Introduction

Parkinson’s disease (PD) is neuropathologically characterized by two defining hallmarks: (1) the presence of α-synuclein (αSyn)-containing Lewy bodies and Lewy neurites, and (2) a significant loss of dopaminergic neurons in the substantia nigra pars compacta (SNc)1,2. Although neither hallmark is exclusive to PD, their co-occurrence is required for a definitive diagnosis. Missense mutations and duplication/triplication in the SNCA gene, encoding αSyn, results in PD and the related disorder dementia with Lewy bodies, strongly implicating that that αSyn pathobiology is a crucial factor in disease pathogenesis.

αSyn is a small, natively unfolded protein that associates with curved membranes, particularly synaptic vesicles, where it modulates neurotransmitter release3,4,5. We and others have shown that sustained AAV‑TH–hαSyn (hαSyn) over‑expression or αSyn preformed fibrils (PFF) exposure can lead to the degeneration of nigrostriatal dopamine neurons6,7,8,9, suggesting that in a disease state, αSyn misfolds and aggregates into oligomeric and fibrillar forms, contributing to dopaminergic neuron dysfunction, and ultimately–cell death6,9,10,11,12. These aberrant αSyn conformations can disrupt multiple intracellular processes, including vesicle trafficking, synaptic transmission, neuronal firing, and mitochondrial function12,13,14,15,16. Furthermore, αSyn aggregates can propagate between neurons via several proposed mechanisms, including calcium-dependent pathways, potentially spreading pathology throughout the brain17,18. αSyn’s role in dopamine signaling is complex, exhibiting both facilitatory and inhibitory effects that depend on neuronal activity, the specific molecular targets engaged, and the duration of αSyn deposition5. Previous studies have shown that, prior to neurodegeneration, excess αSyn impairs dopamine transporter (DAT)-mediated dopamine clearance and promotes DAT-dependent, non-vesicular dopamine efflux, leading to increased extracellular dopamine levels19,20,21. Beyond dopamine homeostasis, αSyn overexpression disrupts the intrinsic firing properties of dopamine neurons and alters their calcium dynamics15,22,23. These perturbations are hypothesized to destabilize neuronal energy balance and increase neuronal susceptibility to degeneration16,24,25. Notably, αSyn pathology and associated toxicity are more pronounced in the SNc, which is highly susceptible to dopaminergic neuron loss, compared to the relatively resilient, adjacent ventral tegmental area (VTA)6.

A critical question in PD research is why SNc dopamine neurons are selectively vulnerable compared to VTA neurons. SNc neurons possess a distinct vulnerability profile, including αSyn-reduced tyrosine hydroxylase (TH) expression, greater dendritic arborization (and thus higher metabolic demands), extensive axonal projections, and differential expression of proteins such as calbindin, L-type Ca2+ channels, and vesicular glutamate transporter 2 (VGLUT2)26,27,28,29,30,31,32. Our previous work, using primary cultures of dopamine neurons overexpressing αSyn and human iPSC-derived dopamine neurons with SNCA triplications, demonstrated that αSyn overexpression alters dopamine neuron firing patterns and dopamine release22,23. However, the differential effects of αSyn deposition on the intrinsic activity of SNc versus VTA dopamine neurons within an intact midbrain circuit remain poorly understood. Elucidating these pre-degenerative changes in dopaminergic neuronal activity and network connectivity in these adjacent dopaminergic nuclei is crucial for developing therapeutic strategies to slow disease progression and restore neuronal function and resilience.

Multiple studies have shown that sustained AAV‑TH–hαSyn over‑expression or αSyn PFF exposure leads to the degeneration of nigrostriatal dopamine neurons6,7,8,9. Our objective in this study was to capture the functional changes that arise before overt cell loss; consequently, we restricted our analyses to our experimentally identified pre‑degenerative window in these two model systems. We found no evidence of cell loss 4 weeks after viral-mediated hαSyn overexpression or preformed fibril (PFF) deposition in the SNc and VTA. This indicates that, in these two animal models of synucleinopathies, we captured an early, pre-degenerative stage of αSyn deposition in dopamine neurons, prior to neurodegeneration. We therefore leveraged these two models to examine the effects of αSyn deposition on the intrinsic electrophysiological properties of SNc and VTA dopamine neurons, their resilience to hyperpolarization-induced challenge, and the connectivity of their dopaminergic networks. Our data demonstrate that both αSyn overexpression and PFF deposition alter the electrophysiological and network connectivity of SNc dopamine neurons, significantly impairing their capacity to maintain homeostatic function, while sparing VTA dopamine neurons. These findings shed light on the early pathophysiological changes contributing to SNc neuron vulnerability in synucleinopathies and provide a foundation for developing interventions aimed at preventing or slowing degeneration of dopamine neurons.

Results

Establishing a pre-degenerative model of synucleinopathy

To establish a pre-degenerative model of synucleinopathy in these two animal models, we characterized dopaminergic neuron survival and changes in αSyn following unilateral microinjection of either a virus expressing human α-synuclein (hαSyn; 1 µL of AAV-TH-hαSyn) or αSyn PFF (2 µg/µL; 0.5µL per region) into the VTA and the SNc (Fig. 1). Our use of mouse αSyn PFFs and AAV-TH–driven human αSyn expression was an intentional, evidence-based decision9,22. Human αSyn expression permits monitoring of transgene distribution with antibodies that selectively recognize the human protein, and we have previously validated this rAAV construct in published work22. Conversely, extensive data show that mouse αSyn PFFs more efficiently seed endogenous mouse αSyn and propagate pathology than human αSyn PFFs in murine models9. This dual strategy therefore maximizes both experimental sensitivity and pathophysiological relevance. At 4 weeks post-injection, a timepoint chosen to precede significant neuronal loss, we assessed neuronal viability in midbrain slices encompassing the VTA and SNc using TUNEL staining and immunohistochemistry. The positive control for the TUNEL assay exhibited clear DNA fragmentation, confirming assay validity. However, no detectable TUNEL-positive cells were observed in either the VTA or SNc of either hemisphere following hαSyn or αSyn PFF injection, indicating the absence of significant apoptosis (Fig. 1B). Immunohistochemical analysis was also performed for anti-TH, pan-neuronal marker FOX3, and either phosphorylated αSyn (pSer129 αSyn; antibody 81 A) to detect aggregated forms of αSyn following αSyn PFF injection, or hαSyn (antibody Syn211) following AAV injection. While the non-injected SNc and VTA of the contralateral hemisphere showed no 81 A or Syn211 immunoreactivity (Fig. 2C, E, G, I), the SNc of the injected hemisphere exhibited staining for hαSyn or punctate 81 A staining, confirming successful AAV-mediated expression of hαSyn or the presence of aggregated αSyn from PFF seeding, respectively (Fig. 2D, H). Notably, despite identical injection parameters, the VTA showed minimal hαSyn or 81 A immunoreactivity compared to the SNc (Fig. 2F, J). Quantification of TH-positive (TH+) neurons revealed no significant difference between the injected hemisphere and the contralateral non-injected hemisphere in either the SNc or VTA (Fig. 2K, M), although a non-significant trend towards reduced TH+ cell number was observed in PFF-injected animals (Fig. 2M). Furthermore, the number of TH+ and FOX3+ neurons remained unchanged (Figs. 2L, N), indicating preserved dopaminergic neuron density. Collectively, these data demonstrate that neither hαSyn expression nor the presence of αSyn PFF induces measurable dopaminergic neuron loss 4 weeks post-injection, confirming that this experimental paradigm represents an early, pre-degenerative stage of synucleinopathy. Multiple studies have shown that sustained AAV‑TH–hαSyn over‑expression or αSyn PFF exposure leads to neurodegeneration at later time points6,7,8. Given the absence of overt neuronal loss in our animal models of synucleinopathy, we next investigated whether hαSyn or αSyn PFF-induced deposition alters the intrinsic electrophysiological properties of VTA and SNc dopamine neurons at this pre-degenerative stage.

Fig. 1: Assessment of neuronal viability in the VTA and SNc after AAV1-TH-hαSyn and αSyn PFF injections using TUNEL assay.
figure 1

A Schematic of the experimental design: Mice received injections of AAV1-TH-hαSyn (1 µL) or αSyn PFF (2 µg/µL; 1µL per hemisphere; 0.5µL in the VTA and 0.5µL in the SNc) into the midbrain. Four weeks post-injection, brain sections were processed for TUNEL staining to assess apoptotic cell death. B Representative fluorescent images of DAPI+ nuclei and TUNEL+ cells in positive control and experimental sections. Midbrain sections from the contralateral (non-injected) and the ipsilateral (injected) sides of the AAV1-TH-hαSyn and αSyn PFF injection sites display DAPI+ nuclei but are entirely TUNEL-negative, indicating no detectable apoptotic cell death. This suggests that neurons in the VTA and SNc remain viable at this time point. Scale bar = 20 µm.

Full size image
Fig. 2: AAV-TH-hαSyn or αSyn pre-formed fibrils (PFF) increase corresponding immunostaining in the substantia nigra pars compacta (SNc) and ventral tegmental area (VTA).
figure 2

A Experimental design: Mice received unilateral injections of either AAV1-TH-hαSyn (1 µL) or αSyn PFF (2 µg/1 µL; 1µL per hemisphere; 0.5µL in the VTA and the SNc) into the midbrain. B Representative coronal section of the ventral midbrain (SNc and VTA), showing the AAV1-TH-hαSyn injected (ipsilateral; right) and non-injected (contralateral; left) hemispheres. Four weeks post-injection (C–F), immunostaining with Syn211 (detecting hαSyn) shows colocalization in TH+ neurons on the injected side in the SNc (D) and VTA (F), but not on the non-injected side respectively (C, E). The same timepoint but following injection with αSyn PFF (G-J), 81 A (detecting phosphorylated αSyn aggregates) shows punctate signals localized within TH+ neurons on the injected side in the SNc (H) and VTA (J), but not on the non-injected side respectively (G, I). These data confirm the presence of hαSyn and aggregated αSyn in both the VTA and SNc. K, M Four weeks after injection of either AAV1-TH-hαSyn or αSyn PFF, there was no significant change in the number of TH+ neurons in the SNc or VTA. (L, N) Similarly, no differences were observed in the number of TH+ and FOX3+ neurons in the SNc or VTA. Error bars represent mean ± SEM. (n = 3 independent biological replicates, one-way ANOVA and t-test, *p < 0.05) Scale bars: 250 µm (B); 20 µm (C–J).

Full size image

hαSyn increases basal firing activity of SNc dopamine neurons but not VTA dopamine neurons

Prior research indicates that SNc dopamine neurons are more vulnerable to neurodegeneration, while VTA dopamine neurons exhibit relative resilience26,33,34,35,36. To investigate the impact of αSyn pathology (using hαSyn or αSyn PFF models) on these adjacent neuronal populations, we conducted whole-cell patch-clamp electrophysiology on midbrain slices (Figs. 3A, 4A). Spontaneous firing rates were measured in both the SNc and VTA regions from the injected ipsilateral and non-injected contralateral hemispheres (Figs. 3B, C, 4B, C). Single-neuron recordings examining hαSyn expression were performed 4 weeks following hαSyn expression (Fig. 3B, C). Consistent with previous studies22,23,37, hαSyn expression in the SNc significantly increased the basal firing frequency of dopamine neurons on the injected side compared to the non-injected hemisphere (t-test, p < 0.05; Fig. 3D). Conversely, PFF-induced deposition did not alter the basal firing activity of SNc dopamine neurons (t-test, p > 0.05; Fig. 4D). Basal firing rates of VTA dopamine neurons remained unchanged by either haSyn or αSyn PFF (Figs. 3G, 4G). Furthermore, action potential morphology, including half-width and membrane capacitance, was unaltered in both SNc and VTA neurons (Figs. 3E, F, H, I, 4E, F, and 4H-I). This suggests that the hαSyn-induced increase in firing activity of SNc dopamine neurons does not stem from alterations in intrinsic membrane properties or cell size at 4 weeks post-hαSyn expression or αSyn PFF injection. These findings offer insights into the early pathophysiology of synucleinopathies, preceding neuronal degeneration. Specifically, hαSyn has been shown to increase calcium activity through D2 receptor and dopamine transporter (DAT)-mediated mechanisms20,22. However, the underlying mechanisms of αSyn PFF-induced dysregulation/degeneration remain unknown. Moreover, VTA neurons maintained stable firing patterns (Figs. 3G, 4G). While VTA dopamine neuron activity is preserved at 4 weeks post-hαSyn expression and αSyn PFF injection, prolonged hαSyn expression may eventually impair the function of these neurons. The “non‑injected” data in Figs. 3 and 4 were collected from independent cohorts. The non-significant shifts in baseline firing, especially in VTA dopamine neurons could reflect typical variabilities across biological replicates. Because all analyses compare the injected side with its own contralateral control, these cohort‑specific baselines did not affect the main conclusions. These results underscore the importance of investigating how synucleinopathies affect neuronal resilience, particularly in response to hyperpolarization-mediated disruptions of cellular homeostasis.

Fig. 3: hαSyn alters the electrophysiological properties of dopaminergic neurons in the SNc but not in the VTA dopamine neurons.
figure 3

A Schematic representation of the experimental design for single-neuron recordings conducted 4 weeks post hαSyn injection (1 µL) into the midbrain. B Representative electrophysiological traces from SNc dopamine neurons in the non-injected (no- inj) and hαSyn-injected (inj) sides. D The basal firing frequency of SNc dopamine neurons is significantly increased on the ipsilateral side post-hαSyn injection. C Representative traces from VTA dopamine neurons on the contralateral and ipsilateral sides, and quantification in G demonstrating that hαSyn injection does not alter the basal firing frequency of VTA dopamine neurons. E, F Bar graphs depicting the half-width and membrane capacitance of SNc dopamine neurons, respectively. H, I Bar graphs depicting the half-width and membrane capacitance of VTA dopamine neurons, respectively. Error bars represent the mean ± SEM. Data were obtained from 3 - 4 mice, resulting in a sample size of n = 8 - 15 neurons. Statistical significance was determined using a t-test (*p < 0.05).

Full size image
Fig. 4: The presence of αSyn PFF does not change the electrophysiological properties of dopaminergic neurons in the SNc or VTA dopamine neurons.
figure 4

A Schematic representation of the experimental design for single-neuron recordings conducted 4 weeks post-unilateral αSyn PFF deposition (2 µg/µL; 1µL per hemisphere; 0.5µL in the SNc and 0.5µL in the VTA) into the midbrain. B Representative electrophysiological traces from SNc dopamine neurons in the non-injected (no-inj) and PFF-injected (inj) sides. D The basal firing frequency of SNc dopamine neurons is similar in αSyn PFF-injected and non-injected mice. C Representative traces from VTA dopamine neurons from PFF-injected and non-injected side with quantification in G demonstrate that αSyn PFF injection does not alter the basal firing frequency of VTA dopamine neurons. E, F, H, I Bar graphs depicting the half-width and membrane capacitance of SNc and VTA dopamine neurons. Error bars represent the mean ± SEM. Data were obtained from 4-6 mice, resulting in a sample size of n = 5 - 21 neurons. Statistical significance was determined using a t-test.

Full size image

αSyn PFF and hαSyn increase SNc dopaminergic neuron vulnerability to hyperpolarization challenge, while VTA neurons remain resilient

Midbrain (SNc and VTA) dopamine neurons maintain spontaneous pacemaking activity, relying on finely tuned homeostatic mechanisms. To assess how hαSyn or αSyn PFF affects neuronal resilience under stress, we applied a hyperpolarization protocol38 using whole-cell patch-clamp electrophysiology in slices from mice injected 4 weeks prior with AAV1-TH-hαSyn (Fig. 5A), or αSyn PFF (Fig. 6A) in the midbrain. This approach allowed us to compare the pre- and post-hyperpolarization firing activity of SNc and VTA dopamine neurons from injected ipsilateral hemisphere and the non-injected contralateral hemisphere. Consistent with recent studies, we found that on the non-injected side, firing properties of dopamine neurons in both VTA and SNc regions returned to their pre-hyperpolarization states38,39. Therefore, we used a 30 second hyperpolarization protocol to challenge the neuronal homeostasis in a time-resolved manner. Interestingly, SNc dopamine neurons on the side injected with either AAV1-TH-hαSyn (Fig. 5), or αSyn PFF (Fig. 6) displayed significantly reduced firing rates following hyperpolarization-induced stress, while those on the contralateral non-injected hemisphere quickly recovered to baseline firing activity (Figs. 5F and 6F). In contrast, VTA dopamine neurons demonstrated comparable resilience on both injected ipsilateral hemisphere and the non-injected contralateral hemisphere, maintaining stable firing rates after the hyperpolarization challenge (Figs. 5G and 6G). These data uncover a previously unrecognized phenotype: unlike the VTA, SNc dopamine neurons with increased intracellular hαSyn (overexpression model) fail to regain their baseline firing rate, many of which would become entirely silent with exception of a few non-pacemaking action potentials. These results underscore the SNc’s susceptibility to αSyn alterations. Over time, this increased vulnerability may contribute to the loss of SNc dopamine neurons seen in PD. Collectively, our data show that before any detectable cell loss (our defined pre‑degenerative phase in these two models; Fig. 3), elevated intracellular hαSyn accelerates the pacemaking activity of SNc dopamine neurons, pushing them beyond their normal homeostatic neuronal activity. This heightened excitability leaves SNc neurons vulnerable to further homeostatic perturbation, suggesting that these early functional disturbances may prime them for subsequent neurodegeneration. In contrast, VTA neurons remain relatively unaffected at 4 weeks post-hαSyn expression or αSyn PFF deposition, supporting their reported resilience33,35,40. Understanding these initial functional changes provides a critical foundation for developing therapeutic strategies aimed at preserving dopaminergic function and preventing the progression of αSyn-associated pathology. In addition, although understanding αSyn pathology at the single-neuron level is important, it is equally important to investigate how αSyn pathology affects dopamine neuronal networks through their integrated activity. Therefore, to bridge the gap from single-neuron physiology to ensemble-neuronal network level, we next examined how hαSyn expression or αSyn PFF deposition influence network connectivity in the SNc and VTA.

Fig. 5: hαSyn overexpression compromises the resilience of SNc dopamine neurons to hyperpolarization challenge.
figure 5

A Experimental design: Four weeks after unilateral hαSyn injection (1 µL), basal firing frequency of SNc dopamine neurons was recorded both ipsilaterally (injected side; inj) and contralaterally (non-injected side; no-inj). Neurons were then hyperpolarized to –100 mV for 30 second. Following restoration of the resting membrane potential, post stress firing frequency was assessed to determine resilience to hyperpolarization. B, C Representative electrophysiological traces from SNc dopamine neurons before and after the 30-second hyperpolarization step. F The accompanying graph shows the percentage change in firing frequency, revealing a diminished capacity of SNc neurons to recover following hyperpolarization stress in the presence of hαSyn. D, E Representative traces from VTA dopamine neurons at baseline and post-hyperpolarization. G The associated graph demonstrates that VTA neurons effectively buffer the hyperpolarization-induced perturbation, maintaining firing rates without significant changes after hαSyn overexpression. Error bars represent mean ± SEM. Data was obtained from 3-4 mice, with a sample size of n = 4 - 13 neurons. Statistical significance was assessed by t-test (*p < 0.05).

Full size image
Fig. 6: αSyn PFF deposition impairs the resilience of SNc dopamine neurons to hyperpolarization challenge.
figure 6

A Experimental design: Four weeks post unilateral αSyn PFF deposition (2 µg/µL; 1µL per hemisphere; 0.5µL in the SNc and 0.5µL in the VTA), the basal firing frequency of SNc dopamine neurons was recorded both ipsilaterally (injected side; inj) and contralaterally (non-injected side; no-inj). Neurons were then hyperpolarized to –100 mV for 30 s. After returning to resting membrane potential, post-stress firing frequency was measured to assess resilience. B, C Representative electrophysiological traces from SNc dopamine neurons before and after the 30-second hyperpolarization step. F The corresponding graph quantifies the percentage change in firing frequency, showing a reduced capacity of SNc neurons to recover following hyperpolarization stress when exposed to αSyn PFF. (D, E) Representative traces from VTA dopamine neurons at baseline and post-hyperpolarization. G The associated graph indicates that VTA neurons maintain stable firing frequencies and remain unaffected by PFF deposition under these conditions. Error bars represent mean ± SEM. Data were obtained from  4–6 mice, resulting in a sample size of n = 6 - 22 neurons. Statistical significance was determined by t-test (**p < 0.01).

Full size image

During the pre-degeneration phase, hαSyn and αSyn PFF deposition increased network hyperconnectivity in the SNc, but not the VTA

To investigate the impact of αSyn PFF or hαSyn accumulation on dopamine network dynamics in the SNc and VTA, we injected DAT:cre mice with a combination of AAV5-FLEx-GCaMP8f (0.3 µL, see Table 1) and either AAV1-TH-hαSyn or αSyn PFF into the SNc or VTA. GCaMP8f is a genetically encoded calcium indicator, serving as a surrogate for studying neuronal activity41. Changes in GCaMP8f fluorescence intensity provide a real-time, non-invasive approach to monitor the activity of individual neurons and neuronal populations. Zhang et al. has characterized GCaMP8f sensor notating it has a half rise time of 6.6 ms and half decay time of 87 ms. The response time of GCaMP8f allows for capturing neuronal firing, improving the temporal resolution41. Post- αSyn PFF or AAV1-TH-hαSyn-injection, midbrain slices were prepared and imaged to monitor GCaMP8f fluorescence signal as an indicator of neuronal activity (Fig. 7A–C). Calcium imaging videos were processed using MIN1PIPE (Miniscope 1-Photon–Based Calcium Imaging Signal Extraction Pipeline; see Methods), which extracts neural signals and identifies regions of interest (ROIs) with spatial footprints. From these data, calcium traces were isolated for individual neurons (Fig. 7C), and raster plots of neuronal activity were generated for the SNc and VTA, binned at 100-second intervals (Fig. 7D). All traces were then correlated using Spearman’s rank correlation, applying a significance threshold of p < 0.05 for network analysis. Neurons (nodes) with significant connections are labeled “1” in yellow, whereas nodes with self-connections or non-significant connections are labeled “0” in green (Fig. 7E), with the corresponding network graph shown in Fig. 7F.

Fig. 7: Investigating hαSyn- and PFF-induced regulation of dopamine neuronal network connectivity in the SNc and VTA.
figure 7

A Schematic representation of the experimental design: DAT:cre mice were bilaterally injected with AVV5:FLEx-GCaMP8f (300 nL) for live cell calcium imaging and unilaterally received AAV-TH-hαSyn (1 µL) or αSyn PFF (2 µg/µL; 1µL per hemisphere; 0.5µL in the SNc and 0.5µL in the VTA). B Representative GCaMP8f-labeled dopamine neurons. C Normalized calcium traces from dopamine neurons in the SNc (top) and VTA (bottom). D Raster plots of neuronal firing activity show temporal patterns of calcium signaling in individual dopamine neurons. E Adjacency matrices illustrating neuronal connectivity, where “1” denotes a connection between two neurons and “0” denotes no connection. F Network connectivity graphs derived from adjacency matrices. Each node represents an individual neuron, with the size and intensity of color indicating the number of connections for that neuron (greater size, larger number of connections and darker color reflect higher connectivity). These analyses highlight the differential network connectivity of dopamine neurons in the SNc and VTA under conditions of hαSyn or αSyn PFF exposure.

Full size image
Table 1 Reagents List
Full size table

To confirm that the measured neuronal activity stemmed from fluctuations in the fluorescent signals of individually isolated neurons within the network, we examined the relationship between inter-neuron distance and correlated activity in all cells. Across all experimental conditions (hαSyn and αSyn PFF, both injected hemisphere and non-injected hemisphere) in the SNc and VTA, we observed no relationship between distance and correlation (Supplemental Fig. 1A–G). This finding suggests that changes in network connectivity parameters were neither due to fluorescent bleed-over from nearby neurons nor neuronal death (inactivity). Among the network connectivity parameters measured were node degree, clustering coefficient, network density, and global efficiency42,43.

Node degree describes the number of links (interconnections) a node has with other nodes. Since the total number of connections is influenced by the total number of neurons in the region of interest, node degree is normalized by dividing by the total number of nodes, yielding a normalized node degree. This metric represents the average number of connections per neuron. Consistent with our electrophysiological data, both hαSyn and αSyn PFF enhanced network connectivity (node degree) among SNc dopamine neurons relative to control conditions. Specifically, SNc dopamine neurons injected with AAV1-TH-hαSyn or αSyn PFF exhibited higher normalized node degree distributions, as well as increased average normalized node degree compared to controls (p < 0.001 and p < 0.05, Fig. 8A–C, E–F). In contrast, VTA network connectivity measures (normalized node degree) remained unchanged (Fig. 8D, G–H). These findings suggest that hαSyn or αSyn PFF accumulation leads to a hyperconnected dopamine neuronal network specifically in the SNc, correlating with the increased firing rates observed both in the present study and in previous reports22,23. Thus, prior to cell death in synucleinopathies, SNc dopamine neurons bear a heightened physiological burden, evident at both the single-neuron and network levels.

Fig. 8: Both hαSyn expression and αSyn PFF deposition increase hyperconnectivity in the SNc, but not the VTA.
figure 8

A Left panel: Network graphs of the SNc for non-injected and the hαSyn-injected hemispheres (AAV-TH-hαSyn, 1 µL). B Network graphs of the SNc for non-injected and the αSyn PFF-injected hemispheres (PFF, 2 µg/µL; 1µL per hemisphere; 0.5µL in the SNc and 0.5µL in the VTA). C Bar graphs of the frequencies of normalized node strengths in non-injected versus hαSyn- or αSyn PFF-injected SNc. D Bar graphs of the frequencies of normalized node degree in non-injected versus hαSyn- or αSyn PFF-injected VTA. E–H Pie graphs of the frequencies of the node strength above the mean of non-injected and hαSyn- or αSyn PFF-injected side in the SNc and VTA. Data was collected from 3 - 5 mice with a sample size ranging from n = 14–46 nodes (t-test, *p < 0.05, ***p < 0.001).

Full size image

We also examined higher-order network metrics, namely clustering coefficient, network density, and global efficiency (Supplemental Fig. 2). The clustering coefficient is a measure of functional segregation that quantifies the proportion of triangular connections around each node. A high clustering coefficient indicates networks in which neurons share many mutual neighbors (i.e., subpopulations of highly interconnected nodes), whereas a lower value indicates a more decentralized or random structure. We observed no change in the mean clustering coefficient following hαSyn or αSyn PFF injection in either the SNc or VTA (Supplemental Fig. 2A) indicating that there was a similar level of highly interconnected subpopulations.

Network density measures the fraction of existing connections out of all possible connections within a network, and global efficiency refers to the average inverse shortest path length—an index of how readily information travels across the network. Higher global efficiency indicates the potential for parallel information processing and better communication between distant nodes. Notably, only hαSyn deposition increased network density and global efficiency in the SNc (but not in the VTA; p < 0.05, t-test, Supplemental Fig. 2B, C). These parameters remained unchanged after αSyn PFF deposition in both the SNc and VTA (Supplemental Fig. 2B, C).

Consistent with single-neuron recordings, hαSyn significantly increased node degree, network density, global efficiency, and spontaneous firing in dopamine neurons. Conversely, αSyn PFF only increased node degree in the SNc; other network properties and basal firing remained unchanged. These findings highlight that hαSyn expression has a substantially greater impact 4 weeks post-injection on basal firing and network properties compared to PFF deposition. It is important to note that we encountered significant challenges when patch-clamping dopamine neurons with PFF deposition, suggesting a potential imminent failure state. Thus, prolonged PFF deposition might lead to a significant reduction in both network density and global efficiency. Collectively, these findings indicate that αSyn accumulation drives early alterations in neuronal and network activity prior to neurodegeneration in the two synucleinopathy models examined here that could generalize across other preclinical models of PD.

Discussion

A central question in Parkinson’s disease research is why dopamine neurons in the SNc are more vulnerable than those in the VTA. This study reveals that αSyn alterations differentially impacts dopamine neuronal activity and network connectivity, causing changes in the SNc before neuronal loss occurs, but not in the VTA. This study provides new insights into the differential early-stage vulnerability of dopaminergic neurons in the SNc compared to those in the VTA in our two synucleinopathy models. Although mounting evidence implicates αSyn aggregates in the selective degeneration of SNc dopamine neurons characteristic of PD, the mechanisms triggering their early neuronal dysfunction remain poorly understood. In the current study, by employing mouse αSyn PFF deposition or viral-mediated hαSyn overexpression (AAV-TH-hαSyn) before overt cell death, we demonstrate that αSyn pathobiology differentially impacts the physiological activity and network connectivity of dopaminergic neurons in the SNc and the VTA. These findings provide additional insights about the contrasting resilience of these neighboring dopamine neuronal populations.

Several key findings emerge from our work. First, we observed that 4 weeks post-injection, alteration in αSyn (hαSyn or αSyn PFF) did not induce measurable loss of dopamine neurons in either the SNc or VTA in the two synucleinopathy models examined here. This allowed us to capture a “pre-degenerative” stage, one in which functional deficits precede structural collapse. At this early stage, we found that although hαSyn expression occurred in the midbrain, SNc dopamine neurons increased spontaneous firing rates and neuronal network hyperconnectivity. Such changes were absent in the VTA dopamine neurons, highlighting a regional difference. Critically, we also show that SNc dopamine neurons become less resilient to a hyperpolarization challenge, suggesting that αSyn-mediated alterations render these neurons more vulnerable to subsequent insults.

These electrophysiological and network-level perturbations in SNc dopamine neurons before cell death resonates with prior studies. SNc neurons are known for their large, complex axonal arbors and an intrinsic pacemaking phenotype reliant on L-type calcium channels, metabolic homeostasis, and sustained bioenergetic demands16,28,29,36,44,45. By contrast, VTA dopamine neurons, though similar in many respects, express distinct molecular markers (e.g., calbindin) and operate under different metabolic and synaptic constraints34,40,46,47,48. These interregional differences have been proposed as contributing factors to the selective vulnerability observed in the SNc dopamine neurons in PD. Secondly, αSyn can bind to several membrane proteins such as, dopamine transporter20, Na+, K+-ATPase, Na+ channels, and K+ channels. αSyn has been shown to increase basal Na+ levels, preventing return to baseline membrane potential in neurons49.

Our data align with the hypothesis that αSyn alterations, even before outright neuronal degeneration, can affect neuronal activity in a manner that promotes instability at both the single-neuron and network levels. The known roles of αSyn in synaptic vesicle trafficking and neurotransmitter release could be altered by its pathological forms, leading to dysregulated dopamine handling, increased basal extracellular dopamine, and an overall hyper-excitable state20,50,51. Indeed, previous work suggests that αSyn triplication or overexpression leads to increased baseline excitability and altered dopamine dynamics19,20,38 potentially increasing the SNc neurons’ energetic and synaptic stress. The increased neuronal activity and network hyperconnectivity may tax cellular homeostatic mechanisms, including mitochondrial function and calcium buffering, thereby setting the stage for neuronal demise.

In contrast, VTA dopamine neurons in our models remained stable, showing no significant changes in firing properties or network connectivity, at least at 4 weeks following αSyn PFF or hαSyn accumulation. This resilience may reflect intrinsic physiological differences between these dopamine neuron populations. VTA dopamine neurons are less reliant on the same calcium handling pathways as SNc dopamine neurons and may retain more robust adaptive responses to stress or αSyn burden. Furthermore, the distinct molecular signatures of VTA dopamine neurons might confer protective advantages, such as differential αSyn clearance mechanisms, or synaptic plasticity profiles that limit the impact of pathological αSyn forms.

Our network connectivity analyses reveal that hαSyn or αSyn PFF accumulation induce a more synchronized, hyperconnected environment in SNc, but not VTA, dopamine neurons. Enhanced neuronal connectivity could arise from augmented synaptic coupling or altered inhibitory-excitatory balances, ultimately rendering the SNc microcircuit more vulnerable to disruptions. The shift to a hyperconnected state may make the network less capable of compensating for perturbations, like metabolic stress. As αSyn aggregates spread and escalate in complexity, this fragile network may fail under sustained burden, leading to progressive synaptic dysfunction and loss of dopamine neurons. These results have important implications for understanding the early pathogenic events in PD. Examining a pre-degenerative window when neuronal dysfunction is present, yet potentially reversible22,23, provides a critical time frame for the development and assessment of therapeutic interventions. Targeting αSyn aggregation, modulating firing rates, or normalizing network connectivity during these early stages may prevent or delay the cascade of events leading to irreversible loss of dopamine neurons. Additionally, exploring why VTA dopamine neurons resist these early changes may help identify protective mechanisms that could be harnessed to bolster SNc dopamine neurons against αSyn pathology.

We acknowledge the limitations of the current study and in our subsequent work will incorporate expanded experimental design, and additional analytical approaches to strengthen the interpretation of our findings. Our comparison of aggregated αSyn (αSyn PFFs) with virally expressed hαSyn (AAV-TH-hαSyn) was necessarily confined to a single dose and a single pre-degenerative time point, 4 weeks after intervention in SNc and VTA dopamine neurons (Fig. 2). Although higher transgene expression or longer exposure generally produces more pronounced pathology and cell loss9,52,53,54, the stringent requirements for stable whole-cell patch clamp recordings limited us to this early window. Future studies should 1) map dose- and time-dependent trajectories, for example, by combining our electrophysiological approach with longitudinal two-photon or mini-scope in vivo calcium imaging to track αSyn-driven dysfunction and neurodegeneration over time. 2) Dissect cellular and molecular mechanisms underlying the disparate vulnerability of SNc versus VTA neurons, including metabolic burden, synaptic inputs, and intrinsic excitability, to identify actionable therapeutic targets. 3) Link physiology to behavior by determining whether the same four-week post-injection window already produces motor deficits, e.g., amphetamine-induced ipsilateral rotations, gait abnormalities, or other Parkinsonian phenotypes, and whether these emerge in parallel with, or subsequent to, the electrophysiological and network disturbances reported here. Despite these constraints, our work demonstrates that before overt cell death, αSyn pathology differentially perturbs SNc and VTA dopamine neurons, selectively destabilizing SNc firing patterns and network integrity and thereby priming these neurons for later degeneration.

Methods

Animals: DATIREScre and Ai95(RCL-GCaMP6f)-D knock-in mice were obtained from Jackson Laboratory (Stock number: 006660 (DATIREScre), 024105 (Ai95D), Bar Harbor, ME, USA). DAT:cre, DAT:cre-GCaMP6, and C57Bl/6 mice aged between 1.5 and 2 months and averaging 22-24 grams in weight, were utilized for this study. Male and female mice were used in this study and no sex-linked differences were found, at 4 weeks post-αSyn deposition. It should be noted that multiple studies have shown that sustained AAV‑TH–hαSyn over‑expression or αSyn PFF exposure leads to the degeneration of nigrostriatal dopamine neurons6,7,8. Our objective in this study was to capture the functional changes that arise before overt cell loss; consequently, we restricted our analyses to a clearly defined pre‑degenerative window in these two model systems. The mice were maintained in a controlled environment at the University of Florida’s animal care facility, with a 12:12-hour light-dark cycle and ad libitum food/water access. The care and experimental procedures involving these animals were conducted in adherence to the guidelines approved by the University of Florida’s Institutional Animal Care and Use Committee (IACUC).

Recombinant αSyn proteins and preparation of αSyn preformed fibrils (PFFs)

Recombinant mouse αSyn protein was expressed in BL21 (DE3) E. coli (New England Biolabs Inc) using the pRK172 bacterial expression vector cloned with the mouse αS cDNA55. High salt (750 mM NaCl, 50 mM Tris, pH 7.5, 1 mM EDTA) and heat-resistant bacterial lysates were purified utilizing size exclusion chromatography followed by Mono Q anion exchange chromatography as described previously55,56. Protein concentrations were determined using the bicinchoninic acid (BCA) assay and bovine serum albumin (BSA) as the standard. For the generation of fibrils, mouse αSyn protein (5 mg/ml) was incubated in sterile PBS (Life Technologies) at 37 °C with continuous shaking at 1050 rpm for 5 days (Thermomixer C, Eppendorf), and fibril formation was assessed by turbidity and K114 fluorometry57. Fibrils were diluted to 2 mg/ml in sterile PBS and sonicated for 60 min in a water bath sonicator and stored at −80 °C until use58.

Stereotaxic viral injection

Unless otherwise specified, 6- to 7-week-old mice were used for intracranial injections of αSyn PFF or AAV1-TH- hαSyn virus in all experiments. In previous studies, we validated the expression and distribution of this rAAV22. In addition, it is well-established that mouse αSyn PFFs are better than human αSyn PFFs at recruiting endogenous mouse PFFs in the seeding models9. Following anesthesia induction at 3% isoflurane, the animal was shaved, placed in the stereotaxic frame, and maintained at 1.3–1.8% isoflurane throughout surgery, with body temperature maintained at 37 °C. Ophthalmic ointment was applied to prevent the eyes from drying. The surgical area was disinfected with alternating chlorhexidine followed by sterile saline. A 1 cm incision was made, and the skin was moved aside to expose lambda and bregma. The cranium was leveled to within 0.05 mm between lambda and bregma. At the appropriate coordinates relative to bregma a 0.5 mm burr hole was made, the dura was punctured using a scalpel and irrigated in the case of superficial bleeding. A glass micropipette backfilled with mineral oil was attached to a Drummond Nanoject III nanoliter injector and fixed to the stereotaxic apparatus. 1 μL AAV1-TH- hαSyn (4.92 × 1013 U/μL) was aspirated into the glass micropipette, guided to the injection coordinates (ML = +1.4 mm; AP = -3.07 mm; DV = -4.5 mm). 0.5 µL of αSyn PFF(2 µg/µL) and was injected into the SNc (ML = +1.4 mm; AP = −3.07 mm; DV = −4.5 mm) and the VTA (ML = +0.4 mm; AP = −3.4 mm; DV = −4.4 mm) for a total of 1µL/ hemisphere of αSyn PFF in each animal. The nanoject remained at the injection site for 10 min before being removed. Sterile saline was applied to the surface of the cranium and burr hole to avoid drying throughout the surgery. Postoperative analgesia (20 mg/kg Meloxicam) and hydration (1 mL sterile injection saline 0.9%) were provided via subcutaneous injection. For 72 h, postoperative care mice were monitored daily for body weight and provided with analgesia as recommended by veterinary staff.

Histological validation AAV1-TH-hαSyn virus and αSyn PFF depositions

For all histological analyses, C57Bl/6 mice were microinjected with either AAV1-TH-hαSyn or αSyn PFF, both of which have been validated in previous studies9,22. Human αSyn enabled us to monitor expression and distribution using syn211, an antibody specific to human αSyn, while mouse αSyn PFF seeding efficiency was examined for phospho-Ser129 using 81 A. Luk et al. demonstrated that elevated αSyn expression becomes detectable 4 weeks after injection6. Guided by this finding, we examined AAV-TH:αSyn injected mice at 3, 4, and 5 weeks using the anti-human αSyn antibody (syn211). Robust αSyn immunoreactivity was evident at 3 weeks in both SNc and VTA. A parallel time course was carried out in αSyn PFF injected mice with the anti-pSer129-αSyn antibody (81 A). The quantity of mouse αSyn PFF used in this study was based on our previous report showing expression and distribution and eventual neuronal loss after persistent expression9. We found that whole-cell recordings from dopamine neurons on the injected side were technically reliable up to 4 weeks after either AAV-TH-hαSyn or αSyn PFF delivery; by 5 weeks the seal often ruptured within minutes, precluding stable measurements. We therefore performed all electrophysiological experiments 4 weeks post-injection. This window also allowed us to probe αSyn-induced functional changes before overt cell loss (see Electrophysiological Recordings and Analysis section). Four weeks post-injection, mice were perfused with 1X PBS followed by 4% paraformaldehyde (PFA) in 1X PBS. Brains were dissected, stored in 4% PFA for 24–48 hours, and then transferred to 1X PBS until sectioning. Coronal midbrain sections (40 µm thick) were prepared using a Vibratome (VT 100, Leica) and stained using a free-floating protocol. Sections were first incubated in blocking solution 1 (5% fetal bovine serum (FBS) and 5% BSA in 0.3% Triton X-100 in 1X PBS) at 37 °C for 1 h (Table 1). They were then incubated overnight at 4°C with primary antibodies diluted in the same blocking solution (Table 2). After incubation, sections were washed 4 times (20 min each) with blocking solution 2 (5% FBS and 5% BSA in 1X PBS). They were subsequently incubated with secondary antibodies diluted in blocking solution for 1 h at room temperature in the dark (See Table 1). Sections were counterstained with 4′,6-diamidino-2-phenylindole (DAPI). Following final washes, sections were mounted and cover-slipped using Fluoromount-G (SouthernBiotech). Images were captured using a Nikon A1 laser-scanning confocal microscope (Nikon Instruments, Melville, NY) with a Plan Fluor 20× 0.75 NA objective. Z-stack images were acquired at 1 µm intervals throughout the specimen, and maximum intensity projection images were generated. Nikon Elements NIS Analysis software was used for data acquisition and data analysis.

Table 2 Antibody list
Full size table

TUNEL Assay

The Elabscience® One-step TUNEL In Situ Apoptosis Kit (E-CK-A32 series) was used to detect apoptosis in tissue samples, following the manufacturer’s instructions. Tissue samples were mounted on gelatin-coated slides. A positive control was prepared by inducing DNA strand breaks with DNase I to verify the assay’s effectiveness and reagent performance. Briefly, 100 µL of 1× DNase I buffer was added to the tissue sections and incubated at room temperature for 5 minutes. This was followed by the addition of 100 µL of DNase I solution (200 U/mL), which was incubated at 37°C for 10-30 minutes. After treatment, the sections were washed three times with PBS for 5 minutes each. The positive control was processed alongside experimental samples, using the same labeling and staining protocol to ensure procedural consistency. All tissue sections, including controls, were washed with PBS and permeabilized with 1× Proteinase K working solution for 20 minutes at 37°C. For labeling, slides were incubated with Terminal Deoxynucleotidyl Transferase (TdT) equilibration buffer, followed by the addition of a labeling solution containing TdT enzyme and fluorescein-labeled deoxyuridine triphosphate (dUTP). Samples were incubated at 37°C for 60 minutes in a humidified, light-protected chamber. Slides were then washed with PBS and counterstained with DAPI for nuclear visualization. Images were captured using a Nikon A1 laser-scanning confocal microscope (Nikon Instruments, Melville, NY) equipped with Plan Fluor 20× 0.75 NA and 40× 1.3 NA objectives. Z-stack images were acquired at 1 µm intervals across the entire specimen, and maximum intensity projection images were generated.

Preparation of midbrain slices for single neuron recording and calcium imaging

Mice were perfused with ice-cold sucrose-sodium replacement solution containing (in mM): 250 sucrose, 26 NaHCO3, 2 KCl, 1.2 NaH2PO4, 11 dextrose, 7 MgCl2, and 0.5 CaCl2. Following perfusion, the mice were decapitated, and their brains were rapidly removed. For sectioning, the brains were cut rostrally to the prefrontal cortex (PFC) in the coronal plane to create a flat surface for attachment to the cutting stage. Coronal midbrain slices (250 µm thick), containing the substantia nigra and the ventral tegmental area, were prepared using a vibratome (VT1200S, Leica). The slices were transferred to a recovery chamber containing oxygenated artificial cerebrospinal fluid (ACSF) at 32°C. The ACSF composition was (in mM): 126 NaCl, 2.5 KCl, 2 CaCl2, 26 NaHCO3, 1.25 NaH2PO4, 2 MgSO4, and 10 dextrose, equilibrated with 95% O2 and 5% CO2. To minimize glutamate-induced toxicity during recovery, 10 µM MK-801 was added to the ACSF. The osmolarity of both the sucrose-sodium replacement solution and ACSF was adjusted to 300–310 mOsm, and the pH was maintained at 7.35–7.4. Slices were allowed to recover in the chamber for at least 30 minutes at 32°C before use in subsequent ex vivo electrophysiological and calcium imaging experiments.

Electrophysiological Recordings and Analysis

Midbrain slices of DATcre:GCaMP6 transgenic mice (male or female) were transferred to a low-profile open diamond bath chamber (RC-26GLP, Warner Instruments, Hamden, CT, USA) and continuously perfused with oxygenated ACSF at 36–37°C. Electrophysiological recordings were performed on DATcre:GCaMP6 transgenic mice. Dopamine neurons were identified by their somatic GCaMP6f fluorescence and further verified by their characteristic electrophysiological features such as broad action potentials and Ih sag59,60,61. A 40× water immersion objective mounted on an Eclipse FN1 upright microscope (Nikon Instruments, Melville, NY, USA) was used. The microscope was equipped with a Spectra X light engine (Lumencor, Beaverton, OR, USA) and a 470/24 nm solid-state illumination source. Images were captured using a 12-bit Zyla 4.2 sCMOS camera (Andor Technology, Belfast, Northern Ireland). Neuronal morphology was visualized with infrared differential interference contrast (IR-DIC) imaging. Borosilicate glass capillaries (1.5 mm O.D.; Sutter Instrument Company, Novato, CA, USA) were pulled into electrodes with a tip resistance of 4-8 MΩ using a P-2000 laser puller. The electrodes were filled with a potassium gluconate-based internal solution containing (in mM): 120 K-gluconate, 20 KCl, 2 MgCl2, 10 HEPES, 0.1 EGTA, 2 Na2ATP, and 0.25 NaGTP, with osmolarity adjusted to 290–295 mOsm and pH set to 7.25–7.30. Electrophysiological recordings were performed using an Axon Axopatch 200B microelectrode amplifier and digitized via a Digidata 1440 A system using ClampEx 10.2 software (Molecular Devices, San Jose, CA, USA).

We found that whole-cell patch-clamp recordings from dopamine neurons on the injected side were technically reliable up to 4 weeks after either AAV-TH-hαSyn or αSyn PFF delivery; by 5 weeks the seal often ruptured, precluding stable recording. We therefore performed all electrophysiological experiments 4 weeks post AAV-TH-hαSyn or αSyn PFF injection that was also before overt cell loss, as identified by TUNNEL assay and histological analysis. Baseline recordings were obtained, followed by a hyperpolarization protocol as described38. During this protocol, baseline spontaneous firing activity was recorded for 30 second without current injection, followed by 30 second of hyperpolarization via injected current, and a subsequent 30-second recovery period to observe spontaneous firing activity. Electrophysiological capacitance was measured using a voltage clamp ramp with a 10 mV deviation over 100 ms (10 mV down −50 ms, 10 mV up −50 ms) waveform. The holding potential was −70 mV; therefore, the ramp was from −70 mV to −80 mV and returned to −70 mV. The difference in current during the rising and falling transient was isolated to the central portion of each ramp, where the rate of change was most stable. Since the change is gradual rather than abrupt, as would occur from a voltage step protocol, the resulting measurement is insensitive to the access resistance since the charge and discharge of the cell has already been counteracted. To determine the capacitance, we utilized the following calculation of Cm = dI/(dV/dT), where Cm = membrane capacitance, dI = change in current between rising and falling transient, dV = change in voltage, and dT = ramp duration62. Data analysis was performed using Origin software, enabling precise quantification of neuronal responses under experimental conditions.

Calcium imaging and video processing

For calcium imaging experiments, AAV-syn-FLEX-jGCaMP8f-WPRE (1.1 × 1013 U/µL), a gift from GENIE Project, 0.3 µL was injected into was injected into the SNc (ML = +1.4 mm; AP = −3.07 mm; DV = −4.5 mm) and the VTA (ML = +0.4 mm; AP = -3.4 mm; DV = −4.4 mm) of DAT:cre mice. Coronal slices containing the VTA and SNc were prepared and illuminated as previously described38. Images were acquired at a frame rate of 20–25 frames per second with no inter-frame delay. Baseline recordings were collected for 250 second for each experiment. Calcium imaging videos were processed by setting intensity thresholds to the observed minimum and maximum values for each video. Videos were converted to 8-bit TIFF format using ImageJ, and subsequently down-sampled for analysis. Trace extraction was performed using Miniscope 1-Photon-Based Calcium Imaging Signal Extraction Pipeline (MIN1PIPE), a semi-automated cell detection and activity quantification software, as referenced in Lu et al. 63.

Network connectivity analysis and network graph construction

Network connectivity was assessed using pairwise Spearman’s rank correlation coefficients (MATLAB function: corr, type: Spearman). A significance threshold of p < 0.05 was applied, and networks with fewer than four neurons were excluded from the analysis. All network analyses utilized the weighted function implementation from the Brain Connectivity Toolbox. In all analyses, a “node” represents an individual neuron, and the normalized network degree is defined as the sum of all connections divided by the total number of neurons in the network, as previously described38,42. Clustering coefficient, global efficiency, and network density are previously described in Rubinov & Sporns et al. 42. Network graphs were constructed using undirected network connections. Data were imported into Gephi v0.9.2 for visualization.

Statistical analysis

Statistical analyses were performed using MATLAB (version 2020a, MathWorks, Cambridge, MA, USA) and GraphPad Prism. Two-sample t-tests or one-way ANOVA were conducted where appropriate, with significance set at an alpha level of 0.05.