Direct current stimulation induced reduction in α-synuclein in primary neurons: targeting Parkinson’s disease
Sophie Bechkos, Scott D Ryan, Alysia Ross, Hongyu Sun, Shawn Hayley
TL;DR
This study shows that direct current stimulation can reduce harmful alpha-synuclein buildup in neurons linked to Parkinson's disease.
Contribution
The study demonstrates that direct current stimulation reduces alpha-synuclein aggregation in primary neurons, a novel finding for Parkinson’s disease treatment.
Findings
Direct current stimulation reduced both wild-type and A53T alpha-synuclein aggregation in neurons.
DCS increased neuronal viability and promoted clearance of intracellular alpha-synuclein.
Extracellular alpha-synuclein levels increased, suggesting enhanced neuronal activity and clearance.
Abstract
Targeted electrical approaches to the treatment for Parkinson’s disease include deep brain stimulation, which is effective for core motor symptoms, such as essential tremor. Interestingly, treating comorbid depressive symptoms in Parkinson’s disease, using electroconvulsive therapy, also appears to help motor disability. But it is unclear whether such electrical strategies have any impact on the underlying disease processes of Parkinson’s disease. Since aggregation of misfolded alpha-synuclein fibrils is a pathological hallmark of Parkinson’s disease, this may be an important therapeutic target. To this end, we presently assessed whether direct current stimulation (DCS) of cortical neurons that were seeded with wild-type or A53T alpha-synuclein mutant pre-formed fibrils (PFFs) would reduce their aggregation. We found that both wild-type and A53T alpha-synuclein PFFs readily induced…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8| Embryonic day | Antibody | Species | Primary concentration | Secondary conjugate | Secondary concentration |
|---|---|---|---|---|---|
| E14 and E17 | Recombinant anti-alpha-synuclein (phospho S129) (ab51253) | Rabbit | 1: 1000 | Alexa Fluor 594 and Alexa Fluor 647 | 1: 1000 |
| E14 and E17 | Recombinant anti-alpha-synuclein aggregate-conformation-specific (ab209538) | Rabbit | 1: 1000 | Alexa Fluor 594 and Alexa Fluor 647 | 1: 1000 |
| E14 and E17 | Anti-MAP2 (SYSY 188 004) | Guinea pig | 1:4000 | Alexa Fluor 488 | 1:1000 |
| E17 | Glia stain (combined) | Mouse | 1:1000 | Alexa Fluor 568 | 1:1000 |
| α-Syn labelling | Min, max | Threshold | Size |
|---|---|---|---|
| Aggregate | 108, 130 | Huang 108 255 | 0.0–0.80 |
| Phospho | 40, 255 | Huang 30 255 | 0.0–1.00 |
- —Canadian Institutes of Health Research10.13039/501100000024
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsNeurological disorders and treatments · Neuroscience and Neural Engineering · Parkinson's Disease Mechanisms and Treatments
Introduction
Parkinson’s disease is characterized by the loss of dopaminergic neurons in the substantia nigra pars compacta (SNpc) and the presence of intracellular Lewy bodies (LB), which are largely composed of misfolded alpha-synuclein (α-syn) protein.^1^ Both characteristics lead to motor and non-motor symptoms including, bradykinesia, tremors and rigidity, as well as impairments in mood, sleep and olfaction.^2^ The native α-syn monomer is predominately found at presynaptic terminals and normally functions to modulate synaptic vesicle maturation and trafficking, along with neurotransmitter release.^3-6^ However, in Parkinson’s disease, the α-syn monomers undergo conformational changes to become unfolded monomers, soluble oligomers, protofibrils and eventually insoluble fibrils.^6,7^ In vivo and in vitro studies suggest that α-syn oligomers initiate pathology through the seeding of abnormal protein aggregates, which become fibrils and form the dense core intracellular Lewy bodies in Parkinson’s disease.^8^
Recent animal-based approaches have utilized recombinant α-syn monomer proteins that have been converted to α-syn pre-formed fibrils (PFFs).^9-11^ Exogenously delivered α-syn PFFs typically spread between cells and induce phosphorylation of the endogenous protein at S129, which has been associated with Parkinson’s disease–like pathology, including Lewy body–like α-syn aggregates.^3,12^ Indeed, α-syn PFFs trigger a progressive aggregation of α-syn and dopaminergic pathology associated with behavioural deficits.^13-15^
Treatment options for Parkinson’s disease are limited with deep brain stimulation (DBS) improving motor symptoms, but this is an invasive technique involving brain surgery and the clinical improvements dissipate over time.^16^ DBS is believed to induce depolarization blockage resulting in synaptic inhibition that disrupts aberrant basal ganglia network activity to normalize firing rate.^17,18^ In contrast to DBS, electroconvulsive therapy (ECT) was first established for the treatment of psychiatric illness, but has also shown benefits in Parkinson’s disease.^19^ Advanced Parkinson’s disease patients, who are unresponsive to L-DOPA, often respond well to ECT with mood, cognitive functions and motor deficits being improved.^19^ Finally, transcranial magnetic stimulation of the motor cortex in Parkinson’s disease patients also results in clinical improvement that is associated with increased cortical plasticity^20,21^ and changes in calcium (Ca^2+^) channels.^22^
We recently found that functional membrane depolarization, by picrotoxin or elevated extracellular K^+^ levels, prevented α-syn accumulation in SH-SY5Y cells.^23^ Our present in vitro experiments go further to provide evidence that direct current stimulation (DCS) can directly promote the clearance of α-syn aggregates induced by either wild-type (WT) or A53T mutant PFFs in primary cortical neurons. These DCS effects were evident in both E14 and E17 neurons across several different time periods and increased overall cellular viability. These data may have important implications for re-purposing existing (e.g. ECT) or creating novel electrical stimulation methods to reduce α-syn burden in patients through widespread depolarizations.
Materials and methods
Animals and study design
This study was approved by the Carleton University Committee for Animal Care and the CCAC ethical board. The mice used in this study were from a C57BL/6 background and embryos were collected at embryonic Day 14 (E14) or embryonic Day 17 (E17) and primary cortical neurons obtained. All mice were born in the sixth-floor vivarium room of the Health Science building at Carleton University. Mice were housed with a 12 h light-dark cycle, with food and water access throughout. Two days before dams underwent breeding, they were introduced to the male’s scent through the addition of the male bedding into the female cage. This was done to help induce oestrus and improve the chances of pregnancy. Female and male mice were bred together for 2 days before separation.
The study follows a 2 (Sham versus DCS) × 3 (PBS versus WT PFF versus A53T PFF) between-subjects experimental design. Hence, E14 and E17 primary cortical neurons are first cultured and then treated with either WT or A53T mutant PFFs or vehicle (PBS). Subsequently, the PFF/PBS-treated neurons received either DCS or Sham electrical current stimulation in order to ascertain the impact of electrical depolarization on α-syn aggregation. All neuronal culture treatments were randomly assigned to treatment groups.
Cell culture
Cortical architecture forms between E11 and E17, with interneurons forming around Day 13.5 and more superficial layers forming around Day 17.^24,25^ It is possible that the developmental state of neurons might influence their ability to incorporate PFFs, as well as their vulnerability to the aggregated protein. For instance, differences between cortical E14 and E17 responses to β-amyloid have previously been noted.^26^ Hence, the time points used represent immature (E14) and mature (E17) neurons to study the difference in synucleinopathy and clearance following electrical stimulation. We did not collect neurons earlier than E14 to ensure interneurons were present, since they regulate motor activity.
Pregnant dams underwent rapid decapitation at the E14 or E17 stage of pregnancy. The embryos were placed into a 6-well plate containing 1 mL of dissection solution [(490 mL HBSS (Sigma), 5 mL HEPES (Sigma), 5 mL Pen-strep (Gibco), 3 g glucose (Sigma)] and run through a sterile filter. All embryos were washed, placed on ice and then underwent dissection to obtain cortical tissue. The embryos were transferred to a 10 cm petri dish containing 6 mL of dissection solution. A cut was made along the orbital to the neck and the brain was gently removed and placed superior-up position in a new 10 cm petri dish containing 6 mL of chilled, fresh dissection solution. Following removal of the olfactory bulbs and meninges, cortical tissue was obtained, cut into smaller pieces and pipetted into a 15 mL falcon tube containing the dissection solution.
The dissection solution was decanted and replaced with 4 mL of warmed TrypLE express (Gibco), and cortical tissue was then mixed and shaken on a rotator at 150 rpm for 25 min at 37°C. Once shaken, 4 mL of completed neurobasal media [Neurobasal media (NBM) (Gibco), B27-supplement (B27) (Gibco), penicillin-streptomycin (Pen-Step) (Gibco), Glutamax (Gibco)] was added and tissue was spun down by centrifugation at 2500 rpm for 5 min at 4°C to form a pellet. The supernatant was then aspirated without disturbing the pellet and replaced with 1 mL of warmed complete neurobasal media. To create a homogenized single-cell suspension for counting, a three-stage titration procedure was used involving flamed glass pipettes that progressively decreased in size.
Following titration, the cell suspension was transferred to a tube containing 15 µL of trypan blue (Sigma) and then counted using a haemocytometer to calculate the concentration of cells in suspension using the following equation: (average cell count) × (dilution factor) × 10^4^ = _________ cells/mL Convert to = ______ × 10^5^cells/mL for plating purposes. The cell suspension was diluted in warmed completed neurobasal media to the desired final concentration for culturing (1 × 10^5^ cells/mL).
Primary cortical neurons were maintained in NBM supplemented with B27 and Pen-Strep at 37°C in a humidified 5% CO2 incubator. Cells were cultured for 2 weeks total in cell media (NDM + B27 + Pen-Strep), with experimental treatments beginning on Day 10. Half media changes were performed every 5 days.
Fibril preparation and administration
WT and A53T PFF were made following established protocols.^27^ Briefly, human α-syn monomer protein for making PFFs (1 mg aliquots, Proteos Cat. RP003) was generated over a 7-day period. Aliquots were thawed on ice and then centrifuged at 4°C and 15 000 g for 10 min. The supernatant was obtained and transferred to a 1.5 mL microcentrifuge tube. The protein concentration of each aliquot was determined by Bio-Rad DC assay. The PFFs were then diluted in Dulbecco’s PBS (Corning) to a final 5 µg/mL concentration. Aliquots were then vortex for 3 s. Tubes were placed in an orbital shaker (Eppendorf Thermomixer R) at 37°C and were shaken for 7 days at 1000 rpm. Following the shaking period, PFFs were aliquoted into 10 µL samples and aliquots were frozen on dry ice and stored at −80°C until experimental use.
The PFFs and vehicle were diluted in cell media (NDM + B27 + Pen-Strep) and were sonicated at approximately 1 pulse/second for 60 s. Primary cortical neurons were then exposed to WT α-syn PFFs (5 µg/mL), A53T α-syn PFFs (5 µg/mL) or vehicle (PBS) for 3 days in order to provoke neuronal aggregation. Each experiment was performed three times.
Direct current stimulation
DCS or Sham stimulation was delivered to cells in a custom-built 24-well plate with two L-shaped Ag/AgCl electrodes connected to a Grass S8800 stimulator. The DCS parameters used in this study were previously optimized and validated in our earlier work.^23^ Electrodes were washed with 70% ethanol, then incubated in NBM with Pen-Step and B27 supplement for 15 min prior to use. The electric field strength was calibrated by adjusting stimulation intensity and directly measuring the potential between electrodes. Membrane depolarization was monitored in real time using voltage-sensitive dye imaging, while cytotoxicity was assessed via NucGreen™ staining. We determined that 10 mV/mm reliably induced significant membrane depolarization without compromising cell viability, making it the optimal condition for our studies. Hence, the stimulation intensity was set to reach 10 mV with a 99 ms train duration and pulse duration of 1 ms for 10 ms intervals for a total duration of 30 min. This was applied following the 3-day treatment incubation and then cell viability was performed 6, 9, 24, and 48 h after stimulation, followed by supernatant collection for α-syn ELISA and by paraformaldehyde (PFA) fixing for immunohistochemistry.
Live cell voltage-sensitive dye imaging
Stimulation intensity was determined based upon the completion of a live cell voltage-sensitive dye imaging to ensure voltage was sufficient to depolarize the neurons. Cells were incubated for 30 min in oxygenated ACSF containing (mM): 124 NaCL, 5 KCl, 1.25 NaH_2_PO_4_, 1.2 MgSO_4_, 26 NAHCO_3_, 2 CaCl_2_ and 10 glucose. Thereafter, cells were stained with 0.2 nM Di-4-ANEPPS (Biotium, CA) for 30 min then transferred into the recording chamber. A shuttered green LED (LEX2, Brainvision, Tokyo, Japan) excitation light was reflected towards the cells through a 531 nm excitation filter. The fluorescence signals were emitted through a 580 nm absorption filter and captured by a MiCam05 CMOS-based camera (SciMedia) with a Leica Plan APO 5× objective (NA: 0.5, Leica Microsystems, Wtzlar, Germany) at a frame rate of 0.5 ms and over a 6–8 min time lapse period. Stimulation intensity was set to reach 10 mV, and the detection of membrane potential changes was recorded. Red colour indicated a membrane depolarization, and blue colour indicated a membrane hyperpolarization.
Evaluation of cell viability
Cortical neurons treated with WT-PFF, A53T-PFF or vehicle alone or in combination with DCS were analysed at 6, 9, 24, and 48 h after DCS. Briefly, the cortical neurons had all media removed, were rinsed twice with HEPES-Buffered Salt Solution, and then replaced with 250 µL of 0.2% trypan blue solution for 2 min at room temperature. Thereafter, the trypan blue solution was removed and cells were rinsed and imaged immediately using the EVOS FL Imaging System. Three images (two middle and one outside of the coverslip) were taken for each treatment. The cellular nuclei counted for both trypan blue-positive (‘dead’; blue cytoplasm) and trypan blue-negative (‘live’; white cytoplasm) were averaged for three replicates and presented as a percentage of dead/live cells.
Immunohistochemistry
Cells were first fixed with 500 µL of 4% PFA, followed by washes with PBS. The plates were then wrapped in parafilm and kept at 4°C until ready for immunolabelling (completed within 2 weeks of fixation). On the day of immunofluorescence labelling, neurons were washed twice with 500 µL of PBB wash (PBS + 0.5% BSA + 10% Triton X) for 5 min. They were then blocked and permeabilized with 2% BSA (Sigma), 10% NGS (Sigma) and 0.3% Triton X (Sigma) for 45 min. Following incubation, cells were washed for a further 2 × 5 min in PBB wash. Primary antibodies were diluted in PBB to a desired concentration (Table 1) and neurons were incubated with antisera for 1 h. Following primary incubation, the neurons were again PBB washed and incubated with a specific secondary antibody conjugated to an Alexa Fluorophore at a desired concentration (Table 1), for 45 min in the dark. Two more 5 min PBB washes were completed followed by incubation with Hoechst dye at 1:10 000 diluted in PBS for 3 min in the dark. Immediately thereafter, the neurons were washed for 2 × 5 min in PBS and mounted with gelvatol on imaging coverslips. Coverslips were left in a horizontal position to dry for 2 days and then imaged on a Zeiss AxioImager LM800 at 40× magnification. Three images were taken per treatment group for each antibody and time point. The images were sampled from the middle and outer portions of the coverslip area in order to obtain a reasonable estimate of the entire cellular field. This was repeated in three separate experiments with cells from three separate dams, giving an n = 3/group.
Alpha-synuclein imaging
α-Syn fibrils were analysed based on immunofluorescent labelling of the three different antibodies (for the specific conformations). Using ImageJ (NIH) software, images within each specific immunofluorescent labelling were run through a unique auto thresholding script we created to obtain area measurement of cell density in microns squared (µm^2^). The script was modified to tailor α-syn immunofluorescent labelling as shown in Table 2.
Alpha-synuclein ELISA
Extracellular α-syn was assessed using a commercially available sandwich ELISA kit according to the manufacturer’s instructions (Ana Spec, Fremont, CA). Using this same kit, we previously found that membrane depolarization causes the extracellular release of α-syn in PFF-treated SH-SY5Y cells.^23^ Following the experimental treatments, 30 µL per well of neuronal cell media was collected in quadruplicates from the primary cortical neurons and stored at −80°C until being assayed. Samples were thawed and gently vortexed and then diluted samples (1:750) were added to a pre-coated anti-α-syn monoclonal antibody 96-well strip plate. The plate also included nine standards, which were run in duplicates. A diluted detection antibody (rabbit polyclonal anti-α-synuclein IgG-HRP, 50 µL) was applied and incubated for 4 h at room temperature in the dark. Following incubation, wells were aspirated and washed with 1× wash buffer 350 µL/well six times. A 5–10 s lag time was allowed between washes and the plate was dried. The TMB colour substrate solution was added at 100 µL/well and incubated at room temperature until a blue gradient was observed (approximately 5–15 min). Then, a stop solution was added at 50 µL/well, and within 20 min, the colour reaction was measured using a microplate Spectra Max 190 absorbance reader at 450 nm. A standard curve was created and the concentration of α-syn was calculated in the samples using Soft-Max Pro Software.
Statistical analyses
The data were quantified and graphs produced using Prism (version 9) statistical software and followed a 3 (PBS versus WT PFF versus A53T PFF) × 2 (Sham versus DCS) between-subjects analysis of variance (ANOVA) design. Differences were considered statistically significant when P < 0.05, and any significant interactions were followed by post hoc comparisons. A biological sample size of three (three different experiments using three different sets of embryos) was used for all analyses. Data analyses were conducted by experimenters blinded to the experimental treatment.
Results
Neuronal depolarization and viability
We first verified that the DCS treatment was able to elicit action potentials in the primary cortical neurons, as we hypothesize this is a key mechanism for its impact on α-syn aggregative potential. Indeed, the DCS was clearly able to stimulate membrane depolarization as indicated by the voltage-sensitive fluorescent dye spread (Fig. 1). We next assessed neuronal viability using trypan blue as a marker of damaged/degenerating cells with a compromised membrane. There was no significant impact of the α-syn treatments upon cellular viability of either embryonic Day 14 or Day 17 cortical neurons (Fig. 1). This is consistent with early (within 48 h) α-syn accumulation prior to when significant cell death would be expected. Interestingly, at E17 (but not E14), the DCS treatment caused a modest, but statistically significant, increase in overall cell viability, as indicated by a greater number of neurons being trypan blue negative F(1.64) = 14.1, P < 0.05). This effect was apparent across all time points and irrespective of PFF treatment, suggesting that the DCS-induced depolarization promoted cellular changes that limited cellular damage.
*Live cell voltage-sensitive dye imaging and cell viability over four time points. Cell viability, enzyme linked immunosorbent assay (ELISA) and immunohistochemistry (IHC) were conducted following the α-synuclein fibril and direct current stimulation (DCS) procedures. A and B present standard live cell voltage-sensitive dye (VSD) imaging data, visualized with a pseudo-colour scale (red for depolarization, blue for hyperpolarization). The screen captures are from a video recording following DCS applied to embryonic Day 17 (E17) primary neurons at 10 days in vitro (DIV). Voltage dye labelling clearly shows a red signal without a blue component from the beginning (A) to the end (B) of the 6–8 min recording session following DCS, indicating that DCS evoked membrane depolarization but not hyperpolarization. Overall cell viability was assessed as a percentage of trypan blue negative cells. The PFF treatment (neurons were exposed for 3 days prior to DCS) had no significant impact on cell viability (with trypan blue being a measure of non-viable neurons) of E17 neurons, irrespective of the time they were harvested following DCS (6 and 48 h are shown; N = 3 different cultures from separate litters, with each data point representing cells from a unique biological litter) (C). The two-way ANOVA revealed no effect of time, but that DCS treatment overall did promote a small, but statistically significant main effect [F(1.64) = 14.1, P < 0.05], with an increase in viable cells across all times at E17 (D). Similar outcomes were observed at E14 (data not shown). P < 0.05, relative to Sham treatment. N = 9 different cultures from separate litters were pooled together (i.e. N = 3 control, WT_PFF and A53T_PFF treatments, with each data point representing cells from a unique biological litter) from at each time.
Cell type confirmation
The cortical primary neuronal culture should be pure, but there is always the possibility that a very small number of glial cells are present. Since these cells could potentially influence outcomes, we conducted IBA1 and GFAP immunolabelling, to assess the presence of microglia and/or astrocytes, respectively. We quantified the different cell types at the earliest and latest time points in E17-derived cells exposed to the PFFs and DCS treatment. To this end, we found that an exceedingly small number of glial cells were present (∼1–2 per well, as a composite count of all IBA1 + GFAP positive cells), relative to the number of neurons present (∼30–40 MAP2^+^ per visible well) (Fig. 2). There was no significant difference between treatment groups with regard to the number of neurons or glia in the culture, indicating that the PFF and DCS treatments had no impact on overall cell number.
Cell type confirmation. Cell type was assessed based on percentage of neuronal cells labelled with microtubule-associated protein 2 (MAP2) (A) or glial cells labelled with ionized calcium-binding adapter molecule 1 (IBA1) and glial fibrillary acidic protein (GFAP) (B). There were an exceedingly small number of glial cells present (<4%) and neither neuronal nor glia cell overall numbers were impacted by the PFF or DCS treatments. N = 3 different cultures from separate litters.
Immunohistochemical assessment of α-syn
Embryonic Day 14
E14: aggregated α-syn
Embryonic Day 14 primary cortical neurons were treated with either WT or A53T PFFs for 3 days and then exposed to DCS (10 mV; 99 ms train duration; 1 ms pulse duration; 10 ms intervals for a total duration of 30 min) before being fixed. We found robust labelling of aggregated α-syn using the aggregate-conformation-specific antibody, such that levels varied as a function of the PFF and DCS treatments at both 6 and 48 h (P < 0.05). As expected, a significant elevation of aggregated α-syn labelled cells was evident at both times with WT and A53T PFFs, relative to non-PFF controls (P < 0.05). No differences were found at 6 h in the α-syn signal when comparing between WT versus A53T PFFs; but, after 48 h the WT PFFs induced greater α-syn labelling than the A53T form (P < 0.05). At both 6 and 48 h, the DCS treatment clearly reduced the aggregate specific form of α-syn for both the WT and A53T treatments (P < 0.05, relative to Sham) (Fig. 3). Thus, time-dependent elevations of aggregated α-syn were evident after PFF exposure and DCS reduced this accumulation.
*Aggregated α-syn labelling 6 and 48 h after DCS in E14 primary cortical neurons. Primary cortical neurons previously treated with PBS, WT or A53T PFFs were fixed (A) 6 h and (B) 48 h after Sham or DCS conditions. All neurons were stained for DAPI (cell nucleus; blue), MAP2 (neuron marker; green), α-syn (aggregated form; red) and a composite image presented. Non-PFF neurons at 6 (top panels: AA1–A4 and B1–B4) and 48 (top panels: BA1–A4 and B1–B4) hours did not show any α-syn labelling. However, WT PFF-treated neurons at 6 h (middle panels: AC3–C4) showed high levels of aggregated α-syn labelling and further still after 48 h (middle panels: BC3–C4). The DCS treatment reduced α-syn labelling at both times (A and BD3–D4). The A53T PFF-treated neurons also displayed robust α-syn labelling at both 6 (bottom panel: AE3–E4) and 48 h (bottom panel: BE3–E4) and this was reduced by the DCS treatment (A and BF3–F4). The dotted square outlines show regions that were magnified in order to better depict the relative pattern of α-syn labelling in and around neurons. As shown in these squared images (C and D), the α-syn overlapped with MAP2 labelling in Sham-treated cells (especially at 48 h), but after DCS, any remaining α-syn labelling was outside MAP2+ neurons. Finally, the bar graphs show the quantification of the WT and A53T PFF (hatched and white bars, respectively) induced aggregate-filament α-syn labelling and the reduction promoted by DCS. The two-way ANOVA revealed significant DCS x PFF interactions for both 6 (C) and 48 (D) hours [Fs(2.16) = 5.8 and 10.4, P < 0.05, respectively]. Additionally, follow-up comparisons demonstrated significantly higher levels in the WT PFF-treated neurons (P < 0.05). P < 0.05, relative to controls (Ctrl) and #P < 0.05, relative to Sham treatment [2 (DCS versus Sham) × 3 (Control versus WT Vs A53T PFFs)] ANOVAs; N = 3 different cultures from separate litters, with each data point representing cells from a unique biological litter.
E14: phosphorylated α-syn
The E14 neurons treated with either WT or A53T PFFs showed moderate phosphorylated α-syn labelling after 6 h and slightly less at 48 h. As previously observed in primary cortical neurons,^28^ the PBS-treated controls had virtually no detectable phosphorylated α-syn labelling (Fig. 4). But the PFF and DCS treatments collectively interacted to significantly alter phospho-α-syn labelling at 6 h (P < 0.05). Specifically, the DCS treatment ameliorated the PFF-elevated levels of phosphorylated α-syn at 6 h (P < 0.05). In contrast, the interaction just missed significance at 48 h (P = 0.055), likely owing to less phospho-α-syn labelling overall being evident. Overall, levels of the phosphorylated α-syn protein were less than the aggregate form and the PFF-induced elevations were somewhat more variable. Nevertheless, the DCS exposure still had its expected impact, suggesting that it was affecting posttranslational modifications of endogenous α-syn.
*Alpha-synuclein treatments exhibit differences in phosphorylated α-syn labelling 6 and 48 h after DCS condition in embryonic Day 14 primary cortical neurons. Primary cortical neurons treated with PBS, WT PFF and A53T were fixed 6 (A) and 48 (B) hours after Sham or DCS conditions. All neurons were stained for DAPI (cell nucleus; blue), MAP2 (neuron marker; green), α-syn (phosphorylated form; red) and a composite image. Neurons showed MAP2 labelling (green), which was not affected by DCS conditions (A and BB2, D2, F2). Control neurons (A and BA1–A4 and B1–B4) did not show any observable α-syn labelling. The WT PFF-treated neurons showed phosphorylated α-syn labelling in a subset of neurons at both times (A and BC3–C4), which was greatly decreased after the DCS administration (A and BD3–D4). Similarly, A53T PFF-treated neurons demonstrated phosphorylated α-syn labelling in many neurons, but only at the 6 h time (AE3–E4), which was also reduced following the DCS treatment (AF3–F4). The dotted square outlines depict cells that were magnified to better show the relation between α-syn and MAP2 labelling (yellow arrows depict α-syn colocalized with MAP2) and bar graphs show quantification (C and D). The two-way ANOVA revealed significant DCS x PFF interaction at 6 h [F(2.16) = 8.1, P < 0.05], but only a main effect for PFF treatment at 48 h [F(2.16) = 6.3, P < 0.05]. P < 0.05, relative to controls (Ctrl) and #P < 0.05, relative to Sham treatment [2 (DCS versus Sham) × 3 (Control versus WT Vs A53T PFFs)] ANOVAs; N = 3 different cultures from separate litters, with each data point representing cells from a unique biological litter.
Embryonic Day 17
E17: aggregated α-syn
We next chose to assess E17 neurons since they are developmentally more mature than E14 neurons and increased PFF seeding efficiency is expected as neurons mature.^29^ We also included assessment of glial cells (using IBA1 and GFAP antibodies) to confirm their absence and that the observed effects were neuronal based. We did indeed find extremely minimal glial presence, indicating relatively pure neuronal cultures. As predicted, the embryonic Day 17 primary cortical neurons treated with WT or A53T PFFs showed robust aggregated α-syn at both the 6 and 48 h time points (P < 0.05; Fig. 5). Both WT and A53T PFF treatments were equally effective in provoking α-syn aggregation, with no significant difference between the two. We subsequently assessed the effectiveness of DCS and found that it significantly reduced α-syn levels at 48 h (P < 0.05), but did not have a statistically significant effect at 6 h (though it moderately diminished levels at this time).
*Alpha-synuclein treatments exhibit differences in aggregated α-syn labelling 6 and 48 h after DCS condition in embryonic Day 17 primary cortical neurons. Primary cortical neurons previously treated with PBS, WT or A53T PFFs were fixed 6 (A) or 48 (B) hours after Sham or DCS conditions. All neurons were stained for DAPI (cell nucleus; blue), MAP2 (neuron marker; green), α-syn (aggregated form; red), IBA1/GFAP (glial markers; orange) and a composite image presented. Non-PFF-treated cells did not show any α-syn labelling at either 6 (top panels: AA1–A5 and B1–B5) or 48 (top panels: BA1–A5 and B1–B5) hours. Likewise, there was virtually no IBA1 or GFAP labelling (A and BA3, B3, C3, D3, E3, F3), confirming an absence of glia. The WT PFF-treated neurons showed high levels of aggregated α-syn labelling after 6 (middle panels: AC4–C5) and 48 (middle panels: BC4–C5) hours. The DCS treatment reduced α-syn labelling at both times (A and BD4–D5). The A53T PFF-treated neurons also displayed robust α-syn labelling at both 6 (bottom panel: AE4–E5) and 48 (bottom panel: BE4–E5) hours and this was also reduced by the DCS treatment (A and BF4–F5). The dotted square outlines show regions that were magnified in order to better depict the relative pattern of α-syn labelling in and around neurons. These images (C and D) show that the α-syn overlapped with MAP2 labelling in Sham-treated cells, but that after DCS any remaining α-syn labelling was outside MAP2+ neurons. Finally, the bar graphs (C and D) show the quantification of the WT and A53T PFF (hatched and white bars, respectively) induced aggregate-filament α-syn labelling and the reduction promoted by DCS. In the absence of an interaction, the two-way ANOVAs revealed significant main effects for both PFF and DCS treatments at the 6 h time point [Fs(1.16) and (2.16) = 4.2 and 23.7, Ps < 0.05, respectively] and at the 48 h time point [Fs(1.16) and (2.16) = 6.1 and 4.5, Ps < 0.05, respectively]. P < 0.05, relative to controls (Ctrl) and #P < 0.05, relative to Sham treatment [2 (DCS versus Sham) × 3 (Control versus WT Vs A53T PFFs)] ANOVAs; N = 3 different cultures from separate litters, with each data point representing cells from a unique biological litter.
E17: phosphorylated α-syn
Primary cortical E17 neurons treated with both WT and A53T PFFs had phosphorylated α-syn present at the 6 and 48 h intervals (Fig. 6). The levels were elevated compared to PBS-treated controls (P < 0.05), and again, the DCS treatment reduced phosphorylated α-syn, relative to the Sham treatment, but only at the 48 h time (P < 0.05).
*Alpha-synuclein treatments exhibit differences in phosphorylated α-syn labelling 6 and 48 h after DCS condition in embryonic Day 17 primary cortical neurons. Primary cortical neurons previously treated with PBS, WT or A53T PFFs were fixed 6 (A) or 48 (B) hours after Sham or DCS conditions. All neurons were stained for DAPI (cell nucleus; blue), MAP2 (neuron marker; green), α-syn (aggregated form; red), IBA1/GFAP (glial markers; orange) and a composite image presented. Non-PFF-treated cells did not show any α-syn labelling at either 6 (top panels: AA1–A5 and B1–B5) or 48 (top panels: BA1–A5 and B1–B5) hours. There was also virtually no IBA1 or GFAP labelling at either time (A and BA3, B3, C3, D3, E3, F3). WT PFF-treated neurons showed modest levels of phospho-α-syn labelling after 6 (middle panels: AC4–C5) and 48 (middle panels: BC4–C5) hours. The DCS treatment reduced WT PFF-induced phospho-α-syn labelling at the 48 h (middle panels: BD4–D5), but not 6 h time (AD4–D5). The A53T PFF-treated neurons also displayed α-syn labelling at both 6 and 48 h (bottom panel: A and BE4–E5), and this was reduced by the DCS treatment at 48 h (A and BF4–F5). The dotted square outlines show regions that were magnified showing that α-syn was dramatically reduced by DCS (C and D). The bottom bar graphs show the quantification of the WT and A53T PFF (hatched and white bars, respectively) induced α-syn labelling and the reduction promoted by DCS. In the absence of a significant interaction, the two-way ANOVAs revealed that the PFF treatment significantly increased phospho-α-syn levels at both 6 (C) and 48 h (D) [Fs(1.16) and (2.16) = 9.3 and 6.3, Ps < 0.05, respectively]. The main effect for DCS reached significance at 48 h [F(1.16) = 9.7, P < 0.05] but not at the earlier 6 h time [F(1.16) = 1.5, P = 0.24]. P < 0.05, relative to controls (Ctrl) and #P < 0.05, relative to Sham treatment [2 (DCS versus Sham) × 3 (Control versus WT Vs A53T PFFs)] ANOVAs; N = 3 different cultures from separate litters, with each data point representing cells from a unique biological litter.
Collectively, our immunocytochemical findings indicated that DCS was highly effective in reducing α-syn burden in neurons, especially at the E17 development stage and with the 48 h later exposure time. This was evident for the aggregate and phosphorylation specific α-syn antibodies we utilized, suggesting that the electrical stimulation caused the overall elimination of all intracellular forms of the protein. However, to better assess morphological changes that might have been induced by the DCS and the type of PFFs (WT versus A53T), we also used an antibody that detects full length or total levels of α-syn (Anti-Alpha-Synuclein; Abcam: ab212184; 1:1000). This antibody should detect all forms α-syn. As shown in Fig. 7, the PFFs induced differing patterns of aggregation at 48 h in E17 primary cortical neurons. The WT PFFs induced a speckled and punctate pattern along neuronal projections, as well as some obvious accumulation in the soma. In contrast, the A53T PFFs were mostly localized to the cell body regions and their labelling was much more robust than evident with the WT form. As was apparent for the aggregate and phosphorylation specific antibodies, we confirmed that the DCS treatment caused a dramatic reduction in total α-syn. The remaining α-syn labelling in the DCS treated cells was only evident in a few scattered MAP2+ and MAP2− DAPI+ neurons around the soma and was now absent from any projections (Fig. 7).
Alpha-synuclein treatments exhibit differences in total α-syn labelling 48 h after DCS in embryonic Day 17 primary cortical neurons. E17 primary cortical neurons were treated with PBS, WT or A53T PFFs were fixed 48 h after Sham or DCS conditions. WT PFFs clearly induced total α-syn labelling (red) in the cell body and projections of MAP2+ (green) neurons. There was also some labelling punctate labelling outside of MAP2+ cells. The A53T PFFs provoked more robust α-syn labelling, which was predominately tightly packed and located around the soma region. The DCS treatment greatly reduced α-syn levels in both PFF treatment cases. Interestingly, some scattered α-syn remained in MAP2+ cells, as well as in DAPI + (blue) cells that were MAP2-. Scale bar is 100 mm.
Alpha-synuclein extracellular quantification
Given that the DCS treatment was clearly reducing intracellular α-syn, we next sought to assess whether neurons might be releasing levels of the protein into the extracellular fluid. To this end, we detected a robust α-syn signal in the extracellular media from the PFF-treated primary neurons. We found that the A53T PFFs caused a greater rise in extracellular levels of α-syn at the 6 h time point than did the WT PFFs (P < 0.05; Fig. 8). Paralleling the intracellular reductions, we then found that the DCS was affective in diminishing this rise in extracellular levels of α-syn at 6 h (P < 0.05). Likewise, at 24 h the DCS reduced α-syn levels below that of A53T and WT PFF treatments in the absence of DCS (P < 0.05). However, there were no significant differences between the groups by 48 h. Finally, we were interested in the possibility that some of the extracellular α-syn might come from intracellular neuronal sources via transport within exosomes. Indeed, recent studies have indicated that α-syn might spread from cells by way of exosome transport.^30,31^ Yet, we found no evidence that this was occurring in the present study. There was no change in total or aggregate α-syn within isolated exosomes, nor did we find any difference in the exosome constituent protein, Alix, which is important for exosome secretion (Fig. 8).
*Extracellular α-syn levels following PFF and DCS treatments. Extracellular α-syn concentrations were assessed at 6, 24 and 48 h in E17 media following WT and A53T PFF exposure (A). The two-way ANOVA revealed a significant interaction between the DCS and PFF treatments [F(6.24) = 4.67, P < 0.05]. Specifically, the A53T PFF treatment provoked an elevation of extracellular α-syn levels at the 6 h time point and this was reduced by the DCS treatment. The DCS administration also reduced all extracellular α-syn levels after 24 h, but no significant difference was evident at 48 h. P < 0.05; N = 3 different cultures from separate litters, with each data point representing cells from a unique biological litter. The DCS treatment, however, did not impact total or aggregated α-syn, or Alix protein levels (arbitrary units: N = 2 different biological samples) in blots from isolated exosomes (B).
Discussion
Overwhelming evidence supports a fundamental role for misfolded α-syn and its spread through the brain in the neuropathology of Parkinson’s disease.^32^ Indeed, it is known that the protein becomes toxic once the monomer is conformationally changed to a fibril form, leading to neuropathology in dopaminergic and cortical neurons.^8,22,33^ As a consequence, many different lines of investigation have focused upon the possibility of treating Parkinson’s disease by reducing α-syn burden in the brain. This may involve using specific pharmacological agents to either prevent α-syn uptake into neurons or alternatively to augment its breakdown and clearance once aggregates have formed.^7,16^ There is also the possibility of utilizing immune targeting approaches to promote microglia-peripheral immune cell breakdown of extracellular α-syn oligomers of fibrils.^34^ Finally, another possible approach is the utilization of electrical stimulation to aid in α-syn degradation and/or clearance. This prompted us to evaluate the role of α-syn in primary cortical neurons and whether direct current electrical stimulation may interfere with aggregation. This has clinical implications since transcranial DCS is already being used clinically in several conditions, including in Parkinson’s disease, wherein it improved motor functioning.^35^
The PFF model provides a seeding mechanism for the formation of toxic α-syn aggregation, likely mediated by its N-terminal region, that affects neurons intracellularly and extracellularly.^36,37^ In addition, the PFF model resembles characteristics seen in human Parkinson’s disease including the formation of Lewy-like α-syn fibrillar inclusions.^36^ We used WT and A53T mutant PFFs over a relatively short period (3 days in culture) to study early pathology. The A53T PFFs are believed to enter neurons faster and typically are large 16–19 nm twisted filaments, whereas WT PFF fibrils have 10–15 nm twisted filaments with a slower spread.^38,39^
We found that α-syn labelling overall generally increased over time and this was evident with both the WT and A53T PFF treatments, suggesting progressive aggregation. We also found obvious morphological difference between WT PFF and A53T PFF, such that the mutant exposed neurons tended to exhibit α-syn more nuclear labelling and longer clustered filaments. This could reflect the differing kinetics and aggregative ability of the two species of PFFs. It is also notable that we presently observed obvious α-syn labelling in both axons and cell bodies. Indeed, there is ample evidence to suggest that PFFs might be taken up by neural projections and then transported retrogradely to the cell soma.^40-42^ Our cell viability test however showed no impact of either of the PFF types on cell survival, which is probably related to the relatively short exposure time. Indeed, an in vitro study that used human cortical neurons found that concentrations of PFFs similar to those presently used did not induce neuronal death but did impact mitochondrial functioning, calcium oscillations and overall neuronal activity.^43^ Likewise, PFF administration induced a reduction in synaptic activity rapidly (within 10 min) in hippocampal neurons, which was believed to be relevant for synaptic dysfunction that occurs in the early stages of Parkinson’s disease.^44^ Similarly, α-syn has been reported to inhibit calcium channels resulting in reduced neuronal activity.^41^ These findings are consistent with the possibility that diminished neuronal activity can foster α-syn aggregation and that electrical simulation, via DCS, could prevent such pathological outcomes.
Our aim was to inform a potential therapeutic targeted approach using electrical stimulation that could help with the clearance of toxic intracellular and extracellular α-syn. We presently did indeed find that direct electrical stimulation of either E14 or E17 primary neurons reduced intracellular and extracellular levels of α-syn following either WT or A53T PFF exposure. We also found that DCS treatment promoted a small but significant overall increase in cell viability, suggesting some degree of neurotrophic stimulation or protection might have been afforded by the depolarizing stimulus. This is consistent with previous in vitro and in vivo studies showing that mild electrical stimulation protected cortical neurons from ischaemic insults.^45^ Likewise, electrical stimulation also increased neurite number and axonal length, along with inducing spontaneous neuronal firing and the production of growth factors.^46^
Since pathological α-syn processes involve protein monomers undergoing conformational changes resulting in intermediate to progressively misfolded toxic forms,^6,7^ we evaluated different α-syn forms. Specifically, we utilized antibodies thought to target the aggregated (NAC domain responsible for the formation of aggregates) and phosphorylated (C-terminal involved in phosphorylation on Ser129) forms of α-syn. To this end, we found that there was a greater degree of labelling for the aggregated (‘filament specific antibody’), compared to the phosphorylated α-syn forms and this was evident for both A53T and WT PFF-treated neurons. The phosphorylated form displayed the lowest signal, but the A53T mutant PFFs had a much greater impact than WT PFFs on phospho-α-syn. This presumably reflects phosphorylation status of endogenous α-syn and indicates that the A53T mutant PFFs were preferentially causing posttranslational modification of α-syn, which is thought to be critical for Parkinson’s disease pathogenesis.^47^ Indeed, Ser129 phosphorylation is believed to augment toxicity α-syn aggregation formation and spread.^41^ and PFFs have been reported to recruit and corrupt endogenous α-syn.^9,48^
Overall, E14 and E17 primary neurons yielded a similar pattern of results; however, E17 displayed a greater sensitivity to the PFFs and DCS treatment. This was particularly evident when assessing the phosphorylated form of α-syn, which was most robustly modulated by the PFFs and DCS at the E17 developmental time. Hence, neuronal maturity was relevant for the phosphorylation status of the presumably endogenous protein. This is consistent with the many studies using phospho-α-syn as a readout for endogenous pathology being induced by PFF spreading.^49,50^
Our data may have particular clinical implication since it is already known that electrical stimulation in the form of ECT, DBS or TMS has been clinically effective in reducing motor symptoms in Parkinson’s disease patients,^19^ but little is known about their impact on α-syn. Our findings clearly indicate that electrical stimulation can induce α-syn clearance from neurons. If α-syn is being ejected out of neurons, we found that it was not being packaged within exosomes and, hence, involves alternate mechanisms. The fact that the α-syn was still cleared after 48 h following DCS suggests that the protein was not reinvading the neurons and that a single session was sufficient. But it is also possible that the α-syn was being directly degraded (by lysosomal or proteasomal processes) within neurons. The possibility also cannot be discounted that the DCS treatment might have also had enduring effects on extracellular matrix enzymes or had unknown consequences on neuronal structure that impacted α-syn distribution. Intriguingly, one study showed that adenoviral forced overexpression of α-syn induced extracellular (but not intracellular) α-syn fragments (4–14 kDa) that was thought to occur through extracellular proteases, plasmin, neurosin and several metalloproteinases.^51^ We also cannot explicitly say exactly how much of the measured α-syn in the current study was exogenous versus endogenous. Whatever the case, it is clear that DCS can interfere with intracellular aggregation of PFFs.
One recent in vitro report did find that DCS administration to naïve untreated SH-SY5Y dopaminergic cells increased the monomeric and reduced the oligomeric form of α-syn.^52^ The DCS also reversed the intracellular elevation of α-syn in SH-SY5Y cells that were exposed to the dopaminergic toxin, rotenone and this was related to its regulation of autophagic signalling (e.g. Beclin-1, LC3 and LAMP2A).^52^ Similarly, DCS was also demonstrated to modulate the mTOR pathway (which is also a regulator of autophagy) and this effect was dependent upon mGlur5, since it was reversed by the mGlur5 negative allosteric regulatory drug, CTEP.^53^ Accordingly, we recently found that CTEP and the mTOR inhibitor, rapamycin, both reduced α-syn aggregation in PFF-treated SH-SY5Y cells.^54^ Finally, it is also noteworthy that DCS was shown to promote neuroplasticity at the single cell level,^55^ which could impact α-syn handling and neuronal viability. This last finding is consistent with the modestly increased number of neurons that was presently evident with DCS. Together with these previous studies, our current findings raise the possibility that DCS can modulate autophagic clearance mechanisms to degrade α-syn aggregates.
Our previous work also found that DCS of SH-SY5Y cells reduced PFF-induced aggregates.^23^ But the current study is the first-time DCS has been extensively tested in PFF-treated primary neurons. Taken together, our data suggest that electrical stimulation may be a viable therapeutic mechanism to interrupt α-syn spreading and as a potential early treatment option for Parkinson’s disease. At the very least, our findings raise the possibility that altering the electrical field of α-syn affected neurons and impact their aggregation load.
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Eriksen JL, Wszolek Z, Petrucelli L. Molecular pathogenesis of Parkinson disease. Arch Neurol. 2005;62:353–357.15767499 10.1001/archneur.62.3.353 · doi ↗ · pubmed ↗
- 2Kalia LV, Lang AE. Parkinson’s disease. Lancet. 2015;386:896–912.25904081 10.1016/S 0140-6736(14)61393-3 · doi ↗ · pubmed ↗
- 3Stefanis L . α-Synuclein in Parkinson’s disease. Cold Spring Harb Perspect Med. 2012;2:a 009399.22355802 10.1101/cshperspect.a 009399 PMC 3281589 · doi ↗ · pubmed ↗
- 4Burré J . The synaptic function of α-synuclein. J Parkinsons Dis. 2015;5:699–713.26407041 10.3233/JPD-150642 PMC 4927875 · doi ↗ · pubmed ↗
- 5Cheng F, Vivacqua G, Yu S. The role of alpha-synuclein in neurotransmission and synaptic plasticity. J Chem Neuroanat. 2011;42:242–248.21167933 10.1016/j.jchemneu.2010.12.001 · doi ↗ · pubmed ↗
- 6Delenclos M, Burgess JD, Lamprokostopoulou A, Outeiro TF, Vekrellis K, Mc Lean PJ. Cellular models of alpha-synuclein toxicity and aggregation. J Neurochem. 2019;150:566–576.31265132 10.1111/jnc.14806 PMC 8864560 · doi ↗ · pubmed ↗
- 7Kouli A, Torsney KM, Kuan W-L. Parkinson’s disease: Etiology, neuropathology, and pathogenesis. In: Parkinson’s disease: Pathogenesis and clinical aspects. Codon Publications; 2018:3–26.
- 8Winner B, Jappelli R, Maji SK, et al In vivo demonstration that α-synuclein oligomers are toxic. Proc Natl Acad Sci U S A. 2011;108:4194–4199.21325059 10.1073/pnas.1100976108 PMC 3053976 · doi ↗ · pubmed ↗