NADPH oxidase 2 activity disrupts Calmodulin/CaMKIIα complex via redox modifications of CaMKIIα-contained Cys30 and Cys289: Implications in Parkinson's disease

2.1. Reagents

To develop our study, we used the following antibodies.

2.2. Molecular dynamics simulations

Computational simulations of CaMKIIα were performed using the AMBER15 software package, with all of them - except for the minimizations - using the GPU version of the PMEMD program. The crystal structure of CaMKIIα in a closed/auto-inhibited conformation (PDB structure 3SOA) was used as the starting structure [31] of human CaMKII in its α isoform, with a β7 linker (the shortest linker in all known splice variants). Some of the subunits in this dodecamer were converted computationally to open conformations based on the 2WEL PDB crystal structure of the CaMKIIδ isoform, which is in the open conformation. This was feasible since the kinase domains of both isoforms have a high sequence similarity of about 93 %. The final structure contained eight closed and four open sub-units of CaMKIIα in a dodecameric configuration (Fig. 1).

In order to bring the modeled molecule to a stable conformation, we ran several cycles of minimization and equilibration. We employed the Amber12 force field and the TIP3P water model. The systems were kept at a temperature of 298 K, using Langevin dynamics with a collision frequency of 2 ps⁻¹, and part of the protocol used pressure control, via a weak-coupling Berendsen barostat, with a relaxation time of 2 ps. The SHAKE algorithm was adopted, allowing the use of a 2-fs time step. The protocol followed for minimization and equilibration was: 1) 100 cycles of minimization, using the XMIN method, followed by 5000 cycles using steepest descent and another 5000 steps using conjugate gradient; 2) 1 ns of heating to 298 K, followed by another ns at constant T and P (1 atm), and finally 20 ns with constant T. The simulation box consisted of one CaMKIIα dodecamer complex with eight subunits in the closed conformation and four subunits in the open conformation. All-atom simulations were performed on CaMKIIα in a box with explicit water plus ions to neutralize electrostatic charges. Systems were >1.3x10⁶ atoms and all runs were >300ns. For all the runs we used Nvidia GeForce TITAN-X GPUs.

2.3. Ventral midbrain primary cultures

Neuronal cultures were prepared from Sprague-Dawley (Charles River, Wilmington, MA, USA) rats’ embryos at day 17 (E−17) obtained from two to three pregnant dams. Brains extracted from E−17 were immersed in cold MEM containing 2 % FBS, 2 % HS, 1 mM Sodium Pyruvate, 1 % NEAA, 1 g/l Glucose, 200 I·U./ml Penicillin and 200 l U./ml Streptomycin (Cellgro). Following anatomical landmarks, the SN, without the VTA, was dissected from a coronal midbrain slice. The cells obtained by enzymatic dissociation using trypsin followed by mechanical trituration, were plated at a density of 5X104 cells/cm2. The cultures were maintained at 37 OC in a humidified atmosphere of 5 % CO2 and 95 %. After 48 h, the medium was replaced with Neurobasal medium supplemented with B27 (1:50 from stock), Glutamax (1:100 from stock), Albumax (1:100 from stock), 200 I·U./ml Penicillin and 200 l U./ml Streptomycin (Cellgro). The medium was, subsequently replaced two times/week. Experiments were performed at day-in vitro 14 (DIV-14). As previously reported, primary ventral midbrain cultures, under the described culture conditions, contain about 2–3% glial cells, 5–10 % DA neurons and a majority of other neurons (MAP2-positive) [32].

All procedures were performed with the approval of the University of Pittsburgh Animal Care and Use Committee.

2.4. Mutagenesis experiments in Drosophila S2R + adherent cells

CaMKIIα and calmodulin expression in Drosophila S2R + cells was accomplished as follows: CaMKIIα and CaM coding sequences were copied from human cDNA clones CAMK2A transcript variant 2 (catalog SC109000, Origene, Rockville MD) and CALM1 Human calmodulin 1 transcript variant 1 (catalog SC115829, Origene, Rockville MD), using PCR primers to add flanking restriction sites, with the following primers: EcoRI-CAMK2A-FWD; GAGAGAATTCATGGCCACCATCACCTGCAC, CAMK2A-STP-SalI-REV; TCTCGT- CGACTTAGTGGGGCAGGACGGAGGGCG, MfeI-CALM1-FWD; GAGACAATTGATGGCT GATCAGCTGACCGA, CALM1-STP-SalI-REV; TCTCGTC GACTCATTTTGCAGTCATCAT- CT. Products were ligated into appropriately gapped vector pRmHA3 (Drosophila Genomics Research Center, Catalog 1145) to generate pRmHA3- CaMKII and pRmHA3-CALM1. Point mutation variants (pRmHA3-CaMKII^C30A, pRmHA3-CaMKII^C289A, and pRmHA3-CaMKII^{C30A, C289A}) were generated by site directed mutagenesis of pRmHA3-CaMKII using QuickChange Lighting Kit (Agilent, Catalog 210518.) The full coding sequence of all pRmHA3-CaMKII constructs was verified by Sanger sequencing. pRmHA3- CaMKII constructs were introduced to Drosophila S2R + cells (Drosophila Genomics Research Center, catalog 181) cultured under standard conditions in Schneider's media with 10 % FBS and Pen-Strep at densities between 0.5 and 2.0 x 10⁶, by transient “Effectene” transfection (Qiagen 301425) according to manufacturer recommendations. Expression was induced with 500 μM CuSO₄ at the time of transfection and transfected cells were incubated for 48 h under standard conditions. pRmHA3-mCherry was co-transfected with pRmHA3-CaMKII constructs to evaluate transfection efficiency and rule out morphological indications of toxicity in transfected ^{cells. Western blot analysis was performed to assess the expression levels of human} CaMKIIα and CaM in cells (Supplemental Fig. 1). ^{Oxidative stress was induced by treating S2R+ cells with 5 nM H}₂^O₂ ^{in normal culture} media for 30 min before immunocytochemistry and lysis for immunoblot assays as described below.

2.5. Immunocytochemistry

At the endpoint, cells were fixed in 4 % paraformaldehyde (PFA), washed with phosphate buffered saline (PBS) to remove the excess of PFA. The cells were also permeabilized with 0.03 % Triton X-100 in 10 % NDS for 1 h and incubated overnight with primary antibodies, including tyrosine hydroxylase (TH), as marker of dopaminergic neurons (refer to Table 1). Fluorescently labeled secondary antibodies were then added for a 1-h incubation at room temperature. After washing with PBS to remove secondary antibodies, the coverslips were mounted on microscopy glass slides using Gelvatol. Proximity ligation assay (PLA; see in the methods section) was also performed in cell cultures to assess levels of CaMKIIα:CaM interaction.

2.6. Calcium live imaging

Primary ventral midbrain cultures were grown in 35 mm petri dishes with a glass coverslip inserted at the bottom (Thermo Fisher). At DIV 15, the medium was replaced with HEPES-buffered salt solution (standard HBSS) of the following composition (in mM): NaCl 137, KCl 5, NaHCO₃ 10, KH₂PO₄ 0.6, Na₂HPO₄ 0.6, MgSO₄ 0.9, CaCl₂ 1.4, HEPES 20, and glucose 5.5 (pH adjusted to 7.4 with NaOH). The cultures were then exposed to (i) rotenone (50 nM), (ii) rotenone + Autocamtide 2 related inhibitor peptide myristoylated (AIP; 50 nM), or (iii) rotenone + Nox2ds-tat (10 μM) for 1, 2, or 4 h.

Thirty minutes before the end points, Fluo 8-AM (Abcam) was added at a concentration of 5 μM and incubated at 37 °C with the cells. Subsequently, cells were washed with standard HBSS. Fluo-8-AM related fluorescence was detected at excitation/emission wavelengths of 490/525 nm using a Nikon Eclipse Ti2 inverted confocal microscope. Fluorescence intensity was measured using Nikon software in 15–25 individual neurons for each microscopic field at 60X magnification (4 fields per coverslip). Background fluorescence, determined from three or four cell-free regions of the coverslips, was subtracted from all signals.

2.7. Animal studies

All experiments utilizing animals were approved by the Institutional Animal Care and Use Committee of the University of Pittsburgh in accordance with published National Institutes of Health guidelines. For the in vivo experiments, we used retired male breeder Lewis rats (7–9 months old - Envigo). The animals were maintained under standard conditions of 12 h light/dark cycle in a 22 ± 1 °C temperature-controlled room with 50%–70 % humidity. Water and food were provided ad libitum. Animals were adapted for one week before the experiments, and randomly divided into the following groups of treatments: (i) Vehicle, (ii) Rotenone and (iii) Rotenone + CPP11H.

Before each daily treatment, body weights were recorded for each animal. Rotenone was administered intraperitoneally once a day at a dose of 2.8 mg/kg until the end of the treatment (5 days). The solution was prepared as a 50X stock dissolved in pure dimethyl sulfoxide at final concentration of 2 %, then diluted in Miglyol 812 N (Sasol North America, Inc, Houston, TX; distributed by Warner Graham, Baltimore, MD, USA) at final concentration of 98 %, and administered at 1 ml/kg. This five-day rotenone regimen does not produce nigrostriatal lesions, but significantly enhances several key PD-related pathogenic events [[33], [34], [35]]. Control animals received an equivalent amount of vehicle (2 % dimethyl sulfoxide 98 % Miglyol; i.p.). For CPP11H treatment, rats were dosed with 15 mg/kg by oral gavage concomitantly with rotenone. Control animals received an equal amount of vehicle (PEG 400; gavage). At the end point (5 days) animals were euthanized and Brains were collected from each animal.

2.8. Histology and immunohistochemistry

Animals were euthanized by CO2 inhalation followed by decapitation. The brains were removed following an intracardial perfusion with saline solution (NaCl 0.9 %) and a fixing perfusion with 4 % PFA. After 48 h, the brains were placed in 30 % sucrose in PBS for cryoprotection until infiltration was complete (at least 3 days).

For immunohistochemistry, brain slices (35 μm) containing Substantia nigra pars compacta were washed in PBS, treated with a blocking/permeabilizing PBS solution containing 10 % FBS and 3 % Triton X-100 for 1 h. The tissue was then exposed to a primary antibody PSS/FBS solution (for antibody dilutions, refer to Table 1). Finally, the slices were exposed to secondary antibodies for canonical immunohistochemistry or placed on glass microscopy slides to perform PLA [36].

2.9. Human tissue

Paraffin-embedded midbrain sections were obtained from the University of Pittsburgh Brain Bank. All banked specimens underwent standardized premortem neurological and postmortem neuropathological assessment. Diagnoses were confirmed and staging performed by the study neuropathologist (JKK) by examination of H&E, alpha-synuclein, tau, silver and ubiquitin stains of key sections needed for Braak staging (37). The study design was reviewed and approved by the University of Pittsburgh Committee for Oversight of Research Involving the Dead. Midbrain sections from 6 PD patients and 7 control subjects, matched for age and postmortem intervals, were used for the analysis; 10–15 cells were imaged per brain section, 3 sections for each patient.

To eliminate endogenous fluorescence, human tissue was pre-treated with Sudan Black (Chemicon, Temecula, CA), an autofluorescence eliminating reagent according to the manufacturer's instructions.

2.10. Fluorescence measurements

Quantitative fluorescence measurements were conducted using a Nikon Eclipse Ti2 inverted confocal microscope, ensuring that images did not contain saturated pixels. To enable accurate comparisons, all imaging parameters (such as laser power, exposure, and pinhole) were kept constant across specimens. Fluorescence intensity was measured in predefined regions of interest drawn around the single cells (i.e. tyrosine hydroxylase (TH)-positive dopaminergic neurons).

In ventral midbrain (VMB) cultures or S2R + cell cultures, analyses were performed in three independent experiments. In VMB cultures, fluorescence signals were measured in 100–150 TH positive neurons per group of treatment per each independent experiment, while in S2R + cell cultures, 200–250 cells were analyzed per group of treatment per each independent experiment. The final analysis represents the normalized intensity average (assuming Vehicle group as 100 %) for each group of treatment in a single independent experiment.

For analyses conducted in rat substantia nigra pars compacta (SNpc), fluorescence in 60–100 TH-positive nigrostriatal neurons were assessed in a slice (with four slices per animal), totaling six animals per treatment group.

2.11. Proximity Ligation Assay

Proximity Ligation Assay (PLA) (Duolink; Sigma Aldrich) was performed as previously described [36] in 4 % PFA-fixed brain tissue (rat or human) or cell cultures (Drosophila S2R + or rat ventral midbrain cell cultures) to assess the level of interaction CaMKIIα:CaM under our experimental conditions. Samples were incubated with specific primary antibodies (refer to Table 1). PLA probes consisting of secondary antibodies (anti-Rabbit and anti-Mouse) conjugated with complementary oligos (plus and minus respectively) were added to the reaction and incubated. Ligation solution, consisting of two oligos and ligase, was added. In this assay, the oligos hybridize to the two PLA probes and join to a closed loop if they are in close proximity. An amplification solution, consisting of nucleotides and fluorescently labeled oligos, was added together with polymerase. The oligonucleotide arm of one of the PLA probes acts as a primer for “rolling-circle amplification” (RCA) using the ligated circle as a template, and this generates a concatemeric product.

Fluorescently labeled oligonucleotides hybridize to the RCA product. The PLA signal is visible as distinct fluorescent spots. Fluorescent images were acquired with a Nikon Eclipse Ti2 inverted confocal microscope and quantification was carried out at 60 -100X magnification. PLA validation was performed by primary antibody deletion (Suppl. Fig. 2).

2.12. Redox western blot

Levels of oxidized thiols (S–S) in CaMKII were measured with a redox Western blot technique. VMB primary cultures or Drosophila S2R + cells were lysed in 50 mM Tris-HCl pH 7.0, 2 % sodium dodecyl sulfate (SDS), 1 mM EDTA, protease inhibitor cocktail and 100 μM N-ethyl- maleimide (NEM), to stably alkylate free thiol residues (SH). The lysate was heated at 65 °C for 5 min to denature proteins and incubated for 20 min at room temperature. Proteins were precipitated in ice-cold acetone to remove un-reacted reagents, resuspended in 50 mM Tris-HCl pH 7, 2 % SDS, and 10 mM TCEP and heated at 65 °C for 5 min and incubated 20 min at room temperature to reduce oxidized thiol residues (e.g. S–S bridges) to free SH groups. Proteins were precipitated again, and the pellet was re-suspended in 50mMTris-HCl pH 7, 2 % SDS, and 100 μM polyethylene glycol-maleimide (PEG-NEM; 10 KDa) to label the formerly oxidized thiols. After 20 min of incubation, the proteins were precipitated and a Western blot for CaMKIIα was performed. PEG-NEM labeling of formerly oxidized thiols increases the molecular weight of the protein by 10 KDa for each labeled thiol. Blotted membranes were imaged with an Odyssey infrared scanner (LiCor), and the signal was quantified with the scanner's software. S-Nitrosation was performed replacing TCEP with the selective S–NO reductant solution, ascorbate 1 mM, and CuCl₂ 1 μM [37].

The “total protein” S–NO assay shown in Supplementary Fig. 3 was performed on SHSY-5Y cell lysates using the same procedure previously described, with the exception of the NEM-mediated alkylating reactions. Briefly, the lysates were first exposed to Alexa Fluor 680-NEM to label the pool of reduced thiols. After a reduction step with 1 mM ascorbate and 1 μM CuCl2, former S–NO residues were labeled with Alexa Fluor 800-NEM. The samples were then run on 4–12 % polyacrylamide gels, transferred to a PVDF membrane, and imaged with an Odyssey scanner. The fluorescent signals at 680 nm and 800 nm were quantified and analyzed as the ratio of S–NO to SH.

2.13. Dihydroethidium (DHE) staining

A 3 mM stock solution of DHE (D11347, Thermo Fisher) was prepared in DMSO and stored at −20 °C. Before use, the 3 mM DHE solution was thawed and kept protected from light due to DHE's light sensitivity. The 3 mM solution was then diluted with cell culture media to a final concentration of 3 μM DHE. The old media was aspirated from each well and replaced with fresh media containing 3 μM DHE. Cells were incubated for 20 min at 37 °C, protected from light. Subsequently, the cells were gently washed with 1 × PBS at room temperature for 5 min and then fixed with 4 % PFA for 20 min. After fixation, the cells were washed three times for 10 min each with 1 × PBS. Single coverslips were then extracted from the wells and mounted on glass slides for confocal microscopy imaging. Imaging parameters were set to avoid signal saturation under oxidative stress conditions and were maintained consistently across all experimental conditions. The final analysis represents the normalized intensity average, with the Vehicle group set as 100 %, for each treatment group in a single independent experiment and repeated for 3 independent experiments.

2.14. Statistical analyses

Each result presented here was derived from three (in vitro) to six independent experiments (in vivo). For simple comparisons of two experimental conditions, two-tailed, unpaired t-tests were used. Where variances were not equal, Welch's correction was used. For comparisons of multiple experimental conditions, one-way ANOVA was used, and if significant overall, post hoc corrections (Bonferroni) for multiple pairwise comparisons were made. P-values less than 0.05 were considered significant.

3.1. Redox changes of Cys30 and Cys289 are critical in the formation of CaMKIIα-Ca²⁺/CaM complex

Our computational investigations have revealed a transient proximity between Cys30 and Cys298 residues of CaMKIIα, suggesting the potential formation of a S–S bridge that could impact the formation of CaMKIIα:Ca²⁺/CaM complex. To assess the significance of these cysteine residues in the CaMKIIα:Ca²⁺/CaM interaction and to circumvent potential cross-immunoreactivity issues with native proteins in mammalian cells, we expressed various mutant forms of human CaM and CaMKIIα in Drosophila S2R + adherent cells (Validation: Suppl. Fig. 1). Utilizing PLA, we examined the CaMKIIα:CaM interaction (Validation: Suppl. Fig. 2) in response to 5 nM H₂O₂ treatment in the mutant cell lines. Notably, wild-type CaM/CaMKIIα cells exhibited disrupted CaMKIIα:CaM interaction upon H₂O₂ exposure, whereas this effect was prevented in CaMKIIα^C30A, CaMKIIα^C289A, and CaMKIIα^C30A−C289A mutant cell lines (Fig. 3 A-B).

Furthermore, employing redox Western blot assays, which rely on the alkylation of thiol residues with a 10 kDa PEG-NEM tag to detect oxidized thiols, we observed distinct molecular weight shifts indicative of thiol oxidation. Specifically, exposure to 5 nM H₂O₂ induced a 20 kDa molecular weight shift in wild-type CaMKIIα, while in C30A mutants, a 10 kDa shift was observed. However, no such shift was detected in C289A mutant lines, suggesting that Cys289 may undergo alternative forms of thiol oxidation rather than participating in S–S bridge formation (Fig. 3 B–C). Remarkably, despite observing redox changes in Cys289 in the absence of Cys30, these alterations were not sufficient to disrupt the CaMKIIα:CaM interaction, as evidenced in Fig. 3 A-B. Moreover, the double mutant C30A/C289A prevented molecular weight shifts in CaMKIIα, indicating that both Cys30 and Cys289 are actively involved in the redox perturbations leading to impaired CaMKIIα:CaM interaction.

Rotenone-induced oxidative stress affects CaMKIIα:CaM complex integrity in Ventral Midbrain DA neurons.

To investigate the relevance of this phenomenon in PD, we designed experiments applied to the rotenone model of PD in rat primary VMB dopamine (DA) neuronal cultures. Proximity ligation assay was used to assess in situ interaction between CaMKIIα and CaM in DA (TH-positive) neurons. Under basal conditions, a robust fluorescent signal indicative of CaMKIIα:CaM interaction was observed. However, upon 24 h of exposure to rotenone (50 nM), a significant reduction in this signal was noted. Notably, co-administration of the ROS scavenger, N-acetylcysteine (NAC; 250 μM) prevented rotenone-induced CaMKIIα:CaM complex disruption, suggesting the involvement of redox-sensitive mechanisms in CaMKIIα:Ca²⁺/CaM complex formation (Fig. 4 A-B).

To further explore the redox state of CaMKIIα cysteine residues in response to oxidative conditions, we conducted redox Western blot assays in the same experimental model. Using two different thiol reduction steps - TCEP for non-specific thiol reduction and ascorbic acid for selective reduction of S–NO residues (Validated: Suppl. Fig. 3) - we observed a 20 kDa shift in the CaMKIIα band in TCEP-treated lysates from cells exposed to rotenone. However, this shift was not observed in lysates treated with ascorbic acid, indicating the presence of two oxidized cysteine residues in CaMKIIα under oxidative stress conditions and excluding the formation of S–NO residues (Fig. 4 C-D).

These findings underscore the involvement of redox-sensitive mechanisms in the regulation of CaMKIIα:Ca²⁺/CaM interaction and highlight the potential relevance of such mechanisms in the pathogenesis of Parkinson's disease.

NOX2 modulates CaM-independent CaMKIIa kinase activity and causes Ca²⁺ dysregulation in rotenone treated ventral midbrain neurons.

We investigated the role of NOX2 in disrupting the CaMKIIα:CaM complex and triggering hyperactivation of CaM-independent CaMKIIα kinase due to rotenone-induced mitochondrial dysfunction in VMB DA neurons.

Our findings, illustrated in Fig. 5 A-B, reveal a significant reduction in the CaMKIIα:CaM complex in DA (TH positive) neurons following 4 h of rotenone exposure. Notably, co-treatment with the cell-permeable selective NOX2 inhibitor peptide, Nox2ds-tat (10 μM) prevented the rotenone induced disruption of the CaMKIIα:CaM complex. Furthermore, experiments conducted in CRISPR/Cas9 gene-edited NOX2 knockout (NOX2^−/−) HEK-293 cells exposed to rotenone, revealed no NOX2 activity [17] and a significant prevention of rotenone-induced disruption of the CaMKIIα:CaM complex compared to wild type cells (Suppl. Fig. 4). Overall, the data suggest a critical role for NOX2 activity in promoting the formation of an oxidized post-translational modification (PTM) form of CaMKIIα, rendering it unresponsive to Ca²⁺/CaM modulation. This is supported by evidence from redox Western blot analyses conducted in Drosophila S2R + implicating Cys30 and Cys289 residues in the binding capacity of CaMKIIα with CaM under oxidative stress conditions. Moreover, DHE staining performed under the same experimental conditions, showed a significant increase of cellular oxidant species production in neurons after 4 h of rotenone exposure, that was prevented by the co-administration of Nox2ds-tat (Suppl. Fig. 5). Consistently with previous study [17], this evidence supports the role of NOX2 activity as major early oxidative event in PD.

Additionally, the observed increased levels in CaMKIIα^(Thr286) phosphorylation in rotenone exposed VMB cultures, indicative of prolonged CaM-independent CaMKIIα kinase activity, were consistently prevented by Nox2ds-tat co-treatment, suggesting a role of NOX2 activity in sustaining elevated CaM-independent CaMKIIα kinase activity in the context of cellular oxidative stress.

Interestingly, the selective cell-permeable CaMKIIα kinase inhibitor AIP; (50 nM) did not prevent rotenone-induced disruption of the CaMKIIα:CaM complex but did prevent the elevated levels of CaMKIIα^(Thr286) phosphorylation induced by rotenone exposure. This suggests a direct inhibitory action of AIP on CaMKIIα in a redox-independent manner.

Furthermore, we investigated the downstream effects of CaMKIIα hyperactivity on Ca²⁺ homeostasis. Using the green, fluorescent calcium binding dye Fluo-8-AM (5 μM; Abcam) and live imaging confocal microscopy, we observed a time-dependent increase in intracellular calcium levels in VMB neurons upon rotenone treatment. Importantly, this abnormal calcium accumulation was prevented by both Nox2ds-tat and AIP co-treatment (Fig. 6).

Together, this experimental evidence highlights the involvement of CaM-independent CaMKII kinase activity in calcium dysregulation in a PD-relevant model, shedding light on potential novel mechanisms underlying PD pathogenesis.

CaMKIIα-Ca²⁺/CaM complex is disrupted in DA nigrostriatal neurons in the rotenone model of PD in rat.

The mitochondrial complex I inhibitor rotenone recapitulates many features of PD-related pathogenic events including mitochondrial dysfunction, oxidative stress, altered Ca²⁺ homeostasis [40], and accumulation of toxic forms of α-synuclein [19,41]. We further assayed the CaMKIIα:CaM interaction in this experimental paradigm to examine if the disruption of binding is manifested in the rotenone model of PD in rat. A robust CaMKIIα:CaM interaction was observed in DA nigrostriatal neurons under basal conditions, whereas rotenone treated animals showed a significant reduction of PLA signal for CaMKIIα:CaM complex (by over 70 %) in DA nigrostriatal neurons of substantia nigra pars compacta (SNpc) (Fig. 7 A-B). Interestingly, the integrity of the CaMKIIα:CaM complex was preserved in rotenone-exposed animals co-treated with the highly selective NOX2 inhibitor, CPP11–H, suggesting a critical role of NOX2 – as major

source of ROS in DA neurons - in inducing redox changes of CaMKIIα. Previous evidence in PD models report CaMKIIα hyperactivity related to high levels of phosphorylation at the Thr286 residue [21,22], an event linked to an uncontrolled Calmodulin-independent CaMKIIα activity. Similarly, as shown in Fig. 7 C-D, we detected a significant increase of CaMKIIα^(Thr286) phosphorylation in rotenone treated animals. This event suggests that, in absence of Ca²⁺/CaM modulation, CaMKIIα undergoes uncontrolled kinase activation. Consistently with the PLA-based results, levels of CaMKIIα^(Thr286) phosphorylation were reduced to control in rotenone/CPP11–H co-treated rats.

3.2. Altered CaMKIIα-Ca²⁺/CaM interaction is associated with idiopathic PD

To determine whether this process is relevant to idiopathic human PD, we performed PLA for CaMKIIα:CaM interaction in blinded postmortem substantia nigra sections from PD patients (N = 5) and from controls (N = 4). Compared to controls, nigrostriatal dopamine neurons from all PD cases showed very low PLA signal, indicating loss of interaction between CaMKIIα and CaM (Fig. 8). This evidence strongly suggests that CaM-mediated modulation of CaMKIIα is impaired in the human disease and may be a relevant factor in PD pathophysiology.