Systemic inflammation activates coagulation and immune cell infiltration pathways in brains with propagating α-synuclein fibril aggregates

Synucleinopathies are a group of diseases characterized by brain aggregates of α-synuclein (α-syn). The gradual accumulation of α-syn and the role of inflammation in early-stage pathogenesis remain poorly understood. We explored this interaction by inducing chronic inflammation in a common pre-clinical synucleinopathy mouse model. Three weeks post unilateral intra-striatal injections of human α-syn pre-formed fibrils (PFF), mice underwent repeated intraperitoneal injections of 1 mg/ml lipopolysaccharide (LPS) for 3 weeks. Histological examinations of the ipsilateral site showed phospho-α-syn regional spread and LPS-induced neutrophil recruitment to the brain vasculature. Biochemical assessment of the contralateral site confirmed spreading of α-syn aggregation to frontal cortex and a rise in intracerebral TNF-α, IL-1β, IL-10 and KC/GRO cytokines levels due to LPS. No LPS-induced exacerbation of α-syn pathology load was observed at this stage. Proteomic analysis was performed contralateral to the PFF injection site using LC-MS/MS. Subsequent downstream Reactome Gene-Set Analysis indicated that α-syn pathology alters mitochondrial metabolism and synaptic signaling. Chronic LPS-induced inflammation further lead to an overrepresentation of pathways related to fibrin clotting as well as integrin and B cell receptor signaling. Western blotting confirmed a PFF-induced increase in fibrinogen brain levels and a PFF + LPS increase in Iba1 levels, indicating activated microglia. Splenocyte profiling revealed changes in T and B cells, monocytes, and neutrophils populations due to LPS treatment in PFF injected animals. In summary, early α-syn pathology impacts energy homeostasis pathways, synaptic signaling and brain fibrinogen levels.


Introduction
Synucleinopathies are a group of neurodegenerative diseases, such as Parkinson's disease (PD) and Multiple System Atrophy (MSA), characterized by the presence of pathological aggregates of the protein α-synuclein (α-syn) in the brain.In PD, α-syn aggregates are found in neurons, termed Lewy bodies (LBs) [1], whereas in MSA, α-syn aggregates are present to a lesser extent in neurons, as neuronal cytoplasmic inclusions (NCIs), and more extensively in oligodendrocytes, called glial cytoplasmic inclusions (GCIs) [2].
It is by now well established that neuroinflammation is a pathological feature associated with synucleinopathies [3], and especially brains from MSA patients are characterized by extensive neuroinflammation and blood-brain barrier (BBB) disruption [4][5][6].Still, how neuroinflammation contributes to the pathological process remains elusive.While some suggest that inflammation is a physiological response to degenerating cells or aggregated α-syn in PD [7,8], others report that it is already present at early disease stages [9,10] and suggest that it contributes to the pathological progression [11].It has also been hypothesized that infections may be a triggering factor for α-syn aggregation, whether it is the infectious agent itself or the inflammatory reaction it elicits [11][12][13][14][15].
Epidemiological studies have linked inflammatory bowel disease to a higher risk of developing PD [16,17] with a protective effect of anti-inflammatory medication [17].This suggests that an inflammatory condition could be an important factor in the early disease development.
To address this, we have used a well-established animal model of synucleinopathy; the pre-formed fibril (PFF) mouse model.PFF-injected mice reproduce the propagation of α-syn pathology under normal physiological conditions, both ipsi-and contralateral to the PFF injected hemisphere, as the endogenous α-syn aggregates, triggered by a relatively small amount of exogenous PFF seeds [18][19][20].Three weeks after unilateral PFF-injection, before α-syn pathology load peaks in most regions [20], we exposed mice to repeated low doses of intraperitoneally injected bacterial endotoxin lipopolysaccharide (LPS), which is a well characterized agent to induce inflammation, both systemically and within the brain [21].
In this study, we used the injected hemisphere to validate, by immunohistochemistry, the extent of regional PFF-induced phosphorylated (p-) α-syn spreading as well as neutrophil activation triggered by LPS, as neutrophil response to LPS have shown involvement in the interaction between the brain and peripheral immune system [22].To explore the molecular processes and pathways affected in the initial stages of α-syn pathology spread, and to understand how infection-induced inflammation might influence these processes, we conducted a proteomics analysis.This analysis was performed on the hemisphere contralateral to the PFF injection site to eliminate any confounding effects of the mechanical injury caused by the injection itself.We assessed the spread of α-syn pathology to the contralateral side of the PFF injection and the LPS-induced neuroinflammation through biochemical assays for α-syn aggregation and cytokine levels.Lastly, to evaluate the systemic immune response, we performed flow cytometry on isolated spleenocytes.

Animals
Animal experiments were performed in accordance with and approved by the Danish Animal Experimentation Council (license no.2018-15-0201-01468).Seven-week-old male C57BL6/NJr mice (Janvier Lab) were housed 6-8 animal pr.cage under a 12:12 h reversed light/dark cycle with access to food (Altromin 1324) and water ad libitum in the vivarium at the Center for Translational Research, Bispebjerg Hospital (DK).

PFF production
Full length, human α-syn (UniProt ID: P37840) was produced in-house at Lundbeck A/S.The PFFs were produced as fibrils type 91 as described by Makky and colleagues [23], fragmented by sonication to produce 30-80 nm fragments, aliquoted and stored at -80 o C. It was verified that the batch contained <0.5 EU endotoxin/mg.
The needle was left for 3 min before removal.Animals were injected subcutaneously with 3.3 mL/kg Temgesic just after the surgery, in the evening, and the next morning (Indivior, #521634).

LPS administration
Each group, receiving PBS-or PFF-injection, was separated in two; one half receiving LPS-treatment and the other an equal volume of saline (10 mL/kg) twice a week for three weeks.The first dose of LPS (E.coli, O111:B4, #L2630, Sigma, diluted in sterile saline) was injected intraperitoneally (IP) (1 mg/kg) 20-25 days after the stereotaxic surgery.

Tissue collection
Less than 24 h after the last LPS-or saline-injection (at week 6 after the stereotaxic surgery), the animals were euthanized by cervical dislocation followed by decapitation.The brain was split in two; the right hemisphere (ipsilateral to the injection) was placed in 10% formalin buffer (CellPath, #1000.5000)and the left hemisphere snap-frozen in powdered dry ice and stored at -80 o C.After two weeks, the formalin fixed hemisphere was cryoprotected in 30% sucrose, then snap-frozen in powdered dry ice, and stored at -80 o C until sectioning.For a randomly selected subset of the mice, the spleen was removed and kept in ice-cold PBS until splenocyte isolation (n(PFF and saline)=3, n(PFF and LPS)=5, n(PBS and saline)=5, n(PBS and LPS)=3).Details on splenocyte isolation is found in Supplementary file B.

Volume fraction estimation of p-α-syn staining
A point grid method inspired by the Cavalieri method [24] was used for estimation of the volume fraction of p-α-syn immunopositive structures in three areas: striatum (STR), pre-frontal cortex (PFC), and the basolateral amygdala (AMYG).For each brain area, 5-12 sections were included per brain.Definitions of brain areas are found in Supplementary file A, Fig. A. 4, and details on microscope and magnification are described in Supplementary file B. Volumes were estimated using the CAST software (v.2020.01,Visiopharm, Hørsholm, Denmark).In short, one grid was used for estimating the total volume (reference volume) of the region of interest (ROI), and another for the volume of p-α-syn immunopositive structures.In each frame of the grid, a point was counted when hitting an immunoreactive speciesvisually recognized as brownin focus.The sections were examined through the entire height of the section.
The volumes were calculated according to following formula.
where a/p is the area per point in the point grid.SSF is the section sampling fraction; that is, the number of sections used for estimations, divided by the total number of sections covering the ROI.
The volume fraction was then calculated by:   =  −−

Neutrophil counts in brain
Brain neutrophils were imaged by an Olympus BX60 microscope equipped with a motorized stage, and with an oil-immersion ×60 objective (numerical aperture = 1.40

Protein Extraction
The frontal region (illustrated in Supplementary file A, Fig. A.2) of the frozen hemisphere was homogenized as described previously [25] with the addition of 1% Phosphatase Inhibitor Cocktail (ThermoFisherScientific, #78420).We prepared protein extracts for the four groups of mice: 1) PFF and saline (n=12); 2) PFF and LPS (n=11); 3) PBS and saline (n=9); 4) PBS and LPS (n=8).Total protein concentrations were measured with the Bradford protein assay and extracts were stored at -α-syn aggregation assay α-syn aggregation was assessed on the homogenates with a Fluorescence Resonance Energy Transferbased assay for α-syn aggregation (PerkinElmer, #6FASYPEG).The standard protocol provided by the manufacturer was followed.In short, protein extracts were diluted 1:8 in 1X lysis buffer to ensure the signal was in the linear range of the assay, and the α-syn antibodies from the kit (d2 and Tb-Cryptate) were diluted 1:1:50 in detection buffer #1. 10 µL of sample and 10 µL of antibody dilution were added to each well.Negative and positive controls were added as suggested in the standard protocol.Samples were run in duplicates.After 20 h of incubation in darkness at RT, the plate was read at excitation λ 337 nm and emission λ 620 nm and 665 nm on the ClarioSTAR Plus (BMG Labtech) apparatus.Relative aggregation signal was calculated as Delta F% in the standard protocol (raw signal is adjusted to well-to-well noise as well as negative control to increase comparability).
Raw data was analyzed with the MaxQuant software (version 1.6.5.0) [26] as label-free and using standard parameters.The UniProt entry UP000589 was used as a reference genome (downloaded on May 23, 2022).Initial analysis of the MaxQuant protein intensity output was conducted using the Perseus software (version 2.0.3.1)[27].A total of 3,794 proteins were detected in the raw MaxQuant output.Contaminants and poor signals marked as "Only identified by site", "Reverse" or "Potential contaminant" were excluded.The intensities were log2 transformed to obtain a normal distribution.
Sample outliers in profile plotsindicating low quality MS outputwere removed from further analysis.One sample in each group, except the LPS-group, was excluded.Proteins were annotated with the mus musculus UniProt data set.Proteins present in less than 50% of the samples in at least one of the experimental groups were removed.A total of 2,005 proteins were left for final analysis.
Pathway enrichment analysis was performed on the filtered protein dataset using the Reactome Gene-Set Analysis (GSA) tool [28] using the PADOG algorithm.Linear models were fitted for each individual protein (protein level ~ PFF + LPS + PFF:LPS) using the R function lm.False discovery

J o u r n a l P r e -p r o o f
Journal Pre-proof 13 rate (FDR) was controlled by adjusting all p-values with the Benjamini-Hochberg correction.The protein levels which were significantly predicted by PFF-treatment were analyzed using STRING Enrichment in Cytoscape [29,30].

PFF-injected mice display α-syn pathology spreading across brain regions
We were able to successfully induce α-syn pathology as shown by p-α-syn + fibers or intracellular inclusions, when looking at the ipsilateral hemisphere of the intrastriatal PFF-injection site (Fig. 1A).
No signal was detectable in PBS-injected mice (Fig. 1B).Distinct regions, such as the PFC, STR, and AMYG presented high pathology load (Fig. 1A).The volume fraction of p-α-syn pathology load was quantified for PFC, STR and AMYG (CV and CE values shown in Supplementary file A, Table A.4) and showed no statistical difference between the LPS-and saline-treated PFF mice for any of these brain regions (two-way ANOVA: Area p>0.05;LPS p>0.05; interaction p>0.05) (Fig. 1C).

Intrastriatal PFF-injections induces aggregated α-syn in the contralateral frontal cortex
We corroborated that the PFF-injection also resulted in α-syn pathology in the contralateral side, by measuring α-syn aggregation levels with the α-syn aggregation assay in the frontal cortex (Fig. 1D).
There was a main effect of PFF injection on the aggregation signal, and as also seen in the ipsilateral site, no effect of LPS (two-way ANOVA: PFF p<0.01;LPS p>0.05).Post-hoc analysis of the aggregation assay signals showed a significant increase in α -syn aggregation levels between control and PFF-injected mice (p<0.01).To ascertain that this was indeed endogenous α-syn aggregation, we validated results by western blot on the same protein extracts using mouse p-α-syn, corroborating the PFF effect (Supplementary file A, Fig. A8).

PFF-injection and LPS-induced neuroinflammation lead to decreased mitochondrial respiration pathways
Proteomics analysis of the frontal cortex contralateral to the PFF-or PBS-injected hemisphere resulted in a total of 2,005 uniquely identified proteins.When fitting a linear model and after FDRcorrection, the levels of 51 proteins were found to be significantly affected by the PFF-injection (Table 1, Fig. 3A), and significantly related to each other according to a protein-protein interaction (PPI) network analysis using STRING (p-value < 1.0*10 -6 , Fig. 4B).Several pathways related to cellular respiration were found in the top 10 most significant GO Process pathways based on STRING Enrichment analysis (Fig. 3B).When taking LPS as the predictive variable in the linear model, no proteins passed FDR correction (Supplementary file C), but when including both PFF, LPS and PFF x LPS, 32 proteins were significantly regulated after FDR correction (Table 2).For 9 proteins, the interaction between PFF and LPS significantly contributed to the model (Supplementary file C) with most of them involved in electron transport, mitochondria respiratory chain and oxidative phosphorylation: ATPase Na+/K+ transporting subunit Alpha2 (Atp1a2) (p<0.01) and subunit Beta Post-hoc analysis with Tukey adjustment showed a reduction of these proteins by the LPS-treatment corresponding to the levels in PFF-injected mice (Fig. 3C).Linear models were not fitted to the four proteins encoded by the genes Fau, C1qb, Ighm, and Slc7a5 as they were completely absent in either PFF-injected groups or LPS-treated groups (Supplementary file A, Fig. A.6).

Overrepresentation of pathways related to complements and coagulation cascade and B-cell receptor signaling in LPS-treated PFF-injected mice
In addition to the enriched pathways related to cellular respiration found in the STRING analysis, a ReactomeGSA pathway analysis on the entire set of proteins further revealed that pathways enriched in PFF-injected mice involved processes related to, among other, mitochondrial apoptotic signaling and neurotransmitter signaling.Furthermore, we found upregulation of pathways related to cell surface interactions at the vascular wall (Fig. 4C and Supplementary file A, Table A.5). Pathways downregulated in LPS-treated non-PFF-injected mice included complement-related pathways as well as processes involving mitochondrial respiration (Fig. 4B and Supplementary file A, Table A.6).
When looking at the effect of LPS treatment on the PFF-injected mice, we found upregulated pathways related to B cell receptor (BCR) signaling, integrin signaling, biosynthesis of inflammationresolving molecules, as well as the complement and coagulation cascade (Fig. 4A and Supplementary file A, Table A.7).Only a few of the enriched pathways overlapped across groups (Fig. 4D).
We further corroborated the enrichment of pathways related to coagulation and integrin signaling, by performing Western blotting for fibrinogen.According to a two-way ANOVA, there was a significant increase in fibrinogen levels in the PFF-injected animals (p<0.01,Fig. 5A).There was further a correlation between α-syn pathology, measured with the aggregation assay, and fibrinogen (p=0.02,r=0.36,Supplementary file A, Fig.

A.7).
To investigate whether overrepresentation of these pathways could be associated with an increase in microglia activation we measured protein levels of Iba1, the production of which is known to increase in LPS-activated microglia [33].Quantitative western blotting showed a significant increase in Iba1 protein levels, induced by both LPS and PFF (Two-way ANOVA, LPS p<0.001,PFF p<0.05) (Fig. 5B).Post-hoc analysis showed significant increase in Iba1 levels upon LPS-treatment of PFF-injected animals compared to saline-treated PFF-injected animals (p<0.05).

Differences in peripheral B-and T-cell compositions after LPS-treatment of PFF-injected animals
When analyzing splenocytes and thereby the systemic immune response, there was a general effect of LPS-treatment on the fraction of B cells (LPS p<0.05,PFF p>0.05), T cells (LPS p<0.05,PFF p>0.05), neutrophils (LPS p<0.001,PFF p>0.05) and monocytes (LPS p<0.05,PFF p>0.05) according to two-way ANOVAs (Fig. 6A).Post-hoc analysis revealed that specifically for PFFinjected mice, LPS induced a significantly increased fraction of B cells (p<0.01) and a significant decrease in the fraction of T cells (p<0.05).For neutrophils, there was a significant LPS-indudced increased fraction in both PBS-(p<0.001)and PFF-injected animals (p<0.0001).The monocyte fraction was unchanged in the PFF-injected animals but decreased in LPS-treated controls (p<0.05)(Fig. 6A).
The neutrophil population was further subclassified according to the presence or abscence of ICAM-1 (also known as CD54) and CXCR1.For neutrophils expressing CXCR1, but not ICAM1, two-way ANOVA showed a trend of PFF-injection to increase levels, independent of the LPS-treatment (LPS p>0.05,PFF p=0.063).For neutrophils expressing ICAM1, but not CXCR1, there was a general effect of the LPS-treatment for increased levels (LPS p<0.01,PFF p>0.05), with post-hoc testing showing a significant increase for PFF-injected animals only (p<0.05) (Fig. 6B).

Discussion
The purpose of this study was to investigate the molecular processes affected by early brain α-syn pathology and the interactive effect of environmental insults such as an infection or sustained systemic inflammation.This is based on previous studies suggesting that chronic low-grade systemic inflammation is implicated in disease initiation and progression in PD patients [16,34].Our findings indicate that PFF-induced α-syn pathology is associated with a shift in the proteome, leading to reduced mitochondrial respiration, which we also observed in the LPS-treated animals.Further, our

J o u r n a l P r e -p r o o f
Journal Pre-proof 20 results suggest presence of immune cell crosstalk between the brain and periphery.The time interval used in our study -6 weeks -was chosen to mimic early stage α-syn pathology, based on previous studies reporting that the pathology load peaks at 6 months after intrastriatal PFF-injection [19,20].
To eliminate any potential masking effects of the PFF lesion, we performed unilateral striatal injections, focusing specifically on the contralateral side.Previous studies mapping temporal and regional spreading of α-syn have demostrated that pathology does spread to the contralateral site, albeit with a lower pathology load [35].Therefore, we assume that we are observing the effects of newly spread, early stage α-syn pathology.Indeed, we confirmed a significantly higher aggregation signal in the contralateral hemispheres of the PFF-injected mice compared to the mice injected with PBS.It has been reported that microglia activation occurs in both the injected and contralateral hemisphere, peaking seven days after the PFF-injection and subsequently decreasing [36], though studies reporting neuroinflammation in the PFF model have not always be consistent [reviewed by 37,38].In our study, we observe a significant effect of PFF on Iba1 protein level in the contralateral hemisphere.An increase in Iba1 protein levels is one of the most common markers of microglial activation [39,40] and has been associated with support of microglial phagocytosis [40], although it should be noted that Iba1 is also expressed in other cells such as infiltrating macrophages [40].
In our model, we induced a neuroinflammatory condition by LPS-treatment.It is well-established that systemic LPS-injections lead to neuroinflammation [39,41,42].According to a recent report, the magnitude of the inflammatory stimulus may determine whether the effect is beneficial or detrimental in terms of the neuropathological development [43].Very low doses of LPS, ranging from 0.2 to 0.9 mg/kg, may induce neuroprotective effects if administered as pre-conditioning, since this shifts the microglia into a neuroprotective phenotype instead of a pro-inflammatory and detrimental one [43].Activated pro-inflammatory microglia release IL-1β [40], and here, we observed increased levels of brain IL-1β accompanied by an increase in Iba1 levels, supporting a pro-inflammatory response.We also observed increased levels of IL-10, that seemed more pronounced in the PFF-injected animals.According to Mizobuchi & Soma [43] this is characteristic of the neuroprotective microglia phenotype that promotes phagocytosis, tissue repair and recovery from pathological inflammation [43].Additionally, we observed that pathways concerning biosynthesis of resolvins and protectins are upregulated in the LPS-treated PFF-injected mice, which may also indicate a process of inflammation resolution in these mice [44].
This may explain why we did not see an aggravating effect of the LPS-treatment on the spreading of α-syn pathology.Similarly, another study on α-syn overexpressing transgenic mice found no potentiating effect of LPS on α-syn pathology one month post-injection [45].However, others reported a sustained neuroinflammation in the α-syn expressing mice five months after a single LPSinjection, an effect which was not observed in wild type mice [46].It appears that the inflammatory insult has long-lasting effecs on brains that are already primed by α-syn pathology.
In accordance with this, we observed that the LPS-treatment induced a peripheral immune response, which was more pronounced for B and T cells, and neutrophils in the PPF-injected animals.Further, treatment with α-syn PFFs appeared to counteract the downregulatory effect of LPS on the monocyte population.Consistent with this, a previous study has reported that intrastriatal PFF-injections impact the numbers and frequencies of leukocyte subtypes in the spleen five months post-injection, suggesting interaction between the central and peripheral immune system [47].Interestingly, LPS induced an increase in the CXCR1 low ICAM1 high -population, indicative of activated neutrophils [48], particulary in the PFF-injected animals.ICAM1 promotes adhesion and transmigration of circulating neutrophils across the BBB [49], and in combination with low CXCR1 expression, it has been suggested to signify reverse transmigrated cells, namely neutrophils that have returned to circulation after infiltration into tissue, including crossing the BBB [22,50].We furthermore observed increased brain levels of the neutrophil-attractant CXCL1, that binds to CXCR1 [51], indicating neutrophil intracerebral recruitment in response to LPS.This was supported by an increase in numbers of neutrophils in the hemisphere ipsilateral to the injection site.Other studies have reported neutrophil infiltration following acute systemic LPS-treatment [22,52].
The proteomics analysis revealed that multiple proteins involved in the citric acid cycle and the electron transport chain were significantly downregulated in the brains of PFF-injected mice, suggesting that mitochondrial respiration is affected by α-syn pathology.Two of the downregulated proteins, Ndufs4 and Uqcrc2 (components of the mitochondrial complex I and III, respectively), are essential regulators of oxidative phosphorylation (OXPHOS), which is the primary pathway for ATP generation [53-55].These results support other studies linking deficits in the OXPHOS-pathway to α-syn pathology [56,57].Moreover, downregulation of the OXPHOS-pathway has previously been reported in brains of PD patients [58,59] and is suggested as a likely indicator of an induced bioenergetic deficiency due to a pathological insult [60].
Another interesting observation is the significant downregulation of pathways related to synaptic signaling in response to PFF-injection.This is evident by a significant decrease in the levels of proteins involved in neurotransmitter signaling and regulation (Discs large homolog 2 (Dlg2) and 4 (Dlg4), Homer protein homolog 1 (Homer1), Sodium-and chloride-dependent GABA transporter 1 (Slc6a1) and 3 (Slc6a11), Stx1b, and excitatory amino acid transporter 1 (Slc1a3)).According to a proteomics study on the insoluble fraction of PFF-treated neurons, PFF-induced aggregates sequester not only mitochondrial but also synaptic proteins [61].Likewise, post-mortem PD brains also show signs of synaptic disturbances [62] with loss of glutamatergic synapses [reviewed by 63], suggesting that the presence of α-syn fibrils may disrupt network connectivity.Indeed, PFF seeding of neurons in vitro leads to disrupted synchronization and frequency of neuronal firing [64].

J o u r n a l P r e -p r o o f
Interestingly, when looking at the PFF and LPS interaction, there was an overrepresentation of pathways related to the complement and coagulation cascade, integrin signaling, and B cell signaling.
This points towards a recruitment of peripheral immune cells in the PFF-and LPS-treated mice.
Consistent with this, it is reported that peripheral administration of LPS stimulate recruitment of CD45 high leukocytes to the brain, and this is enhanced by the presence of α-syn fibrils [65].
Enrichment of complement and coagulation pathways in the PFF-injected model, associated with pathological α-syn induced neurodegeneration, have previously been reported six months postinjection [66].Here we show that upregulation of complement and coagulation pathways is already taking place six weeks after the PFF-injection.
An important component of the coagulation and complement pathways is the blood borne protein fibrinogen.Fibrinogen and fibrin clot formation are implicated in regulating innate immune responses and neuroinflammation in the brain [67] and plasminogen, another component of the fibrin clot pathway, is involved in mediating the communication between the peripheral and central immune systems [68].In our study, PFF-injected mice displayed increased levels of fibrinogen in the brain, independent of the LPS-treatment.This is interesting in light of previous results from our group showing increased levels of fibrinogen in brains of MSA patients [25].Together with these results, we may assume that increased fibrinogen in MSA is caused by the α-syn pathology, and this may be involved in the crosstalk with the peripheral immune system.Accordingly, we see an increase in Iba1 protein levels, suggesting either microglia activation, proliferation or infiltration of peripheral macrophages, in the PFF injected animals exposed to chronic systemic inflammation.Previously, it has been shown that weekly dosing of 1 mg/kg LPS in mice leads to microglial activation, assessed by morphological changes, without observing an increase in Iba1+ cells, and thereby excluding macrophage infiltration [69].Based on this, we can assume that the increase in Iba1 levels we observe most likely is due to microglial activation and not proliferation.
In conclusion, our results suggest that PFF-induced α-syn pathology affects mitochondrial respiration, synaptic signaling and fibrinogen levels in the brain early in the spreading of the pathology.Three weeks with systemic inflammation did not affect brain pathology load, but resulted in upregulation of pathways related to the coagulation cascade and immune cell infiltration in the brain, and increased Iba1 protein levels, suggesting microglia activation.These results support the believe that there is a crosstalk between the peripheral and neuroimmune system and that a conjunction of independent events may have implications for the immune brain response in synucleinopathies.

Highlights
• Oxidative phosphorylation is affected at early stages of α-synuclein pathology • Systemic inflammation induces immune infiltration pathways in α-synuclein brains • α-synuclein pathology leads to increased brain fibrinogen levels • α-synuclein and systemic inflammation increases brain Iba1 levels

Figure 2 .
Figure 2. Effect of LPS on the brain.(A) Forebrain sections ipsilateral to the PFF injection site

Figure 4 .
Figure 4. Pathway analysis with ReactomeGSA.Differentially regulated pathways significant after

Figure 5 .
Figure 5. Fibrinogen and Iba1 levels in brain tissue.(A) Total fibrinogen level in brain protein

Figure 6 .
Figure 6.Systemic immune response.Flow cytometry on splenocytes showed that the relative

Figure 6
Figure6 An overview of linear model formulas and transformations of data is found in Supplementary file A, TableA.3.For cytokines with non-detectable measurement in the non-LPS-treated groups (TNFα and IL10), the data was pooled into LPS-and saline-treated groups and the difference in detection frequency tested with χ2 test of independence (R function chisq.test).
[32]s were incubated on ice with 10 µL FcR Blocking Reagent (Miltenyi Biotec, #130-092-575) and stained with fixable viability dye (ThermoFisherScientific, #65-0865-14) and fluorochrome-conjugated antibodies: CD11b-Alexa-Fluor 488, Ly-6G-PE-eFluor 610, ICAM1/CD54-APC, CXCR1/IL-8R/CD181-PE, CD19-Super Bright 436, and CD3-PE-Cyanine7 (Supplementary file A, Table A.1). Controls were incubated with isotype-control antibodies (Supplementary file A, Table A.1).Samples were incubated for 30 min on ice and centrifuged at 570×g for 5 min at 4°C, and resuspended in PBS + 1% formalin buffer.Flow cytometry was performed with a Beckman Coulter Gallios flow cytometer.Data were collected and analyzed with the Kaluza software (version 2.1).The gating strategy is described in Supplementary file A, Fig. A.1.StatisticsTwo-way ANOVA tests were used to test for the effect of PFF-injection and/or LPS-treatment on the p-α-syn volume fraction and neutrophil counts, flow-cytometry data, cytokine measurements, Western blot analysis, Cisbio aggregation measurements and levels of selected proteins identified by mass spectrometry.In RStudio (v.2022.07.1), the lm function was used to generate linear models, followed by the anova function for statistical test of main effects.The emmeans function from the R package emmeans v.1.8.3[31]was used for post-hoc pairwise comparisons with Tukey adjustment for multiple comparisons.Normality of residuals was examined with qq plots.Data with non-normal residuals were transformed.significancelevel(α) was set at 0.05.No outliers were removed except as stated above for the proteomics data.All graphs show the individual values and the mean ± SEM and were created using the R package ggplot2 v.3.3.6[32]orGraphPad Prism v.9.3.1.

Table 1 .
PFF-injection as a predictor of protein level in the linear model

Table 2 .
PFF, LPS and PFF x LPS as predictors of protein level in the linear model