Lewy body-like alpha-synuclein inclusions trigger reactive microgliosis prior to nigral degeneration

Lewy body-like alpha-synuclein inclusions trigger reactive microgliosis prior to nigral degeneration Background: Converging evidence suggests a role for microglia-mediated neuroinflammation in Parkinson’s disease (PD). Animal models of PD can serve as a platform to investigate the role of neuroinflammation in degeneration in PD. However, due to features of the previously available PD models, interpretations of the role of neuroinflammation as a contributor to or a consequence of neurodegeneration have remained elusive. In the present study, we investigated the temporal relationship of neuroinflammation in a model of synucleinopathy following intrastriatal injection of pre-formed alpha-synuclein fibrils (α-syn PFFS). Methods: Male Fischer 344 rats (N = 114) received unilateral intrastriatal injections of α-syn PFFs, PBS, or rat serum albumin with cohorts euthanized at monthly intervals up to 6 months. Quantification of dopamine neurons, total neurons, phosphorylated α-syn (pS129) aggregates, major histocompatibility complex-II (MHC-II) antigen-presenting microglia, and ionized calcium-binding adaptor molecule-1 (Iba-1) immunoreactive microglial soma size was performed in the substantia nigra. In addition, the cortex and striatum were also examined for the presence of pS129 aggregates and MHC-II antigen-presenting microglia to compare the temporal patterns of pSyn accumulation and reactive microgliosis. Results: Intrastriatal injection of α-syn PFFs to rats resulted in widespread accumulation of phosphorylated α-syn inclusions in several areas that innervate the striatum followed by significant loss (~ 35%) of substantia nigra pars compacta dopamine neurons within 5–6 months. The peak magnitudes of α-syn inclusion formation, MHC-II expression, and reactive microglial morphology were all observed in the SN 2 months following injection and 3 months prior to nigral dopamine neuron loss. Surprisingly, MHC-II immunoreactivity in α-syn PFF injected rats was relatively limited during the later interval of degeneration. Moreover, we observed a significant correlation between substantia nigra pSyn inclusion load and number of microglia expressing MHC-II. In addition, we observed a similar relationship between α-syn inclusion load and number of microglia expressing MHC-II in cortical regions, but not in the striatum. (Continued on next page) * Correspondence: caryl.sortwell@hc.msu.edu Department of Translational Science and Molecular Medicine, Michigan State University, 400 Monroe Avenue NW, Grand Rapids, MI 49503-2532, USA Mercy Health Hauenstein Neuroscience Medical Center, Grand Rapids, MI, USA Full list of author information is available at the end of the article © The Author(s). 2018, corrected publication May/2018. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 2 of 18 (Continued from previous page) Conclusions: Our results demonstrate that increases in microglia displaying a reactive morphology and MHC-II expression occur in the substantia nigra in close association with peak numbers of pSyn inclusions, months prior to nigral dopamine neuron degeneration, and suggest that reactive microglia may contribute to vulnerability of SNc neurons to degeneration. The rat α-syn PFF model provides an opportunity to examine the innate immune response to accumulation of pathological α-syn in the context of normal levels of endogenous α-syn and provides insight into the earliest neuroinflammatory events in PD. Keywords: Neuroinflammation, Parkinson’s disease, Animal models, Synucleinopathy, Microglia, Major-histocompatibility complex-II, Neurodegeneration, Selective vulnerability Background risk for development of PD after pesticide exposure [12]. The etiology of Parkinson’s disease (PD) is stochastic: a Increased MHC-II is often concurrently upregulated with culmination of aging-related changes in brain environ- genes for proinflammatory cytokines such as tumor necro- ment, genetic predispositions, and environmental insults sis factor (TNF) and interleukin-1 beta (IL-1β)[13]. More- that result in accumulation of alpha-synuclein (α-syn) over, decreased MHC-II expression was shown to attenuate inclusions (i.e., Lewy bodies) and degeneration of nigros- downstream secretion of proinflammatory cytokines [14, triatal dopamine neurons [1, 2]. Converging evidence 15]. Taken together, it is likely that MHC-II is most closely suggests a role for microglia-mediated neuroinflamma- associated with a proinflammatory phenotype in microglia tion in human PD. This theory is supported by observa- and may play a contributory role in nigral degeneration in tions of increased inflammatory cytokines in both PD PD. However, while the concept that MHC-II expression patient cerebral spinal fluid (CSF) and plasma [3, 4] and on microglia is increased in PD patients is not novel [9], in the patient brain as longitudinal PET imaging has the temporal pattern of observed increases in MHC-II in demonstrated early and sustained microglial activation relation to α-syn aggregation and/or nigrostriatal degener- in the basal ganglia [5]. Furthermore, postmortem ana- ation has been unable to be systematically examined. lyses in PD patients revealed increased expression of in- Animal models of PD can serve as platforms to inves- flammatory markers such as human leukocyte antigen tigate the role of neuroinflammation in PD-related cell (HLA-DR), major histocompatibility complex-II (MHC- death and dysfunction. The neuroinflammatory conse- II), phagocytic marker CD68, intercellular adhesion quences of nigral degeneration and/or α-syn aggregation molecule-I (ICAM-1), and integrin adhesion molecule have been examined previously in various models, (LFA-1) in the substantia nigra [6, 7]. However, a draw- including but not limited to, neurotoxicant models back of biofluid and postmortem PD brain samples is (6-hydroxydopamine (6-OHDA) [16, 17]; 1-methyl-4- that they only provide a static snapshot of events within phenyl-1,2,3,6-tetrahydropyridine (MPTP) [18, 19]); a longitudinal cascade of PD pathophysiology. This is es- transgenic models expressing human wild-type or mutant pecially problematic as the overwhelming majority of PD α-syn (A503T, A30P [20–22]) and viral vector-mediated patient samples are collected from individuals who have overexpression of human wild-type or mutated α-syn in likely harbored PD-related pathology for decades before, the nigrostriatal system [22–28]. However, certain charac- if also not after, diagnosis [8]. This confounds interpreta- teristics of these models limit interpretations regarding tions of the role of neuroinflammation in degeneration the specific initiator of the neuroinflammation observed— in PD and prevents the understanding as to whether synuclein inclusions and/or degeneration. Neurotoxicant neuroinflammation participates as a contributor to nigral models (6-OHDA, MPTP) rarely exhibit α-syn pathology degeneration or is simply an artifact of cell death. [18, 29]. Transgenic models generally do not recapitulate One of the most consistent observations in postmor- marked nigrostriatal degeneration despite widespread, tem PD tissue is an increase in the number of microglia α-syn pathology [21, 30]. Whereas a robust inflamma- expressing MHC-II (HLA-DR in humans [7, 9, 10]), a tory response is observed in association with the ele- cell surface protein on antigen-presenting cells which is vated α-syn levels, aggregates, and nigral degeneration necessary for CD4+ T cell infiltration. More recently, in viral vector-based α-syn overexpression models gene expression changes related to inflammation, includ- [22–28, 31–33], the contribution of supraphysiological α- ing an upregulation of MHC-II, have also been noted in syn levels or the α-syn species difference (human α-syn incidental Lewy body disease subjects (Braak stages 1–3 expressed in rat or mouse) to the neuroinflammatory [10, 11]). Additionally, a variant in the HLA-DR gene response is unclear. Importantly, in human sporadic PD, which encodes for MHC-II is associated with amplified total α-syn levels are not increased; rather, phosphorylation Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 3 of 18 and the ratio of soluble to insoluble α-syn increases over described [39–41]. Prior to sonication, α-syn fibrils were time [34–36]. assessed to verify lack of contamination (LAL Assay An alternative model of the key features of human (~ 1 endotoxin units/mg), high molecular weight sporadic PD such as (1) protracted development of (sedimentation assay), beta sheet conformation (thiofla- α-syn inclusions under conditions of (2) normal expres- vin T), and structure (electron microscopy). Prior to injec- sion levels of endogenous α-syn during an interval that (3) tion, PFFs were thawed, diluted in sterile Dulbecco’sPBS precedes significant nigrostriatal degeneration would offer (DPBS, 2 μg/μl), and sonicated at room temperature using distinct advantages and allow the time course and poten- an ultrasonicating homogenizer (300VT; Biologics, Inc., tial impact of neuroinflammation to be delineated. Re- Manassas, VA) with the pulser set at 20% and power out- cently, our lab has characterized a rat model of PD that put at 30% for 60 pulses at 1 s each. Following sonication, recapitulates this sequence of events, extending previous a sample of the PFFs was analyzed using transmission findings in mice [37, 38]. In this model, nigrostriatal synu- electron microscopy (TEM). Formvar/carbon-coated cop- cleinopathy is induced by intrastriatal injection of soni- per grids (EMSDIASUM, FCF300-Cu) were washed twice cated pre-formed α-syn fibrils (α-syn PFFs) into wild-type with ddH O and floated for 1 min on a 10-μldrop of soni- rats [37, 39]. The fibrils act as seeds to template and cated α-syn fibrils diluted 1:20 with DPBS. Grids were trigger normal levels of endogenous α-syn to accumulate stained for 1 min on a drop of 2% uranyl acetate aqueous into misfolded hyperphosphorylated, pathological α-syn solution; excess uranyl acetate was wicked away with filter (Fig. 1a). The initial injection of α-syn PFFs per se does paper and allowed to dry before imaging. Grids were imaged not directly cause toxicity, given that α-syn pathology and on a JEOL JEM-1400 transmission electron microscope. nigral degeneration do not occur in α-syn knockout ani- The length of over 500 fibrils per sample was measured to mals injected with PFFs [38]. In this model, we observe a determine average fibril size. The mean length of sonicated widespread accumulation of intraneuronal Lewy neurite- mouse α-syn PFFs was estimated to be 51.22 ± 1.31 nm, well like and Lewy body-like inclusions of phosphorylated within the optimal fibril length previously reported to result α-syn (pSyn) in areas that innervate the striatum. Import- in seeding of endogenous phosphorylated α-syn inclusions antly, the accumulation of intracellular pSyn is gradual in vitro and in vivo (Fig. 1b, c)[42]. and results in loss of striatal dopamine and metabolites in addition to ~ 40% loss of SNc dopamine neurons over Intrastriatal injections 6 months [37]. Thus, the synucleinopathy produced in the Sonicated PFFs were kept at room temperature during α-syn PFF model provides a unique opportunity to exam- the duration of the surgical procedures. All rats were ine the neuroinflammatory consequences of α-syn inclu- deeply anesthetized with isoflurane received two 2-μl sion accumulation in the context of normal levels of unilateral intrastriatal injections (4 μl total; AP + 1.6, endogenous, intracellular α-syn. In the present study, we ML + 2.4, DV − 4.2; AP − 1.4, ML + 2.0, DV − 7.0 from systematically investigated the temporal profile of Lewy the skull) either of sonicated mouse α-syn PFFs (2 μg/μlas body-like phosphorylated α-syn inclusion load, reactive described previously [37]) or an equal volume of DPBS at a microglial morphology, MHC-II antigen presentation, and rate at 0.5 μl/min (n = 6 per treatment per time point). In- degeneration in the SN. Importantly, we observe reactive jections were administered made using a pulled glass needle microglia and increased microglial MHC-II expression in attached to a 10-μl Hamilton syringe. After each injection, association with peak load of SNc pSyn inclusions months the needle was left in place for 1 min, retracted 0.5 mm, left prior to degeneration, suggesting that neuroinflammation in place for an additional 2 min, and then slowly with- may contribute to nigrostriatal degeneration. drawn. Animals were monitored post-surgery and eutha- nized at predetermined time points (14, 30, 60, 90, 120, Methods 150, and 180 days; Fig. 1). In a subsequent experiment, rats Animals received two 2-μl unilateral intrastriatal injections either of Young adult (2 months), male Fischer344 rats (n = 114) mouse α-syn PFFs 2 μg/μl, DPBS, or rat serum albumin were used in this study. All animals were provided food (RSA, Sigma-Aldrich, St. Louis, MO; 9048-46-8; 2 μg/μl) at and water ad-libitum and housed at the AAALAC- the identical coordinates and were euthanized at 2 months approved Van Andel Research Institute vivarium. All postinjection (n = 6 per treatment). procedures were approved and conducted in accordance with Institute for Animal Use and Care Committee Immunohistochemistry (IACUC) at Michigan State University. All animals were euthanized via pentobarbital overdose (60 mg/kg) and intracardially perfused with heparinized Preparation of α-syn PFFs and verification of fibril size 0.9% saline followed by cold 4% paraformaldehyde in Purification of recombinant, full-length mouse α-syn 0.1 M PO . Brains were extracted and postfixed in 4% and in vitro fibril assembly was performed as previously PFA for 48 h and placed in 30% sucrose until they sunk. Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 4 of 18 Fig. 1 Experimental design and PFF quality control. a Experimental design: 2-month-old male Fischer344 rats (N = 114) received two unilateral intrastriatal injections of either sonicated α-syn PFFs, Dulbecco’s PBS (PBS), or rat serum albumin (RSA; follow-up study only). Cohorts of rats were euthanized at an early time point (2 weeks) and monthly intervals thereafter. Brains were removed and processed for immunohistochemical measures of pathology as detailed. b Electron micrographs of unsonicated (left) α-syn PFFs and sonicated α-syn PFFs (right); scale bars = 100 nm. c Measurement distribution of ~ 500 sonicated PFFs prior to injection; mean fibril size = 51.22 ± 1.31 nm. d Schematic of PFF model of synucleinopathy. Sonicated α-syn fibrils are injected into the striatum and taken up by nigrostriatal terminals (1), after which they template and convert endogenous α-syn to a hyperphosphorylated, pathological form (2), ultimately accumulating into Lewy neurite- and Lewy body-like inclusions (3). Abbreviations: α-syn = alpha-synuclein; PFFs = pre-formed alpha-synuclein fibrils; PBS = phosphate-buffered saline; DMS = dorsal medial striatum; VLS = ventrolateral striatum; IHC = immunohistochemistry; pSyn = α-syn phosphorylated at serine 129; MHC-II = major histocompatibility complex-II; Iba-1 = ionized calcium-binding adaptor molecule 1; TH = tyrosine hydroxylase; NeuN = neuronal nuclei; CD68 = cluster of differentiation 68; QC = quality control For sectioning, brains were frozen on a sliding microtome blocking, sections were immunolabeled with primary anti- and cut at 40 μm. Free-floating sections (1:6 series) were bodies: mouse anti-α-syn fibrils/oligomers (O2; 1:5000 transferred to 0.1 M tris-buffered saline (TBS). Following [43]) or mouse anti α-syn fibrils (F2; 1:5000 [43]), pan the washes, endogenous peroxidases were quenched in 3% rabbit-anti α-syn (Abcam, Cambridge, MA; AB15530, H O for 1 h and rinsed in TBS. Sections were blocked in 1:1000), mouse anti-phosphorylated α-syn at serine 2 2 10% normal goat serum/0.5% Triton X-100 in TBS (NGS, 129 (pSyn, 81A; Abcam, Cambridge, MA; AB184674; Gibco; Tx-100 Fischer Scientific) for 1 h. Following the 1:10,000), rabbit anti-tyrosine hydroxylase (TH; Millipore, Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 5 of 18 Temecula, CA; MAB152, 1:4000), rabbit anti-ionized reagent (Vector Laboratories, Burlingame, CA) was used calcium-binding adaptor molecule-1 (Iba-1; Wako, as the chromogen. Slides were rinsed in TBS and cover- Richmond, VA; 019-19741, 1:1000), mouse anti-neuronal slipped with Cytoseal 60. Images were taken on a Nikon nuclei (Neu-N; Millipore, Temecula, CA; MAB 377, 1:5000), Eclipse 90i microscope with a QICAM camera (QImaging, or mouse anti-rat major histocompatibility complex-II for Surrey, British Colombia, Canada). antigen-presenting microglia (MHC Class II RT1B clone OX-6, Bio-Rad, Hercules, CA; MCA46G, 1:5000) over- Quantification of TH, NeuN, pSyn, and MHC-II night in 1% NGS/0.5% Tx-100/TBS at 4 °C. Following the immunoreactive profiles washes, sections were incubated in biotinylated secondary Microbrightfield (MBF) Stereoinvestigator (MBF Bioscience, antibodies (1:500) against mouse (Millipore, Temecula, Williston, VT) was used to estimate the total population CA; AP124B) or rabbit IgG (Millipore, Temecula, CA; of THir and NeuNir neurons to determine the time course AP132B) followed by washes in TBS and 2-h incubation of TH phenotype loss and overt nigral degeneration. Con- with Vector ABC standard detection kit (Vector Labora- tours were drawn around the SNc using the ×4 objective tories, Burlingame, CA; PK-6100). Labeling for pSyn, on every sixth section through the rostrocaudal axis (9–10 MHC-II, and TH was visualized by development in 0. sections). A series of counting frames (50 μm×50 μm) 5 mg/ml 3,3′ diaminobenzidine (DAB; Sigma-Aldrich St. was systematically and randomly distributed over grid Louis, MO; D5637-10G) and 0.03% H O . For dual bright- (183 μm × 112 μm) placed over the SNc, allowing for 2 2 field visualization of Neu-N and Iba-1, sections were de- quantification of approximately 20% of the SNc. An inves- veloped according to the manufacturer’sinstructions tigator blinded to experimental conditions counted THir using the Vector ImmPACT DAB Peroxidase (Vector and NeuNir cells using the optical fractionator probe with Labs, Burlingame, CA; SK-4605) and ImmPACT VIP a ×60 oil immersion objective. Markers were placed on Peroxidase (Vector Labs, Burlingame, CA; SK-4105) kits, each THir or NeuNir cell in a 1–2-μm z-stack within the respectively. Slides were dehydrated in ascending ethanol counting frame. Between 50 and 500 objects were counted series and then xylenes before coverslipping with Cytoseal to generate stereological estimates of the total cell popula- (Richard-Allan Scientific, Waltham, MA). A subset of tion. The total population estimate was calculated using pSyn-labeled sections were also counterstained with cresyl optical fractionator estimates, and variability within ani- violet for quantification of intraneuronal pSyn inclusions mals was assessed via the Gunderson coefficient of error in the SNc. (< 0.1). Due to heterogeneity in the distribution of both pSyn and MHC-II immunoreactive profiles within the SN, RNAscope in situ hybridization for Iba-1 and MHC-II IHC total enumeration rather than counting frames was used Forty-micrometer-thick striatal tissue sections were in- for quantification. Neurons with intraneuronal pSyn in- cubated in pretreat 1 from the RNAscope Pretreatment clusions were defined as profiles of dark, densely Kit (Advanced Cell Diagnostics, Hayward, CA; 310020) stained pSyn immunoreactivity within cresyl violet- for 1 h. Sections were washed in TBS and then mounted positive neurons. Contours were drawn around the on VistaVision HistoBond slides (VWR, Randor, PA; SNcusing the×4objectiveon everysixth section 16004-406) and placed on slide warmer at 60 °C over- through the entire rostrocaudal axis of the SNc (9–10 night. Slides were then incubated for 10 min in pretreat sections). pSyn inclusions and MHC-IIir microglia 2 at 99 °C and washed twice in water. Tissue was out- were then systematically counted within each contour lined with Pap Pen (Abcam, Cambridge, UK; ab2601), using the ×20 objective. Numbers represent the raw incubated with pretreat 3 in a hybridization oven at 40 °C total number of pSyn inclusions or MHC-IIir microglia for 15 min, washed twice in water, and incubated with the per animal multiplied by 6 to extrapolate the popula- probe for AIF1 (Iba1; Advanced Cell Diagnostics, tion estimate. Hayward, CA; 457731) for 2 h in the hybridization oven at 40 °C. Six amplification steps with the amplification Microglial soma area analysis buffers (Advanced Cell Diagnostics, Hayward, CA; Forty-micrometer-thick nigral tissue sections (1:6 series) 320600) were then performed in alternating 30- and from animals injected with α-syn PFFs, RSA, or and 15-min incubation intervals in the hybridization oven per DPBS 2 months and 6 months following injection were manufacturer instructions. Tissue was developed using dual labeled for NeuN and Iba-1 as described above to the supplied DAB reagent (Advanced Cell Diagnostics, distinguish the SNc from the SNr. The three nigral sec- Hayward, CA; 320600). Tissue was then counterstained tions adjacent to the sections containing the most pSyn for MHC-II (RT1B clone OX-6, Bio-Rad, Hercules, CA; inclusions were identified. z-stack images of the ipsilat- MCA46G, 1:500) in a hybridization chamber, following eral and contralateral SNr bordering the SNc were taken the same procedures as detailed for other immunohisto- on a Nikon Eclipse 90i microscope with a QICAM cam- chemical stains with the exception that the Vector SG era (QImaging, Surrey, British Colombia, Canada) using Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 6 of 18 the ×20 objective and analyzed with Nikon Elements AR Results (version 4.50.00, Melville, NY). Using the auto-detect Sonicated α-syn PFFs are the optimal size for pathology feature, each Iba-1ir soma’s border was outlined and ad- induction in vivo justed accordingly to obtain an accurate quantification Prior to intrastriatal injection of mouse α-syn PFFs, we in- of area of the soma, excluding any processes. All micro- vestigated the size of the PFFs following sonication using glia in the field of view of each z-stack per section, per transmission electron microscopy (TEM; Fig. 1b, c). The rat were quantified with total number of microglia per mean length of sonicated mouse α-syn PFFs was estimated rat calculated (100–250). Data are expressed as mean to be 51.22 ± 1.31 nm, well within the optimal fibril length Iba-1ir soma area per treatment group. Soma measure- previously reported to result in seeding of endogenous ments for all microglia per treatment were also grouped phosphorylated α-syn inclusions (Fig. 1d,schematic)in into 10 μm bins and expressed as a percentage of total vitro and in vivo [42]. microglia counted. Unilateral intrastriatal injection of α-syn PFFs induces widespread Lewy-like pathology Thioflavin-S staining In our previous work [37], we reported that unilateral 1:12 series was washed in TBS and subsequently intrastriatal injection of mouse α-syn PFFs results in phos- mounted on subbed slides to dry (~ 1 h). Slides were in- phorylated α-syn (pSyn) intraneuronal accumulations in cubated in 0.5% KMnO in TBS for 25 min, followed by several areas that innervate the striatum [45], most prom- five washes in TBS. Sections were destained in 0.2% inently the frontal (primary motor and somatosensory, K S O /0.2% oxalic acid in TBS for 3 min followed by 2 2 5 layer 5) and insular cortices, amygdala, and SNc. Over incubation in 0.0125% thioflavin-S in 40% EtOH/TBS for time, accumulations increase in number in these regions. 3 min and differentiated in 50% EtOH for 15 min. In the present study, we observed an identical pattern of Sections were rinsed first in TBS and then ddH 0be- pSyn accumulation in rats injected with mouse α-syn fore coverslipping with VECTASHIELD Mounting PFFs. Specifically, we observe abundant pSyn pathology Medium for fluorescence. bilaterally in cortical regions (layers 2/3 of the secondary motor area, insular cortex, and orbital areas; Fig. 2a). In contrast, we observed unilateral pSyn accumulation in the Proteinase-K digestion SNc ipsilateral to the injected striatum and complete ab- 1:12 nigral series was washed in TBS. A subset of free sence of pSyn aggregates in animals injected with an equal floating tissue sections was treated with 10 μg/ml pro- volume of PBS or equal volume and concentration of RSA teinase K (Invitrogen, Carlsbad, CA; 25530015) for (Additional file 1: Figure S1). 30 min at room temperature, followed by three washes Accumulation of pSyn inclusions followed a distinct in TBS and four washes in TBS-Tx. Sections were then temporal pattern depending on the region examined. At processed for pan α-syn immunohistochemistry (rabbit 2 months postinjection (p.i.), we observed abundant anti-α-syn, Abcam, Cambridge, UK; AB15530) as de- soma and neuritic pSyn inclusions bilaterally in the agra- scribed above, mounted on subbed slides, dehydrated to nular insular cortex that persisted over the course of xylenes, and coverslipped. 6 months (Additional file 2: Figure S2A). Abundant pSyn accumulations were observed within the ipsilateral SNc Statistics at 2 months p.i., which remained ipsilateral and decreased Statistical analyses were performed using IBM SPSS in number over the course of 6 months (Additional file 2: Statistics (IBM, Armonk, NY) or GraphPad Prism (La Figure S2D–F). The abundance of pSyn inclusions in the Jolla, CA). Statistical significance for all cases was set striatum followed an opposite pattern (Additional file 2: at p < 0.05. Statistical outliers were assessed using Figure S2G–I). We observed relatively sparse pSyn inclu- the Absolute Deviation from the Median (ADAM) sions in the striatum at 2 months p.i. that were primarily method using the “very conservative” criterion [44]. restricted to neurites. At 4 and 6 months p.i., the number To compare numbers of O2 vs. F2 immunoreactive of pSyn accumulations in striatal somata increased in cells (Fig. 4), THir and NeuNir neurons (Fig. 5), abundance and also were observed in the contralateral pSyn α-syn inclusions (Fig. 6), MHC-IIir microglia striatal hemisphere (Additional file 2: Figure S2H–I). (Fig. 5), and Iba-1ir microglia number and size (Fig. 7), a one-way ANOVA with Tukey’s post hoc analyses was α-Syn inclusions in the SNc exhibit oligomeric, fibrillary used. Correlation analysis was conducted to investigate conformations, and Lewy body-like characteristics the relationship between ipsilateral and contralateral THir The oligomeric form of α-syn is proposed to be one of neurons (Fig. 5) and between MHC-IIir and pSyn α-syn the toxic species [43, 46–48]. We further characterized inclusions (Fig. 6). the nature of pSyn inclusions within the SNc at 1 month Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 7 of 18 Fig. 2 α-Syn inclusions in the SNc exhibit oligomeric and fibrillary conformations and Lewy body-like characteristics. a–c Representative images of Lewy-body-like intraneuronal pSyn inclusions in the substantia nigra pars compacta (SNc) at 1 month p.i. show pathology is localized to the ipsilateral SNc. d–f Adjacent SN tissue sections stained for O2 (oligomeric/fibrillar α-syn conformation specific) and F2 (g–i) (fibrillar-specific conformation) α-syn reveal that many intraneuronal inclusions possess mature, fibrillar inclusions of α-syn. j Percent of pSyn inclusions with either oligomeric/fibrillar (O2) or predominantly fibrillar (F2) immunoreactivity and estimated proportion of pSyn inclusions that are oligomeric only. Data represent mean ± SEM. Scale bars (b, e, h)= 50 μm, (c, f, i)= 10 μm. k Endogenous α-syn immunoreactivity in the SNc and SNr. l Adjacent tissue sections exposed to proteinase-K reveal the absence of soluble α-syn in the SNr and the presence of insoluble, neuronal inclusions of α-syn in the SNc. m Thioflavin-S fluorescence of amyloid structure present in SNc neurons. Scale bars (k, l, m)= 50 μm, (insets) = 25 μm. Abbreviations: α-syn = alpha-synuclein; p.i. = postinjection; pSyn = α-syn phosphorylated at serine 129; O2 = oligomeric/fibrillar α-syn antibody; F2 = fibrillar α-syn only antibody; SNc = substantia nigra pars compacta; SNr = substantia nigra pars reticulata; SEM = standard error of the mean p.i. using conformation-specific antibodies for oligo- only) with an estimated 31.7 ± 5.9% of inclusions sug- meric/fibrillar α-syn (O2) or fibrillar-predominant α-syn gested to be in an oligomeric conformation (O2 only (F2) and compared that with immunoreactivity to pSyn minus F2 only; Fig. 2j). α-Syn inclusions in the SNc (Fig. 2a–i [43]). When adjacent sections were quantified 2 months p.i. displayed Lewy body-like characteristics using unbiased stereology, we observed that 88.2 ± 6.4% [49, 50], including resistance to proteinase-K digestion of pSyn immunoreactive inclusions exhibited either an (Fig. 2i, k, l) as well as markers for ß-sheet structure as oligomeric or fibrillary conformation (O2 only). Further- detected by thioflavin-S (Fig. 2m). Collectively, these re- more, 56.4 ± 6.04% of nigral pSyn inclusions was de- sults suggest that intrastriatal injection of mouse α-syn tected as predominately mature, fibrillar aggregates (F2 PFFs triggers pathological conversion of endogenous Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 8 of 18 α-syn to phosphorylated, oligomeric, and fibrillary In a control experiment, we examined whether intras- conformations in the SNc that ultimately result in triatal injection of an exogenous protein taken up by insoluble, amyloid inclusions resembling Lewy bodies. neurons [51, 52], rat serum albumin (RSA), induced an inflammatory response in the absence of intracellular PFF-induced synucleinopathy induces significant bilateral pSyn accumulation. To rule out acute toxicity induced loss of SNc neurons by RSA, THir SNc neurons were quantified at 2 months We previously observed that unilateral intrastriatal mouse after injection. No significant differences in THir SNc α-syn PFF injections to rats resulted in bilateral nigrostria- neurons were observed due to PBS, RSA, or PFF injec- tal degeneration of THir SNc neurons within 6 months tion in either the ipsilateral or contralateral hemisphere [37]. To validate this finding in our present cohort, we con- (F = 0.3731, p > 0.05, Fig. 3j). These results (5, 26) ducted unbiased stereology of THir SNc neurons at 2, 4, 5, demonstrate that RSA injection, used as an additional and 6 months p.i. in α-syn PFF- and PBS-injected rats. In- control treatment, did not compromise the survival of jection of PBS did not result in significant loss of THir SNc THir SNc neurons. neurons at any time point (F = 1.991, p > 0.05); thus, (7, 18) PBS-injected time points were combined for comparison Phosphorylated α-syn inclusions peak in the SNc at to PFF-injected rats between identical hemispheres (ipsilat- 2 months and significantly decrease in number during eral PBS = 12,518 ± 554; contralateral PBS 11,577 ± 536). the 5–6-month interval of SNc degeneration Similar to our previous studies, we observed significant, bi- We determined the time course (1–6 months p.i.) of lateral reduction (~ 35%) in SNc THir neurons (Fig. 3a–e). phosphorylated α-syn (pSyn) accumulation in the SNc Specifically, the number of SNc THir neurons ipsilateral to following intrastriatal α-syn PFF injection at monthly in- α-syn PFF injection at both 5 (8227 ± 1015) and 6 months tervals. pSyn inclusions were observed in the SNc ipsilat- (8851 ± 1148) p.i. was significantly reduced compared to eral to injection in all α-syn PFF-injected rats, with the PBS control rats (F = 4.297, p < 0.027, Fig. 5e). Within (5, 36) number of inclusions varying based on time point after the contralateral SNc, significantly fewer THir neurons injection (Fig. 4a–d). Inclusions were most abundant at were observed 5 months following α-syn PFF injec- months 1, 2, and 3, with all three time points exhibiting tion (F =5.782, p < 0.013) with a non-significant re- (5, 36) significantly higher α-syn inclusions compared to the duction in the contralateral SNc observed at 6 months interval of SNc degeneration at months 4, 5, and 6 (p > 0.05). A positive correlation existed between the (Figs. 3e, g and 4d; F =2.251, p ≤ 0.001). The num- (5, 18) extent of ipsilateral THir SNc neuron loss and the extent ber of intraneuronal α-syn inclusions in the SNc was sig- of contralateral loss of THir SNc neurons (r = 0.8855, nificantly greater at 2 months p.i. compared to all other p = 0.0007, R = 0.7842, Fig. 3f). time points except the 1-month time point (Fig. 4d; Lastly, to confirm whether reductions in THir neurons p ≤ 0.006). At 2 months, approximately 2220 ± 148.6 SNc induced by PFF injection represented phenotype loss or neurons possessed pSyn inclusions. By comparison, a loss overt degeneration, unbiased stereology of NeuN-ir neu- of ≈ 3804 THir SNc neurons ipsilateral to PFF injection rons in the SNc was conducted in PFF- or PBS-treated was observed at 5–6 months. These results suggest that groups at 5 and 6 months p.i. No significant differences pSyn inclusion formation in the SNc between 1 and were observed within the corresponding hemisphere be- 3 months after PFF injection precedes degeneration of the tween 5 and 6 months due to either PBS or PFF injec- SNc neurons at 5–6months p.i. tion (PBS: F = 1.238, p > 0.05; PFF: F = 0.3986, (3, 4) (3, 8) p > 0.05). Therefore, the 5- and 6-month time points were combined into one time point. The number of SNc MHC-II immunoreactive (MHC-IIir) microglia increase in NeuN-ir neurons ipsilateral to PFF injection was signifi- the SNc in association with accumulation of α-syn cantly reduced compared to either the ipsilateral or inclusion but are decreased during the interval of contralateral hemisphere of PBS-injected rats (F = degeneration (3, 16) 7.089, p < 0.02). The number of NeuN-ir neurons in the MHC-II expression on microglia is associated with contralateral SNpc of PFF injected rats was significantly co-expression of pro-inflammatory genes such as reduced compared to the ipsilateral SNc of PBS injected TNF, IL-1β, and CD80 as well as proinflammatory cyto- rats (p < 0.0319). When compared to the contralateral kine secretion [13–15]. We quantified MHC-II immuno- SNc of PBS-injected rats, NeuN-ir neurons were reduced reactive (MHC-IIir) microglia within an adjacent series of yet did not reach significance (p = 0.0563, Fig. 3g–i). Over- SNc tissue sections at months 1, 2, 3, 4, 5, and 6 after uni- all, our results replicate our previous findings that intras- lateral α-syn PFF or PBS intrastriatal injection in order to triatal α-syn PFF injection results in significant bilateral examine neuroinflammation. Double labeling for MHC-II reductions in THir and NeuN-ir SNc neurons over the proteinand Iba-1mRNA confirmed the identity of course of 6 months [37]. MHC-IIir cells to be microglia (Fig. 4e). Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 9 of 18 Fig. 3 α-syn PFF-seeded synucleinopathy induces protracted, significant bilateral loss of SNc dopamine neurons. a–d Unilateral intrastriatal α-syn PFF injection induces visible loss of THir neurons (brown) in the SN at 6 months compared to an age-matched, PBS-injected control. Scale bars (a, c)=25 μm. e Stereological assessment of THir neuron loss at 2, 4, 5, and 6 months following α-syn PFF or saline injection. Significant ipsilateral reduction in THir neurons was observed at 5 and 6 months postinjection compared to saline-injected controls with significant contralateral loss at 5 months *p < 0.027 compared to respective PBS hemisphere, **p < 0.013 compared to respective PBS hemisphere. f Correlation between extent of ipsilateral THir neuron loss and contralateral loss as compared to PBS control (r = 0.8855, p = 0.0007, R = 0.7842). g Stereological assessment of NeuNir neurons reveals overt degeneration distinct from loss of TH phenotype, *p < 0.03 compared to ipsilateral PBS. h Representative IHC of NeuNir neurons (brown, arrows) in the SNc in PBS (left) and PFF injected (right) animals 6 months p.i. i Stereological assessment of THir neurons at 2 months p.i. reveals no significant acute toxicity from injection of rat serum albumin (RSA). Data represent mean ± SEM. Abbreviations: PFF = pre-formed alpha-synuclein fibrils; p.i. = postinjection; ipsilateral = ipsilateral hemisphere relative to injection; contralateral = contralateral hemisphere relative to injection; SNc = substantia nigra pars compacta; SNr = substantia nigra pars reticulata; Neu-Nir = neuronal nuclei immunoreactive; SEM = standard error of the mean; THir = tyrosine hydroxylase immunoreactive; PFFs = pre-formed alpha-synuclein fibrils; PBS = phosphate-buffered saline; RSA = rat serum albumin Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 10 of 18 Fig. 4 Antigen-presenting MHC-II immunoreactive (MHC-IIir) microglia increase in the SNc in association with peak accumulation of α-syn inclusions but are limited during the interval of degeneration. a–c Representative images of pSyn inclusions in the SNc at 2, 4, and 5 months p.i.; scale bar (a–c)=10 μm. d Stereological assessment of pS129 containing neurons in the substantia nigra in PFF animals at 1, 2, 3, 4, 5, and 6 months p.i.; *p ≤ 0.001 compared to 4, 5, and 6 months. p ≤ 0.006 compared to 3, 4, 5, and 6 months. pSyn inclusions decrease over time in association with neuronal loss. e Major-histocompatibility complex-II (MHC-II; blue) protein colocalizes with ionized calcium-binding adaptor molecule 1 mRNA (brown) within microglia. f, g Representative images of MHC-II antigen-presenting microglia in the SN at 2 months in PBS- and PFF-injected rats and 6 months post-PFF injection (h); scale bar (f–h)= 50 μm, insets = 10 μm. i Stereological assessment of MHC-IIir microglia in the SN reveals MHC-IIir microglia are significantly higher in PFF vs. PBS animals at 2, 4, and 5 months *p < 0.006. More MHC-IIir microglia are evident in 2-month PFF animals vs. all other PFF time points p < 0.02. Notably, MHC-IIir microglia peak at the same time pSyn aggregation peaks (d, i). j Number of MHC-II immunoreactive microglia correlated with number of SNc neurons with intraneuronal pSyn inclusions (r = 0.8858, p =0.0015, R = 0.7846). k In a follow-up study, intrastriatal injection of rat serum albumin (RSA) does not impact numbers of MHC-IIir microglia compared to PBS p > 0.05. Injection of PFFs in this second cohort confirmed previous observations of a significant increase in MHC-IIir microglia compared to PBS or RSA at 2 months p.i. *** p ≤ 0.0006. Abbreviations: p.i. = postinjection; PFFs = pre-formed alpha-synuclein fibrils; PBS = phosphate-buffered saline; RSA = rat serum albumin; MHC-II = major-histocompatibility complex-II; Iba-1 = ionized calcium binding adaptor molecule 1; ir = immunoreactive; SNc = substantia nigra pars compacta; SEM = standard error of the mean MHC-IIir microglia were observed in the SNc ipsilat- MHC-IIir microglia varied over time and followed a eral to injection in both α-syn PFF and PBS control rats nearly identical pattern to that observed with pSyn at all time points. No MHC-IIir microglia were observed inclusion accumulation. At the one-month time point, in the contralateral SNc. However, the magnitude of no significant differences were observed between the Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 11 of 18 number of MHC-IIir microglia in the ipsilateral SNc time point (F =0.256, p = 0.855, Fig. 5b). At the (3, 16) of PBS controls compared to α-syn PFF-injected rats 2-month time point coinciding with the peak of pSyn (F = 17.45, p > 0.05), presumably reflecting a non- α-syn inclusion accumulation in the SNc, we observed an (11, 45) specific response to injection (Fig. 4i). However, signifi- appreciable increase in the soma size and thickness and cantly higher numbers of MHC-IIir microglia were ob- number of microglial processes in the SNr of PFF-injected served in the ipsilateral SN of α-syn PFF-injected rats rats compared to a more classically quiescent microglial compared to PBS-injected control rats at months 2, 4, and morphology observed in control injected rats. Specifically, 5(p <0.006, Fig. 4f, g, i). The peak of MHC-IIir microglia in the ipsilateral SNr of rats 2 months following α-syn PFF occurred in the SN 2 months following α-syn PFF injec- injection, the average microglia cell body area was signifi- tion (p < 0.02 compared to PFF-injected rats all other time cantly larger compared to PBS-injected rats (F =4.016, (3, 16) points), corresponding to the time point when the greatest p =0.02, Fig. 5c–e). Microglia soma area varied in all con- number of SNc neurons possesses α-syn aggregates ditions between ≈ 10–200 μm , with a significantly greater (Fig. 4d, g, i). In contrast, significantly fewer MHC-IIir percentage of microglia > 70 μm observed in the SNr microglia were observed in PFF-injected rats at months 5 of rats possessing SNc pSyn α-syn inclusions at and 6, corresponding to the interval of SNc THir neuron 2 months compared to rats injected with PBS (Fig. 5f–h, loss, although these numbers were still significantly higher F =4.613, p =0.03). (2, 12) than PBS-injected controls (p <0.006, Figs. 3e and 4i). At 6 months p.i., a time point corresponding to the There was a positive correlation between the number of time point of very few pSyn α-syn inclusions and imme- MHC-IIir microglia and the number of SNc neurons diately following loss of SNc neurons, no significant possessing pS129 α-syn inclusions in the SN (r = 0.8858, differences in microglia soma size were observed be- p = 0.0015, R = 0.7846, Fig. 4j,months2, 4,and 6). tween α-syn PFF- and PBS-injected rats (F = 2.089, (3, 10) To confirm these findings, we repeated injections in a p > 0.05, Fig. 5i–m). Of note, the average microglial soma separate cohort of animals with rat serum albumin area in PBS-injected rats at 6 months (rats 8 months of (RSA) as an additional control group for neuronal up- age) was significantly larger than PBS-injected rats at take of exogenous protein in the absence of pSyn accu- 2 months (4 months of age) suggesting an age-related mulation, as PBS injection only controls for needle increase (F = 37.00, p < 0.001). The distribution of (3, 12) insertion into the parenchyma. As in previous cohorts, microglia soma areas between PFF and PBS rats α-syn PFF injection resulted in a significant increase in 6 months following injection also appeared similar MHC-IIir microglia in the SN at 2 months p.i. (Fig. 4k, (Fig. 5l, m) with an apparent age-related effect [53–55] p ≤ 0.0006). Injection of RSA resulted in similar numbers reflected in a greater percentage of microglia > 70 μm in of MHC-IIir microglia as observed in PBS-injected con- 8-month-old rats compared to 4-month-old rats. trol rats. No acute neurotoxicity was observed in RSA- Overall, our finding that the peak time point of injected animals at 2 months p.i. (Fig. 3i). Collectively, SNc pSyn α-syn inclusions is associated with a signifi- these results reveal that the preponderance of MHC-II ex- cant increase in microglia soma size suggests that pression in SN microglia is associated with pSyn α-syn in- synucleinopathy in the SNc triggers early disturbances clusions at early time points, however is significantly in local microglia. The interval in which we observe attenuated during the interval of THir SNc degeneration. this synucleinopathy-induced reactive microglial morph- ology is 3 months prior to loss of SNc neurons (Fig. 3e, g) pSyn inclusions in the SNc are associated with a reactive suggesting that reactive microglia have the potential microglial morphology in the adjacent SNr to contribute to vulnerability of SNc neurons to The number and distribution of MHC-IIir microglia in degeneration. the SN suggested that not all microglia were expressing We also examined a series of sections throughout the MHC-II. We next used Iba-1 immunoreactivity to exam- SN and striatum at 2, 4, and 6 months p.i. in α-syn PFF- ine the entire microglia population within an adjacent and PBS-injected rats for the presence of cluster of series of SN tissue sections at 2 and 6 months after α- differentiation 68 (CD68) which labels both phagocytic syn PFF, RSA, or PBS intrastriatal injection (Fig. 5). microglia and infiltrating macrophages. While a few Quantitation of the number of Iba-1 immunoreactive CD68-ir cells were observed in blood vessels, no CD68-ir (Iba-1ir) microglia in the adjacent SNr revealed no sig- cells were observed in the parenchyma during any of the nificant differences in microglial number due to α-syn time points examined (data not shown). The lack of CD68 PFF, RSA, or PBS injection at either 2 months (Fig. 5a) immunoreactivity in the parenchyma of the SN or stri- or 6 months p.i. (2 months: F = 0.2637, p > 0.05; atum at any time point suggests that the magnitude of (3, 16) 6 months: F = 0.2427, p > 0.05). No significant dif- synucleinopathy and subsequent degeneration produced (3, 10) ferences were observed in microglial soma area in the in the α-syn PFF model does not trigger microglial SNr due to intrastriatal RSA injections at the 2-month phagocytic activity. Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 12 of 18 Fig. 5 SNr microglia exhibiting reactive morphology are associated with pSyn inclusion-bearing neurons in the SNc. a Total number of microglia did not differ between PBS-, RSA-, and PFF-injected animals, p > 0.05. b Microglia soma area in the ipsilateral and contralateral SNr did not differ significantly between animals receiving intrastriatal injections of PBS or RSA, p > 0.05. c Microglia soma area was significantly increased at 2 months in the ipsilateral SNr, when peak numbers of pSyn aggregates are present in nearby SNc neurons as compared to PBS-injected animals, *p = 0.02. d, e Representative images of SN sections dual labeled for Iba-1 immunoreactive microglia (purple) and NeuN-ir neurons (brown) 2 months following either intrastriatal PBS or α-syn PFF-injected rats. α-Syn PFF-injected rats exhibit larger cell bodies and increased number and thickness of processes. f, inset Distribution of microglia soma area measurements 2 months following intrastriatal PBS or RSA injection illustrated as a percent of total microglia quantified. g Distribution of microglia soma area measurements 2 months following α-syn PFF injection illustrated as a percent of total microglia quantified. h Percent of total microglia quantified for soma area analysis with cell body areas > 70 μm at 2 months. Microglia in PFF-injected rats possessed significantly more microglia with cell bodies larger than > 70 μm compared to PBS-injected rats, *p = 0.03. i At 6 months p.i. microglia soma area in the ipsilateral and contralateral SNr did not differ significantly between rats receiving either PBS or α-syn PFF intrastriatal injections, during the interval of ongoing degeneration in the SNc of PFF-injected animals, p > 0.05. j, k Representative images of SN sections dual labeled for Iba-1 immunoreactive microglia (purple) and NeuN-ir neurons (brown) at 6 months p.i. exhibit a hyper ramified morphology, regardless of treatment. l, m Distribution of microglia soma area measurements 6 months p.i in PBS- and PFF-injected rats as a percent of total microglia quantified. Scale bars (d, e, j, k)=25 μm. Data represent mean ± SEM. Abbreviations: p.i. = postinjection; PFFs = pre-formed alpha-synuclein fibrils; PBS = phosphate-buffered saline; RSA = rat serum albumin; Iba-1 = ionized calcium-binding adaptor molecule 1; ir = immunoreactive; NeuN-ir = neuronal nuclei immunoreactive; ipsilateral = ipsilateral hemisphere relative to injection; contralateral = contralateral hemisphere relative to injection; SNc = substantia nigra pars compacta; SNr = substantia nigra pars reticulata; SEM = standard error ofthe mean Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 13 of 18 MHC-IIir microglia in the agranular insular cortex are observed between treatment groups at the 1-month time associated with the accumulation of α-syn inclusions point, but the magnitude of the MHC-II response ap- We also examined the time course of pSyn inclusion ac- peared slightly larger in PFF-injected rats compared to the cumulation and MHC-II expression on microglia in the PBS-injected rats at the 2-month time point. However, at agranular insular cortex, as this region possesses abun- 4 and 6 months, during the interval of continuing accu- dant Lewy-body like pathology in our model and is mulation of pSyn α-syn inclusions in the striatum, the implicated in non-motor symptoms in PD [56]. At number of MHC-IIir microglia decreased dramatically 2 months p.i., we observe abundant pSyn inclusions pri- with no differences observed in the small number of marily localized to the somata (Fig. 6a), with increased MHC-IIir microglia observed in both treatment groups neuritic pathology evident by 4 months (Fig. 6b). Inter- (Additional file 1: Figure S1E, F, H, I). These results sug- estingly, an observable decrease in both neuritic and gest that the acute microglial response to the PFF injectate somata inclusions is evident by 6 months (Fig. 6c). We may differ from the acute response to the surgical injec- observe a similar temporal pattern of MHC-IIir micro- tion alone but that the subsequent increase α-syn inclu- glia as described for the SN: the highest number of sion load within neurons in the striatum does not trigger MHC-IIir microglia is observed at 2 months p.i., when a second wave of microglial MHC-II immunoreactivity. pSyn inclusions first peak, and a decrease in MHC-IIir microglia in association with reductions in the number Discussion of pSyn inclusions. Interestingly, a decrease in MHC-IIir In the present study, intrastriatal injection of α-syn PFFs microglia is observed between 2 and 4 months p.i. when to rats resulted in widespread accumulation of phos- pSyn pathology becomes more abundant with the phorylated α-syn inclusions in several areas that innerv- appearance of neuritic inclusions. These observations ate the striatum, as previously reported in rats and mice suggest that MHC-II is upregulated as a first response to [37, 57]. Further examination of the inclusions formed formation of pSyn inclusions and is not sustained over in the SNc revealed that they share many key features time, despite a secondary increase in synuclein burden. with Lewy bodies and were most abundant between Few to no MHC-IIir microglia were observed in PBS- months 1–3 after intrastriatal α-syn PFF injection, peak- injected animals (Fig. 6g–l), strengthening the concept ing at 2 months. The magnitude of ipsilateral SNc neu- that MHC-IIir on microglia is induced by initial accu- rons bearing α-syn inclusions 1–3 months after α-syn PFF mulation of pSyn. injection approximated the magnitude of loss of ipsilateral SNc neurons observed at 5–6 months, suggesting a direct MHC-IIir microglia in the striatum are not associated with relationship between α-syn inclusion accumulation and the accumulation of α-syn inclusions degeneration of SNc neurons. Synucleinopathy-specific Lastly, we examined the time course of accumulation of MHC-II expression in the ipsilateral SNc similarly peaked pSyn inclusions and number of MHC-IIir microglia in in the SN at 2 months and was associated with a reactive the striatum in rats that received unilateral α-syn PFF or microglial morphology, characterized by significantly lar- PBS intrastriatal injection. As reported previously [37], ger soma size, 3 months prior to degeneration. Surpris- the pattern of pSyn α-syn inclusion accumulation in the ingly, although the period of nigral degeneration was striatum is strikingly different from accumulation in the associated with an increased MHC-II signal relative to SNc. At 2 months, α-syn inclusions in cell bodies in the controls, MHC-II immunoreactivity during the period of ipsilateral striatum are relatively sparse and pSyn α-syn degeneration was significantly decreased relative to immunoreactivity appeared primarily localized to neurites, MHC-II immunoreactivity during the earlier peak of synu- presumably in terminals from the SNc (Additional file 1: cleinopathy. Overall, the temporal pattern of peak Lewy Figure S1A). Over the course of the 4 months, pSyn α-syn body-like inclusion formation was associated with peak inclusions in cell bodies in the striatum increase in num- neuroinflammation in the SN, both of which appear ber, involving the contralateral hemisphere as well, with months prior to loss of SNc neurons. These results suggest the greatest number of pSyn α-syn inclusions observed in that an increase in MHC-II may be a first-response mech- the ipsilateral striatum at the 6-month time point anism to initial accumulation of intracellular α-syn and (Additional file 1: Figure S1C). In adjacent striatal tissue that reactive microglia have the potential to contribute to sections, we examined the temporal pattern of MHC-IIir vulnerability of SNc neurons to degeneration (Fig. 7, left). microglia following unilateral α-syn PFF or PBS intrastria- Within the agranular insular cortex, we observe pSyn tal injection. Early after injection, at 2 weeks, 1 month, primarily localized to the somata at 2 months, followed and 2 months, abundant MHC-IIir microglia were by an increase in neuritic pathology by 4 months, and a observed in the ipsilateral striatum in proximity to the significant decrease overall in pSyn immunoreactivity at injection sites in both PFF- and PBS-injected rats 6 months (Fig. 7, right). This suggests that similar to the (Additional file 1: Figure S1D, G). No differences were SN, neurons in the agranular insular cortex harboring Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 14 of 18 Fig. 6 Microglia expressing MHC-II are associated with pSyn inclusions in the agranular insular cortex. a–c Representative images of pSyn accumulation in the agranular insular cortex at 2, 4, and 6 months p.i. Pathology is primarily localized to the soma at 2 months (a), with increased neuritic pathology by 4months(b) and an observable decrease overall by 6 months, possibly due to death of neurons. c No pSyn inclusions were observed in PBS-injected animals (g–i). d–f A similar pattern was observed in MHC-II expression on microglia. Peak numbers of MHC-IIir microglia were observed at 2 months p.i. (d), decreased at 4 months (e), and virtually absent by 6 months (f). j–l Few to no MHC-IIir microglia were observed in PBS injected animals at any time point, suggesting that MHC-II expression occurs in response to accumulation of pSyn inside neurons. Abbreviations: p.i. = postinjection; pSyn = α-syn phosphorylated at serine 129; PFFs = pre-formed alpha-synuclein fibrils; PBS = phosphate-buffered saline; MHC-II = major-histocompatibility complex-II; ir = immunoreactive inclusions die over the course of 6 months. Moreover, a striatum (Additional file 3: Figure S3). The procedure of similar pattern of MHC-II expression on microglia is ob- intrastriatal injection itself triggered an acute increase in served: numbers of MHC-IIir microglia peak when pSyn MHC-II immunoreactive microglia that appeared to be pathology is most abundant and decrease over time. Al- slightly enhanced by the presence of the PFF injectate; though this has not been systematically quantified in the however, the presence of MHC-II decreased dramatically present study, future studies investigating inflammation over time in all conditions despite an ever increasing α- and degeneration in this area following intrastriatal α- syn inclusion load. Compared to the SNc, the accumula- syn PFF injection are warranted, as the insula has been tion of α-syn inclusions in the striatum is delayed with no implicated in the manifestations of non-motor symp- loss of striatal neurons observed at 6 months [37]and it is toms in human PD [56, 58, 59]. unknown whether degeneration of striatal neurons ultim- Within the striatum, the site of α-syn PFF injection, we ately occurs at later time points. It is unclear whether α- observed a distinctly different pattern of accumulation of syn inclusion load increases and peaks past the 6-month α-syn inclusions compared to the SNc (Additional file 3: time point and whether the presence of MHC-II immuno- Figure S3J). We observed a dissociation between α-syn reactive microglia may have similarly tracked with a future inclusion load and MHC-II immunoreactivity in the peak. The results in the striatum illustrate the necessity of Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 15 of 18 Fig. 7 Regional timelines of synucleinopathy, neuroinflammation, and degeneration in the substantia nigra and agranular insular cortex following intrastriatal α-syn PFF injection. (Left): Early accumulation of phosphorylated inclusions of α-syn (peak at 2 months) in the substantia nigra leads to loss of TH phenotype and eventual loss of nigrostriatal dopamine neurons by 5–6 months p.i. In the SN, the pattern of microgliosis similarly follows that of pSyn: microglia in the adjacent SNr exhibit a reactive morphology at 2 months p.i. when nearby SNc neurons possess the greatest number of SNc pSyn inclusions. Interestingly, MHC-IIir antigen-presenting microglia in the SNc also peak at 2 months p.i., again coinciding with the greatest number of pSyn intraneuronal inclusions and decrease over time to near non-detectable levels during the interval of degeneration, suggesting a relationship between pathological α-syn and inflammation. (Right): Early accumulation of pSyn inclusions occurs between 2 and 4 months, with inclusions primarily localized to the somata at 2 months, an increase in neuritic inclusions at 4 months, and an observable decrease of overall pSyn pathology at 6 months, suggesting that similar to the SN, neurons in the agranular insular cortex harboring inclusions eventually die off. MHC-II immunoreactivity follows a similar pattern to that observed in the SN, with peak expression observed at 2 months p.i., and decreased over the course of 6 months determining the role of an acute inflammatory response to secretion of proinflammatory and anti-inflammatory cyto- local injection and that the magnitude of this response may kines in the SN at time points prior to and following nigral prevent the reactive microgliosis associated with patho- degeneration is warranted. logical α-syn inclusions. Further, these striatal results suggest Human PD studies examining neuroinflammation sug- that α-syn inclusions do not automatically trigger reactive gest involvement of both the local brain immune response microgliosis and that other factors including rate of inclu- and the adaptive immune system in PD [73–75]. In the sion formation, impending cytotoxicity, local environment, present study, we did not examine the possibility of per- or microglia and astrocyte density [42] may be involved in ipheral immune cell infiltration beyond phagocytic CD68+ determining the neuroinflammatory cascade of events. macrophages, which were not detected in the parenchyma In response to persistent neuronal stress and protein in PFF-injected animals at any stage, and data from hu- accumulation such as α-syn aggregation, microglia can man tissue regarding the presence of CD68+ macrophages become chronically activated, proliferate, migrate, se- in human PD is limited [9]. However, microglia are con- crete pro-inflammatory cytokines and reactive oxygen sidered to be the principle antigen-presenting cell within species, and ultimately contribute to neuronal injury in the brain and MHC-II expression is associated with the an uncontrolled, feed-forward manner [20, 60–63]. recognition of CD4+ T-helper cells. It is possible that Microglia can also phagocytose both living and dead CD4+ T cells participate in the response to α-syn inclu- neurons [64–66]. The observation that α-syn inclusion sions in the SN; however, whether the net effect of CD4+ triggered MHC-II expression on microglia in the SN T cells is neurodegenerative or neuroprotective requires prior to degeneration indicates that neuroinflammation further systematic evaluation [76]. In addition, MHC-I ex- may contribute to the mechanism of pathophysiology. pression by SNc neurons was observed in association with However, it is unlikely that microglial activation is the α-syn overexpression [77] and may similarly be expressed sole arbiter of degeneration given that neuronal death by α-syn inclusion-bearing SNc neurons. In addition, it can result from α-syn PFF-induced intraneuronal inclu- was recently shown that T lymphocytes isolated from PD sions in cultures in the absence of microglia [40]. A more patients recognize specific α-syn peptides [78], strength- likely scenario involves neuroinflammation contributing ening the concept that neuroinflammation can be induced to or accelerating nigrostriatal degeneration with proper- by α-syn and potentially involved early in PD progression. ties unique to the nigral environment [67–72]adding to While the concept that MHC-II is involved in PD and the cascade of events. Future studies investigating the correlates with α-syn burden [9] has existed for several Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 16 of 18 decades, our results are the first to systematically evaluate Additional file 3: Figure S3. Antigen-presenting MHC-IIir microglia are the time course of endogenous pSyn accumulation and not associated with peak of intraneuronal inclusions of pSyn in the striatum. Progression of pSyn pathology and MHC-IIir microglia in the microglial MHC-II expression prior to and after nigros- striatum. (a) At 2 months p.i., pSyn inclusions are localized to neurites, triatal degeneration has occurred. Our results indicate that presumably representing terminals from the SNc. (b–c) Over time pSyn MHC-II expression in the SNc is increased at early time inclusions become primarily localized to the soma of striatal neurons. (d) Abundant MHC-IIir microglia in the striatum primarily localized around points in which SNc neurons possess pSyn inclusions, and the α-syn PFF injection site at 2 months. (e–f) MHC-IIir microglia in the is relatively sparse during the interval of nigral degen- striatum are largely absent during continuing accumulation of intraneuronal eration, suggesting that MHC-II is a response to pSyn inclusions at 4 months (e) and 6 months (f) p.i. (g) Intrastriatal injection of PBS results abundant MHC-IIir microglia in the striatum localized near the initial inclusion formation. Future studies investigating site of injection at 2 months p.i., although appearing less abundant than the direct impact of increased or decreased MHC-II MHC-IIir microglia in the striatum of α-syn PFF rats at the same time point expression on the magnitude of degeneration and (d). (h) MHC-IIir microglia are similarly absent from the parenchyma by 4 months (h) and 6 months p.i (i). Scale bars A–I= 50 μm. Abbreviations: what role, if any, peripheral T cells play in disease p.i. = postinjection; PFFs = pre-formed alpha-synuclein fibrils; progression are warranted. PBS = phosphate-buffered saline; pSyn = α-syn phosphorylated at serine 129, Our results and the well-characterized rat α-syn PFF MHC-IIir = major-histocompatibility complex-II immunoreactive. (TIF 112368 kb) model will facilitate future studies to provide key mech- anistic insights into the specific relationship between Abbreviations 6-OHDA: 6-Hydroydopamine; AAV: Adenoassociated virus; CD68: Cluster of pathological α-syn inclusions, neuroinflammation, and differentiation molecule 68; DMS: Dorsomedial striatum; Iba-1: Ionized degeneration in sporadic PD. calcium-binding adaptor molecule-1; IHC: Immunohistochemistry; MHC-II: Major histocompatibility complex II; MPTP: 1-Methyl-4-phenyl-1,2,3, 6-tetrahydropyridine; NeuN: Neuronal nuclei; PBS: Phosphate-buffered saline; Conclusions PD: Parkinson’s disease; PFFs: Alpha-synuclein pre-formed fibrils; pSyn: Alpha-synuclein phosphorylated at serine 129; RSA: Rat serum albumin; Accumulation of intraneuronal inclusions of phosphory- SNc: Substantia nigra pars compacta; SNr: Substantia nigra pars reticulata; lated α-syn induces increased MHC-II expression and re- TH: Tyrosine hydroxylase; VLS: Ventrolateral striatum; α-syn: Alpha-synuclein active microgliosis in the substantia nigra months prior to Funding dopaminergic cell death, suggesting that microglia may be A support was provided by the Department of Translational Science and a contributor to rather than only a consequence of nigral Molecular Medicine, the Neuroscience Graduate Program, National Institute degeneration. These results will facilitate future studies to of Neurological Disorders and Stroke (NS099416), and the Edwin A. Brophy Endowment at Michigan State University. provide key mechanistic insights into the specific relation- ship between pathological α-syn inclusions, neuroinflam- Authors’ contributions mation, and degeneration in sporadic PD. This research project was conceived and organized by MFD, TJC, MGT, KCL, NMK, and CES. It was executed by MFD, CES, TJC, JRP, CJK, KCL, KLP, NMK, DLF, NKP, OLB, JWH, NNV, NKM, and OMAE. The data were analyzed by MFD, CES, and TJC. The manuscript was first written and revised by MFD and CES, Additional files and it was reviewed and critiqued by all authors. All authors read and approved the final manuscript. Additional file 1: Figure S1. Unilateral intrastriatal injection of α-syn PFFs, but not RSA or PBS, induces bilateral cortical and unilateral SNc Ethics approval Lewy-body like inclusions of phosphorylated α-syn (pSyn). (a) pSyn All procedures performed in studies involving animals were in accordance pathology is observed bilaterally in cortical areas after unilateral injection with the ethical standards of the Institute for Animal Use and Care of α-syn PFFs, namely in layers 2/3 and orbital and agranular insular Committee (IACUC) at Michigan State University. cortices. (b) Injection of PBS or (c) RSA did not induce pSyn accumulation. (d) pSyn accumulation in the ipsilateral substantia nigra pars compacta Competing interests (SNc) at 2 months postinjection, with no evidence of pSyn inclusions in the The authors declare that they have no competing interests. contralateral SNc. Scale bars (A–D) = 50 μm. Abbreviations: α-syn = alpha- synuclein; PFFs = pre-formed alpha-synuclein fibrils; PBS = phosphate- Publisher’sNote buffered saline; RSA = rat serum albumin; pSyn = α-syn phosphorylated at Springer Nature remains neutral with regard to jurisdictional claims in serine 129. (TIF 117729 kb) published maps and institutional affiliations. Additional file 2: Figure S2. Unilateral intrastriatal injection of α-syn PFFs induces widespread accumulation of Lewy-body like inclusions of Author details phosphorylated α-syn (pSyn). Representative images illustrating the time Department of Translational Science and Molecular Medicine, Michigan course of pSyn pathology in regions innervating the striatum. (a–c) pSyn State University, 400 Monroe Avenue NW, Grand Rapids, MI 49503-2532, USA. pathology in the ipsilateral agranular insular cortex localized to both Neuroscience Graduate Training Program, Michigan State University, Grand the soma and neurites at 2 months p.i. (postinjection) that over time Rapids, MI, USA. MD/PhD Program, Michigan State University, Grand Rapids, becomes primarily localized to the soma; scale bar = 50 μm, inset = 10 μm. MI, USA. Mercy Health Hauenstein Neuroscience Medical Center, Grand (d–f) Ipsilateral accumulation of pSyn in the substantia nigra peaks at Rapids, MI, USA. Center for Neurodegenerative Disease Research, 2 months and becomes less abundant over time as neurons degenerate; Department of Pathology and Laboratory Medicine, University of scale bar = 200 μm, inset = 25 μm. (g–i) In contrast to other areas, pSyn in Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA. the striatum is primarily localized to neurites at 2 months and becomes Department of Physiology, Emory University School of Medicine, Atlanta, more abundant and localized to the soma over time, scale bar = 50 μm, GA, USA. Neurological Disorders Research Center, Qatar Biomedical Research inset = 10 μm. Abbreviations: α-syn = alpha-synuclein; PFFs = pre-formed Institute (QBRI), Hamad Bin Khalifa University (HBKU), Education City, Qatar. alpha-synuclein fibrils; pSyn = α-syn phosphorylated at serine 129; Life Sciences Division, College of Science and Engineering, Hamad Bin p.i. = postinjection. (TIF 33472 kb) Khalifa University (HBKU), Education City, Qatar. Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 17 of 18 Received: 15 March 2018 Accepted: 20 April 2018 phosphorylation of α‐synuclein and activation of caspase‐9: resemblance to pathogenetic changes in Parkinson's disease. J Neurosci. 2004; 91:451–61. 24. Ulusoy A, et al., Chapter 5 - Viral vector-mediated overexpression of α- synuclein as a progressive model of Parkinson’s disease. In: Björklund A and Cenci MA, editors. Progress in Brain Research. Elsevier; 2010. p. 89–111. References 25. Theodore S, et al., Targeted Overexpression of Human α-Synuclein Triggers 1. Collier TJ, Kanaan NM, Kordower JH. Aging and Parkinson's disease: different Microglial Activation and an Adaptive Immune Response in a Mouse Model sides of the same coin? Mov Disord. 2017;32(7):983–90. of Parkinson Disease. J Neuropathol Exp Neurol. 2008; 67(12): p. 1149–58. 2. Collier TJ, Kanaan NM, Kordower JH. Ageing as a primary risk factor for 26. Oliveras-Salvá M, et al. rAAV2/7 vector-mediated overexpression of Parkinson’s disease: evidence from studies of non-human primates. Nat Rev alpha-synuclein in mouse substantia nigra induces protein aggregation Neurosci. 2011;12(6):359–366 and progressive dose-dependent neurodegeneration. Mol Neurodegener. 3. Mogi M, et al. Interleukin (IL)-1 beta, IL-2, IL-4, IL-6 and transforming growth 2013;8:44. factor-alpha levels are elevated in ventricular cerebrospinal fluid in juvenile 27. Gombash SE, et al. Morphological and behavioral impact of AAV2/5-mediated parkinsonism and Parkinson's disease. Neurosci Lett. 1996;211(1):13–6. overexpression of human wildtype alpha-synuclein in the rat nigrostriatal 4. Lindqvist D, et al. Cerebrospinal fluid inflammatory markers in Parkinson’s system. PLoS One. 2013;8(11):e81426. disease—associations with depression, fatigue, and cognitive impairment. 28. Fischer DL, et al., Viral Vector-Based Modeling of Neurodegenerative Brain Behav Immun. 2013;33:183–9. Disorders: Parkinson’s Disease. In: Manfredsson FP, Editor. Gene Therapy for 5. Gerhard A, et al. In vivo imaging of microglial activation with [11C](R)-PK11195 Neurological Disorders: Methods and Protocols. New York: Springer New PET in idiopathic Parkinson's disease. Neurobiol Dis. 2006;21(2):404–12. York; 2016. p. 367–382. 6. McGeer PL, et al. Reactive microglia are positive for HLA-DR in the 29. Simola N, Morelli M, Carta A. The 6-hydroxydopamine model of Parkinson's substantia nigra of Parkinson's and Alzheimer's disease brains. Neurology. Disease. 2007;11:151–67. 1988;38(8):1285–91. 30. Matsuoka Y, et al. Lack of nigral pathology in transgenic mice expressing 7. Imamura K, et al. Distribution of major histocompatibility complex class human alpha-synuclein driven by the tyrosine hydroxylase promoter. II-positive microglia and cytokine profile of Parkinson's disease brains. Neurobiol Dis. 2001;8(3):535–9. Acta Neuropathol. 2003;106(6):518–26. 31. Ip CW, et al. AAV1/2-induced overexpression of A53T-α-synuclein in the 8. Kordower JH, et al. Disease duration and the integrity of the nigrostriatal substantia nigra results in degeneration of the nigrostriatal system with system in Parkinson's disease. (1460–2156 (Electronic)). Lewy-like pathology and motor impairment: a new mouse model for 9. Croisier E, et al. Microglial inflammation in the parkinsonian substantia nigra: Parkinson’s disease. Acta Neuropathol Commun. 2017;5:11. relationship to alpha-synuclein deposition. J Neuroinflammation. 2005;2:14. 32. Sanchez-Guajardo V, et al. Microglia acquire distinct activation profiles 10. Dijkstra AA, et al. Evidence for immune response, axonal dysfunction and depending on the degree of alpha-synuclein neuropathology in a rAAV reduced endocytosis in the substantia nigra in early stage Parkinson’s based model of Parkinson's disease. PLoS One. 2010;5(1):e8784. disease. PLoS One. 2015;10(6):e0128651. 33. Thakur P, Breger LS, Lundblad M, et al. Modeling Parkinson’s disease 11. Braak H, et al. Staging of brain pathology related to sporadic Parkinson's pathology by combination of fibril seeds and α-synuclein overexpression in disease. Neurobiol Aging. 2003;24(2):197–211. the rat brain. Proc Natl Acad Sci USA. 2017;114(39):E8284–93 12. Kannarkat GT, et al. Common genetic variant association with altered HLA 34. Farrer M, et al. Comparison of kindreds with parkinsonism and alpha-synuclein expression, synergy with Pyrethroid exposure, and risk for Parkinson's genomic multiplications. Ann Neurol. 2004;55(2):174–9. disease: an observational and case-control study. NPJ Parkinsons Dis. 2015;1: 35. Su X, et al. Alpha-Synuclein mRNA Is Not Increased in Sporadic PD and 359–66. Alpha-Synuclein Accumulation Does Not Block GDNF Signaling in 13. Spiller KL, et al., The role of macrophage phenotype in vascularization of Parkinson's Disease and Disease. Models Mol Ther.2017; 25(10): p. 2231– tissue engineering scaffolds. Biomaterials. 2014;35(15): 4477–88. 14. Harms AS, et al. MHCII is required for alpha-synuclein-induced activation of microglia, CD4 T cell proliferation, and dopaminergic neurodegeneration. 36. Zhou J, et al., Changes in the solubility and phosphorylation of α-synuclein J Neurosci. 2013;33(23):9592–600. over the course of Parkinson’s disease. Acta Neurochir. 2011;121(6): p. 695– 15. Liu X, et al. Intracellular MHC class II molecules promote TLR-triggered 37. Paumier KL, et al. Intrastriatal injection of pre-formed mouse alpha-synuclein innate immune responses by maintaining activation of the kinase Btk. fibrils into rats triggers alpha-synuclein pathology and bilateral nigrostriatal Nat Immunol. 2011;12:416. degeneration. Neurobiol Dis. 2015;82:185–99. 16. Cicchetti F, et al. Neuroinflammation of the nigrostriatal pathway during 38. Luk KC, et al. Intracerebral inoculation of pathological α-synuclein initiates a progressive 6-OHDA dopamine degeneration in rats monitored by rapidly progressive neurodegenerative α-synucleinopathy in mice. immunohistochemistry and PET imaging. Eur J Neurosci. 2002;15(6):991–8. J Exp Med. 2012;209(5):975–86. 17. Koprich JB, et al. Neuroinflammation mediated by IL-1beta increases 39. Polinski NK, et al., Best Practices for Generating and Using Alpha-Synuclein susceptibility of dopamine neurons to degeneration in an animal model of Pre-Formed Fibrils to Model Parkinson's Disease in Rodents. J Parkinsons Parkinson's disease. J Neuroinflammation. 2008;5:8. Dis. 2018. 18. Forno LS, et al. Similarities and differences between MPTP-induced parkinsonism and Parkinson's disease. Neuropathologic considerations. 40. Volpicelli-Daley LA, Luk KC, Lee VM. Addition of exogenous alpha-synuclein Adv Neurol. 1993;60:600–8. preformed fibrils to primary neuronal cultures to seed recruitment of 19. Kurkowska-Jastrzebska I, et al. The inflammatory reaction following endogenous alpha-synuclein to Lewy body and Lewy neurite-like 1-methyl-4-phenyl-1,2,3, 6-tetrahydropyridine intoxication in mouse. aggregates. Nat Protoc. 2014;9(9):2135–46. Exp Neurol. 1999;156(1):50–61. 41. Volpicelli-Daley LA, et al. Exogenous alpha-synuclein fibrils induce Lewy 20. Gao HM, et al. Neuroinflammation and alpha-synuclein dysfunction body pathology leading to synaptic dysfunction and neuron death. Neuron. 2011;72(1):57–71. potentiate each other, driving chronic progression of neurodegeneration in a mouse model of Parkinson's disease. Environ Health Perspect. 42. Tarutani A, Suzuki G, Shimozawa A, et al. The Effect of Fragmented 2011;119(6):807–14. Pathogenic α-Synuclein Seeds onPrion-like Propagation. The Journal of Biological Chemistry. 2016;291(36):18675–18688 21. Gomez-Isla T, et al. Motor dysfunction and gliosis with preserved 43. Vaikath NN, et al., Generation and characterization of novel conformation- dopaminergic markers in human alpha-synuclein A30P transgenic mice. specific monoclonal antibodies for α-synuclein pathology. Neurobiol Dis. Neurobiol Aging. 2003;24(2):245–58. 2015; 79: p. 81–99. 22. Koprich JB, et al. Expression of human A53T alpha-synuclein in the rat substantia nigra using a novel AAV1/2 vector produces a rapidly evolving 44. Leys C, et al. Detecting outliers: do not use standard deviation around pathology with protein aggregation, dystrophic neurite architecture and the mean, use absolute deviation around the median. J Exp Soc Psychol. nigrostriatal degeneration with potential to model the pathology of 2013;49(4):764-6. Parkinson's disease. Mol Neurodegener. 2010;5:43. 45. Leys C, et al., Detecting outliers: Do not use standard deviation around the 23. Yamada M, Iwatsubo T, Mizuno Y, Mochizuki H. Overexpression of α‐ mean, use absolute deviation around the median. J Exp Soc Psychol. 2013; synuclein in rat substantia nigra results in loss of dopaminergic neurons, 49(4): p. 764-766. Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 18 of 18 46. Kim C, et al. Neuron-released oligomeric alpha-synuclein is an endogenous formation in human synucleopathies. J Neuropathol Exp Neurol. agonist of TLR2 for paracrine activation of microglia. Nat Commun. 2003;62(10):1060–75. 2013;4:1562. 72. Urrutia PJ, Mena NP, Núñez MT. The interplay between iron accumulation, 47. Wan OW, Chung KK. The role of alpha-synuclein oligomerization and mitochondrial dysfunction, and inflammation during the execution step of aggregation in cellular and animal models of Parkinson's disease. PLoS One, neurodegenerative disorders. Front Pharmacol. 2014;5:38. 2012; 7(6): p. e38545. 73. Brochard V, et al., Infiltration of CD4+ lymphocytes into the brain contributes to neurodegeneration in a mouse model of Parkinson disease. J 48. Winner B, et al., In vivo demonstration that alpha-synuclein oligomers are Clin Invest. 2009. 119(1): p. 182–92. toxic. Proc Natl Acad Sci USA, 2011. 108(10): p. 4194–9. 74. Jiang S, et al., The correlation of lymphocyte subsets, natural killer cell, and 49. Li JY, et al., Characterization of Lewy body pathology in 12- and 16-year-old Parkinson's disease: a meta-analysis. Neurol Sci. 2017; 38(8): p. 1373–1380. intrastriatal mesencephalic grafts surviving in a patient with Parkinson's 75. Kustrimovic N, et al., Dopaminergic Receptors on CD4+ T Naive and disease. Mov Disord. 2010. 25(8): p. 1091–6. Memory Lymphocytes Correlate with Motor Impairment in Patients with 50. Tanji K, et al., Proteinase K-resistant alpha-synuclein is deposited in Parkinson's Disease. Sci Rep. 2016; 6: p. 33738. presynapses in human Lewy body disease and A53T alpha-synuclein 76. Carson MJ, Thrash JC, Walter B. The cellular response in neuroinflammation: transgenic mice. Acta Neuropathol. 2010; 120(2): p. 145–54. The role of leukocytes, microglia and astrocytes in neuronal death and 51. Liu Z, et al., Neuronal uptake of serum albumin is associated with survival. Clin Neurosci Res. 2006; 6(5): p. 237–45. neuron damage during the development of epilepsy. Exp Ther Med. 77. Cebrian C, et al. MHC-I expression renders catecholaminergic neurons 2016; 12(2): p. 695–701. susceptible to T-cell-mediated degeneration. Nat Commun. 2014;5:3633. 52. Rey NL., et al., Widespread transneuronal propagation of alpha- 78. Sulzer D., et al., T cells from patients with Parkinson's disease recognize synucleinopathy triggered in olfactory bulb mimics prodromal Parkinson's alpha-synuclein peptides. Nature. 2017; 546(7660): p. 656–61. disease. J Exp Med. 2016; 213(9): p. 1759–78. 53. Conde JR, Streit WJ. Effect of aging on the microglial response to peripheral nerve injury. Neurobiol Aging. 2006; 27(10): p. 1451–61 54. Streit WJ, Microglia and neuroprotection: implications for Alzheimer's disease. Brain Res Brain Res Rev, 2005; 48(2): p. 234–9. 55. Streit WJ, et al., Dystrophic microglia in the aging human brain. Glia. 2004; 45(2): p. 208–12. 56. Christopher L, et al., Uncovering the role of the insula in non-motor symptoms of Parkinson's disease. Brain. 2014; 137(Pt 8): p. 2143–54. 57. Luk KC, et al. Pathological alpha-synuclein transmission initiates Parkinson-like neurodegeneration in nontransgenic mice. Science. 2012;338(6109):949–53. 58. Maillet A, et al., The prominent role of serotonergic degeneration in apathy, anxiety and depression in de novo Parkinson's disease. Brain. 2016; 139(Pt 9): p. 2486–502. 59. Cerasa A., et al., Cortical volume and folding abnormalities in Parkinson's disease patients with pathological gambling. Parkinsonism Relat Disord. 2014. 20(11): p. 1209–14. 60. Block ML, Hong JS. Microglia and inflammation-mediated neurodegeneration: multiple triggers with a common mechanism. Prog Neurobiol. 2005;76(2):77–98. 61. Gao HM, et al. Neuroinflammation and oxidation/nitration of alpha-synuclein linked to dopaminergic neurodegeneration. J Neurosci. 2008;28(30):7687–98. 62. Hwang O. Role of oxidative stress in Parkinson's disease. Exp Neurobiol. 2013;22(1):11–7. 63. Wilshusen RA, Mosley RL. Innate and adaptive immune-mediated neuroinflammation and neurodegeneration in Parkinson’s disease. In: Peterson PK, Toborek M, editors. Neuroinflammation and neurodegeneration. New York: Springer New York; 2014. p. 119–42. 64. Hong S, Beja-Glasser VF, Nfonoyim BM, et al. Complement and Microglia Mediate Early Synapse Loss in Alzheimer Mouse Models. Science (New York, NY). 2016;352(6286):712–716. 65. Solga AC, et al., RNA-sequencing reveals oligodendrocyte and neuronal transcripts in microglia relevant to central nervous system disease. Glia. 2015; 63(4): p. 531–548. 66. Park JY, et al., Microglial phagocytosis is enhanced by monomeric alpha- synuclein, not aggregated alpha-synuclein: implications for Parkinson's disease. Glia. 2008; 56(11): p. 1215–23. 67. Farooqui T, Farooqui AA. Lipid-mediated oxidative stress and inflammation in the pathogenesis of Parkinson's disease. Parkinsons Dis. 2011;2011:247467. 68. Kim WG, et al. Regional difference in susceptibility to lipopolysaccharide- induced neurotoxicity in the rat brain: role of microglia. J Neurosci. 2000;20(16):6309–16. 69. Zhang K, Kaufman RJ. From endoplasmic-reticulum stress to the inflammatory response. Nature. 2008;454(7203):455–62. 70. Zhang W, et al. Human neuromelanin: an endogenous microglial activator for dopaminergic neuron death. Front Biosci (Elite Ed). 2013;5:1–11. 71. Andringa G, et al. Mapping of rat brain using the synuclein-1 monoclonal antibody reveals somatodendritic expression of alpha-synuclein in populations of neurons homologous to those vulnerable to Lewy body http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Neuroinflammation Springer Journals

Loading next page...
 
/lp/springer_journal/lewy-body-like-alpha-synuclein-inclusions-trigger-reactive-MqITlBa8UP
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s).
Subject
Biomedicine; Neurosciences; Neurology; Neurobiology; Immunology
eISSN
1742-2094
D.O.I.
10.1186/s12974-018-1171-z
Publisher site
See Article on Publisher Site

Abstract

Background: Converging evidence suggests a role for microglia-mediated neuroinflammation in Parkinson’s disease (PD). Animal models of PD can serve as a platform to investigate the role of neuroinflammation in degeneration in PD. However, due to features of the previously available PD models, interpretations of the role of neuroinflammation as a contributor to or a consequence of neurodegeneration have remained elusive. In the present study, we investigated the temporal relationship of neuroinflammation in a model of synucleinopathy following intrastriatal injection of pre-formed alpha-synuclein fibrils (α-syn PFFS). Methods: Male Fischer 344 rats (N = 114) received unilateral intrastriatal injections of α-syn PFFs, PBS, or rat serum albumin with cohorts euthanized at monthly intervals up to 6 months. Quantification of dopamine neurons, total neurons, phosphorylated α-syn (pS129) aggregates, major histocompatibility complex-II (MHC-II) antigen-presenting microglia, and ionized calcium-binding adaptor molecule-1 (Iba-1) immunoreactive microglial soma size was performed in the substantia nigra. In addition, the cortex and striatum were also examined for the presence of pS129 aggregates and MHC-II antigen-presenting microglia to compare the temporal patterns of pSyn accumulation and reactive microgliosis. Results: Intrastriatal injection of α-syn PFFs to rats resulted in widespread accumulation of phosphorylated α-syn inclusions in several areas that innervate the striatum followed by significant loss (~ 35%) of substantia nigra pars compacta dopamine neurons within 5–6 months. The peak magnitudes of α-syn inclusion formation, MHC-II expression, and reactive microglial morphology were all observed in the SN 2 months following injection and 3 months prior to nigral dopamine neuron loss. Surprisingly, MHC-II immunoreactivity in α-syn PFF injected rats was relatively limited during the later interval of degeneration. Moreover, we observed a significant correlation between substantia nigra pSyn inclusion load and number of microglia expressing MHC-II. In addition, we observed a similar relationship between α-syn inclusion load and number of microglia expressing MHC-II in cortical regions, but not in the striatum. (Continued on next page) * Correspondence: caryl.sortwell@hc.msu.edu Department of Translational Science and Molecular Medicine, Michigan State University, 400 Monroe Avenue NW, Grand Rapids, MI 49503-2532, USA Mercy Health Hauenstein Neuroscience Medical Center, Grand Rapids, MI, USA Full list of author information is available at the end of the article © The Author(s). 2018, corrected publication May/2018. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 2 of 18 (Continued from previous page) Conclusions: Our results demonstrate that increases in microglia displaying a reactive morphology and MHC-II expression occur in the substantia nigra in close association with peak numbers of pSyn inclusions, months prior to nigral dopamine neuron degeneration, and suggest that reactive microglia may contribute to vulnerability of SNc neurons to degeneration. The rat α-syn PFF model provides an opportunity to examine the innate immune response to accumulation of pathological α-syn in the context of normal levels of endogenous α-syn and provides insight into the earliest neuroinflammatory events in PD. Keywords: Neuroinflammation, Parkinson’s disease, Animal models, Synucleinopathy, Microglia, Major-histocompatibility complex-II, Neurodegeneration, Selective vulnerability Background risk for development of PD after pesticide exposure [12]. The etiology of Parkinson’s disease (PD) is stochastic: a Increased MHC-II is often concurrently upregulated with culmination of aging-related changes in brain environ- genes for proinflammatory cytokines such as tumor necro- ment, genetic predispositions, and environmental insults sis factor (TNF) and interleukin-1 beta (IL-1β)[13]. More- that result in accumulation of alpha-synuclein (α-syn) over, decreased MHC-II expression was shown to attenuate inclusions (i.e., Lewy bodies) and degeneration of nigros- downstream secretion of proinflammatory cytokines [14, triatal dopamine neurons [1, 2]. Converging evidence 15]. Taken together, it is likely that MHC-II is most closely suggests a role for microglia-mediated neuroinflamma- associated with a proinflammatory phenotype in microglia tion in human PD. This theory is supported by observa- and may play a contributory role in nigral degeneration in tions of increased inflammatory cytokines in both PD PD. However, while the concept that MHC-II expression patient cerebral spinal fluid (CSF) and plasma [3, 4] and on microglia is increased in PD patients is not novel [9], in the patient brain as longitudinal PET imaging has the temporal pattern of observed increases in MHC-II in demonstrated early and sustained microglial activation relation to α-syn aggregation and/or nigrostriatal degener- in the basal ganglia [5]. Furthermore, postmortem ana- ation has been unable to be systematically examined. lyses in PD patients revealed increased expression of in- Animal models of PD can serve as platforms to inves- flammatory markers such as human leukocyte antigen tigate the role of neuroinflammation in PD-related cell (HLA-DR), major histocompatibility complex-II (MHC- death and dysfunction. The neuroinflammatory conse- II), phagocytic marker CD68, intercellular adhesion quences of nigral degeneration and/or α-syn aggregation molecule-I (ICAM-1), and integrin adhesion molecule have been examined previously in various models, (LFA-1) in the substantia nigra [6, 7]. However, a draw- including but not limited to, neurotoxicant models back of biofluid and postmortem PD brain samples is (6-hydroxydopamine (6-OHDA) [16, 17]; 1-methyl-4- that they only provide a static snapshot of events within phenyl-1,2,3,6-tetrahydropyridine (MPTP) [18, 19]); a longitudinal cascade of PD pathophysiology. This is es- transgenic models expressing human wild-type or mutant pecially problematic as the overwhelming majority of PD α-syn (A503T, A30P [20–22]) and viral vector-mediated patient samples are collected from individuals who have overexpression of human wild-type or mutated α-syn in likely harbored PD-related pathology for decades before, the nigrostriatal system [22–28]. However, certain charac- if also not after, diagnosis [8]. This confounds interpreta- teristics of these models limit interpretations regarding tions of the role of neuroinflammation in degeneration the specific initiator of the neuroinflammation observed— in PD and prevents the understanding as to whether synuclein inclusions and/or degeneration. Neurotoxicant neuroinflammation participates as a contributor to nigral models (6-OHDA, MPTP) rarely exhibit α-syn pathology degeneration or is simply an artifact of cell death. [18, 29]. Transgenic models generally do not recapitulate One of the most consistent observations in postmor- marked nigrostriatal degeneration despite widespread, tem PD tissue is an increase in the number of microglia α-syn pathology [21, 30]. Whereas a robust inflamma- expressing MHC-II (HLA-DR in humans [7, 9, 10]), a tory response is observed in association with the ele- cell surface protein on antigen-presenting cells which is vated α-syn levels, aggregates, and nigral degeneration necessary for CD4+ T cell infiltration. More recently, in viral vector-based α-syn overexpression models gene expression changes related to inflammation, includ- [22–28, 31–33], the contribution of supraphysiological α- ing an upregulation of MHC-II, have also been noted in syn levels or the α-syn species difference (human α-syn incidental Lewy body disease subjects (Braak stages 1–3 expressed in rat or mouse) to the neuroinflammatory [10, 11]). Additionally, a variant in the HLA-DR gene response is unclear. Importantly, in human sporadic PD, which encodes for MHC-II is associated with amplified total α-syn levels are not increased; rather, phosphorylation Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 3 of 18 and the ratio of soluble to insoluble α-syn increases over described [39–41]. Prior to sonication, α-syn fibrils were time [34–36]. assessed to verify lack of contamination (LAL Assay An alternative model of the key features of human (~ 1 endotoxin units/mg), high molecular weight sporadic PD such as (1) protracted development of (sedimentation assay), beta sheet conformation (thiofla- α-syn inclusions under conditions of (2) normal expres- vin T), and structure (electron microscopy). Prior to injec- sion levels of endogenous α-syn during an interval that (3) tion, PFFs were thawed, diluted in sterile Dulbecco’sPBS precedes significant nigrostriatal degeneration would offer (DPBS, 2 μg/μl), and sonicated at room temperature using distinct advantages and allow the time course and poten- an ultrasonicating homogenizer (300VT; Biologics, Inc., tial impact of neuroinflammation to be delineated. Re- Manassas, VA) with the pulser set at 20% and power out- cently, our lab has characterized a rat model of PD that put at 30% for 60 pulses at 1 s each. Following sonication, recapitulates this sequence of events, extending previous a sample of the PFFs was analyzed using transmission findings in mice [37, 38]. In this model, nigrostriatal synu- electron microscopy (TEM). Formvar/carbon-coated cop- cleinopathy is induced by intrastriatal injection of soni- per grids (EMSDIASUM, FCF300-Cu) were washed twice cated pre-formed α-syn fibrils (α-syn PFFs) into wild-type with ddH O and floated for 1 min on a 10-μldrop of soni- rats [37, 39]. The fibrils act as seeds to template and cated α-syn fibrils diluted 1:20 with DPBS. Grids were trigger normal levels of endogenous α-syn to accumulate stained for 1 min on a drop of 2% uranyl acetate aqueous into misfolded hyperphosphorylated, pathological α-syn solution; excess uranyl acetate was wicked away with filter (Fig. 1a). The initial injection of α-syn PFFs per se does paper and allowed to dry before imaging. Grids were imaged not directly cause toxicity, given that α-syn pathology and on a JEOL JEM-1400 transmission electron microscope. nigral degeneration do not occur in α-syn knockout ani- The length of over 500 fibrils per sample was measured to mals injected with PFFs [38]. In this model, we observe a determine average fibril size. The mean length of sonicated widespread accumulation of intraneuronal Lewy neurite- mouse α-syn PFFs was estimated to be 51.22 ± 1.31 nm, well like and Lewy body-like inclusions of phosphorylated within the optimal fibril length previously reported to result α-syn (pSyn) in areas that innervate the striatum. Import- in seeding of endogenous phosphorylated α-syn inclusions antly, the accumulation of intracellular pSyn is gradual in vitro and in vivo (Fig. 1b, c)[42]. and results in loss of striatal dopamine and metabolites in addition to ~ 40% loss of SNc dopamine neurons over Intrastriatal injections 6 months [37]. Thus, the synucleinopathy produced in the Sonicated PFFs were kept at room temperature during α-syn PFF model provides a unique opportunity to exam- the duration of the surgical procedures. All rats were ine the neuroinflammatory consequences of α-syn inclu- deeply anesthetized with isoflurane received two 2-μl sion accumulation in the context of normal levels of unilateral intrastriatal injections (4 μl total; AP + 1.6, endogenous, intracellular α-syn. In the present study, we ML + 2.4, DV − 4.2; AP − 1.4, ML + 2.0, DV − 7.0 from systematically investigated the temporal profile of Lewy the skull) either of sonicated mouse α-syn PFFs (2 μg/μlas body-like phosphorylated α-syn inclusion load, reactive described previously [37]) or an equal volume of DPBS at a microglial morphology, MHC-II antigen presentation, and rate at 0.5 μl/min (n = 6 per treatment per time point). In- degeneration in the SN. Importantly, we observe reactive jections were administered made using a pulled glass needle microglia and increased microglial MHC-II expression in attached to a 10-μl Hamilton syringe. After each injection, association with peak load of SNc pSyn inclusions months the needle was left in place for 1 min, retracted 0.5 mm, left prior to degeneration, suggesting that neuroinflammation in place for an additional 2 min, and then slowly with- may contribute to nigrostriatal degeneration. drawn. Animals were monitored post-surgery and eutha- nized at predetermined time points (14, 30, 60, 90, 120, Methods 150, and 180 days; Fig. 1). In a subsequent experiment, rats Animals received two 2-μl unilateral intrastriatal injections either of Young adult (2 months), male Fischer344 rats (n = 114) mouse α-syn PFFs 2 μg/μl, DPBS, or rat serum albumin were used in this study. All animals were provided food (RSA, Sigma-Aldrich, St. Louis, MO; 9048-46-8; 2 μg/μl) at and water ad-libitum and housed at the AAALAC- the identical coordinates and were euthanized at 2 months approved Van Andel Research Institute vivarium. All postinjection (n = 6 per treatment). procedures were approved and conducted in accordance with Institute for Animal Use and Care Committee Immunohistochemistry (IACUC) at Michigan State University. All animals were euthanized via pentobarbital overdose (60 mg/kg) and intracardially perfused with heparinized Preparation of α-syn PFFs and verification of fibril size 0.9% saline followed by cold 4% paraformaldehyde in Purification of recombinant, full-length mouse α-syn 0.1 M PO . Brains were extracted and postfixed in 4% and in vitro fibril assembly was performed as previously PFA for 48 h and placed in 30% sucrose until they sunk. Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 4 of 18 Fig. 1 Experimental design and PFF quality control. a Experimental design: 2-month-old male Fischer344 rats (N = 114) received two unilateral intrastriatal injections of either sonicated α-syn PFFs, Dulbecco’s PBS (PBS), or rat serum albumin (RSA; follow-up study only). Cohorts of rats were euthanized at an early time point (2 weeks) and monthly intervals thereafter. Brains were removed and processed for immunohistochemical measures of pathology as detailed. b Electron micrographs of unsonicated (left) α-syn PFFs and sonicated α-syn PFFs (right); scale bars = 100 nm. c Measurement distribution of ~ 500 sonicated PFFs prior to injection; mean fibril size = 51.22 ± 1.31 nm. d Schematic of PFF model of synucleinopathy. Sonicated α-syn fibrils are injected into the striatum and taken up by nigrostriatal terminals (1), after which they template and convert endogenous α-syn to a hyperphosphorylated, pathological form (2), ultimately accumulating into Lewy neurite- and Lewy body-like inclusions (3). Abbreviations: α-syn = alpha-synuclein; PFFs = pre-formed alpha-synuclein fibrils; PBS = phosphate-buffered saline; DMS = dorsal medial striatum; VLS = ventrolateral striatum; IHC = immunohistochemistry; pSyn = α-syn phosphorylated at serine 129; MHC-II = major histocompatibility complex-II; Iba-1 = ionized calcium-binding adaptor molecule 1; TH = tyrosine hydroxylase; NeuN = neuronal nuclei; CD68 = cluster of differentiation 68; QC = quality control For sectioning, brains were frozen on a sliding microtome blocking, sections were immunolabeled with primary anti- and cut at 40 μm. Free-floating sections (1:6 series) were bodies: mouse anti-α-syn fibrils/oligomers (O2; 1:5000 transferred to 0.1 M tris-buffered saline (TBS). Following [43]) or mouse anti α-syn fibrils (F2; 1:5000 [43]), pan the washes, endogenous peroxidases were quenched in 3% rabbit-anti α-syn (Abcam, Cambridge, MA; AB15530, H O for 1 h and rinsed in TBS. Sections were blocked in 1:1000), mouse anti-phosphorylated α-syn at serine 2 2 10% normal goat serum/0.5% Triton X-100 in TBS (NGS, 129 (pSyn, 81A; Abcam, Cambridge, MA; AB184674; Gibco; Tx-100 Fischer Scientific) for 1 h. Following the 1:10,000), rabbit anti-tyrosine hydroxylase (TH; Millipore, Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 5 of 18 Temecula, CA; MAB152, 1:4000), rabbit anti-ionized reagent (Vector Laboratories, Burlingame, CA) was used calcium-binding adaptor molecule-1 (Iba-1; Wako, as the chromogen. Slides were rinsed in TBS and cover- Richmond, VA; 019-19741, 1:1000), mouse anti-neuronal slipped with Cytoseal 60. Images were taken on a Nikon nuclei (Neu-N; Millipore, Temecula, CA; MAB 377, 1:5000), Eclipse 90i microscope with a QICAM camera (QImaging, or mouse anti-rat major histocompatibility complex-II for Surrey, British Colombia, Canada). antigen-presenting microglia (MHC Class II RT1B clone OX-6, Bio-Rad, Hercules, CA; MCA46G, 1:5000) over- Quantification of TH, NeuN, pSyn, and MHC-II night in 1% NGS/0.5% Tx-100/TBS at 4 °C. Following the immunoreactive profiles washes, sections were incubated in biotinylated secondary Microbrightfield (MBF) Stereoinvestigator (MBF Bioscience, antibodies (1:500) against mouse (Millipore, Temecula, Williston, VT) was used to estimate the total population CA; AP124B) or rabbit IgG (Millipore, Temecula, CA; of THir and NeuNir neurons to determine the time course AP132B) followed by washes in TBS and 2-h incubation of TH phenotype loss and overt nigral degeneration. Con- with Vector ABC standard detection kit (Vector Labora- tours were drawn around the SNc using the ×4 objective tories, Burlingame, CA; PK-6100). Labeling for pSyn, on every sixth section through the rostrocaudal axis (9–10 MHC-II, and TH was visualized by development in 0. sections). A series of counting frames (50 μm×50 μm) 5 mg/ml 3,3′ diaminobenzidine (DAB; Sigma-Aldrich St. was systematically and randomly distributed over grid Louis, MO; D5637-10G) and 0.03% H O . For dual bright- (183 μm × 112 μm) placed over the SNc, allowing for 2 2 field visualization of Neu-N and Iba-1, sections were de- quantification of approximately 20% of the SNc. An inves- veloped according to the manufacturer’sinstructions tigator blinded to experimental conditions counted THir using the Vector ImmPACT DAB Peroxidase (Vector and NeuNir cells using the optical fractionator probe with Labs, Burlingame, CA; SK-4605) and ImmPACT VIP a ×60 oil immersion objective. Markers were placed on Peroxidase (Vector Labs, Burlingame, CA; SK-4105) kits, each THir or NeuNir cell in a 1–2-μm z-stack within the respectively. Slides were dehydrated in ascending ethanol counting frame. Between 50 and 500 objects were counted series and then xylenes before coverslipping with Cytoseal to generate stereological estimates of the total cell popula- (Richard-Allan Scientific, Waltham, MA). A subset of tion. The total population estimate was calculated using pSyn-labeled sections were also counterstained with cresyl optical fractionator estimates, and variability within ani- violet for quantification of intraneuronal pSyn inclusions mals was assessed via the Gunderson coefficient of error in the SNc. (< 0.1). Due to heterogeneity in the distribution of both pSyn and MHC-II immunoreactive profiles within the SN, RNAscope in situ hybridization for Iba-1 and MHC-II IHC total enumeration rather than counting frames was used Forty-micrometer-thick striatal tissue sections were in- for quantification. Neurons with intraneuronal pSyn in- cubated in pretreat 1 from the RNAscope Pretreatment clusions were defined as profiles of dark, densely Kit (Advanced Cell Diagnostics, Hayward, CA; 310020) stained pSyn immunoreactivity within cresyl violet- for 1 h. Sections were washed in TBS and then mounted positive neurons. Contours were drawn around the on VistaVision HistoBond slides (VWR, Randor, PA; SNcusing the×4objectiveon everysixth section 16004-406) and placed on slide warmer at 60 °C over- through the entire rostrocaudal axis of the SNc (9–10 night. Slides were then incubated for 10 min in pretreat sections). pSyn inclusions and MHC-IIir microglia 2 at 99 °C and washed twice in water. Tissue was out- were then systematically counted within each contour lined with Pap Pen (Abcam, Cambridge, UK; ab2601), using the ×20 objective. Numbers represent the raw incubated with pretreat 3 in a hybridization oven at 40 °C total number of pSyn inclusions or MHC-IIir microglia for 15 min, washed twice in water, and incubated with the per animal multiplied by 6 to extrapolate the popula- probe for AIF1 (Iba1; Advanced Cell Diagnostics, tion estimate. Hayward, CA; 457731) for 2 h in the hybridization oven at 40 °C. Six amplification steps with the amplification Microglial soma area analysis buffers (Advanced Cell Diagnostics, Hayward, CA; Forty-micrometer-thick nigral tissue sections (1:6 series) 320600) were then performed in alternating 30- and from animals injected with α-syn PFFs, RSA, or and 15-min incubation intervals in the hybridization oven per DPBS 2 months and 6 months following injection were manufacturer instructions. Tissue was developed using dual labeled for NeuN and Iba-1 as described above to the supplied DAB reagent (Advanced Cell Diagnostics, distinguish the SNc from the SNr. The three nigral sec- Hayward, CA; 320600). Tissue was then counterstained tions adjacent to the sections containing the most pSyn for MHC-II (RT1B clone OX-6, Bio-Rad, Hercules, CA; inclusions were identified. z-stack images of the ipsilat- MCA46G, 1:500) in a hybridization chamber, following eral and contralateral SNr bordering the SNc were taken the same procedures as detailed for other immunohisto- on a Nikon Eclipse 90i microscope with a QICAM cam- chemical stains with the exception that the Vector SG era (QImaging, Surrey, British Colombia, Canada) using Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 6 of 18 the ×20 objective and analyzed with Nikon Elements AR Results (version 4.50.00, Melville, NY). Using the auto-detect Sonicated α-syn PFFs are the optimal size for pathology feature, each Iba-1ir soma’s border was outlined and ad- induction in vivo justed accordingly to obtain an accurate quantification Prior to intrastriatal injection of mouse α-syn PFFs, we in- of area of the soma, excluding any processes. All micro- vestigated the size of the PFFs following sonication using glia in the field of view of each z-stack per section, per transmission electron microscopy (TEM; Fig. 1b, c). The rat were quantified with total number of microglia per mean length of sonicated mouse α-syn PFFs was estimated rat calculated (100–250). Data are expressed as mean to be 51.22 ± 1.31 nm, well within the optimal fibril length Iba-1ir soma area per treatment group. Soma measure- previously reported to result in seeding of endogenous ments for all microglia per treatment were also grouped phosphorylated α-syn inclusions (Fig. 1d,schematic)in into 10 μm bins and expressed as a percentage of total vitro and in vivo [42]. microglia counted. Unilateral intrastriatal injection of α-syn PFFs induces widespread Lewy-like pathology Thioflavin-S staining In our previous work [37], we reported that unilateral 1:12 series was washed in TBS and subsequently intrastriatal injection of mouse α-syn PFFs results in phos- mounted on subbed slides to dry (~ 1 h). Slides were in- phorylated α-syn (pSyn) intraneuronal accumulations in cubated in 0.5% KMnO in TBS for 25 min, followed by several areas that innervate the striatum [45], most prom- five washes in TBS. Sections were destained in 0.2% inently the frontal (primary motor and somatosensory, K S O /0.2% oxalic acid in TBS for 3 min followed by 2 2 5 layer 5) and insular cortices, amygdala, and SNc. Over incubation in 0.0125% thioflavin-S in 40% EtOH/TBS for time, accumulations increase in number in these regions. 3 min and differentiated in 50% EtOH for 15 min. In the present study, we observed an identical pattern of Sections were rinsed first in TBS and then ddH 0be- pSyn accumulation in rats injected with mouse α-syn fore coverslipping with VECTASHIELD Mounting PFFs. Specifically, we observe abundant pSyn pathology Medium for fluorescence. bilaterally in cortical regions (layers 2/3 of the secondary motor area, insular cortex, and orbital areas; Fig. 2a). In contrast, we observed unilateral pSyn accumulation in the Proteinase-K digestion SNc ipsilateral to the injected striatum and complete ab- 1:12 nigral series was washed in TBS. A subset of free sence of pSyn aggregates in animals injected with an equal floating tissue sections was treated with 10 μg/ml pro- volume of PBS or equal volume and concentration of RSA teinase K (Invitrogen, Carlsbad, CA; 25530015) for (Additional file 1: Figure S1). 30 min at room temperature, followed by three washes Accumulation of pSyn inclusions followed a distinct in TBS and four washes in TBS-Tx. Sections were then temporal pattern depending on the region examined. At processed for pan α-syn immunohistochemistry (rabbit 2 months postinjection (p.i.), we observed abundant anti-α-syn, Abcam, Cambridge, UK; AB15530) as de- soma and neuritic pSyn inclusions bilaterally in the agra- scribed above, mounted on subbed slides, dehydrated to nular insular cortex that persisted over the course of xylenes, and coverslipped. 6 months (Additional file 2: Figure S2A). Abundant pSyn accumulations were observed within the ipsilateral SNc Statistics at 2 months p.i., which remained ipsilateral and decreased Statistical analyses were performed using IBM SPSS in number over the course of 6 months (Additional file 2: Statistics (IBM, Armonk, NY) or GraphPad Prism (La Figure S2D–F). The abundance of pSyn inclusions in the Jolla, CA). Statistical significance for all cases was set striatum followed an opposite pattern (Additional file 2: at p < 0.05. Statistical outliers were assessed using Figure S2G–I). We observed relatively sparse pSyn inclu- the Absolute Deviation from the Median (ADAM) sions in the striatum at 2 months p.i. that were primarily method using the “very conservative” criterion [44]. restricted to neurites. At 4 and 6 months p.i., the number To compare numbers of O2 vs. F2 immunoreactive of pSyn accumulations in striatal somata increased in cells (Fig. 4), THir and NeuNir neurons (Fig. 5), abundance and also were observed in the contralateral pSyn α-syn inclusions (Fig. 6), MHC-IIir microglia striatal hemisphere (Additional file 2: Figure S2H–I). (Fig. 5), and Iba-1ir microglia number and size (Fig. 7), a one-way ANOVA with Tukey’s post hoc analyses was α-Syn inclusions in the SNc exhibit oligomeric, fibrillary used. Correlation analysis was conducted to investigate conformations, and Lewy body-like characteristics the relationship between ipsilateral and contralateral THir The oligomeric form of α-syn is proposed to be one of neurons (Fig. 5) and between MHC-IIir and pSyn α-syn the toxic species [43, 46–48]. We further characterized inclusions (Fig. 6). the nature of pSyn inclusions within the SNc at 1 month Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 7 of 18 Fig. 2 α-Syn inclusions in the SNc exhibit oligomeric and fibrillary conformations and Lewy body-like characteristics. a–c Representative images of Lewy-body-like intraneuronal pSyn inclusions in the substantia nigra pars compacta (SNc) at 1 month p.i. show pathology is localized to the ipsilateral SNc. d–f Adjacent SN tissue sections stained for O2 (oligomeric/fibrillar α-syn conformation specific) and F2 (g–i) (fibrillar-specific conformation) α-syn reveal that many intraneuronal inclusions possess mature, fibrillar inclusions of α-syn. j Percent of pSyn inclusions with either oligomeric/fibrillar (O2) or predominantly fibrillar (F2) immunoreactivity and estimated proportion of pSyn inclusions that are oligomeric only. Data represent mean ± SEM. Scale bars (b, e, h)= 50 μm, (c, f, i)= 10 μm. k Endogenous α-syn immunoreactivity in the SNc and SNr. l Adjacent tissue sections exposed to proteinase-K reveal the absence of soluble α-syn in the SNr and the presence of insoluble, neuronal inclusions of α-syn in the SNc. m Thioflavin-S fluorescence of amyloid structure present in SNc neurons. Scale bars (k, l, m)= 50 μm, (insets) = 25 μm. Abbreviations: α-syn = alpha-synuclein; p.i. = postinjection; pSyn = α-syn phosphorylated at serine 129; O2 = oligomeric/fibrillar α-syn antibody; F2 = fibrillar α-syn only antibody; SNc = substantia nigra pars compacta; SNr = substantia nigra pars reticulata; SEM = standard error of the mean p.i. using conformation-specific antibodies for oligo- only) with an estimated 31.7 ± 5.9% of inclusions sug- meric/fibrillar α-syn (O2) or fibrillar-predominant α-syn gested to be in an oligomeric conformation (O2 only (F2) and compared that with immunoreactivity to pSyn minus F2 only; Fig. 2j). α-Syn inclusions in the SNc (Fig. 2a–i [43]). When adjacent sections were quantified 2 months p.i. displayed Lewy body-like characteristics using unbiased stereology, we observed that 88.2 ± 6.4% [49, 50], including resistance to proteinase-K digestion of pSyn immunoreactive inclusions exhibited either an (Fig. 2i, k, l) as well as markers for ß-sheet structure as oligomeric or fibrillary conformation (O2 only). Further- detected by thioflavin-S (Fig. 2m). Collectively, these re- more, 56.4 ± 6.04% of nigral pSyn inclusions was de- sults suggest that intrastriatal injection of mouse α-syn tected as predominately mature, fibrillar aggregates (F2 PFFs triggers pathological conversion of endogenous Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 8 of 18 α-syn to phosphorylated, oligomeric, and fibrillary In a control experiment, we examined whether intras- conformations in the SNc that ultimately result in triatal injection of an exogenous protein taken up by insoluble, amyloid inclusions resembling Lewy bodies. neurons [51, 52], rat serum albumin (RSA), induced an inflammatory response in the absence of intracellular PFF-induced synucleinopathy induces significant bilateral pSyn accumulation. To rule out acute toxicity induced loss of SNc neurons by RSA, THir SNc neurons were quantified at 2 months We previously observed that unilateral intrastriatal mouse after injection. No significant differences in THir SNc α-syn PFF injections to rats resulted in bilateral nigrostria- neurons were observed due to PBS, RSA, or PFF injec- tal degeneration of THir SNc neurons within 6 months tion in either the ipsilateral or contralateral hemisphere [37]. To validate this finding in our present cohort, we con- (F = 0.3731, p > 0.05, Fig. 3j). These results (5, 26) ducted unbiased stereology of THir SNc neurons at 2, 4, 5, demonstrate that RSA injection, used as an additional and 6 months p.i. in α-syn PFF- and PBS-injected rats. In- control treatment, did not compromise the survival of jection of PBS did not result in significant loss of THir SNc THir SNc neurons. neurons at any time point (F = 1.991, p > 0.05); thus, (7, 18) PBS-injected time points were combined for comparison Phosphorylated α-syn inclusions peak in the SNc at to PFF-injected rats between identical hemispheres (ipsilat- 2 months and significantly decrease in number during eral PBS = 12,518 ± 554; contralateral PBS 11,577 ± 536). the 5–6-month interval of SNc degeneration Similar to our previous studies, we observed significant, bi- We determined the time course (1–6 months p.i.) of lateral reduction (~ 35%) in SNc THir neurons (Fig. 3a–e). phosphorylated α-syn (pSyn) accumulation in the SNc Specifically, the number of SNc THir neurons ipsilateral to following intrastriatal α-syn PFF injection at monthly in- α-syn PFF injection at both 5 (8227 ± 1015) and 6 months tervals. pSyn inclusions were observed in the SNc ipsilat- (8851 ± 1148) p.i. was significantly reduced compared to eral to injection in all α-syn PFF-injected rats, with the PBS control rats (F = 4.297, p < 0.027, Fig. 5e). Within (5, 36) number of inclusions varying based on time point after the contralateral SNc, significantly fewer THir neurons injection (Fig. 4a–d). Inclusions were most abundant at were observed 5 months following α-syn PFF injec- months 1, 2, and 3, with all three time points exhibiting tion (F =5.782, p < 0.013) with a non-significant re- (5, 36) significantly higher α-syn inclusions compared to the duction in the contralateral SNc observed at 6 months interval of SNc degeneration at months 4, 5, and 6 (p > 0.05). A positive correlation existed between the (Figs. 3e, g and 4d; F =2.251, p ≤ 0.001). The num- (5, 18) extent of ipsilateral THir SNc neuron loss and the extent ber of intraneuronal α-syn inclusions in the SNc was sig- of contralateral loss of THir SNc neurons (r = 0.8855, nificantly greater at 2 months p.i. compared to all other p = 0.0007, R = 0.7842, Fig. 3f). time points except the 1-month time point (Fig. 4d; Lastly, to confirm whether reductions in THir neurons p ≤ 0.006). At 2 months, approximately 2220 ± 148.6 SNc induced by PFF injection represented phenotype loss or neurons possessed pSyn inclusions. By comparison, a loss overt degeneration, unbiased stereology of NeuN-ir neu- of ≈ 3804 THir SNc neurons ipsilateral to PFF injection rons in the SNc was conducted in PFF- or PBS-treated was observed at 5–6 months. These results suggest that groups at 5 and 6 months p.i. No significant differences pSyn inclusion formation in the SNc between 1 and were observed within the corresponding hemisphere be- 3 months after PFF injection precedes degeneration of the tween 5 and 6 months due to either PBS or PFF injec- SNc neurons at 5–6months p.i. tion (PBS: F = 1.238, p > 0.05; PFF: F = 0.3986, (3, 4) (3, 8) p > 0.05). Therefore, the 5- and 6-month time points were combined into one time point. The number of SNc MHC-II immunoreactive (MHC-IIir) microglia increase in NeuN-ir neurons ipsilateral to PFF injection was signifi- the SNc in association with accumulation of α-syn cantly reduced compared to either the ipsilateral or inclusion but are decreased during the interval of contralateral hemisphere of PBS-injected rats (F = degeneration (3, 16) 7.089, p < 0.02). The number of NeuN-ir neurons in the MHC-II expression on microglia is associated with contralateral SNpc of PFF injected rats was significantly co-expression of pro-inflammatory genes such as reduced compared to the ipsilateral SNc of PBS injected TNF, IL-1β, and CD80 as well as proinflammatory cyto- rats (p < 0.0319). When compared to the contralateral kine secretion [13–15]. We quantified MHC-II immuno- SNc of PBS-injected rats, NeuN-ir neurons were reduced reactive (MHC-IIir) microglia within an adjacent series of yet did not reach significance (p = 0.0563, Fig. 3g–i). Over- SNc tissue sections at months 1, 2, 3, 4, 5, and 6 after uni- all, our results replicate our previous findings that intras- lateral α-syn PFF or PBS intrastriatal injection in order to triatal α-syn PFF injection results in significant bilateral examine neuroinflammation. Double labeling for MHC-II reductions in THir and NeuN-ir SNc neurons over the proteinand Iba-1mRNA confirmed the identity of course of 6 months [37]. MHC-IIir cells to be microglia (Fig. 4e). Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 9 of 18 Fig. 3 α-syn PFF-seeded synucleinopathy induces protracted, significant bilateral loss of SNc dopamine neurons. a–d Unilateral intrastriatal α-syn PFF injection induces visible loss of THir neurons (brown) in the SN at 6 months compared to an age-matched, PBS-injected control. Scale bars (a, c)=25 μm. e Stereological assessment of THir neuron loss at 2, 4, 5, and 6 months following α-syn PFF or saline injection. Significant ipsilateral reduction in THir neurons was observed at 5 and 6 months postinjection compared to saline-injected controls with significant contralateral loss at 5 months *p < 0.027 compared to respective PBS hemisphere, **p < 0.013 compared to respective PBS hemisphere. f Correlation between extent of ipsilateral THir neuron loss and contralateral loss as compared to PBS control (r = 0.8855, p = 0.0007, R = 0.7842). g Stereological assessment of NeuNir neurons reveals overt degeneration distinct from loss of TH phenotype, *p < 0.03 compared to ipsilateral PBS. h Representative IHC of NeuNir neurons (brown, arrows) in the SNc in PBS (left) and PFF injected (right) animals 6 months p.i. i Stereological assessment of THir neurons at 2 months p.i. reveals no significant acute toxicity from injection of rat serum albumin (RSA). Data represent mean ± SEM. Abbreviations: PFF = pre-formed alpha-synuclein fibrils; p.i. = postinjection; ipsilateral = ipsilateral hemisphere relative to injection; contralateral = contralateral hemisphere relative to injection; SNc = substantia nigra pars compacta; SNr = substantia nigra pars reticulata; Neu-Nir = neuronal nuclei immunoreactive; SEM = standard error of the mean; THir = tyrosine hydroxylase immunoreactive; PFFs = pre-formed alpha-synuclein fibrils; PBS = phosphate-buffered saline; RSA = rat serum albumin Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 10 of 18 Fig. 4 Antigen-presenting MHC-II immunoreactive (MHC-IIir) microglia increase in the SNc in association with peak accumulation of α-syn inclusions but are limited during the interval of degeneration. a–c Representative images of pSyn inclusions in the SNc at 2, 4, and 5 months p.i.; scale bar (a–c)=10 μm. d Stereological assessment of pS129 containing neurons in the substantia nigra in PFF animals at 1, 2, 3, 4, 5, and 6 months p.i.; *p ≤ 0.001 compared to 4, 5, and 6 months. p ≤ 0.006 compared to 3, 4, 5, and 6 months. pSyn inclusions decrease over time in association with neuronal loss. e Major-histocompatibility complex-II (MHC-II; blue) protein colocalizes with ionized calcium-binding adaptor molecule 1 mRNA (brown) within microglia. f, g Representative images of MHC-II antigen-presenting microglia in the SN at 2 months in PBS- and PFF-injected rats and 6 months post-PFF injection (h); scale bar (f–h)= 50 μm, insets = 10 μm. i Stereological assessment of MHC-IIir microglia in the SN reveals MHC-IIir microglia are significantly higher in PFF vs. PBS animals at 2, 4, and 5 months *p < 0.006. More MHC-IIir microglia are evident in 2-month PFF animals vs. all other PFF time points p < 0.02. Notably, MHC-IIir microglia peak at the same time pSyn aggregation peaks (d, i). j Number of MHC-II immunoreactive microglia correlated with number of SNc neurons with intraneuronal pSyn inclusions (r = 0.8858, p =0.0015, R = 0.7846). k In a follow-up study, intrastriatal injection of rat serum albumin (RSA) does not impact numbers of MHC-IIir microglia compared to PBS p > 0.05. Injection of PFFs in this second cohort confirmed previous observations of a significant increase in MHC-IIir microglia compared to PBS or RSA at 2 months p.i. *** p ≤ 0.0006. Abbreviations: p.i. = postinjection; PFFs = pre-formed alpha-synuclein fibrils; PBS = phosphate-buffered saline; RSA = rat serum albumin; MHC-II = major-histocompatibility complex-II; Iba-1 = ionized calcium binding adaptor molecule 1; ir = immunoreactive; SNc = substantia nigra pars compacta; SEM = standard error of the mean MHC-IIir microglia were observed in the SNc ipsilat- MHC-IIir microglia varied over time and followed a eral to injection in both α-syn PFF and PBS control rats nearly identical pattern to that observed with pSyn at all time points. No MHC-IIir microglia were observed inclusion accumulation. At the one-month time point, in the contralateral SNc. However, the magnitude of no significant differences were observed between the Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 11 of 18 number of MHC-IIir microglia in the ipsilateral SNc time point (F =0.256, p = 0.855, Fig. 5b). At the (3, 16) of PBS controls compared to α-syn PFF-injected rats 2-month time point coinciding with the peak of pSyn (F = 17.45, p > 0.05), presumably reflecting a non- α-syn inclusion accumulation in the SNc, we observed an (11, 45) specific response to injection (Fig. 4i). However, signifi- appreciable increase in the soma size and thickness and cantly higher numbers of MHC-IIir microglia were ob- number of microglial processes in the SNr of PFF-injected served in the ipsilateral SN of α-syn PFF-injected rats rats compared to a more classically quiescent microglial compared to PBS-injected control rats at months 2, 4, and morphology observed in control injected rats. Specifically, 5(p <0.006, Fig. 4f, g, i). The peak of MHC-IIir microglia in the ipsilateral SNr of rats 2 months following α-syn PFF occurred in the SN 2 months following α-syn PFF injec- injection, the average microglia cell body area was signifi- tion (p < 0.02 compared to PFF-injected rats all other time cantly larger compared to PBS-injected rats (F =4.016, (3, 16) points), corresponding to the time point when the greatest p =0.02, Fig. 5c–e). Microglia soma area varied in all con- number of SNc neurons possesses α-syn aggregates ditions between ≈ 10–200 μm , with a significantly greater (Fig. 4d, g, i). In contrast, significantly fewer MHC-IIir percentage of microglia > 70 μm observed in the SNr microglia were observed in PFF-injected rats at months 5 of rats possessing SNc pSyn α-syn inclusions at and 6, corresponding to the interval of SNc THir neuron 2 months compared to rats injected with PBS (Fig. 5f–h, loss, although these numbers were still significantly higher F =4.613, p =0.03). (2, 12) than PBS-injected controls (p <0.006, Figs. 3e and 4i). At 6 months p.i., a time point corresponding to the There was a positive correlation between the number of time point of very few pSyn α-syn inclusions and imme- MHC-IIir microglia and the number of SNc neurons diately following loss of SNc neurons, no significant possessing pS129 α-syn inclusions in the SN (r = 0.8858, differences in microglia soma size were observed be- p = 0.0015, R = 0.7846, Fig. 4j,months2, 4,and 6). tween α-syn PFF- and PBS-injected rats (F = 2.089, (3, 10) To confirm these findings, we repeated injections in a p > 0.05, Fig. 5i–m). Of note, the average microglial soma separate cohort of animals with rat serum albumin area in PBS-injected rats at 6 months (rats 8 months of (RSA) as an additional control group for neuronal up- age) was significantly larger than PBS-injected rats at take of exogenous protein in the absence of pSyn accu- 2 months (4 months of age) suggesting an age-related mulation, as PBS injection only controls for needle increase (F = 37.00, p < 0.001). The distribution of (3, 12) insertion into the parenchyma. As in previous cohorts, microglia soma areas between PFF and PBS rats α-syn PFF injection resulted in a significant increase in 6 months following injection also appeared similar MHC-IIir microglia in the SN at 2 months p.i. (Fig. 4k, (Fig. 5l, m) with an apparent age-related effect [53–55] p ≤ 0.0006). Injection of RSA resulted in similar numbers reflected in a greater percentage of microglia > 70 μm in of MHC-IIir microglia as observed in PBS-injected con- 8-month-old rats compared to 4-month-old rats. trol rats. No acute neurotoxicity was observed in RSA- Overall, our finding that the peak time point of injected animals at 2 months p.i. (Fig. 3i). Collectively, SNc pSyn α-syn inclusions is associated with a signifi- these results reveal that the preponderance of MHC-II ex- cant increase in microglia soma size suggests that pression in SN microglia is associated with pSyn α-syn in- synucleinopathy in the SNc triggers early disturbances clusions at early time points, however is significantly in local microglia. The interval in which we observe attenuated during the interval of THir SNc degeneration. this synucleinopathy-induced reactive microglial morph- ology is 3 months prior to loss of SNc neurons (Fig. 3e, g) pSyn inclusions in the SNc are associated with a reactive suggesting that reactive microglia have the potential microglial morphology in the adjacent SNr to contribute to vulnerability of SNc neurons to The number and distribution of MHC-IIir microglia in degeneration. the SN suggested that not all microglia were expressing We also examined a series of sections throughout the MHC-II. We next used Iba-1 immunoreactivity to exam- SN and striatum at 2, 4, and 6 months p.i. in α-syn PFF- ine the entire microglia population within an adjacent and PBS-injected rats for the presence of cluster of series of SN tissue sections at 2 and 6 months after α- differentiation 68 (CD68) which labels both phagocytic syn PFF, RSA, or PBS intrastriatal injection (Fig. 5). microglia and infiltrating macrophages. While a few Quantitation of the number of Iba-1 immunoreactive CD68-ir cells were observed in blood vessels, no CD68-ir (Iba-1ir) microglia in the adjacent SNr revealed no sig- cells were observed in the parenchyma during any of the nificant differences in microglial number due to α-syn time points examined (data not shown). The lack of CD68 PFF, RSA, or PBS injection at either 2 months (Fig. 5a) immunoreactivity in the parenchyma of the SN or stri- or 6 months p.i. (2 months: F = 0.2637, p > 0.05; atum at any time point suggests that the magnitude of (3, 16) 6 months: F = 0.2427, p > 0.05). No significant dif- synucleinopathy and subsequent degeneration produced (3, 10) ferences were observed in microglial soma area in the in the α-syn PFF model does not trigger microglial SNr due to intrastriatal RSA injections at the 2-month phagocytic activity. Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 12 of 18 Fig. 5 SNr microglia exhibiting reactive morphology are associated with pSyn inclusion-bearing neurons in the SNc. a Total number of microglia did not differ between PBS-, RSA-, and PFF-injected animals, p > 0.05. b Microglia soma area in the ipsilateral and contralateral SNr did not differ significantly between animals receiving intrastriatal injections of PBS or RSA, p > 0.05. c Microglia soma area was significantly increased at 2 months in the ipsilateral SNr, when peak numbers of pSyn aggregates are present in nearby SNc neurons as compared to PBS-injected animals, *p = 0.02. d, e Representative images of SN sections dual labeled for Iba-1 immunoreactive microglia (purple) and NeuN-ir neurons (brown) 2 months following either intrastriatal PBS or α-syn PFF-injected rats. α-Syn PFF-injected rats exhibit larger cell bodies and increased number and thickness of processes. f, inset Distribution of microglia soma area measurements 2 months following intrastriatal PBS or RSA injection illustrated as a percent of total microglia quantified. g Distribution of microglia soma area measurements 2 months following α-syn PFF injection illustrated as a percent of total microglia quantified. h Percent of total microglia quantified for soma area analysis with cell body areas > 70 μm at 2 months. Microglia in PFF-injected rats possessed significantly more microglia with cell bodies larger than > 70 μm compared to PBS-injected rats, *p = 0.03. i At 6 months p.i. microglia soma area in the ipsilateral and contralateral SNr did not differ significantly between rats receiving either PBS or α-syn PFF intrastriatal injections, during the interval of ongoing degeneration in the SNc of PFF-injected animals, p > 0.05. j, k Representative images of SN sections dual labeled for Iba-1 immunoreactive microglia (purple) and NeuN-ir neurons (brown) at 6 months p.i. exhibit a hyper ramified morphology, regardless of treatment. l, m Distribution of microglia soma area measurements 6 months p.i in PBS- and PFF-injected rats as a percent of total microglia quantified. Scale bars (d, e, j, k)=25 μm. Data represent mean ± SEM. Abbreviations: p.i. = postinjection; PFFs = pre-formed alpha-synuclein fibrils; PBS = phosphate-buffered saline; RSA = rat serum albumin; Iba-1 = ionized calcium-binding adaptor molecule 1; ir = immunoreactive; NeuN-ir = neuronal nuclei immunoreactive; ipsilateral = ipsilateral hemisphere relative to injection; contralateral = contralateral hemisphere relative to injection; SNc = substantia nigra pars compacta; SNr = substantia nigra pars reticulata; SEM = standard error ofthe mean Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 13 of 18 MHC-IIir microglia in the agranular insular cortex are observed between treatment groups at the 1-month time associated with the accumulation of α-syn inclusions point, but the magnitude of the MHC-II response ap- We also examined the time course of pSyn inclusion ac- peared slightly larger in PFF-injected rats compared to the cumulation and MHC-II expression on microglia in the PBS-injected rats at the 2-month time point. However, at agranular insular cortex, as this region possesses abun- 4 and 6 months, during the interval of continuing accu- dant Lewy-body like pathology in our model and is mulation of pSyn α-syn inclusions in the striatum, the implicated in non-motor symptoms in PD [56]. At number of MHC-IIir microglia decreased dramatically 2 months p.i., we observe abundant pSyn inclusions pri- with no differences observed in the small number of marily localized to the somata (Fig. 6a), with increased MHC-IIir microglia observed in both treatment groups neuritic pathology evident by 4 months (Fig. 6b). Inter- (Additional file 1: Figure S1E, F, H, I). These results sug- estingly, an observable decrease in both neuritic and gest that the acute microglial response to the PFF injectate somata inclusions is evident by 6 months (Fig. 6c). We may differ from the acute response to the surgical injec- observe a similar temporal pattern of MHC-IIir micro- tion alone but that the subsequent increase α-syn inclu- glia as described for the SN: the highest number of sion load within neurons in the striatum does not trigger MHC-IIir microglia is observed at 2 months p.i., when a second wave of microglial MHC-II immunoreactivity. pSyn inclusions first peak, and a decrease in MHC-IIir microglia in association with reductions in the number Discussion of pSyn inclusions. Interestingly, a decrease in MHC-IIir In the present study, intrastriatal injection of α-syn PFFs microglia is observed between 2 and 4 months p.i. when to rats resulted in widespread accumulation of phos- pSyn pathology becomes more abundant with the phorylated α-syn inclusions in several areas that innerv- appearance of neuritic inclusions. These observations ate the striatum, as previously reported in rats and mice suggest that MHC-II is upregulated as a first response to [37, 57]. Further examination of the inclusions formed formation of pSyn inclusions and is not sustained over in the SNc revealed that they share many key features time, despite a secondary increase in synuclein burden. with Lewy bodies and were most abundant between Few to no MHC-IIir microglia were observed in PBS- months 1–3 after intrastriatal α-syn PFF injection, peak- injected animals (Fig. 6g–l), strengthening the concept ing at 2 months. The magnitude of ipsilateral SNc neu- that MHC-IIir on microglia is induced by initial accu- rons bearing α-syn inclusions 1–3 months after α-syn PFF mulation of pSyn. injection approximated the magnitude of loss of ipsilateral SNc neurons observed at 5–6 months, suggesting a direct MHC-IIir microglia in the striatum are not associated with relationship between α-syn inclusion accumulation and the accumulation of α-syn inclusions degeneration of SNc neurons. Synucleinopathy-specific Lastly, we examined the time course of accumulation of MHC-II expression in the ipsilateral SNc similarly peaked pSyn inclusions and number of MHC-IIir microglia in in the SN at 2 months and was associated with a reactive the striatum in rats that received unilateral α-syn PFF or microglial morphology, characterized by significantly lar- PBS intrastriatal injection. As reported previously [37], ger soma size, 3 months prior to degeneration. Surpris- the pattern of pSyn α-syn inclusion accumulation in the ingly, although the period of nigral degeneration was striatum is strikingly different from accumulation in the associated with an increased MHC-II signal relative to SNc. At 2 months, α-syn inclusions in cell bodies in the controls, MHC-II immunoreactivity during the period of ipsilateral striatum are relatively sparse and pSyn α-syn degeneration was significantly decreased relative to immunoreactivity appeared primarily localized to neurites, MHC-II immunoreactivity during the earlier peak of synu- presumably in terminals from the SNc (Additional file 1: cleinopathy. Overall, the temporal pattern of peak Lewy Figure S1A). Over the course of the 4 months, pSyn α-syn body-like inclusion formation was associated with peak inclusions in cell bodies in the striatum increase in num- neuroinflammation in the SN, both of which appear ber, involving the contralateral hemisphere as well, with months prior to loss of SNc neurons. These results suggest the greatest number of pSyn α-syn inclusions observed in that an increase in MHC-II may be a first-response mech- the ipsilateral striatum at the 6-month time point anism to initial accumulation of intracellular α-syn and (Additional file 1: Figure S1C). In adjacent striatal tissue that reactive microglia have the potential to contribute to sections, we examined the temporal pattern of MHC-IIir vulnerability of SNc neurons to degeneration (Fig. 7, left). microglia following unilateral α-syn PFF or PBS intrastria- Within the agranular insular cortex, we observe pSyn tal injection. Early after injection, at 2 weeks, 1 month, primarily localized to the somata at 2 months, followed and 2 months, abundant MHC-IIir microglia were by an increase in neuritic pathology by 4 months, and a observed in the ipsilateral striatum in proximity to the significant decrease overall in pSyn immunoreactivity at injection sites in both PFF- and PBS-injected rats 6 months (Fig. 7, right). This suggests that similar to the (Additional file 1: Figure S1D, G). No differences were SN, neurons in the agranular insular cortex harboring Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 14 of 18 Fig. 6 Microglia expressing MHC-II are associated with pSyn inclusions in the agranular insular cortex. a–c Representative images of pSyn accumulation in the agranular insular cortex at 2, 4, and 6 months p.i. Pathology is primarily localized to the soma at 2 months (a), with increased neuritic pathology by 4months(b) and an observable decrease overall by 6 months, possibly due to death of neurons. c No pSyn inclusions were observed in PBS-injected animals (g–i). d–f A similar pattern was observed in MHC-II expression on microglia. Peak numbers of MHC-IIir microglia were observed at 2 months p.i. (d), decreased at 4 months (e), and virtually absent by 6 months (f). j–l Few to no MHC-IIir microglia were observed in PBS injected animals at any time point, suggesting that MHC-II expression occurs in response to accumulation of pSyn inside neurons. Abbreviations: p.i. = postinjection; pSyn = α-syn phosphorylated at serine 129; PFFs = pre-formed alpha-synuclein fibrils; PBS = phosphate-buffered saline; MHC-II = major-histocompatibility complex-II; ir = immunoreactive inclusions die over the course of 6 months. Moreover, a striatum (Additional file 3: Figure S3). The procedure of similar pattern of MHC-II expression on microglia is ob- intrastriatal injection itself triggered an acute increase in served: numbers of MHC-IIir microglia peak when pSyn MHC-II immunoreactive microglia that appeared to be pathology is most abundant and decrease over time. Al- slightly enhanced by the presence of the PFF injectate; though this has not been systematically quantified in the however, the presence of MHC-II decreased dramatically present study, future studies investigating inflammation over time in all conditions despite an ever increasing α- and degeneration in this area following intrastriatal α- syn inclusion load. Compared to the SNc, the accumula- syn PFF injection are warranted, as the insula has been tion of α-syn inclusions in the striatum is delayed with no implicated in the manifestations of non-motor symp- loss of striatal neurons observed at 6 months [37]and it is toms in human PD [56, 58, 59]. unknown whether degeneration of striatal neurons ultim- Within the striatum, the site of α-syn PFF injection, we ately occurs at later time points. It is unclear whether α- observed a distinctly different pattern of accumulation of syn inclusion load increases and peaks past the 6-month α-syn inclusions compared to the SNc (Additional file 3: time point and whether the presence of MHC-II immuno- Figure S3J). We observed a dissociation between α-syn reactive microglia may have similarly tracked with a future inclusion load and MHC-II immunoreactivity in the peak. The results in the striatum illustrate the necessity of Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 15 of 18 Fig. 7 Regional timelines of synucleinopathy, neuroinflammation, and degeneration in the substantia nigra and agranular insular cortex following intrastriatal α-syn PFF injection. (Left): Early accumulation of phosphorylated inclusions of α-syn (peak at 2 months) in the substantia nigra leads to loss of TH phenotype and eventual loss of nigrostriatal dopamine neurons by 5–6 months p.i. In the SN, the pattern of microgliosis similarly follows that of pSyn: microglia in the adjacent SNr exhibit a reactive morphology at 2 months p.i. when nearby SNc neurons possess the greatest number of SNc pSyn inclusions. Interestingly, MHC-IIir antigen-presenting microglia in the SNc also peak at 2 months p.i., again coinciding with the greatest number of pSyn intraneuronal inclusions and decrease over time to near non-detectable levels during the interval of degeneration, suggesting a relationship between pathological α-syn and inflammation. (Right): Early accumulation of pSyn inclusions occurs between 2 and 4 months, with inclusions primarily localized to the somata at 2 months, an increase in neuritic inclusions at 4 months, and an observable decrease of overall pSyn pathology at 6 months, suggesting that similar to the SN, neurons in the agranular insular cortex harboring inclusions eventually die off. MHC-II immunoreactivity follows a similar pattern to that observed in the SN, with peak expression observed at 2 months p.i., and decreased over the course of 6 months determining the role of an acute inflammatory response to secretion of proinflammatory and anti-inflammatory cyto- local injection and that the magnitude of this response may kines in the SN at time points prior to and following nigral prevent the reactive microgliosis associated with patho- degeneration is warranted. logical α-syn inclusions. Further, these striatal results suggest Human PD studies examining neuroinflammation sug- that α-syn inclusions do not automatically trigger reactive gest involvement of both the local brain immune response microgliosis and that other factors including rate of inclu- and the adaptive immune system in PD [73–75]. In the sion formation, impending cytotoxicity, local environment, present study, we did not examine the possibility of per- or microglia and astrocyte density [42] may be involved in ipheral immune cell infiltration beyond phagocytic CD68+ determining the neuroinflammatory cascade of events. macrophages, which were not detected in the parenchyma In response to persistent neuronal stress and protein in PFF-injected animals at any stage, and data from hu- accumulation such as α-syn aggregation, microglia can man tissue regarding the presence of CD68+ macrophages become chronically activated, proliferate, migrate, se- in human PD is limited [9]. However, microglia are con- crete pro-inflammatory cytokines and reactive oxygen sidered to be the principle antigen-presenting cell within species, and ultimately contribute to neuronal injury in the brain and MHC-II expression is associated with the an uncontrolled, feed-forward manner [20, 60–63]. recognition of CD4+ T-helper cells. It is possible that Microglia can also phagocytose both living and dead CD4+ T cells participate in the response to α-syn inclu- neurons [64–66]. The observation that α-syn inclusion sions in the SN; however, whether the net effect of CD4+ triggered MHC-II expression on microglia in the SN T cells is neurodegenerative or neuroprotective requires prior to degeneration indicates that neuroinflammation further systematic evaluation [76]. In addition, MHC-I ex- may contribute to the mechanism of pathophysiology. pression by SNc neurons was observed in association with However, it is unlikely that microglial activation is the α-syn overexpression [77] and may similarly be expressed sole arbiter of degeneration given that neuronal death by α-syn inclusion-bearing SNc neurons. In addition, it can result from α-syn PFF-induced intraneuronal inclu- was recently shown that T lymphocytes isolated from PD sions in cultures in the absence of microglia [40]. A more patients recognize specific α-syn peptides [78], strength- likely scenario involves neuroinflammation contributing ening the concept that neuroinflammation can be induced to or accelerating nigrostriatal degeneration with proper- by α-syn and potentially involved early in PD progression. ties unique to the nigral environment [67–72]adding to While the concept that MHC-II is involved in PD and the cascade of events. Future studies investigating the correlates with α-syn burden [9] has existed for several Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 16 of 18 decades, our results are the first to systematically evaluate Additional file 3: Figure S3. Antigen-presenting MHC-IIir microglia are the time course of endogenous pSyn accumulation and not associated with peak of intraneuronal inclusions of pSyn in the striatum. Progression of pSyn pathology and MHC-IIir microglia in the microglial MHC-II expression prior to and after nigros- striatum. (a) At 2 months p.i., pSyn inclusions are localized to neurites, triatal degeneration has occurred. Our results indicate that presumably representing terminals from the SNc. (b–c) Over time pSyn MHC-II expression in the SNc is increased at early time inclusions become primarily localized to the soma of striatal neurons. (d) Abundant MHC-IIir microglia in the striatum primarily localized around points in which SNc neurons possess pSyn inclusions, and the α-syn PFF injection site at 2 months. (e–f) MHC-IIir microglia in the is relatively sparse during the interval of nigral degen- striatum are largely absent during continuing accumulation of intraneuronal eration, suggesting that MHC-II is a response to pSyn inclusions at 4 months (e) and 6 months (f) p.i. (g) Intrastriatal injection of PBS results abundant MHC-IIir microglia in the striatum localized near the initial inclusion formation. Future studies investigating site of injection at 2 months p.i., although appearing less abundant than the direct impact of increased or decreased MHC-II MHC-IIir microglia in the striatum of α-syn PFF rats at the same time point expression on the magnitude of degeneration and (d). (h) MHC-IIir microglia are similarly absent from the parenchyma by 4 months (h) and 6 months p.i (i). Scale bars A–I= 50 μm. Abbreviations: what role, if any, peripheral T cells play in disease p.i. = postinjection; PFFs = pre-formed alpha-synuclein fibrils; progression are warranted. PBS = phosphate-buffered saline; pSyn = α-syn phosphorylated at serine 129, Our results and the well-characterized rat α-syn PFF MHC-IIir = major-histocompatibility complex-II immunoreactive. (TIF 112368 kb) model will facilitate future studies to provide key mech- anistic insights into the specific relationship between Abbreviations 6-OHDA: 6-Hydroydopamine; AAV: Adenoassociated virus; CD68: Cluster of pathological α-syn inclusions, neuroinflammation, and differentiation molecule 68; DMS: Dorsomedial striatum; Iba-1: Ionized degeneration in sporadic PD. calcium-binding adaptor molecule-1; IHC: Immunohistochemistry; MHC-II: Major histocompatibility complex II; MPTP: 1-Methyl-4-phenyl-1,2,3, 6-tetrahydropyridine; NeuN: Neuronal nuclei; PBS: Phosphate-buffered saline; Conclusions PD: Parkinson’s disease; PFFs: Alpha-synuclein pre-formed fibrils; pSyn: Alpha-synuclein phosphorylated at serine 129; RSA: Rat serum albumin; Accumulation of intraneuronal inclusions of phosphory- SNc: Substantia nigra pars compacta; SNr: Substantia nigra pars reticulata; lated α-syn induces increased MHC-II expression and re- TH: Tyrosine hydroxylase; VLS: Ventrolateral striatum; α-syn: Alpha-synuclein active microgliosis in the substantia nigra months prior to Funding dopaminergic cell death, suggesting that microglia may be A support was provided by the Department of Translational Science and a contributor to rather than only a consequence of nigral Molecular Medicine, the Neuroscience Graduate Program, National Institute degeneration. These results will facilitate future studies to of Neurological Disorders and Stroke (NS099416), and the Edwin A. Brophy Endowment at Michigan State University. provide key mechanistic insights into the specific relation- ship between pathological α-syn inclusions, neuroinflam- Authors’ contributions mation, and degeneration in sporadic PD. This research project was conceived and organized by MFD, TJC, MGT, KCL, NMK, and CES. It was executed by MFD, CES, TJC, JRP, CJK, KCL, KLP, NMK, DLF, NKP, OLB, JWH, NNV, NKM, and OMAE. The data were analyzed by MFD, CES, and TJC. The manuscript was first written and revised by MFD and CES, Additional files and it was reviewed and critiqued by all authors. All authors read and approved the final manuscript. Additional file 1: Figure S1. Unilateral intrastriatal injection of α-syn PFFs, but not RSA or PBS, induces bilateral cortical and unilateral SNc Ethics approval Lewy-body like inclusions of phosphorylated α-syn (pSyn). (a) pSyn All procedures performed in studies involving animals were in accordance pathology is observed bilaterally in cortical areas after unilateral injection with the ethical standards of the Institute for Animal Use and Care of α-syn PFFs, namely in layers 2/3 and orbital and agranular insular Committee (IACUC) at Michigan State University. cortices. (b) Injection of PBS or (c) RSA did not induce pSyn accumulation. (d) pSyn accumulation in the ipsilateral substantia nigra pars compacta Competing interests (SNc) at 2 months postinjection, with no evidence of pSyn inclusions in the The authors declare that they have no competing interests. contralateral SNc. Scale bars (A–D) = 50 μm. Abbreviations: α-syn = alpha- synuclein; PFFs = pre-formed alpha-synuclein fibrils; PBS = phosphate- Publisher’sNote buffered saline; RSA = rat serum albumin; pSyn = α-syn phosphorylated at Springer Nature remains neutral with regard to jurisdictional claims in serine 129. (TIF 117729 kb) published maps and institutional affiliations. Additional file 2: Figure S2. Unilateral intrastriatal injection of α-syn PFFs induces widespread accumulation of Lewy-body like inclusions of Author details phosphorylated α-syn (pSyn). Representative images illustrating the time Department of Translational Science and Molecular Medicine, Michigan course of pSyn pathology in regions innervating the striatum. (a–c) pSyn State University, 400 Monroe Avenue NW, Grand Rapids, MI 49503-2532, USA. pathology in the ipsilateral agranular insular cortex localized to both Neuroscience Graduate Training Program, Michigan State University, Grand the soma and neurites at 2 months p.i. (postinjection) that over time Rapids, MI, USA. MD/PhD Program, Michigan State University, Grand Rapids, becomes primarily localized to the soma; scale bar = 50 μm, inset = 10 μm. MI, USA. Mercy Health Hauenstein Neuroscience Medical Center, Grand (d–f) Ipsilateral accumulation of pSyn in the substantia nigra peaks at Rapids, MI, USA. Center for Neurodegenerative Disease Research, 2 months and becomes less abundant over time as neurons degenerate; Department of Pathology and Laboratory Medicine, University of scale bar = 200 μm, inset = 25 μm. (g–i) In contrast to other areas, pSyn in Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA. the striatum is primarily localized to neurites at 2 months and becomes Department of Physiology, Emory University School of Medicine, Atlanta, more abundant and localized to the soma over time, scale bar = 50 μm, GA, USA. Neurological Disorders Research Center, Qatar Biomedical Research inset = 10 μm. Abbreviations: α-syn = alpha-synuclein; PFFs = pre-formed Institute (QBRI), Hamad Bin Khalifa University (HBKU), Education City, Qatar. alpha-synuclein fibrils; pSyn = α-syn phosphorylated at serine 129; Life Sciences Division, College of Science and Engineering, Hamad Bin p.i. = postinjection. (TIF 33472 kb) Khalifa University (HBKU), Education City, Qatar. Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 17 of 18 Received: 15 March 2018 Accepted: 20 April 2018 phosphorylation of α‐synuclein and activation of caspase‐9: resemblance to pathogenetic changes in Parkinson's disease. J Neurosci. 2004; 91:451–61. 24. Ulusoy A, et al., Chapter 5 - Viral vector-mediated overexpression of α- synuclein as a progressive model of Parkinson’s disease. In: Björklund A and Cenci MA, editors. Progress in Brain Research. Elsevier; 2010. p. 89–111. References 25. Theodore S, et al., Targeted Overexpression of Human α-Synuclein Triggers 1. Collier TJ, Kanaan NM, Kordower JH. Aging and Parkinson's disease: different Microglial Activation and an Adaptive Immune Response in a Mouse Model sides of the same coin? Mov Disord. 2017;32(7):983–90. of Parkinson Disease. J Neuropathol Exp Neurol. 2008; 67(12): p. 1149–58. 2. Collier TJ, Kanaan NM, Kordower JH. Ageing as a primary risk factor for 26. Oliveras-Salvá M, et al. rAAV2/7 vector-mediated overexpression of Parkinson’s disease: evidence from studies of non-human primates. Nat Rev alpha-synuclein in mouse substantia nigra induces protein aggregation Neurosci. 2011;12(6):359–366 and progressive dose-dependent neurodegeneration. Mol Neurodegener. 3. Mogi M, et al. Interleukin (IL)-1 beta, IL-2, IL-4, IL-6 and transforming growth 2013;8:44. factor-alpha levels are elevated in ventricular cerebrospinal fluid in juvenile 27. Gombash SE, et al. Morphological and behavioral impact of AAV2/5-mediated parkinsonism and Parkinson's disease. Neurosci Lett. 1996;211(1):13–6. overexpression of human wildtype alpha-synuclein in the rat nigrostriatal 4. Lindqvist D, et al. Cerebrospinal fluid inflammatory markers in Parkinson’s system. PLoS One. 2013;8(11):e81426. disease—associations with depression, fatigue, and cognitive impairment. 28. Fischer DL, et al., Viral Vector-Based Modeling of Neurodegenerative Brain Behav Immun. 2013;33:183–9. Disorders: Parkinson’s Disease. In: Manfredsson FP, Editor. Gene Therapy for 5. Gerhard A, et al. In vivo imaging of microglial activation with [11C](R)-PK11195 Neurological Disorders: Methods and Protocols. New York: Springer New PET in idiopathic Parkinson's disease. Neurobiol Dis. 2006;21(2):404–12. York; 2016. p. 367–382. 6. McGeer PL, et al. Reactive microglia are positive for HLA-DR in the 29. Simola N, Morelli M, Carta A. The 6-hydroxydopamine model of Parkinson's substantia nigra of Parkinson's and Alzheimer's disease brains. Neurology. Disease. 2007;11:151–67. 1988;38(8):1285–91. 30. Matsuoka Y, et al. Lack of nigral pathology in transgenic mice expressing 7. Imamura K, et al. Distribution of major histocompatibility complex class human alpha-synuclein driven by the tyrosine hydroxylase promoter. II-positive microglia and cytokine profile of Parkinson's disease brains. Neurobiol Dis. 2001;8(3):535–9. Acta Neuropathol. 2003;106(6):518–26. 31. Ip CW, et al. AAV1/2-induced overexpression of A53T-α-synuclein in the 8. Kordower JH, et al. Disease duration and the integrity of the nigrostriatal substantia nigra results in degeneration of the nigrostriatal system with system in Parkinson's disease. (1460–2156 (Electronic)). Lewy-like pathology and motor impairment: a new mouse model for 9. Croisier E, et al. Microglial inflammation in the parkinsonian substantia nigra: Parkinson’s disease. Acta Neuropathol Commun. 2017;5:11. relationship to alpha-synuclein deposition. J Neuroinflammation. 2005;2:14. 32. Sanchez-Guajardo V, et al. Microglia acquire distinct activation profiles 10. Dijkstra AA, et al. Evidence for immune response, axonal dysfunction and depending on the degree of alpha-synuclein neuropathology in a rAAV reduced endocytosis in the substantia nigra in early stage Parkinson’s based model of Parkinson's disease. PLoS One. 2010;5(1):e8784. disease. PLoS One. 2015;10(6):e0128651. 33. Thakur P, Breger LS, Lundblad M, et al. Modeling Parkinson’s disease 11. Braak H, et al. Staging of brain pathology related to sporadic Parkinson's pathology by combination of fibril seeds and α-synuclein overexpression in disease. Neurobiol Aging. 2003;24(2):197–211. the rat brain. Proc Natl Acad Sci USA. 2017;114(39):E8284–93 12. Kannarkat GT, et al. Common genetic variant association with altered HLA 34. Farrer M, et al. Comparison of kindreds with parkinsonism and alpha-synuclein expression, synergy with Pyrethroid exposure, and risk for Parkinson's genomic multiplications. Ann Neurol. 2004;55(2):174–9. disease: an observational and case-control study. NPJ Parkinsons Dis. 2015;1: 35. Su X, et al. Alpha-Synuclein mRNA Is Not Increased in Sporadic PD and 359–66. Alpha-Synuclein Accumulation Does Not Block GDNF Signaling in 13. Spiller KL, et al., The role of macrophage phenotype in vascularization of Parkinson's Disease and Disease. Models Mol Ther.2017; 25(10): p. 2231– tissue engineering scaffolds. Biomaterials. 2014;35(15): 4477–88. 14. Harms AS, et al. MHCII is required for alpha-synuclein-induced activation of microglia, CD4 T cell proliferation, and dopaminergic neurodegeneration. 36. Zhou J, et al., Changes in the solubility and phosphorylation of α-synuclein J Neurosci. 2013;33(23):9592–600. over the course of Parkinson’s disease. Acta Neurochir. 2011;121(6): p. 695– 15. Liu X, et al. Intracellular MHC class II molecules promote TLR-triggered 37. Paumier KL, et al. Intrastriatal injection of pre-formed mouse alpha-synuclein innate immune responses by maintaining activation of the kinase Btk. fibrils into rats triggers alpha-synuclein pathology and bilateral nigrostriatal Nat Immunol. 2011;12:416. degeneration. Neurobiol Dis. 2015;82:185–99. 16. Cicchetti F, et al. Neuroinflammation of the nigrostriatal pathway during 38. Luk KC, et al. Intracerebral inoculation of pathological α-synuclein initiates a progressive 6-OHDA dopamine degeneration in rats monitored by rapidly progressive neurodegenerative α-synucleinopathy in mice. immunohistochemistry and PET imaging. Eur J Neurosci. 2002;15(6):991–8. J Exp Med. 2012;209(5):975–86. 17. Koprich JB, et al. Neuroinflammation mediated by IL-1beta increases 39. Polinski NK, et al., Best Practices for Generating and Using Alpha-Synuclein susceptibility of dopamine neurons to degeneration in an animal model of Pre-Formed Fibrils to Model Parkinson's Disease in Rodents. J Parkinsons Parkinson's disease. J Neuroinflammation. 2008;5:8. Dis. 2018. 18. Forno LS, et al. Similarities and differences between MPTP-induced parkinsonism and Parkinson's disease. Neuropathologic considerations. 40. Volpicelli-Daley LA, Luk KC, Lee VM. Addition of exogenous alpha-synuclein Adv Neurol. 1993;60:600–8. preformed fibrils to primary neuronal cultures to seed recruitment of 19. Kurkowska-Jastrzebska I, et al. The inflammatory reaction following endogenous alpha-synuclein to Lewy body and Lewy neurite-like 1-methyl-4-phenyl-1,2,3, 6-tetrahydropyridine intoxication in mouse. aggregates. Nat Protoc. 2014;9(9):2135–46. Exp Neurol. 1999;156(1):50–61. 41. Volpicelli-Daley LA, et al. Exogenous alpha-synuclein fibrils induce Lewy 20. Gao HM, et al. Neuroinflammation and alpha-synuclein dysfunction body pathology leading to synaptic dysfunction and neuron death. Neuron. 2011;72(1):57–71. potentiate each other, driving chronic progression of neurodegeneration in a mouse model of Parkinson's disease. Environ Health Perspect. 42. Tarutani A, Suzuki G, Shimozawa A, et al. The Effect of Fragmented 2011;119(6):807–14. Pathogenic α-Synuclein Seeds onPrion-like Propagation. The Journal of Biological Chemistry. 2016;291(36):18675–18688 21. Gomez-Isla T, et al. Motor dysfunction and gliosis with preserved 43. Vaikath NN, et al., Generation and characterization of novel conformation- dopaminergic markers in human alpha-synuclein A30P transgenic mice. specific monoclonal antibodies for α-synuclein pathology. Neurobiol Dis. Neurobiol Aging. 2003;24(2):245–58. 2015; 79: p. 81–99. 22. Koprich JB, et al. Expression of human A53T alpha-synuclein in the rat substantia nigra using a novel AAV1/2 vector produces a rapidly evolving 44. Leys C, et al. Detecting outliers: do not use standard deviation around pathology with protein aggregation, dystrophic neurite architecture and the mean, use absolute deviation around the median. J Exp Soc Psychol. nigrostriatal degeneration with potential to model the pathology of 2013;49(4):764-6. Parkinson's disease. Mol Neurodegener. 2010;5:43. 45. Leys C, et al., Detecting outliers: Do not use standard deviation around the 23. Yamada M, Iwatsubo T, Mizuno Y, Mochizuki H. Overexpression of α‐ mean, use absolute deviation around the median. J Exp Soc Psychol. 2013; synuclein in rat substantia nigra results in loss of dopaminergic neurons, 49(4): p. 764-766. Duffy et al. Journal of Neuroinflammation (2018) 15:129 Page 18 of 18 46. Kim C, et al. Neuron-released oligomeric alpha-synuclein is an endogenous formation in human synucleopathies. J Neuropathol Exp Neurol. agonist of TLR2 for paracrine activation of microglia. Nat Commun. 2003;62(10):1060–75. 2013;4:1562. 72. Urrutia PJ, Mena NP, Núñez MT. The interplay between iron accumulation, 47. Wan OW, Chung KK. The role of alpha-synuclein oligomerization and mitochondrial dysfunction, and inflammation during the execution step of aggregation in cellular and animal models of Parkinson's disease. PLoS One, neurodegenerative disorders. Front Pharmacol. 2014;5:38. 2012; 7(6): p. e38545. 73. Brochard V, et al., Infiltration of CD4+ lymphocytes into the brain contributes to neurodegeneration in a mouse model of Parkinson disease. J 48. Winner B, et al., In vivo demonstration that alpha-synuclein oligomers are Clin Invest. 2009. 119(1): p. 182–92. toxic. Proc Natl Acad Sci USA, 2011. 108(10): p. 4194–9. 74. Jiang S, et al., The correlation of lymphocyte subsets, natural killer cell, and 49. Li JY, et al., Characterization of Lewy body pathology in 12- and 16-year-old Parkinson's disease: a meta-analysis. Neurol Sci. 2017; 38(8): p. 1373–1380. intrastriatal mesencephalic grafts surviving in a patient with Parkinson's 75. Kustrimovic N, et al., Dopaminergic Receptors on CD4+ T Naive and disease. Mov Disord. 2010. 25(8): p. 1091–6. Memory Lymphocytes Correlate with Motor Impairment in Patients with 50. Tanji K, et al., Proteinase K-resistant alpha-synuclein is deposited in Parkinson's Disease. Sci Rep. 2016; 6: p. 33738. presynapses in human Lewy body disease and A53T alpha-synuclein 76. Carson MJ, Thrash JC, Walter B. The cellular response in neuroinflammation: transgenic mice. Acta Neuropathol. 2010; 120(2): p. 145–54. The role of leukocytes, microglia and astrocytes in neuronal death and 51. Liu Z, et al., Neuronal uptake of serum albumin is associated with survival. Clin Neurosci Res. 2006; 6(5): p. 237–45. neuron damage during the development of epilepsy. Exp Ther Med. 77. Cebrian C, et al. MHC-I expression renders catecholaminergic neurons 2016; 12(2): p. 695–701. susceptible to T-cell-mediated degeneration. Nat Commun. 2014;5:3633. 52. Rey NL., et al., Widespread transneuronal propagation of alpha- 78. Sulzer D., et al., T cells from patients with Parkinson's disease recognize synucleinopathy triggered in olfactory bulb mimics prodromal Parkinson's alpha-synuclein peptides. Nature. 2017; 546(7660): p. 656–61. disease. J Exp Med. 2016; 213(9): p. 1759–78. 53. Conde JR, Streit WJ. Effect of aging on the microglial response to peripheral nerve injury. Neurobiol Aging. 2006; 27(10): p. 1451–61 54. Streit WJ, Microglia and neuroprotection: implications for Alzheimer's disease. Brain Res Brain Res Rev, 2005; 48(2): p. 234–9. 55. Streit WJ, et al., Dystrophic microglia in the aging human brain. Glia. 2004; 45(2): p. 208–12. 56. Christopher L, et al., Uncovering the role of the insula in non-motor symptoms of Parkinson's disease. Brain. 2014; 137(Pt 8): p. 2143–54. 57. Luk KC, et al. Pathological alpha-synuclein transmission initiates Parkinson-like neurodegeneration in nontransgenic mice. Science. 2012;338(6109):949–53. 58. Maillet A, et al., The prominent role of serotonergic degeneration in apathy, anxiety and depression in de novo Parkinson's disease. Brain. 2016; 139(Pt 9): p. 2486–502. 59. Cerasa A., et al., Cortical volume and folding abnormalities in Parkinson's disease patients with pathological gambling. Parkinsonism Relat Disord. 2014. 20(11): p. 1209–14. 60. Block ML, Hong JS. Microglia and inflammation-mediated neurodegeneration: multiple triggers with a common mechanism. Prog Neurobiol. 2005;76(2):77–98. 61. Gao HM, et al. Neuroinflammation and oxidation/nitration of alpha-synuclein linked to dopaminergic neurodegeneration. J Neurosci. 2008;28(30):7687–98. 62. Hwang O. Role of oxidative stress in Parkinson's disease. Exp Neurobiol. 2013;22(1):11–7. 63. Wilshusen RA, Mosley RL. Innate and adaptive immune-mediated neuroinflammation and neurodegeneration in Parkinson’s disease. In: Peterson PK, Toborek M, editors. Neuroinflammation and neurodegeneration. New York: Springer New York; 2014. p. 119–42. 64. Hong S, Beja-Glasser VF, Nfonoyim BM, et al. Complement and Microglia Mediate Early Synapse Loss in Alzheimer Mouse Models. Science (New York, NY). 2016;352(6286):712–716. 65. Solga AC, et al., RNA-sequencing reveals oligodendrocyte and neuronal transcripts in microglia relevant to central nervous system disease. Glia. 2015; 63(4): p. 531–548. 66. Park JY, et al., Microglial phagocytosis is enhanced by monomeric alpha- synuclein, not aggregated alpha-synuclein: implications for Parkinson's disease. Glia. 2008; 56(11): p. 1215–23. 67. Farooqui T, Farooqui AA. Lipid-mediated oxidative stress and inflammation in the pathogenesis of Parkinson's disease. Parkinsons Dis. 2011;2011:247467. 68. Kim WG, et al. Regional difference in susceptibility to lipopolysaccharide- induced neurotoxicity in the rat brain: role of microglia. J Neurosci. 2000;20(16):6309–16. 69. Zhang K, Kaufman RJ. From endoplasmic-reticulum stress to the inflammatory response. Nature. 2008;454(7203):455–62. 70. Zhang W, et al. Human neuromelanin: an endogenous microglial activator for dopaminergic neuron death. Front Biosci (Elite Ed). 2013;5:1–11. 71. Andringa G, et al. Mapping of rat brain using the synuclein-1 monoclonal antibody reveals somatodendritic expression of alpha-synuclein in populations of neurons homologous to those vulnerable to Lewy body

Journal

Journal of NeuroinflammationSpringer Journals

Published: May 1, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off