Modified Eagle’s medium (MEM), fetal bovine serum (FBS), L-glutamine, MitoSOX for reactive oxygen species (ROS) assay, IRDye-tagged secondary antibodies, Hoechst nuclear stain, penicillin, streptomycin, and other cell culture reagents were purchased from Invitrogen (Carlsbad, CA, USA). ATF-4, p-IRE1α, p-eIF2α, PKCδ, and p-PKCδ Y311 antibodies were purchased from Cell Signaling Technology (Danvers, MA, USA). A comprehensive list of the antibodies used in this study has been provided in Supplementary Table 1. NLRP3, ASC, and caspase-1 antibodies were purchased from AdipoGen (San Diego, CA, USA). The TrX and TXNIP antibody purchased from the Abcam (Cambridge MA). The β-actin antibody was purchased from Sigma-Aldrich (St. Louis, MO, USA). Mito-apocynin (MitoApo) was obtained from Dr. Kalyanaraman (Medical College of Wisconsin, Milwaukee, WI, USA). Protein A/G magnetic beads were purchased from Thermo Fisher Scientific (Waltham, MA). The CD11b magnetic separation kit was purchased from Stem Cell Technologies (Vancouver, Canada). The Duo-link proximity ligation assay reagents were purchased from Sigma Aldrich (St. Louis, MO). The Bradford protein assay was purchased from Bio-Rad Laboratories (Hercules, CA). The quick western kit was purchased from the LICOR (Lincoln, NE, USA).

Ten milliliters of Luria broth medium with 100 μg/ml kanamycin was inoculated with BL21 (DE3) cells transformed with a pT7–7 plasmid encoding WT human αSyn from frozen stocks and incubated overnight at 37°C with shaking (pre-culture). The next day, the pre-culture was used to inoculate 1 liter of Luria broth/kanamycin medium. When the OD600 of the cultures reached 0.6, protein expression was induced with 1 mM isopropyl β-D-1-thiogalactopyranoside (Invitrogen), and the cells were further incubated at 37°C for 3 h before harvesting by centrifugation. Lysis was performed by resuspending the cell pellet in approximately 30 ml of 500 mM NaCl, 50 mM Tris HCl pH 7.6 using an Omni homogenizer at full speed for 5 min and ultra-sonicated with 30-s pulses followed by a 30-s pause, for a total ultra-sonication time of 3 min. Bacterial proteins were heat precipitated by incubating for 15 min in a 90°C water bath with gentle stirring. The solution was cooled by floating in ice water for 10 min then centrifuged at 10,000× g for 20 min. The supernatant was transferred to another tube and DNA was precipitated with streptomycin, 1 mg/ml of lysate, for 10 min at 4°C, and then ultracentrifuged at 24,500× g. The lysate was filtered through a 0.2-micron syringe filter into 3500 kD dialysis tubing and dialyzed against 20 mM Tris HCl pH 8.0 overnight. The lysate was collected from dialysis tubing and centrifuged at 15,000× g for 10 min, the supernatant was concentrated to a final volume of 10 ml, and the concentrate was filtered through a 0.2-μm syringe filter and applied to a GE Sephacryl S-200-HP column using the AKTA Pure FPLC unit. Fractions containing αSyn were identified by SDS-PAGE with Coomassie blue staining and pooled. Pooled fractions were then injected into a GE HiPrep Anion exchange column via AKTA Pure FPLC unit and eluted with a 20 mM Tris HCl and 2M NaCl solution. Fractions containing pure αSyn monomer were identified by SDS-PAGE and Coomassie blue staining was dialyzed against 20 mM Tris HCl pH 8.0 overnight. Protein concentration was determined via a NanoDrop 2,000 Spectrophotometer and aliquots of 1 mg were stored at −80°C until use.

Recombinant mouse αSyn proteins were purified as previously reported (Mao et al., 2016). The ToxinEraser endotoxin removal kit (GenScript) was used to remove the bacterial endotoxin following the manufacturer’s instructions. First, αSyn_agg was prepared in PBS to 5 mg/ml by agitating the monomers at 1,000 rpm at 37°C for 5 days continuously and then stored at −80°C until use. To obtain the preformed fibrils, the αSyn_agg stock is diluted from 5 mg/ml to 2 mg/ml in sterile PBS in a 1.5 ml Eppendorf tube. This preparation is sonicated using the Bioruptor Plus at high power for 10 cycles (30 s on, 30 s off) at a constant temperature of 10°C. The endotoxin levels post Toxin Eraser treatment were 0.08 EU/μg protein. All batches of αSyn were found to have endotoxin levels of <0.5 EU/μg protein.

For TEM analysis αSyn_agg and αSyn_PFF were resuspended in 20 μl of sterile PBS and subsequently mixed with 2% uranyl acetate and incubated for 5 min at room temperature. Then, EM images were captured by adding 5 μl of the sample to carbon-coated copper grids and subsequently examined in a JEOL 2100 200-kV electron microscope operated at 80 kV. Additional analysis was performed using a Thermo Fisher Noran System 6 elemental analysis system.

Approximately 25,000 primary microglial cells were plated onto PDL-coated coverslips. Primary mouse microglial cells were incubated with 1 μM of αSyn aggregates (rPeptide) for 1 h. At the end of the incubation period, cells were washed 3 times in 1× PBS and subsequently fixed with 4% PFA, washed in PBS, and cells were made permeable by incubating with 0.25% Triton X-100, and 0.05% Tween-20, blocked with 2% BSA for 1 h at RT. Cells were then immunostained with the respective primary antibodies IBA1 (1:1,000, Rabbit polyclonal) and human αSyn (1:500, mouse monoclonal) diluted in PBS containing 1% BSA and incubated overnight at 4°C followed by staining with Alexa Fluor 488 and Alexa Fluor 555 dye-conjugated secondary antibodies respectively. Nuclei were counterstained using Hoechst stain (10 μg/ml), and coverslips were mounted with Fluoromount medium and examined using an inverted fluorescence microscope (Nikon, Tokyo, Japan), and images were analyzed using Keyence BZ-X810.

Six- to eight-week-old C57BL/6 mice were obtained from Charles River Labs (Wilmington, MA) and housed under standard conditions (22^oC, 30% relative humidity, a 12-h light cycle, and ad libitum food and water). Given that males are at a higher risk of developing PD than women (Cerri et al., 2019), we utilized male mice in these studies. Mice were randomly divided into two groups (Control and αSyn_PFF). An intrastriatal αSyn_PFF-induced neuroinflammation model was used for this study as per Duffy et al. (2018). Briefly, the mouse was given continuous anesthesia throughout the procedure, and mouse-ear bars were placed firmly into each ear to secure the skull in place. The Angle 2 stereotaxic instrument was used with a 10-μl Hamilton syringe to inject the αSyn_PFF directly into the striatum (STR) at the following stereotaxic coordinates in relation to Bregma (mm): −2 ML, 0.5 AP, −4 DV. Prior to injection, αSyn_PFF is sonicated using a high-power bath sonicator system. The total volume of αSyn_PFF injected per site was 2 μl per mouse. The injection needle remained in place for an additional 2 min after completing the injection before it was gently withdrawn over a period of at least 30 s to minimize leakage. Mice were euthanized either 2 months after the initial intrastriatal injection for molecular experiments or after 6 months for TH analysis. The STR and SN were collected and stored at −80°C until further experimentation. The use of animals and all animal-related procedures in this study were approved and supervised by the Institutional Animal Care and Use Committee (IACUC) at Iowa State University, Ames, IA, USA.

A mixture of 200 mg/kg ketamine and 20 mg/kg xylazine was used to deeply anesthetize mice for transcardial perfusion with freshly prepared 4% paraformaldehyde (PFA) solution. The perfused brains were immediately post-fixed in 4% PFA for 48 h as per our previous publication (Gordon et al., 2016b). Fixed brains were then cryoprotected in 10% sucrose before being sectioned into 30-μm coronal sections using a freezing microtome (Leica Microsystems, Wetzlar, Germany). DAB immunostaining in the coronal SN sections was performed as per Ghosh et al. (2013) with slight modification. Briefly, on the day of staining, 30-μm sections were washed with PBS, incubated with methanol containing 3% H₂O₂ for 30 min, washed with PBS 6x for 5 min each, and blocked with 10% normal goat serum, 0.5% Triton X-100 and 2% BSA in PBS for 1 h at room temperature (RT). Then, the sections were incubated with an anti-TH antibody (1:1,000, mouse monoclonal) overnight at 4°C and washed in PBS 6× for 10 min each. Biotinylated anti-rabbit secondary antibody was then used for 1 h at RT, washed in PBS 5× for 10 min each followed by incubation with avidin peroxidase and triple washed in PBS for 10 min each. Immunolabeling was observed using a DAB solution, which yielded a brown-colored stain. The stained sections were washed in PBS and carefully mounted on poly-L-lysine (PDL)-coated slides with organic solvent (DPX). Using the Stereo Investigator software (MBF Bioscience, Williston, VT, USA), the total numbers of TH⁺ neurons were counted stereologically using every sixth section of the SN (Ghosh et al., 2010). Samples were visualized using an Eclipse TE2000-U (Nikon, Tokyo, Japan) inverted fluorescence microscope, and images were captured using Keyence BZ-0023 (Keyence, Osaka, Japan).

Treated mice were sacrificed through transcardial perfusion performed using 4% PFA, and perfused brains were post-fixed in 4% PFA for 48 h. Next, 5-μm paraffin-embedded sections were cut on a microtome by the Department of Pathology, College of Veterinary Medicine, ISU. Sections underwent deparaffinization through a series of steps that includes xylene (x2), xylene:ethanol (1:1), 100% ethanol, 95% ethanol, 75% ethanol, ad 50% ethanol for 3 min each. Then, the sections underwent antigen retrieval using citrate buffer (10 mM sodium citrate, pH 8.5) for 30 min at 80°C. Next, the sections were blocked with 2% bovine serum albumin, 0.5% Triton X-100, and 0.05% Tween 20 in PBS for 1 h at RT. Sections were incubated with different primary antibodies such as anti-TXNIP (1:500; rabbit monoclonal), anti-eIF2α (1:500; rabbit monoclonal), and anti-Iba1 (1:500; goat polyclonal) overnight at 4°C. After washing with PBS, sections were incubated in appropriate secondary antibodies (Alexa Fluor 488 or 594 or 555) for 2 h followed by incubation with 10 μg/ml Hoechst 33342 for 5 min at RT to stain the nucleus. Then, the slides were mounted with Fluoromount medium (Sigma-Aldrich) on glass slides for visualization. Sections were viewed under a Keyence inverted fluorescence microscope.

Immunofluorescence studies in primary mouse microglia (PMG) cells were performed based on previously published protocols with few modifications (Ghosh et al., 2013). Approximately 25,000 cells were plated onto PDL-coated coverslips. After treatments, cells were fixed with 4% PFA, washed in PBS, and blocked with buffer (PBS containing 2% BSA, 0.5% Triton X-100, and 0.05% Tween-20) for 1 h at RT. Coverslips containing cells were probed with the respective primary antibodies IBA1 (1:1,000, goat polyclonal) and p-eIF2α (1:500, Rabbit polyclonal) diluted in PBS containing 1% BSA and incubated overnight at 4°C. Coverslipped cells were washed several times in PBS and incubated with Alexa Fluor 488 and Alexa Fluor 555 dye-conjugated secondary antibodies. Nuclei were counterstained using Hoechst stain (10 μg/ml), and coverslips were mounted with Fluoromount medium on glass slides for visualization. Samples were visualized using an inverted fluorescence microscope (Nikon, Tokyo, Japan), and images were captured using Keyence BZ-X810.

As a measure of oxidative stress, mitochondrial superoxide levels were estimated using MitoSox. Primary microglial cells (~20,000 cells per well) were seeded in a 96-well plate and treated with αSyn_agg at various time points after cell attachment. Following treatment, the cells were exposed to 5 μM MitoSOX dye and incubated for 20 min at 37°C, protected from light. The cells were then washed twice with Hank’s Buffered Salt Solution (HBSS) before measuring the fluorescence intensity using a Spectramax™ microplate reader at 510 and 580 nm of excitation and emission wavelengths, respectively.

Mitochondrial function, a key indicator of cell health, can be assessed by monitoring changes in MMP. Membrane potential changes in the mitochondria were monitored by 5,5′,6,6′-tetrachloro-1,1′,3,3′-tetraethyl-benzimidazolcarbocyanineiodide (JC-1) staining as previously described (Ghosh et al., 2010). Briefly, the cells were plated at 20,000 cells/well into a 96-well black-walled, clear-bottom plate. The assay plates were incubated overnight at 37°C for cell adhesion. After the respective incubation times, JC-1 dye solution was added to each well, and the cells were incubated at 37°C, 5% CO₂ for 30 min. The monomeric form of JC-1 dye in the cytosol was detected fluorometrically through the emission of green fluorescence at the excitation/emission wavelengths 485/535 nm and aggregates in healthy mitochondria were detected through the red fluorescence emitted at the excitation/emission wavelengths of 550/600 nm using a Spectramax™ microplate reader. The changes in MMP were calculated as the red/green ratio and expressed as a percentage of control.

Nitric oxide production by primary microglial cells was measured by Griess assay (Sigma-Aldrich) as described previously. The cells were plated at 20,000 cells/well into a 96-well plate. The assay plates were incubated overnight at 37°C for cell adhesion and exposed to αSyn_agg at various time points (6, 12, 18, 24 h). At the end of each treatment, 100 μl of supernatant was added to an equal volume of Griess reagent per well of a 96-well plate. The samples were then incubated at 37°C at RT for 15 min until their color stabilized. The absorbance was measured at 540 nm using a Synergy 2 multimode microplate reader (BioTek Instruments, Winooski, VT, USA). Sodium nitrate was used as the standard to determine the nitrite concentration in the samples.

MTS assay using the Cell Titer 96^® Aqueous One Solution Cell Proliferation (MTS) assay kit was conducted based on the method described in our previous publications (Charli et al., 2016; Singh et al., 2018). Briefly, 20,000 cells/well were seeded in a 96-well tissue culture plate and exposed to 100 μl of treatment media, and the respective conditioned media was collected from the microglia. Cells were then incubated with the MTS dye for 45 min in a humidified, 5% CO₂ atmosphere at 37°C. The supernatant was removed, the resulting formazan crystals were dissolved with 25 μl of dimethyl sulfoxide (Sigma-Aldrich), and the color change was measured spectrophotometrically at 490 nm. The values were expressed as % of control.

SYBR Green qRT-PCR was performed on the harvested cells and tissues as described by Samidurai et al. (2020). RNA was extracted using the TRIZOL reagent (Invitrogen, Carlsbad, CA, USA) as per the manufacturer’s protocol. The isolated RNA was converted into cDNA using a High-Capacity cDNA synthesis kit (Applied Biosystems, Waltham, MA, USA). The qRT-PCR was carried out on an Applied Biosystems QuantStudio 3 system with the SYBR master mix (Invitrogen, Carlsbad, CA, USA). QuantiTect Primer Assays (Qiagen, Germantown, MD, USA) for genes specific to the pro-inflammatory cytokines TNF-α, IL-1β (#330001), and IL-6 were used. Normalization was done with the housekeeping gene 18S rRNA. TNF-α and IL-6 were obtained from the Iowa State University DNA Facility ISU DNA Facility (sequences are provided in Supplementary Table 2), IL-1β (QT01048355), and 18S (#PPM57735E) purchased from the Qiagen. Dissociation curves were run to ensure a single amplicon peak in all the SYBR Green reactions. The results were graphed as fold change in the respective gene expression calculated using the ΔΔCt method.

Cells and tissue were lysed and homogenized using modified RIPA buffer and subjected to fluorescent immunoblotting as previously described (Gordon et al., 2016b; Samidurai et al., 2020). Normalized protein (30 μg) from each sample was loaded into their respective wells and separated using a 12% SDS-PAGE gel. The proteins were then transferred to a nitrocellulose membrane, and the nonspecific-binding sites were blocked for 1 h at RT using an Odyssey blocking buffer (LI-COR, Lincoln, NE, USA). Membranes were then incubated overnight at 4°C in one of the following primary antibodies p-PKCδ, PKCδ, ATF-4, p-IRE1α, p-eIF2α (1:1,000, Rabbit monoclonal), NLRP3 (1:1,000, mouse monoclonal), ASC (1:1,000, rabbit polyclonal) caspase-1 (1:1,000, mouse polyclonal), TXNIP and TrX (1:1,000, Rabbit polyclonal). Next, the membranes were triple-washed with PBS, and then the membranes were visualized using IRDye-tagged secondary antibodies (incubated for 1 h at RT) on an LI-COR Odyssey infrared imaging system. To avoid variation in loading, the same blots were stripped via incubation in NewBlot Stripping Buffer for 30 min in RT and incubated with an anti-actin antibody; β-actin (1:10,000, mouse monoclonal) was used as a loading control to confirm the loading of equal protein concentration in each lane. Full blot images have been provided in the Supplementary Figure 4.

Immunoprecipitation studies were carried out as described previously but with minor modifications (Filhoulaud et al., 2019; Panicker et al., 2019). Approximately 6 × 10⁶ MMCs were seeded and treated with αSyn_agg or vehicle for 18 h. Subsequently, for TXNIP immunoprecipitation experiments, cells were collected and spun down at 5,000× g and the cell pellets were resuspended and lysed with TNE buffer (10 mM Tris-HCl, pH 7.5, 1% Nonidet P-40, 0.15 M NaCl, 1 mM EDTA, and 1:1,00 protease inhibitor mixture), and kept on ice for 30 min. The lysates were then centrifuged at 17,400× g for 35 min at 4°C to remove the debris. The supernatant protein concentration was measured and normalized using the Bradford method. For the input fraction, approximately 50 μg protein was used. Approximately for immunoprecipitation analysis, 500 μg protein per sample in 500 μl TNE buffer was used. The extracted proteins were incubated with 5 μg of anti-TxNIP antibody and placed on an orbital shaker at 4°C overnight. The following day, bound proteins were recovered following the addition of protein A/G magnetic beads (Thermo-fisher) and subsequent placement on an orbital shaker for 1 h at RT. Protein A/G magnetic beads were centrifuged and subsequently subjected to four washes for 2–3 min each with TNE buffer. The bound proteins were eluted from the beads by boiling in Laemmli buffer for 10 min at 97°C. The eluted proteins were collected by centrifuging at 5,000 rpm for 5 min and separated on 12% SDS-PAGE gels and followed by immunoblotting using antibodies against TXNIP. Quick Western detection was used to avoid the visualization of denatured IgG. Membrane images were captured on an LI-COR Odyssey imaging system.

Proximity ligation assay (PLA) was carried out in PMG according to the manufacturer’s protocol. Briefly, 10,000 primary microglial cells were seeded on PDL-coated coverslips in 96 well culture plates. After treatment of PMG with αSyn for 18 h, cells were washed with PBS and fixed using 4% PFA, blocked with blocking buffer (PBS containing 1.5% BSA, 0.5% Triton X-100, and 0.05% Tween-20), and subsequently incubated with antibodies against NLRP3 and TXNIP at 4°C overnight. Secondary antibodies conjugated with positive and negative oligonucleotides (PLA probes) were then incubated with the cells. Next, ligase was added and these oligonucleotides hybridize to the PLA probes, and when close enough, form a closed circle. In the DNA circle, one of the antibody-conjugated DNA probes serves as a primer for rolling-circle amplification (RCA), and a repeated sequence (concatemeric) product was generated when DNA polymerase and nucleotides were added. Fluorescently labeled oligonucleotides proceed to bind to the RCA product, allowing visualization of the protein-protein interaction as single dots via fluorescence microscopy (Wang et al., 2015). The protein-protein interactions were visualized as green puncta and analyzed by fluorescence microscopy using a Keyence microscope (with 60X objective). The PLA signals were further quantified using the ImageJ software according to the previous publication (Gomes et al., 2016).

The PKCδ siRNA was purchased from Santa Cruz Biotech (#SC-36246; Dallas, TX, USA), and TXNIP siRNA was purchased from life technologies (Ambion #4390771; Carlsbad, CA, USA). Lipofectamine 3,000 reagent was used for all the siRNA transfections according to the manufacturer’s protocol. One-day-old MMCs seeded at the density of 1.5 × 10⁶ in 6-well plates were used for transfection studies. Next, 700 pM of PKCδ siRNA, or 1.5 nM of TXNIP siRNA, or an equal amount of scrambled siRNA mixed with 5 μl of Lipofectamine 3,000 was added to the respective wells. The cells were then incubated with siRNA for 48 h, after which cells were treated with αSyn_agg for 24 h. Immunoblots were performed to check for the efficiency of target gene knockdown as well as for the determination of inflammation and ERS markers.

Data were analyzed using a two-tailed t-test (two groups) or one-way or two-way ANOVA (multiple groups) followed by Bonferroni’s post hoc analysis (PRISM 6.0 software, GraphPad, La Jolla, CA, USA), and are represented as mean ± SEM. Differences were considered statistically significant for p-values ≤ 0.05.

Lewy bodies, which represent a classical pathological hallmark of PD and DLB exhibits microglial and neuronal inclusions enriched with αSyn_agg (Fellner et al., 2011; Hoffmann et al., 2019). Given that αSyn_agg positive inclusions appear prior to the development of clinical signs (Braak et al., 2006) and that they are taken up by the microglia and contribute to the templated conversion of endogenous monomeric αSyn (Earls et al., 2019; Pike et al., 2021). We initially characterized the αSyn fibrils. Therefore, in this study, we used the human αSyn_agg which was found to induce microglial activation response consistent with a previous finding from our lab (Panicker et al., 2019). In our previous studies, we examined αSyn fibrils using the thioflavin T assay to determine the extent of fibrillar content in the αSyn_PFF fractions. Consistent with a previous report our studies revealed that a majority of the αSyn_agg were composed of fibrillar species (Manne et al., 2019). We further validated the confirmation of fibrillar αSyn species using transmission electron microscopy (TEM; Figure 1A) prior to microglia treatment studies. Moreover, we found internalization and microglial activation capabilities by human αSyn_agg in mouse primary microglia as previously reported (Panicker et al., 2019). In fact, immunofluorescence staining verified that treatment of primary microglia with 1 μM αSyn_agg led to its internalization (Figure 1B) as determined using an antibody against human total αSyn, consistent with a previous finding from our lab (Panicker et al., 2019). Given that αSyn aggregation reportedly induces mitochondrial dysfunction, thereby contributing to disease pathology (Smith et al., 2005) next, we investigated the temporal pattern of mitoROS generation, mitochondrial membrane potential (MMP) collapse, and PKCδ activation in mouse primary microglia (mouse primary microglia) stimulated with 1 μM αSyn_agg based on previous reports (Boza-Serrano et al., 2014; Lee et al., 2018; Panicker et al., 2019) for increasing durations (6, 12, 18, 24 h; Hoffmann et al., 2016; Sarkar et al., 2020). Following treatment of primary microglial cells with 1 μM of αSyn_agg robust mitochondrial superoxide generation, mitochondrial membrane potential (MMP) generation and nitrite release were evidenced as determined by MitoSOX, JC-1, and Griess spectrophotometric plate reader assays, respectively. The cells treated with αSyn_agg exhibited a significant (p < 0.001) time-dependent increase in mitochondrial (mito)ROS generation with an accompanying dissipation of MMP (Figures 1C,D), as well as nitrite release in a time-dependent manner (Supplementary Figure 1B) as compared to vehicle-treated cells. These results indicate that αSyn_agg induced mitochondrial dysfunction in primary microglia via a mitochondrial oxidative stress mechanism and accompanying collapse of MMP.

Previous reports show that PKCδ is a critical regulator of the proinflammatory microglial activation state in response to diverse inflammogens including αSyn, TNF-α, and LPS (Wen et al., 2011; Gordon et al., 2016a). Next, we examined whether activated PKCδ, a redox-sensitive kinase, is associated with mitochondrial dysfunction. Our WB studies revealed that αSyn_agg stimulation of mouse primary microglia resulted in a pronounced time-dependent activation of PKCδ as evidenced by prominent PKCδ phosphorylation at site Tyr-311 at 12 h which remained elevated for the remainder of the treatment duration as compared to vehicle-treated cells (Figure 1E), suggesting that mitochondrial oxidative stress may serve as a trigger for PKCδ activation consistent with previous reports (Majumder et al., 2001; Steinberg, 2015). These data collectively suggest that PKCδ activation contributes to microglia activation presumably via mitochondrial oxidative stress mechanism. Several reports link misfolded αSyn-induced PD pathology and ERS. For example, the overexpression of A53T αSyn promotes ER dysfunction, thereby eliciting pronounced ER stress (Smith et al., 2005; Karim et al., 2020), and blockage of ERS was found to protect against A53T αSyn-induced cell death (Boyce et al., 2005). Moreover, aberrant generation of mitoROS has been shown to contribute to ERS (Liu et al., 2019). Therefore, we further examined the magnitude of expression of ERS markers in αSyn_agg stimulated mouse primary microglia. As anticipated, the upregulation of ERS markers, including p-IRE1α, p-eIF2α, CHOP, and ATF-4, were observed in αSyn_agg-treated mouse primary microglia as compared with vehicle-treated cells (Figure 1F). Likewise, immunofluorescence analysis showed that αSyn_agg increased the p-eIF2α immunoreactivity in the primary microglia cells as compared to the vehicle treated cells further supporting the central role of ERS in microglial activation response in response to αSyn_agg (Figure 1G). To ensure that the observed effects were not due to cellular toxicity, we performed a cell viability assay using an MTS assay. MTS revealed little or no evidence of cell death at a concentration of 1 μM αSyn_agg (Supplementary Figure 1A). Given that αSyn concentrations ranging between 0.5–5 μM have been widely used in various cell culture studies and that the synaptic concentration of αSyn is believed to fall within the range of 2–4 μM (Konno et al., 2012; Westphal and Chandra, 2013; Boza-Serrano et al., 2014; Domert et al., 2016; Lee et al., 2018; Hoffmann et al., 2019), we used 1 μM αSyn_agg in the remainder of the experiments unless otherwise stated. These results indicate that our αSyn_agg proteins are capable of inducing a reactive microglial activation state via mitochondrial oxidative stress and PKCδ activation mechanism as well as ERS in mouse primary microglia.

Thioredoxin-interacting protein (TXNIP), also known as thioredoxin-binding protein-2 or vitamin D3-upregulated protein 1, has been shown to be an endogenous antagonist of the antioxidant protein thioredoxin (Trx) (Zhao et al., 2017b). In oxidatively stressed cells, in response to ERS, TXNIP has been linked to NLRP3 inflammasome activation (Wang C. Y. et al., 2019). To test whether TXNIP/Trx expression levels are altered by αSyn_agg, we treated primary microglia with αSyn_agg. As predicted, TXNIP expression was significantly increased while Trx expression levels were downregulated in mouse primary microglia stimulated with αSyn_agg (Figure 2A) as compared to controls. We next examined whether TXNIP upregulation is associated with increased expression of NLRP3 inflammasome activation markers as well as inflammatory cytokine generation. Our studies indicate that αSyn_agg treatment increased protein expression of NLRP3 inflammasome components alongside expression of the pro-inflammatory cytokines IL-1β, IL-6, and TNF-α in αSyn_agg-treated mouse primary microglia as compared to vehicle (con) treated cells (Figure 2B). Together these studies suggest that the microglial activation state induced by αSyn_agg may involve a functional interaction between NLRP3 and TXNIP in primary microglia.

Previous studies have identified mitoROS as a critical causative factor in promoting Trx disassociation from TXNIP, and its subsequent association with NLRP3, thereby leading to the generation of inflammasome activation markers (Zhou et al., 2010; Han et al., 2018). Therefore, we next assessed the interaction between NLRP3 and TXNIP using proximity ligation assay (PLA) in αSyn_agg-stimulated cells. Immunofluorescence analysis revealed a significant interaction between endogenous TXNIP and NLRP3 proteins in αSyn_agg-stimulated mouse primary microglia as compared to vehicle-treated cells (Figure 3A). We further validated our PLA findings using IP/IB analysis whereby a significant interaction between TXNIP and NLRP3 was evidenced in αSyn_agg stimulated mouse primary microglia as compared to controls (Figure 3B) suggesting that the TXNIP/NLRP3 association may at least in part trigger a reactive microglial activation state in the context of α-synucleinopathy.

Previous reports suggest that ERS is a critical regulator of the TXNIP/NLRP3 signaling axis and resultant inflammatory response in diverse experimental models of inflammation-related disorders (Chen X. et al., 2019; Feng and Zhang, 2019). Furthermore, pharmacological inhibition of ERS was found to afford neuroprotection in the A53T mouse model of PD (Boyce et al., 2005). However, the link between ERS and TXNIP/Trx, redoxisome dysfunction during an αSyn_agg-induced microglial pro-inflammatory signaling event remains poorly characterized. We hypothesized that the TXNIP-Trx system is a critical intermediate that links ERS to microglia-mediated NLRP3 inflammasome activation subsequent to αSyn_agg stimulation. Therefore, we examined the contribution of ERS in TXNIP-Trx dysfunction utilizing SAL, which is a selective inhibitor of eIF2α and has been shown to exhibit anti-inflammatory effects in a rodent model of traumatic brain injury (Logsdon et al., 2016). We first examined whether SAL inhibited αSyn_agg-induced dysregulation of the TXNIP/Trx system. Using mouse primary microglia, we first determined whether 50 μM of SAL (Huang et al., 2012) modulated the expression of ERS markers, including p-eIF2α, in αSyn_agg-treated cells. Pretreatment with SAL increased αSyn_agg-induced upregulation of p-eIF2α (data not shown). Next, we investigated the effect of SAL on αSyn_agg-induced dysregulation of the TXNIP/Trx pathway in mouse primary microglia treated with αSyn_agg. Our WB analysis revealed that SAL treatment attenuated αSyn_agg-induced TXNIP upregulation while upregulating Trx expression (Figure 4A). Together, these results indicate that ERS blockade via SAL reduced the αSyn_agg-induced dysregulation of the TXNIP/Trx signaling pathway which may, in turn, prevent the reactive microglial activation state. Emerging evidence suggests that ERS plays a critical role in TXNIP and resultant activation of the NLRP3 inflammasome (Lerner et al., 2012; Oslowski et al., 2012). However, to what extent ERS regulates NLRP3 inflammasome activation in response to αSyn_agg stimulation of mouse primary microglia remains poorly understood. Here we identified that NLRP3 inflammasome activation markers are activated in an ERS-dependent manner. Our immunoblot results showed that SAL attenuated the stimulatory effect of αSyn_agg on NLRP3 (Figure 4A), and its activation markers as evidenced by reduced pro-inflammatory cytokine mRNA expression of IL-1β and TNF-α (Figure 4B). These findings indicate that ERS blockade via SAL may in part ameliorate the αSyn_agg-induced neurotoxic microglial activation state via inhibition of TXNIP/NLRP3 inflammasome axis activation in αSyn_agg-stimulated primary microglia.

Since αSyn_agg induces mitochondrial oxidative stress in primary microglia, next we explored the effects of (MitoApo), a mitochondrially targeted antioxidant on the αSyn_agg-induced ERS and TXNIP/NLRP3 signaling cascade. Owing to the fact that mouse microglial cell line, MMC closely mimic neonatal primary microglia (Sarkar et al., 2020), we utilized these cells to determine the activation of the aforementioned microglial pro-inflammatory mediators in MMCs treated with αSyn_agg in the presence or absence of mitoapocynin. A previous report demonstrated that ER-mitochondria proximity promotes NLRP3 localization to mitochondria, triggering its activation (Shimada et al., 2012). Indeed, we previously demonstrated that the NLRP3 inflammasome translocated to the mitochondria in rotenone-stimulated microglial cells, which coincided with an elevated mitochondrial oxidative stress (Lawana et al., 2017). Based on these findings, we hypothesized that blocking mitochondrial oxidative stress via MitoApo would reduce ERS and the associated TXNIP/NLRP3 signaling pathway. To address this hypothesis, MMC microglial cells were treated with MitoApo (10 μM). We found that there was a significant upregulation of ERS markers including eIF2α, ATF-4, TXNIP, and NLRP3 protein expression, as well as pro-inflammatory cytokine mRNA expression (IL-1β and TNF-α) that was accompanied by an upregulation of Trx expression in MMC cells treated with αSyn_agg, which was markedly reduced by treatment of mitoapocynin (Figures 5A–C). Collectively, these results indicate that MitoApo ameliorates the αSyn_agg-induced ERS and TXNIP/NLRP3 signaling axis in a cell culture model of α-synucleinopathy.

We next determined whether siRNA-mediated knockdown of PKCδ altered the αSyn_agg-induced microglial activation response. We have demonstrated that the redox-sensitive kinase PKCδ is linked to microglial NLRP3 inflammasome activation and IL-1β generation under oxidative stress conditions (Lawana et al., 2017; Panicker et al., 2019), raising the possibility that PKCδ may be involved in NLRP3 inflammasome activation via the ERS response. To determine whether or how PKCδ expression influence the αSyn_agg-induced ERS-associated microglial activation response, MMC mouse microglial cells were transfected with either control siRNA or PKCδ siRNA and subsequently treated with αSyn_agg. MMC microglial cells that were transfected with a small interfering RNA (siRNA) against PKCδ for 48 h displayed markedly reduced endogenous PKCδ levels (60–70%) as compared with scrambled siRNA-transfected cells (Supplementary Figure 2). We next evaluated the effects of RNAi-mediated knockdown of PKCδ on the αSyn_agg-induced ER stress response in MMCs. Furthermore, as expected, downregulation of PKCδ remarkably repressed the αSyn_agg-induced ER stress response as exemplified by reduced expression of BIP (Binding immunoglobulin protein), p-eIF2α in MMCs (Figure 6A).

Next, we investigated the impact of PKCδ knockdown on the expression of TXNIP and its binding partner Trx in αSyn_agg-stimulated microglial cells using WB analysis. PKCδ downregulation dramatically decreased the expression of TXNIP while upregulating Trx expression in microglial cells treated with αSyn_agg (Figure 6A). Finally, we tested the causal relationship between PKCδ activation and NLRP3 inflammasome activation in microglial cells stimulated with αSyn_agg. PKCδ knockdown in MMC microglial cells treated with αSyn_agg significantly reduced the expression of NLRP3 (Figure 6A) and mRNA expression of proinflammatory cytokines including TNF-α and IL-1β (Figure 6B). Taken together, these results demonstrate that downregulation of PKCδ reduces the αSyn_agg-induced microglial activation response at least in part via amelioration of ERS and the TXNIP/NLRP3 signaling axis and the associated generation of proinflammatory cytokines in MMC microglial cells.

Having demonstrated that TXNIP associates with NLRP3, we next determined whether TXNIP regulated NLRP3 inflammasome activation. Our approach was to assess NLRP3 activation markers since previous studies have demonstrated a positive association between TXNIP upregulation and activation and NLRP3 inflammasome activation markers (Tseng et al., 2016). Thus, in the initial set of experiments, we downregulated TXNIP expression using siRNA-mediated knockdown to determine its impact on NLRP3 inflammasome activation markers. Our WB analysis revealed a 65% knockdown efficiency (Supplementary Figure 3) of TXNIP that was accompanied by a marked reduction in the protein expression of NLRP3 inflammasome and cleaved caspase-1 expression (Figure 7A) in mouse primary microglia exposed to αSyn_agg. In parallel studies, qPCR analysis revealed that TXNIP siRNA ameliorated αSyn_agg-induced mRNA expression of IL-1β and TNF-α (Figure 7B). Our findings indicate a mechanistic link between TXNIP and the NLRP3 inflammasome activation in promoting a proinflammatory microglial phenotype and that downregulation of TXNIP ameliorates the proinflammatory microglial activation response in αSyn_agg-stimulated mouse primary microglia.

To further test the influence of αSyn_agg-induced microglial ERS on DAergic neuronal survival we treated primary microglia with αSyn_agg in the presence or absence of the ERS inhibitor SAL followed by 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) cell viability assay in MN9D DAergic neuronal cells treated with microglia-conditioned media (MCM) collected from αSyn_agg-stimulated microglial cells treated with or without SAL. Mouse primary microglia were pretreated with SAL for 6 h and subsequently stimulated with αSyn_agg for 12 h, followed by washing and incubation with fresh cell culture media for another 24 h (Figure 8A). The resulting MCM was transferred to MN9D DAergic neuronal cells. After 18 h following MCM addition, cell viability was assessed using an MTT assay. The MCM collected from αSyn_agg-stimulated mouse primary microglia increased MN9D DAergic cell death whilst this effect was markedly reduced in MN9D cells that were treated with MCM from SAL-pretreated, αSyn_agg-stimulated microglial cells (SAL/αSyn_agg-MCM; Figure 8B). These results are in line with previous studies showing that a neurotoxic microglial activation state exerts detrimental effects on DAergic neuronal integrity subsequent to the αSyn-induced microglial inflammatory response (Yun et al., 2018).

Our in vitro studies revealed that treatment of primary microglia induced PKCδ activation with an accompanying induction of the ERS-mediated TXNIP/NLRP3 signaling axis. Next, we sought to investigate the aforementioned signaling events in the αSyn_PFF mouse model of sporadic PD that reproduces several key PD pathological correlates including αSyn pathology, loss of DAergic neurons, and the microglial activation response in vivo. In this model, αSyn_PFF seeds are taken up by the striatum and transmitted to the nigra in a retrograde fashion via the nigrostriatal pathway where they function as a template and promote the seeding of endogenous murine αSyn to accumulate into misfolded phosphorylated pathological aggregates (Luk et al., 2012; Earls et al., 2019). Although microglia has been implicated in neuroinflammation and associated nigral TH neuronal loss (Duffy et al., 2018). To date, there is no published literature detailing TXNIP nor eIF2α-expression within microglia in the context of synucleinopathy. Therefore, we used specific markers of ERS and inflammation to further support the role of microglial ERS in PD like pathology. To this end, we performed double immunolabeling for p-eIF2α or TXNIP a in nigral brain section from αSyn_PFF and PBS (control) infused mice. A previous study reported that TXNIP colocalizes with microglia after subarachnoid hemorrhage (SAH; Zhao et al., 2017a). As expected, control mice displayed scant expression of eIF2α or TXNIP. In contrast, αSyn_PFF-infused mice display considerable colocalization of eIF2α and TXNIP within IBA-1-positive microglia (Figures 9A,B). Thus, our results demonstrate that both p-eIF2α and TXNIP colocalized with microglia in the substantia nigra of αSyn_PFF-infused mice, suggesting that these inflammatory markers may aggravate nigral DAergic neurotoxicity by triggering pro-inflammatory signaling events. Moreover, we cannot exclude the possibility that activation of the aforementioned proinflammatory factors in other CNS cell types might contribute to the nigral DAergic neurotoxicity. Additionally, consistent with our in vitro results, our WB analysis further confirmed that αSyn_PFF intrastriatal infusion significantly upregulated the ERS markers p-eIF2α, CHOP (C/EBP Homologous Protein), BIP, and ATF-4 (Figures 9C,D) in the SNpc, which positively correlated with TXNIP upregulation and the associated downregulation of Trx levels as compared to PBS-infused mice (Figure 9A). Likewise, this effect was accompanied by PKCδ activation and upregulation of the NLRP3 inflammasome in the SNpc that was associated with enhanced generation of proinflammatory cytokine mRNA levels including Il-1β, TNF-α, and IL-6 in the striatum as compared to PBS-infused mice at 60 dpi (Figure 9E). Next, we determined whether TH neuronal loss occurs during the latter part of the disease (180 dpi) in vivo as demonstrated previously (Duffy et al., 2018). As expected, delayed TH⁺ neuron loss in the SN of αSyn_PFF-infused mice was evidenced at 180 dpi as compared to PBS infused mice as determined by unbiased stereological analysis (Figure 9F). These results are consistent with our hypothesis that irremediable microglial ERS increases the susceptibility of nigral DAergic neurons to neurodegeneration in the context of α-synucleinopathy concomitant with the induction of TXNIP/NLRP3 mediated proinflammatory signaling events.