pathway. 46,47Thus, subtle imbalances in the indirect pathway could result in increased M1 inhibition or no signi cant changes since dopaminergic meso-cortical projections might sustain M1 activity. 48,49If dopaminergic transmission is supplemented with levodopa, signi cant changes would not be observed likely because dopaminergic signaling would not be su ciently imbalanced by the lesion.9 Thus, the net effect would be no noticeable changes in M1 activity.However, levodopa treatment would restore de cient dopaminergic transmission and allow for recovery of M1 activity.Interestingly, studies assessing M1 activity changes induced by deep brain stimulation (DBS) of the subthalamic nucleus (STN) report decrease and regularization in M1 ring rate with stimulation, 51 likely due to increased cortical GABAergic interneuron excitability [52][53][54] .The e cacy of both levodopa and STN DBS to alleviate parkinsonian symptoms in experimental animals and PD patients while modulating M1 activity reiterates the essential role of M1 in PD pathogenesis and supports the multi-circuitry regulation of M1 activity. Futur experiments evaluating changes in M1 activity in response to STN DBS are necessary to further elucidate potential mechanisms of M1 activity modulation.
Impaired motor performance and/or bradykinesia were observed in high lesion level rats.However, mild lesion level rats did not show ne motor impairment in the SPRT.This reinforces the notion that motor de cits in PD are observed when more than 70% of nigral dopaminergic neurons are lost and a ne balance of the direct and indirect cortico-basal ganglia-thalamic pathways is no longer achieved. 55In contrast to Metz et al. (2001), who did not detect motor improvement with chronic levodopa treatment, 56 we report a signi cant improvement in the performance of SPRT in the levodopa-treated state compared to the lesioned state with reduced number of attempts to successfully grasp and eat a pellet.Additionally, the attempt duration to reach a pellet increased after levodopa treatment, particularly when the pellet was present in the receptacle.In agreement with our ndings pertaining grasping duration, Hyland and colleagues (2019) showed no changes in the terminal part of the reaches possibly because this is the deceleration part of the movement when velocities in lesioned rats and controls may converge. 34,57This suggests that levodopa-treated animals developed a motor control strategy focused on goal achievement with ne tuning of proximal and distal movements to accurately reach and grasp the pellet.
Limitations of our study include the fact that not every rat included in the calcium imaging analysis was included in the behavioral analysis.Therefore, we were unable to correlate M1 calcium activity with speci c limb movements and draw spatiotemporal maps to evaluate neuronal ensembles encoding for speci c movements.Such information would further elucidate the role of dopamine in M1 neuronal modulation to execute speci c ne movements.Furthermore, it must be acknowledged that calcium imaging measures neuronal activity through the in ux of calcium into neurons, 58 which does not fully correspond to action potentials.Additionally, uorescent calcium imaging relies on signal emitted by active cells.It is possible that neurons become quiescent in the 6-OHDA lesioned state and that calcium imaging picks up the signals only from neurons that remain active.This would result in the study of active neurons only, neglecting the population of inactive neurons, which may not be representative of the entire M1 neuronal population.Future studies seeking to correlate neuronal calcium signaling and action potentials would enhance the accuracy and signi cance of calcium imaging analysis.
In summary, our study demonstrates the feasibility of long-term (over three months) monitoring of M1 neuronal activity via calcium signaling and simultaneous motor behavior in awake, freely behaving hemiparkinsonian rats and contributes to clarify the effects of disrupted midbrain dopaminergic transmission in M1.Furthermore, we demonstrate that while motor behavior is affected by severe changes in dopaminergic transmission, calcium activity in M1 is more susceptible to changes in response to quick uctuations in dopaminergic transmission, rather than to a stabilized level of dopaminergic transmission with a higher or lower amplitude.

Animals
We used a total of 13 adult (8-9 weeks old) Sprague Dawley rats, including 8 males and 5 females, with an approximate weight of 250-280 g.The rats were kept on a standard 12-hour light/dark cycle in single housing at a constant 21°C temperature and 45% humidity with ad libitum access to water and food.After approval by the Mayo Clinic Institutional Animal Care and Use Committee (IACUC), all animal procedures and experiments were conducted following the terms and guidelines of the National Institutes of Health for the use of animals and complied with the ARRIVE guidelines.

General Surgical Procedures
Anesthesia was induced using 5% iso urane and maintained using 1-2% iso urane via a nose cone.
Anesthetized rats were placed on heated stage (37°C) and the skull was secured with blunt ear bars and a nose clamp incisor bar in a Kopf stereotaxic frame (Kopf Systems, Tujunga, USA).Analgesia was provided pre-operatively with a 1.0 mg/kg buprenorphine extended-release injection.

Viral Transduction Procedure
On day one, the rat's head was shaved and disinfected with betadine, and a midline scalp 1.5-2 cm incision was performed starting between the eyes and extending caudally to the level of the ears.The periosteum was removed, and the skull cleaned with 0.9% saline and sterile cotton swabs.A 0.5 mm burr hole was made above the M1 (AP + 2.5 mm, ML + 2.6 mm from bregma, DV -1.8 mm from dura mater).
Coordinates for all brain regions targeted in the surgical procedures were determined using a rat brain atlas. 59The dura mater was incised to allow insertion of a 28-gauge needle and infusion of 500 nL of 110 vg/mL genetically encoded calcium indicator pENN.AAV9-CaMKII-GcaMP6f-WPRE-SV40 (Addgene Inc, Watertown, USA) at a rate of 100 nL/min.The needle was maintained in place for 5 min to prevent re ux and then withdrawn at a rate of 10 µm/s.The burr hole was covered with bone wax (Medtronic, Minneapolis, USA) and the skin incision was closed in one layer with simple interrupted stitches using 4 − 0 vicryl sutures.Rats were continuously monitored in a heated cage and returned to their home cage once they became ambulatory.

GRIN Lens Implantation Procedure
On day 14, rats were implanted with a guide-cannula over the MFB and a GRIN lens was implanted in M1.Anesthesia, analgesia, and surgical approach were identical to the those described for the viral transfection procedure.The burr hole above M1 was widened to approximately 1.5 mm and a 1 mm dorsoventral incision was made in the cortex with a straight-edge #11 dissection knife attached to a stereotactic arm perpendicular to the brain surface to insert a 1 mm GRIN prism lens (Inscopix, Palo Alto, CA) into M1.The exposed areas surrounding the lens were covered with a biocompatible silicone elastomer Kwik-Sil (World Precision Instruments, Sarasota, FL).A second burr-hole (1.5 mm) was drilled to insert a 22G guide-cannula 2.0 mm above the MFB.Two skull screws were placed bilaterally anterior to bregma and two additional screws were placed posterior to bregma on the contralateral side of the lens and guide-cannula to secure a headcap built with Metabond quick adhesive (Parkell, Edgewood, US).

Baseplating Procedure
On day 21, a baseplate (~ 0.5 g) was attached above the GRIN lens to allow for attachment and support of a miniature microscope.Rats were anesthetized and secured on the stereotaxic frame as described previously, but no analgesia was provided since this is a non-invasive procedure.Iso urane concentration was modulated to allow for visualization of calcium activity from M1 neurons via the GRIN lens to ensure correct placement of the baseplate as the baseplate was moved along three degrees of freedom (ML, AP, DV) in order to identify the location with maximal uorescence.Once this location was identi ed, the baseplate was secured in place using Metabond.

Food Restriction
Rats were put on a food restricted diet of 10-15 g standard food chow in order to entice them to grasp and eat sucrose pellets during the SPRT.Food restriction started 3 days prior to the rst training session and was maintained during testing for a maximum of 3 weeks or until rats reached 90% of their initial body weight.Rats were weighed daily during the food restriction period.

Chamber for Assessment of Motor Function
Motor function was assessed via SPRT using a custom built 360 x 80 mm chamber with transparent plastic sidewalls and a metal grid fence on one end, across which a plastic platform (88 x 45 x 25 cm) with receptacles for sucrose pellets was located just outside the fence.The fence openings were just wide enough for the rat to thrust out its forelimbs and grasp a single 45 mg sucrose pellet.

SPRT Assessments of Motor Function Habituation and Training Sessions
Training sessions started 5-7 days following baseplate implantation.One day before training, rats were habituated to the testing chamber for 5 min while receiving sucrose pellets.During the training sessions, rats were gently guided to reach through the fence and grasp sucrose pellets placed on the platform receptacle.The SPRT consisted of 25 trials in which the rat was allowed to grasp and eat a total of 25 sucrose pellets or until the time limit of 15 min expired.Each trial started with the placement of a sucrose pellet in the platform receptacle and ended after the rat ate the pellet.If the rat pushed the pellet beyond its reach, a new sucrose pellet was placed in the platform and was considered the same trial.
Rats were connected to a dummy miniature microscope during the training sessions to habituate the animals to the weight and size of the microscope used during testing.

Testing Sessions
Test sessions began once rats were able to successfully grasp and eat 25 pellets within 15 minutes (usually after 3 to 5 days of training).Testing was conducted for 3 to 5 consecutive days, where rats were connected to a skull-mounted miniature microscope to record calcium activity from M1. Synchronized calcium activity and pellet reach behavioral data were acquired during test sessions.
The number of attempts in a given trial to successfully grasp and eat a sucrose pellet was not limited (Fig. 1B).An investigator blinded to the treatment analyzed and quanti ed the rats' forelimb movements according to the following criteria: full grasp: extension of paw beyond the fence and grasping of a pellet.The success ratio was calculated based on categorization of full grasps in successful if the rat ate the pellet or failure if the rat did not eat or dropped the pellet after grasping it.
reach without grasp: extension of the paw beyond the fence allowing to touch the pellet but not grasping it.
grasp without pellet: extension of paw beyond the fence to attempt to grasp a pellet in the absence of a sucrose pellet in the receptacle.
attempt duration: starts when paw lifts from the oor, extends beyond the fence, and ends when pellet is brought to mouth (full grasp) or when paw is placed back on the oor (reach without grasp).
reaching duration: starts when paw lifts from the oor, extends beyond the fence, and ends when the pellet is grasped.
grasping duration: starts when pellet is grasped and ends when pellet is put in mouth or dropped.

6-OHDA lesion and levodopa treatment
Rats were unilaterally infused with 3 µl of 6-OHDA hydrobromide (4 µg/µl solution of 0.02% sterile ascorbate saline) at a rate of 0.5 µl/min with 5 min for diffusion into the MFB the day after SPRT test sessions in the naïve state.To prevent 6-OHDA uptake by noradrenergic neurons, desipramine hydrochloride (25 mg/kg i.p.) was administered 30 min before 6-OHDA infusion.A period of 3 weeks was allowed following 6-OHDA lesioning for dopaminergic lesion stabilization.After stabilization, rats received a daily 10 mg/mL levodopa (Sigma Millipore, Burlington, MA, CAG1361009) intraperitoneal (i.p.) injection co-administered with a 1.25 mg/mL carbidopa decarboxylase inhibitor (Sigma Millipore, Burlington, MA, CAG1095506) dissolved in 0.9% saline for 14 days to restore dopaminergic levels while preventing peripheral levodopa degradation and increasing central concentration.SPRT habituation and training was initiated after 3-5 days of levodopa/carbidopa treatment.

Histology
At the end of all experiments, rats received an i.p. overdose of pentobarbital sodium (100 mg/kg) and were transcardially perfused with 0.1 M phosphate buffered saline (PBS) followed by 4% paraformaldehyde.Rats were decapitated, their scalp exposed, skull opened, and their brains extracted and xed overnight in 4% paraformaldehyde.Brains were stored in 30% glycerol.Cryosectioning with 40 µm slices was performed on a sliding microtome (Leica Biosystems, Wetzlar, Germany).Tissue sections were stored in 0.1% Sodium Azide in 0.1 M PBS.
Coronal M1 slices were mounted onto glass slides and coverslipped with VectaShield containing DAPI (Vector Laboratories, Newark, CA) to assess PRISM lens placement and GCaMP6f expression.If the location of the PRISM lens or GCaMP6f expression were not in the M1 region, the corresponding rats were excluded from the analysis.To quantify dopaminergic cell loss, we performed immunohistochemical staining using tyrosine-hydroxylase (TH).Coronal slices of the SNc were rinsed in 0.01 M PBS with 0.2% Triton X-100 (PBS-Tx), incubated in 3% H 2 O 2 for 10 min to quench endogenous peroxidase activity and blocked in 10% normal goat serum (NGS) in PBS-Tx for 1 h.Slices were incubated with the primary antibody, anti-TH (rabbit polyclonal, Abcam 1:4000), overnight at 4 O C on a shaker.Following primary antibody incubation, sections were rinsed in PBS-Tx and incubated in 10% NGS with the biotinylated goat anti-rabbit IgG secondary antibody (Vector Laboratories, 1:500) for 1 h at room temperature.Following the rinse in PBS-Tx, slices were incubated with an avidin-biotin enzyme complex (VECTASTAIN® Elite® ABC-HRP kit, PK-6101, Vector Laboratories), rinsed in 0.1 M PBS, and incubated with 3,3'-Diaminobenzidine chromogen (DAB Substrate kit, SK-4100, Vector Laboratories).After nal rinse in 0.1 M PBS, slices were mounted onto glass slides and coverslipped with mounting media (Eukitt® Quick-hardening mounting medium, 03989, Sigma-Aldrich).Slides were imaged under a microscope (Keyence, BZ-X800).

Characterization of Dopaminergic Lesion
Dopaminergic neuron loss was assessed by bilateral quanti cation of TH-positive neurons in the SNc.Two to four SNc-containing midbrain sections from each animal were evaluated and the relative ratio of TH + cells was determined as the percentage of TH + cells in the lesioned SNc relative to the intact SNc.Midbrain sections of R13 were damaged and densitometry of TH + terminals was conducted to evaluate the level of dopaminergic depletion for this animal (Supplementary Figure S2).The number of TH + neurons in the SNc of the 6-OHDA-lesioned hemisphere (right side) was compared to the number of TH + neurons on the non-lesioned control hemisphere (left side).

Calcium data collection
Cortical calcium imaging was performed using an nVista miniature microscope (Inscopix, Palo Alto, CA) weighing approximately 2 g and attached to the baseplate on the rat's head while rats moved freely in the SPRT chamber.Real-time imaging was transmitted to a computer via a data acquisition (DAQ) system connected to the miniature microscope.Each test session rendered one or more videos of calcium activity.Thus, each state (naïve, 6-OHDA-lesioned, and levodopa-treated) produced at least four videos for data analysis.

Calcium data analysis
Calcium data for each state were compared and only the videos with the same eld of view were stitched to allow uniform preprocessing and longitudinal registration of the neuroanatomy.Calcium data across states were not processed longitudinally due to differences in eld of view.Each calcium data set was preprocessed by spatially down sampling each video (by a factor of two), cropping the eld of view to the region with increased calcium signals, and motion correcting using the Turboreg algorithm implemented in the Inscopix Data Processing Software (IDPS) (Inscopix, Palo Alto, CA).Data were then exported as ISXD les and loaded into MATLAB (The MathWorks, Natick, MA) to extract putative neurons or regions of interest (ROIs) using a constrained non-negative matrix factorization (CNMF-E) algorithm. 60he ring model of background uorescence was used for all data sets.Acceptance parameters were selected based on the data and included neuron diameter ≥ 7px, signal-to-noise ratio (SNR) between 3 and 5. Next, the CNMF-E-detected ROIs were loaded into IDPS for visual inspection of the shape and traces of each ROI and additional inclusion criteria were applied by the observer to ensure that (i) a single ROI outlined only one neuron, (ii) a single neuron was not labeled by multiple ROIs, and (iii) selected ROIs were present in all videos within each state.ROIs that didn't meet these criteria were excluded from further analysis.
A peak detection algorithm was used to extract individual calcium events from raw calcium intensity traces for each ROI in MATLAB.Additionally, the area under curve (AUC) was calculated to measure the total calcium in ux using the event onset and offset of each event, indicating the activity level for each speci c neuron.Calcium intensity for selected ROIs was converted to a z-score scale to allow comparisons between rats across states.The calcium event rate was calculated using Eq. ( 1).The average calcium in ux was determined by Eq. ( 2) in each state.Finally, event rate and average calcium in ux data were used for statistical analysis and effect comparisons between states in GraphPad Prism. ( (2)

Statistical Analysis
Statistical analysis of the SPRT was performed using a repeated measures two-way ANOVA with subject and state as the independent variables.Calcium imaging analysis was performed using a mixed-model effects to account for the different number of neurons recorded in each state.Tukey's multiple comparisons test was used to assess signi cant differences between states in both behavioral and calcium activity analyses.Analysis of calcium activity for individual animals was performed using a twoway ANOVA followed by a Tukey's multiple comparisons test to evaluate the effects of each state on M1 calcium activity.The effects of dopaminergic lesion on M1 calcium activity were further assessed by applying a simple linear regression model of the average of calcium event frequency and the average of calcium in ux magnitude for each rat relative to their percentage of TH + neurons.Neuronal calcium activity in the motor cortex of rats.Single-photon calcium recordings were acquired during naïve, 6-OHDA-lesion, and levodopa treatment states.Unilateral injection of 6-OHDA in the MFB and subsequent levodopa treatment produced changes in the M1 calcium activity.Data were analyzed using a linear mixed model with a random effect for animal.The average for each rat is displayed as median and interquartile range with the shape of the violin plots representing the distribution of individual data points.N = 7 rats (ranging within an average of 91 neurons recorded during naïve, 81 after 6-OHDA lesion, and 90 during levodopa treatment).

Supplementary Files
This is a list of supplementary les associated with this preprint.Click to download. SupplementaryInformation.pdf

Figures Figure 1
Figures

Table 1
Description of histological assessment and reasons for removal from data analysis.