Chronic Neurodegeneration Receptor Expression on Microglia During

Systemic Inflammation Modulates Fc Receptor Expression on Microglia during Chronic Neurodegeneration

This information is current as Katie Lunnon, Jessica L. Teeling, Alison L. Tutt, Mark S. of September 27, 2021. Cragg, Martin J. Glennie and V. Hugh Perry J Immunol 2011; 186:7215-7224; Prepublished online 13 May 2011; doi: 10.4049/jimmunol.0903833

http://www.jimmunol.org/content/186/12/7215 Downloaded from

References This article cites 55 articles, 14 of which you can access for free at: http://www.jimmunol.org/content/186/12/7215.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on September 27, 2021

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2011 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Systemic Inﬂammation Modulates Fc Receptor Expression on Microglia during Chronic Neurodegeneration

Katie Lunnon,*,†,1 Jessica L. Teeling,*,1 Alison L. Tutt,‡ Mark S. Cragg,‡ Martin J. Glennie,‡ and V. Hugh Perry*

Chronic neurodegeneration is a major worldwide health problem, and it has been suggested that systemic inflammation can ac- celerate the onset and progression of clinical symptoms. A possible explanation is that systemic inflammation “switches” the phenotype of microglia from a relatively benign to a highly aggressive and tissue-damaging phenotype. The current study investigated the molecular mechanism underlying this microglia phenotype “switching.” We show in mice with chronic neurodegeneration (ME7 prion model) that there is increased expression of receptors that have a key role in macrophage activation and associated signaling pathways, including TREM-2, Siglec-F, CD200R, and FcgRs. Systemic inflammation induced by LPS further increased protein levels of the activating FcgRIII and FcgRIV, but not of other microglial receptors, including the inhibitory Downloaded from FcgRII. In addition to these changes in receptor expression, IgG levels in the brain parenchyma were increased during chronic neurodegeneration, and these IgG levels further increased after systemic inflammation. g-Chain–deficient mice show modified proinflammatory cytokine expression in the brain after systemic inflammation. We conclude that systemic inflammation during chronic neurodegeneration increases the expression levels of activating FcgR on microglia and thereby lowers the signaling threshold for Ab-mediated cell activation. At the same time, IgG influx into the brain could provide a cross-linking ligand resulting in excessive microglia activation that is detrimental to neurons already under threat by misfolded protein. The Journal http://www.jimmunol.org/ of Immunology, 2011, 186: 7215–7224.

n chronic neurodegenerative diseases, there is a highly accumulation of misfolded protein, microglial activation, sys- atypical innate immune response within the brain, which is tematic spread of the disease through the brain, and neuronal death I dominated by microglia, the resident macrophages of the and has many features in common with Alzheimer’s disease pa- brain (1). The microglia adopt an activated morphology that may thology (4–6). last for many years prior to the death of the patient. The contri- In murine prion disease, the microglia adopt a highly branched bution of this innate inflammatory response to the progression activated morphology and upregulation of F4/80, CD11b, and of disease is unclear: epidemiological studies suggest that anti- CD68 but surprisingly do not express a proinflammatory pheno- by guest on September 27, 2021 inflammatory drugs may delay the onset and progression of dis- type: instead, the inflammatory mediator milieu is dominated by ease, but clinical trials have yet to show clear benefit (2). Other TGF-b and PGE2 (7–9). This anti-inflammatory profile appears studies have suggested a neuroprotective role of microglia, me- early in the course of disease and persists despite the increasing diated by phagocytosis and clearance of toxic protein aggregates accumulation of misfolded protein and progressive neuronal loss and apoptotic cell debris (3). To define better the contribution (7, 10). This profile is akin to that described when macrophages of the inflammatory response associated with chronic neuro- phagocytose apoptotic cells and develop an anti-inflammatory degeneration to disease progression, we have investigated the in- phenotype (11, 12). nate immune response associated with prion disease in mice, a It is well established that systemic inflammation communicates laboratory model of chronic fatal neurodegenerative disease. The with the brain leading to transient behavioral changes and cytokine disease, initiated by the ME7 prion agent, is associated with the production (13). We previously showed that a systemic LPS challenge in animals with prion disease leads to a rapid switch in microglia phenotype: from an anti-inflammatory or “primed” phe- *Central Nervous System Inflammation Group, School of Biological Sciences, Uni- notype to an aggressive proinflammatory response associated with versity of Southampton, Southampton SO16 6YD, United Kingdom; †National In- stitute of Health Research Biomedical Research Centre, Institute of Psychiatry, increased IL-1b,TNF-a, and IL-6 production, exaggerated behav- King’s College London, London SE1 7EH, United Kingdom; and ‡Tenovus Research ioral changes, neuronal death, and a significantly faster progression Laboratory, Cancer Sciences, School of Medicine, General Hospital, University of Southampton, Southampton SO16 6YD, United Kingdom of the disease (14, 15). The effect of systemic inflammation on brain pathology and disease progression is not restricted to the prion model 1K.L. and J.L.T. contributed equally to this work. and has now been documented in animal models of normal aging Received for publication December 4, 2009. Accepted for publication April 12, 2011. (16), acute neurodegeneration (17), motor-neuron disease (18), and This work was supported by Biogen and the Wellcome Trust. ischemia (19, 20). Importantly, the impact of systemic infections on The sequences presented in this article have been submitted to the Gene Expression human brain pathology has also been observed in patients with Omnibus database under accession number GSE23182. Alzheimer’s disease, stroke, and multiple sclerosis (21–24). These Address correspondence and reprint requests to Dr. Jessica L. Teeling, School of Biological Sciences, University of Southampton, Southampton General Hospital, studies not only suggest that microglia priming is a general phe- Tremona Road, Southampton SO16 6YD, United Kingdom. E-mail address: jt8@ nomenon of brain neuropathology but, importantly, that systemic soton.ac.uk infections can contribute to disease progression in a wide range of Abbreviations used in this article: DAB, diaminobenzidine; PLP, periodate–lysine– diseases by inducing a switch in microglial phenotype. paraformaldehyde; TKO, triple knockout; WT, wild-type. In this study, we aim to unravel the molecular mechanisms in- Copyright Ó 2011 by The American Association of Immunologists, Inc. 0022-1767/11/$16.00 volved in microglial switching to better understand the regulatory www.jimmunol.org/cgi/doi/10.4049/jimmunol.0903833 7216 SYSTEMIC INFLAMMATION MODULATES FcgR EXPRESSION IN BRAIN

Table I. Differential gene expression during ME7 prion disease before and after systemic LPS challenge

Category ME7 versus NBH ME7 + LPS versus ME7 NBH + LPS versus NBH Immune response inﬂammation 13.7 26.6 28.2 Signal transduction 27.3 31.6 25.1 Cellular adhesion 3.7 3.8 3.5 Cell death 3.4 3.9 3.5 Protein turnover 14.7 11.6 11.1 Transcription translation 6.25 6.57 8.37 Mitochondrial 4.7 1.93 3.08 Neuronal 15.9 6.95 8.36 Other 10.98 6.95 8.8 The percentage (%) of genes that are differentially expressed in different categories is shown for ME7 versus NBH animals, ME7 + LPS versus ME7 animals (i.e., the effect of LPS in ME7 animals), and NBH + LPS versus NBH animals (i.e., the effect of LPS in NBH animals). pathways that control microglia in health and disease. Our results Microarray analysis show that IgG FcgR and increased inﬂux of IgG into the brain play The quality of RNA for microarray analysis was determined on an Agilent an important role in microglial switching that may exacerbate neu- BioAnalyzer 2100 (Agilent Technologies), and samples with a 28S/18S

ronal damage and disease progression after systemic inflammation. ratio of 2.0 were used for microarray analysis. RNA was reverse tran- Downloaded from scribed to cDNA and amplified using a WT ovation kit (Nugen Technol- ogies), and cDNA was hybridized to Mouse 430_2.0 Affymetrix chips. Materials and Methods Rosetta Resolver Gene Expression Data was analyzed by the Rosetta Resolver Gene Expression Analysis system and Bioconductor Open Animals and stereotaxic surgery Source software for bioinformatics. All analysis was performed using Female C57BL/6J mice (Harlan, Bicester, U.K.) were bred and maintained a cutoff p value ,0.01. The p value was generated by a modified t test in local facilities. Fcg-chain–deficient mice were maintained in-house (25). called local pooled error, which is frequently used for identifying differ- Mice were housed in groups of 4 to 10, under a 12-h light/12-h dark cycle entially expressed genes with a small number of replicated microarrays http://www.jimmunol.org/ at 21˚C, with food and water ad libitum. To induce prion disease, mice (26). The microarray data were deposited in the National Center for were anesthetized with Avertin (2,2,2-tribromoethanol in tertiary amyl Biotechnology Information Gene Expression Omnibus database (http:// alcohol), and 1 ml of either ME7-derived brain homogenate (ME7 animals) www.ncbi.nlm.nih.gov/geo/) under GEO Series accession number (10% w/v) or normal brain homogenate (NBH animals) was injected GSE23182. stereotaxically and bilaterally at the coordinates from bregma: ante- roposterior, 22.0 mm; lateral, 21.7 mm; depth, 1.6 mm. All procedures Quantitative PCR were performed in accordance with U.K. Home Office licensing. Total RNA was extracted from hippocampus/thalamus tissue using RNeasy mini kits (Qiagen). Any contaminating genomic DNA was degraded using LPS challenges a DNase I enzyme (Qiagen). cDNA was synthesized using RT-gold At 18 wk postinoculation, ME7 animals and control NBH animals were reagents (Applied Biosystems, Warrington, U.K.). Samples were quanti- by guest on September 27, 2021 injected i.p. with 500 mg/kg LPS (Salmonella abortus; Sigma, Dorset, fied against a standard curve derived from normal mouse spleen or brain U.K.) or an equivalent volume of sterile saline as a control. tissue of ME7 animals that received an intracerebral injection of LPS. All results were normalized to the measurement for GAPDH. Primers and Tissue collection probes were designed using Primer Express software and purchased from Sigma Genosys: FcgRI: forward 59-GGAGATGACATGTGGCTTCTAA- Mice were terminally anesthetized and transcardially perfused with 0.9% CA-39, reverse 59-GCCTTGGTGGCATTAACCA-39, probe 59-CCACCG- heparinized saline. For microarray and quantitative PCR studies, tissue was ACTGGAACCCAAAGTAGCAGA-39;FcgRII: forward 59-CTGTCC- collected at 6 and 24 h after LPS or saline treatment. The left dorsal hip- AAGGGCCCAAGTC-39, reverse 59-GCGACAGCAATCCCAGTGA-39, pocampus and underlying thalamic region (∼20 mg) was punched out, placed probe 59-CAATTGTCAATACTGGTAAAGACCTGCT-39;FcgRIII: for- in RNAlater (Qiagen, Crawley, U.K.), and stored at 4˚C prior to RNA iso- ward 59-CAATGGCTACTTCCACCACTGA-39, reverse 59-AGAGCTG- lation. For immunocytochemistry, unfixed brain or spleen tissue was col- CACTCTGCCTGTCT-39, probe 59-AAAAGCAAACAGCAGCAGAATT- lected 24 h after LPS or saline treatment and directly embedded and frozen in 39;FcgRIV: forward 59-ACCCAGTGCAACTAGAGGTCCAT-39, reverse Tissue-Tek OCT compound (Sakura Finetek Europe B.V, Zoeterwoude, The 59-GGGTCCCCCTCCTGGAA-39, probe 59-ACTTAGTGGTCTGAAGC- Netherlands). For collection of fixed tissue, mice were further perfused with AATAGCCAGCCCA-39; g-chain: forward 59-CGCACCCAGGATGATC- periodate–lysine–paraformaldehyde (PLP) and tissue postfixed in PLP for 4– TCA-39, reverse 59-CAGGGCGGCTGCTTGTT-39,probe59-CACCAAA- 6 h, transferred to 30% sucrose in 0.1 M phosphate buffer, and embedded in AGGAGCAAGAACAAGATCACGG-39. Probes were labeled at the 59 end OCT compound. All tissue was kept at 220˚C until further use. with a 69-carboxyfluorescein reporter dye (FAM) and at the 39 end with a 69-

Table II. Differential gene expression during ME7 prion disease before and after systemic LPS challenge

ME7 + LPS versus NBH + LPS versus Category Subcategory ME7 versus NBH ME7 NBH Immune response and Soluble mediators and receptors 5.2 13.1 11.5 inﬂammation Innate immune response 4.1 5.4 7.9 Lymphocyte genes 1.5 3.1 3.9 Other 2.3 5.0 4.9 Total 13.7 26.6 28.2 Signal transduction Cytokine responsive pathways 5.0 15.8 12.8 pathways Kinases 6.3 3.5 2.6 Second messenger generation 1.8 1.9 0.4 Other 14.1 10.4 9.3 Total 27.3 31.7 25.1 The percentage (%) of genes that are differentially expressed in different categories is shown for ME7 versus NBH animals, ME7 + LPS versus ME7animals (i.e., the effect of LPS in ME7 animals), and NBH + LPS versus NBH animals (i.e., the effect of LPS in NBH animals). The Journal of Immunology 7217

Table III. Fold change in gene expression for receptors and adapter as chromagen or by the fluorescently labeled secondary Abs, goat anti- molecules involved in ITAM and ITIM signaling between treatment rabbit Alexa Fluor 546 or rabbit anti-rat Alexa Fluor 488 (Invitrogen). groups Expression of F4/80, Siglec-F, and FcgR was quantified using Leica software. Two images per brain were obtained at 320 magnification and ME7 versus ME7 + LPS NBH + LPS the number of pixels/field of DAB staining of cells and their processes Signaling NBH versus ME7 versus NBH measured. Data were analyzed by two-way ANOVA followed by Bonfer- roni’s post hoc test, and p , 0.05 was considered significantly different. Activation (ITAM) CD200R4 7.2 — — Behavioral changes TREM-2 20.9 — — Changes in species-specific behaviors (open field activity, burrowing, and Gp49a 21.5 2.3 4.6 glucose consumption) were assessed as previously described (15, 27). Dectin-1 91 21.5 — DAP12 11.8 — — Cytokine measurement CD300D 7.4 — — FcRI 4.3 — 2.5 Hippocampal tissue was collected and homogenized in ice-cold buffer (150 FcRIII 6.0 2.5 3.5 mM NaCl, 25 mM Tris, 1% Triton, pH 7.4) containing a protease inhibitor FcRIV 9.4 2.4 — mixture (Roche Diagnostics, Mannheim, Germany). Homogenates were g-Chain 8.5 — — centrifuged for 30 min at 14,000 rpm, and supernatants were assayed for Inhibitory (ITIM) total protein using a Bio-Rad DC protein assay (Bio-Rad, Hertfordshire, Siglec-3 2.8 — — U.K.). Cytokine levels were measured using MSD multiplex for mouse Siglec-F 26.7 — — proinflammatory cytokines (Mesoscale Discovery, Gaithersburg, MD) ac- Lair 3.8 21.6 — cording to the manufacturer’s instructions.

CD300A 4.1 — — Downloaded from CD72 14 — — Statistical analysis FcRII 11.7 2.5 3.5 Data were analyzed by two-way ANOVA. Post hoc one-tailed t tests The fold change in gene expression for various receptors that use ITAM or ITIM (Bonferroni) were performed to analyze individual comparisons. Results for their cell signaling is shown for ME7 versus NBH animals, ME7 + LPS versus are reported as means 6 SEM. ME7 animals (i.e., the effect of LPS in ME7 animals), and NBH + LPS versus NBH animals (i.e., the effect of LPS in NBH animals). Results Upregulation of innate immunity-associated genes in chronic http://www.jimmunol.org/ carboxy-tetramethyl rhodamine quencher dye (TAMRA). Primers and probe neurodegeneration for the housekeeping gene GAPDH were purchased from Applied Bio- systems. To determine the effect of LPS on gene expression, we calculated To ﬁnd candidate genes involved in microglia phenotype switch- the fold-difference in mRNA transcripts between the LPS-treated groups of ing, we performed microarray analysis on brain tissue from control ME7 or NBH animals and those without LPS challenge. (NBH animals) or prion diseased (ME7 animals) animals before and after i.p. LPS challenge and compared the gene expression Immunocytochemistry proﬁle in the hippocampal and thalamic regions. We chose to study Coronal sections (10 mm) were cut on a cryostat. mAbs against mouse the dorsal hippocampus/thalamic region at 18 wk into disease

FcgRI (IgG2b, clone 290305; R&D Systems) and Siglec-F (a kind gift because previous characterization of the pathology at this stage by guest on September 27, 2021 from Prof. P. Crocker, University of Dundee) were used on PLP fixed demonstrated significant neuronal degeneration and extensive tissue, whereas mAbs against FcgRII (AT130-2; in-house), FcgRII/III (FCR4G8; Serotec), FcgRIV (AT137; in-house), CD68 (FA11; Serotec), microgliosis (10, 15). In this model, behavioral deficits in bur- CD11b (5C6; Serotec), F4/80 (rabbit polyclonal, a kind gift from Prof. rowing and open field activity are observed after systemic LPS S. Gordon [University of Oxford]), GFAP (polyclonal; DAKO), and IgG challenge peaking at 6 h and returning to baseline levels at 48 h [F(ab9)2 sheep anti-mouse IgG-FITC; Sigma] were used on alcohol-fixed, (14). Tissue for microarray was taken at 6 h. Quantification of the fresh-frozen tissue. Sections were blocked with 10% rabbit serum/1% BSA, incubated with primary Abs for 1 h, followed by biotinylated sec- microarray data showed that 2336 genes (5.8% of total of 40,000 ondary Abs (Vector Lab) for 1 h at room temperature. Staining was either genes) were differentially expressed in ME7 animals compared visualized by the ABC method (Vector Lab) and diaminobenzidine (DAB) with the expression of genes in NBH animals. Analysis of the data

FIGURE 1. Heat map of ﬂuorescent intensity of Affymetrix FcgR probe IDs. The ﬂuorescent intensity measured in each sample (n = 3) for each of the probes corresponding with the four murine FcgRs and the common Fcg-chain is displayed. Green color in the heat map indicates relatively low transcript expression, and red color indicates relatively high transcript expression. 7218 SYSTEMIC INFLAMMATION MODULATES FcgR EXPRESSION IN BRAIN

FIGURE 2. Impact of systemic inflammation on CNS expression of selected immune regulatory receptors in NBH animals and ME7 animals. mRNA expression levels for selected receptors were measured by quantitative PCR in hippocampal/thalamic tissue taken from NBH animals or ME7 animals at 18 Downloaded from wk into the disease and 24 h after a systemic challenge with LPS (NBH + LPS and ME7 + LPS; see Materials and Methods for details). Data for FcgRI–IV and CD68 are presented as mean 6 SEM of two independently performed experiments with NBH/ME7 saline n = 8, NBH + LPS n = 7, and ME7 + LPS n = 5 (total number of animals: 28). Data for the neonatal Fc receptor, FcRn, are presented as mean 6 SEM of two independently performed experiments with NBH saline n = 8, ME7 saline n = 6, NBH + LPS n = 6, and ME7 + LPS n = 6 (total number of animals: 26). Data for TREM-2 are presented as mean 6 SEM of one experiment with NBH saline n = 5, ME7 saline n = 6, NBH + LPS n = 6, and ME7 + LPS n = 5 (total number of animals: 22). Data for Siglec-F are presented as mean 6 SEM of one experiment with NBH saline n = 6, ME7 saline n = 5, NBH + LPS n = 4, and ME7 + LPS n = 5 (total number of animals: 20). *p , 0.05,**p , 0.01, ***p , 0.001 (two-way ANOVA followed by Bonferroni posttest). http://www.jimmunol.org/ shows that 41% of the genes that were upregulated are linked to through GEO Series accession number GSE23182 (http://www. the immune response or cell signaling (Table I). A systemic LPS ncbi.nlm.nih.gov/geo/). challenge induced a further 1118 genes in ME7 animals of which We used quantitative PCR to measure FcgR gene expression a large proportion have a function in immune responses or sig- before and after LPS challenge and compared this with TREM-2 naling (58.2%; (Table I) with many genes (28.9%) encoding sol- and Siglec-F gene expression. Tissue taken from the dorsal uble mediators/receptors (13.1%) and cytokines (15.8%) (Table hippocampus/thalamus after 6 h (data not shown) or 24 h (Fig. 2) II). A selection of genes that were upregulated in ME7 animals is was analyzed. At 24 h, ME7 animals showed a significant up- by guest on September 27, 2021 shown in Table III. We found significant upregulation of receptors regulation of FcgRI mRNA levels [F(1,24) = 13.77, p = 0.01], and that signal through an ITAM, including TREM-2 (21-fold), gp49a systemic administration of LPS induced a small but significant (22-fold), Dectin-1 (91-fold), CD200 (7-fold), FcgRI (4-fold), further increase [F(1,24) = 6.208, p = 0.02]. There was no signifi- FcgRIII (6-fold), and FcgRIV (9-fold) and the ITAM-containing cant interaction between disease and LPS treatment [F(1,24) = adapter molecules g-chain (8.5-fold) and DAP12 (12-fold). Re- 2.357, p = 0.138], suggesting that LPS treatment increased FcgRI ceptors that provide inhibitory signals through an ITIM were in brains of control NBH animals and ME7 animals. FcgRII was also increased, including CD72 (14-fold), Siglec-F (27-fold), and significantly upregulated in ME7 animals [F(1,24) = 26.49, p , the inhibitory FcgRII (12-fold). Systemic LPS challenge further 0.0001] and further increased after a systemic LPS challenge increased the expression of all FcgRs, apart from FcgRI, whereas [F(1,24) = 20.29, p = 0.0001]. Similar observations were made for the majority of other immunoregulatory receptors did not change FcgRIII and FcgRIV: ME7 animals showed increased FcgRIII or showed a downregulated expression pattern. A heat map dis- [F(1,24) = 7.692, p =0.011]andFcgRIV mRNA expression [F(1,24) = playing the individual fluorescent intensities for the Affymetrix 5.785, p = 0.024], and expression further increased after LPS probe set IDs corresponding with each of the four murine FcgRs challenge [FcgRIII: F(1,24) = 7.813, p = 0.01; FcgRIV: F(1,24) = and the common Fcg-chain is shown in Fig. 1, demonstrating the 7.563, p = 0.011]. All three low-affinity FcgRs had a significant transcription of genes encoding all FcgRs was higher in brain disease and LPS interaction [FcgRII: F(1,24) = 18.86, p = 0.0002; tissue from ME7 animals than in control NBH animals and veri- FcgRIII: F(1,24) = 7.121, p = 0.013; FcgRIV: F(1,24) = 4.386, p = fying further increase of FcgRII, FcgRIII, and FcgRIV after a 0.047], indicating that ME7 animals have a heightened central systemic LPS challenge. The data discussed in this article have response to LPS challenge. Finally, analysis of the neonatal Fc been deposited in the National Center for Biotechnology In- receptor, FcRn, also showed increased expression during ME7 formation Gene Expression Omnibus database and are accessible prion disease [F(1,22) = 11.90, p = 0.0023], and further

Table IV. Fold change in gene expression for Fcg receptors

ME7 + LPS versus NBH + LPS versus ME7 + LPS versus NBH + LPS versus Receptor ME7 Versus NBH ME7 6 h NBH 6 h ME7 24 h NBH 24 h FcRI 3.6 1.8 3.1 2.4 2.2 FcRII 10.8 2.3 2.3 10.9 3.0 FcRIII 11 5 7.8 48.2 13.2 FcRIV 3.4 5.1 — 13.8 10.8 Data show the fold change in gene expression for FcgRs between ME7 and NBH animals; the effect of LPS in ME7 or NBH animals after 6 h; and the effect of LPS in ME7 and NBH animals after 24 h. The effect of LPS is expressed as fold change of ME7 or NBH alone. The Journal of Immunology 7219

FIGURE 3. Expression of F4/80 in the thalamus of NBH animals and ME7 animals before and after i.p. LPS challenge. F4/80 protein expression was visualized by immunohistochemistry in NBH animals (A), ME7 animals (B), NBH + LPS animals (C), and ME7 + LPS animals (D). Mice were treated with 500 mg/kg LPS and F4/80 expression levels assessed 24 h later as described in Materials and Methods. A representative example of one independent experiment with n = 3 mice per group is shown. Expression levels (DAB staining/ﬁeld) were quantiﬁed and analyzed as described in Materials and Methods (n = 3). **p , 0.01 (two-way ANOVA, followed by Bonferronni posttest).

upregulation after LPS [F(1,22) = 10.55, p = 0.0037]. Analysis of number of microglia. We quantified CD68 gene expression before TREM-2 and Siglec-F mRNA expression levels confirmed the and after LPS challenge: ME7 animals showed a significant in- microarray data: an increase in TREM-2 expression was found crease in CD68 compared with control NBH animals [F(1,24) = in ME7 animals [F(1,18) = 4.353, p , 0.0514], but there was no 22.30, p , 0.0001], with no further changes in CD68 mRNA Downloaded from further increase after LPS challenge [F(1,18) = 0.421, p = 0.525]. levels in ME7 animals after LPS challenge [F(1,24) = 0.1755, p , Similarly, Siglec-F was significantly upregulated in ME7-infected 0.679]. Therefore, changes in FcgR mRNAs after systemic LPS mice [F(1,17) = 72.53, p , 0.0001], and no further increase was challenge are not explained by more microglial cells alone. In seen after LPS challenge [F(1,17) = 0.09, p = 0.77]. summary, Table IV shows that mRNA levels of all four FcgRs The neuropathology in the hippocampus and thalamus of ME7 increase during chronic neurodegeneration, and, with the excep-

animals is characterized by extensive microgliosis, and therefore tion of FcgRI, these changes are markedly increased 6 and 24 h http://www.jimmunol.org/ changes in FcgR mRNA levels may be simply due to increased after a systemic LPS challenge. by guest on September 27, 2021

FIGURE 4. Characterization of novel mouse FcgR Abs. A, Specificity of anti-FcgRII mAb AT130-2. FcgRII2/2 mice were immunized with an FcgRII– CD4 fusion protein and the resulting hybridomas tested by flow cytometry. AT130-2 was assessed for binding to lymphocytes from WT, FcgRII2/2,or FcgRIII2/2 mice. B–D, Specificity of anti-FcgRIV mAb AT137. Rats were immunized with an FcgRIV–CD4 fusion protein and the resulting hybridomas tested by ELISA and flow cytometry. Binding of AT137 to various concentrations of plate-bound CD4 fusion proteins (0.5, 1, and 2 mg/ml) displaying FcgRI, FcgRII, FcgRIII, or FcgRIV extracellular domains (B). Costaining of peripheral blood monocytes from WT or FcgRI, II, III2/2 (TKO) mice with CD11b–PE and either AT137, 2.4G2, or an irrelevant isotype matched mAb (C). Staining of thioglycolate-elicited peritoneal macrophages from WT or g-chain2/2 mice with either AT137, 2.4G2, or an irrelevant isotype matched mAb (D). E, Light microscopy images of spleen sections stained for FcgRs. Fresh-frozen spleen sections (10 mm) from naive mice were stained for FcgRI, FcgRII, FcgRII/III, and FcgRIV expression using immunohistochemistry. Double fluorescence images of spleen sections stained for FcgRs (green) and F4/80 (red) show colocalization of FcgRI and FcgRIV with macrophages/ monocytes. A representative example of one independent experiment with n = 3 mice per group is shown. Scale bars, 20 mm. 7220 SYSTEMIC INFLAMMATION MODULATES FcgR EXPRESSION IN BRAIN

FIGURE 5. Expression of FcgRI in the thalamus of NBH animals and ME7 animals before and after i.p. LPS challenge. FcgRI protein expression was visualized by immunohistochemistry in NBH animals (A), ME7 animals (B), NBH + LPS animals (C), and ME7 + LPS animals (D). Mice were treated with 500 mg/kg LPS and FcgRI expression levels measured 24 h later as described in Materials and Methods. A representative example of one independent experiment with n = 3 mice per group is shown. Expression levels of FcgRI (pixels/ﬁeld) were quantiﬁed and analyzed as described in Materials and Methods (n = 3). ***p , 0.001 (two-way ANOVA).

Protein levels of FcgRIII and FcgRIV are increased in chronic Peripheral blood monocytes from WT or FcgRI, II, III2/2 (triple neurodegeneration and further upregulated after a systemic knockout; TKO) mice were stained with CD11b–PE and either an LPS challenge irrelevant isotype matched mAb, 2.4G2, or AT137 (Fig. 4C).

Because tissue from ME7 animals analyzed 24 h after LPS These data demonstrate that the dual (FcgRIII and II) specificity Downloaded from challenge revealed further increased mRNA levels in FcgR, we mAb 2.4G2 loses all binding in the TKO mouse as expected, analyzed protein expression levels at the same time point. Brain whereas AT137 retains binding on these cells, indicating its sections from NBH animals and ME7 animals were first stained binding to an FcgR other than I, II, or III. Thioglycolate-elicited 2/2 for well-known microglial activation markers. The characteristic peritoneal macrophages from WT or g-chain mice were morphology of quiescent microglia and perivascular macrophages stained with either AT137, 2.4G2, or an irrelevant isotype matched with low a level of F4/80 expression was observed in NBH ani- mAb. All surface binding of AT137 is lost on cells from the http://www.jimmunol.org/ 2/2 mals, whereas ME7 animals showed an increase in both cell g-chain mice, unlike 2.4G2, which retains binding because of number and F4/80 expression levels (Fig. 3A,3B). Quantification its ability to bind FcgRII (Fig. 4D). Finally, immunocytochemistry revealed that F4/80 immunoreactivity was significantly higher in indicated that FcgR expression pattern in the spleen was as g g ME7 animals than that in NBH animals [F(1,8) = 18.8, p = 0.003], expected, with Fc RI and Fc RIV colocalizing with F4/80- with no further significant increase after LPS (Fig. 3). Similar positive macrophages (Fig. 4E, lower panel). Altogether, these findings were obtained when brain sections were stained for data demonstrate specificity of AT137 for FcgRIV. CD11b or CD68 (data not shown). These data confirm an increase Next, we stained brain tissue from NBH animals and ME7 in both microglial number and/or activation in ME7 animals. animals for FcgR expression before and after systemic LPS To analyze FcgR protein expression, we used a panel of anti- challenge. FcgR expression was low or undetectable in the brains by guest on September 27, 2021 mouse FcgR Abs, including in-house–generated anti-FcgRII of control NBH animals, and no changes were observed after LPS, (AT130-2) and FcgRIV (AT137), and tested their specificity in a with the exception of FcgRI, which modestly increased 24 h after number of cell-based assays (Fig. 4). FcgRII2/2 mice were im- LPS (Fig. 5). In contrast, ME7 animals show high levels of FcgRI munized with an FcgRII–CD4 fusion protein and the resulting [F(1,8) = 35.6, p = 0.0003] and FcgIII on cells with characteristic hybridomas tested by flow cytometry for reactivity against morphology of “activated” microglia (Figs. 5, 6) with FcgRIII lymphocytes from wild-type (WT), FcgRII2/2,orFcgRIII2/2 being further increased 24 h after LPS (Fig. 6C,6D). ME7 animals mice. Fig. 4A shows loss of binding of AT130-2 to lymphocytes showed detectable but very low expression levels of both FcgRII from FcgRII2/2 but not FcgRIII2/2 mice. To generate FcgRIV- and FcgRIV, but FcgRIV levels notably increased after systemic specific Abs, rats were immunized with an FcgRIV–CD4 fusion LPS challenge (Fig. 6C,6D). Dual-labeling immunofluorescence protein and the resulting hybridomas tested by ELISA and flow identified these cells as F4/80-positive microglia (Fig. 6D). Siglec- cytometry. Fig. 4B shows that AT137 specifically binds to FcgRIV. F protein levels were not detectable in brains of NBH animals,

FIGURE 6. Fluorescent confocal images of FcgRII, FcgRIII, and FcgRIV expression levels in the thalamus of NBH animals and ME7 animals before and after i.p. LPS challenge. FcgR protein expression was visualized by immunohistochemistry in NBH animals (A), NBH + LPS animals (B), ME7 animals (C), or ME7 + LPS animals (D). FcgR protein expression was visualized by immunoﬂuorescent staining of FcgRII (AT130-2), FcgRII/III (FCR4G8), or FcgRIV (AT137) (red) and costained for F4/80 (green). Images of both FcgR and F4/80 alone are shown, and colocalization between FcgR and microglia is shown in a merged image. A representative example of one independent experiment with n = 3 mice per group is shown. Scale bars, 20 mm. The Journal of Immunology 7221

FIGURE 7. Expression of Siglec-F in the thalamus of NBH animals and ME7 animals before and after i.p. LPS challenge. Siglec-F protein expression was visualized by immunohistochemistry in NBH animals (A), ME7 animals (B), NBH + LPS animals (C), and ME7 + LPS animals (D). Scale bar, 20 mm. Mice were treated with 500 mg/kg LPS and Siglec-F expression levels measured 24 h later as described in Materials and Methods. A representative example of one independent experiment with n = 3 mice per group is shown. Expression levels of Siglec-F (DAB stained pixels/field) were quantified and analyzed as described in Materials and Methods (n = 3). ***p , 0.001 (two-way ANOVA). whereas ME7 animals in both the hippocampus (data not shown) 8.00, p = 0.012] in WT ME7 animals (Fig. 10), confirming and thalamus (Fig. 7) showed increased expression of Siglec-F microglial switching in the experiment. In contrast, ME7-infected [F(1,8) = 208.9, p , 0.0001]. LPS did not significantly increase g-chain–deficient mice do not increase expression of proin-

Siglec-F expression [F(1,8) = 1.575, p = 0.14]. flammatory cytokine IL-1b after systemic infection, and levels are Downloaded from comparable with those of NBH animals receiving the same dose of Increased IgG levels in ME7-infected brain during systemic LPS (Fig. 10). In addition, ME7-infected g-chain–deficient mice inflammation show altered levels of IL-10 and IL-12 compared with those IgG levels are very low in the normal, healthy brain, due to an intact of ME7-infected WT animals. These differences in cytokine blood–brain barrier. Immunocytochemistry demonstrated that IgG responses in the brains of g-chain–deficient mice are not due to

was undetectable in the brain parenchyma of saline-treated NBH lack of systemic response to the LPS challenge, as serum levels of http://www.jimmunol.org/ animals, and the very low levels that were detected were restricted IL-1b were comparable between the two mouse strains (Fig. 10A). to the cerebral blood vessels (Fig. 8A). Increased levels of IgG were found in the cerebral vasculature and on cells in the parenchyma of ME7 animals (Fig. 8B), consistent with previous data showing increased permeability in the blood–brain barrier in ad- vanced prion disease (28). Notably, after a systemic LPS challenge in ME7 animals, IgG was no longer confined to the blood vessels, and 6 h later increased levels of IgG were found diffusely throughout the brain parenchyma (Fig. 8D). IgG levels were still by guest on September 27, 2021 increased 24 h later (Fig. 8F), and this was associated with microglia cells. These results suggest that the blood–brain barrier is compromised in ME7 animals with increased influx of IgG during periods of systemic inflammation.

A role for FcgR in microglial switching? Thus far, our data are consistent with the observation that increased IgG and expression of activating FcgRs during chronic neurodegeneration play a role in microglial phenotype switching after systemic inflammation. To investigate this further, we inoculated WT and g-chain–deficient mice with ME7 prion agent. To test if onset and disease progression of ME7 prion disease depends on the Fcg-chain, we weekly assessed open field activity, burrowing, and glucose consumption. No major differences were found between the two mouse strains, and the increase in open field activity, decline in burrowing, and a decline in glucose consumption after 12 wk is as previously described (29) (Fig. 9A). We collected brain tissue from ME7-infected WT and ME7-infected g-chain– deficient mice and analyzed the tissue for region-specific microgliosis typical of the late-stage ME7-induced prion disease. No differences in CD68 expression levels between ME7-infected WT and ME7-infected g-chain–deficient mice were found (Fig. 9B), suggesting that onset and progression of ME7-induced prion disease does not depend on the Fcg-chain. We next investigated if FIGURE 8. IgG levels are increased in the parenchyma of ME7 animals LPS-induced microglial switching depends on FcgR interaction. after a systemic inflammatory challenge. Levels of IgG were determined At 18 wk postinoculation, we systemically challenged NBH ani- by immunohistochemistry in brains of NBH animals (A), ME7 animals (B), mals, ME7-infected WT animals, and ME7-infected g-chain–de- NBH + LPS animals at 6 h (C), ME7 animals at 6 h (D), NBH + LPS ficient mice with LPS and measured proinflammatory cytokine animals at 24 h (E), and ME7 + LPS animals at 24 h (F). A representative levels in the brain as a measure of microglial switching. We found example of one independent experiment with n = 3 mice per group is that LPS results in increased levels of IL-1b and IL-12 [F(2,8) = shown. Original magnification 340. 7222 SYSTEMIC INFLAMMATION MODULATES FcgR EXPRESSION IN BRAIN

FIGURE 9. ME7 prion disease in WT and g-chain–deficient mice. WT C57/BL6 or g-chain–deficient mice on a C57BL6 background were inoculated with ME7 prion disease as described in Materials and Methods and behavioral changes assessed weekly. A, Open field activity, burrowing, and glucose consumption of WT animals (open symbols) and Fcg- chain–deficient animals (closed symbols). Data are presented as mean 6 SEM with n = 4 (WT) or n =9(g-chain–deficient) mice per group. *p = 0.02 (unpaired t test). B, Microgliosis. Levels of CD68 expression were determined by immunohistochemistry in brains of WT control, ME7- infected WT, and ME7-infected g-chain– deficient mice. A representative example of one independent experiment with n =3

(WT) or n =4(g-chain–deﬁcient) mice Downloaded from per group is shown. Scale bar, 75 mm.

In our study, increased levels of IgG were found on CD68+ cells resulting in increased proinflammatory cytokine production in the parenchyma of WT ME7 animals, whereas IgG levels and irreversible neuronal damage (15, 30). It is now clear that remained largely associated with blood vessels in g-chain–de- “primed” microglia are seen in a variety of neurologic diseases, all ficient mice without colocalization of CD68+ cells after LPS of which are exaggerated by systemic inflammation, but the http://www.jimmunol.org/ challenge (Fig. 10B). These results further suggest a role of ac- mechanism underlying these changes remains unknown (24). tivating FcgR in microglial switching during chronic neuro- In chronic neurodegeneration, microglia have the capacity to clear degeneration. cell debris and apoptotic cells (31) and become vulnerable to activation to an innate immune challenge due to increased expres- Discussion sion of phagocytic receptors (11, 12, 17, 31). The current study is The neuropathology of murine prion disease is characterized by consistent with the hypothesis that microglial phenotype switching extensive neuronal degeneration, the accumulation of misfolded is initiated by immune receptors expressed on macrophages, and

protein, an increase in the number of morphologically activated their subsequent activation depends on the signaling cascades en- by guest on September 27, 2021 or “primed” microglia and astrocytes, but no typical proin- gaged. In this study, we focused on well-described receptors that flammatory cytokines. Systemic inflammation “switches” this potently regulate macrophage function and, depending on their benign phenotype to an aggressive proinflammatory phenotype cytoplasmic domains, either activate or inhibit macrophage function

FIGURE 10. Cytokine expression levels and IgG deposition in the brains measured 24 h after systemic LPS challenge. ME7 prion disease was induced in WT C57/BL6 or g-chain–deficient mice on a C57BL6 background. At 18 wk into the disease, animals received a systemic injection of LPS, and hippocampal/ thalamic tissue was collected 24 h later. A, Cytokine expression levels. IL-1b, IL-10, and IL-12 (total) expression levels were measured in brain homogenates by multiplex ELISA as described in Materials and Methods. Data are presented as mean 6 SEM with n = 3 (WT) or n =5(g-chain–deficient) mice per group. **p , 0.01 (one-way ANOVA followed by Bonferroni posttest). B, IgG deposition. Levels of IgG were determined by immunohistochemistry in brains of ME7-infected WT or g-chain–deficient mice 24 h after LPS challenge and costained with CD68 (microglia) or laminin (blood vessels) and DAPI (blue). A representative example of one independent experiment with n = 3 (WT) or n =5(g-chain–deficient) mice per group is shown. Scale bar, 75 mm. The Journal of Immunology 7223 through phosphorylation of their receptors bearing ITAM or ITIM FcgRII mRNA was found to be increased in ME7 brains and motifs. Various cellular activities are controlled by ITAM and ITIM further increased after systemic LPS. A previous in vitro study signaling pathways (32), and the balance between these opposing showed that macrophages cultured in the presence of TNF-a show motifs determines whether and to what extent cells become acti- a reduced surface expression of FcgRII, while at the same time the vated. For example, proinflammatory cytokines can significantly mRNA of this receptor was strongly increased as a result of in- increase the expression of receptors that signal through ITAM creased mRNA stability. It is possible that a similar post- motifs, lowering the threshold for activation (33). Microglia express transcriptional regulation of FcgRII explains the lack of dif- several members of this family, including FcgR, TREM-2, SIRPb1, ferential protein expression in our model (50). To study if ac- DAP12, and CD200R (34–36), and there is evidence that the loss of tivating FcgRs have a functional role in switching microglial inhibitory signals results in microglial priming, making them more phenotype, we induced prion disease in g-chain–deficient mice. prone to activation (37). The current study shows increased ex- We show that systemic inflammation no longer induces expression pression of both ITAM- and ITIM-bearing receptors in ME7 ani- of IL-1b in the brain, strongly suggesting a lack of microglial mals at the mRNA level and on microglia at the protein level. switching in g-chain–deficient mice in response to a systemic LPS Systemic inflammation induced further changes in the expression challenge. The IL-10 levels were increased to a greater extent in levels of ITAM-associated FcgRs and gp49a expression, whereas g-chain–deficient mice compared with WT ME7 animals, but other receptors and adapter molecules remained unaffected or without information on baseline levels of this cytokine, it is un- downregulated. Gp49a, a type I transmembrane glycoprotein, does clear if this observation is functionally relevant. Serum levels of not have its own signaling motif and is strongly associated with the IL-1b were comparable between WT and g-chain–deficient mice,

ITAM-bearing g-chain for its function. It is expressed on the sur- ruling out a lack of systemic response to LPS. Furthermore, a Downloaded from faces of mast cells, NK cells, and macrophages. Cross-linking of study from Le et al. (51) shows that microglia from WT or Fcg2/2 gp49a results in calcium mobilization, eicosanoid production, and mice release similar levels of TNF-a after LPS stimulation, cytokine gene transcription, induced by signaling through the whereas IgG immune complexes activate WT cells only. We ob- g-chain, similar to activating FcgRs (38). served reduced IgG deposition on parenchymal cells in g-chain– deficient mice, likely due to lack of FcgR expression. These The role of FcgRs in chronic neurodegeneration results suggest a role for activating FcgR in LPS-induced micro- http://www.jimmunol.org/ FcgRs can be divided into two types, and mice express four dif- glial switching; however, as IL-12 and IL-10 are still induced in ferent classes: the activating receptors FcgRI, FcgRIII, and g-chain–deficient mice, there may be additional mechanisms that FcgRIV, which signal through an ITAM, and an inhibitory re- underlie microglial switching. ceptor FcgRIIB, which signals through an ITIM (39, 40). The Implications for immunotherapy inhibitory FcgRIIB is an important modulator of inflammatory effector cells, and FcgRIIB deficiency leads to amplified immune There is growing academic and commercial interest in using the responses and autoimmunity (41). Cross-linking of activating power of Ab-mediated responses to treat neurodegenerative disease FcgR results in a range of downstream effector functions, in- by vaccination strategies, including Alzheimer’s disease and prion cluding cytokine release, whereas coaggregation of inhibitory diseases, but recent studies suggest that immunotherapy targeting by guest on September 27, 2021 FcgR inhibits these functions (39). In the healthy brain, FcgR brain Ags is not without risks (52, 53). Increased expression of expression is very low and restricted to microglia (42). Enhanced FcgR on microglia as a consequence of neurodegeneration may expression is observed after treatment with IFN-g, TNF-a, and partly explain the inflammatory reactions within the brain ob- LPS in vitro (43) and after stereotaxic injection of LPS in vivo served after immunotherapy in animal models (54, 55). Further (44). It was also shown that transient opening of the blood–brain understanding of the role of activating and inhibiting FcgRin barrier by adrenaline in otherwise healthy rats increased FcgRI these processes and controlling FcgR function is likely to increase expression on microglia (45), and multiple sclerosis lesions have the success of immunotherapy for neurodegenerative diseases and an elevated expression of various activating FcgRs (46, 47). These avoid the clinical setbacks experienced to date. studies suggest that FcgRs can be induced in the brain where they can play an important role in controlling Ab-mediated neuro- Acknowledgments inflammation. However, some studies do not support a key ef- We thank Biogen for K.L.’s scholarship and the opportunity to carry out the fector function of FcgRs in neuroinflammation and suggest a key microarray experiments and Prof. Paul Crocker and Iva Fernandinsky for analysis of Siglec-F mRNA levels. We thank Dr. Ayodeji Asuni and Dr. role of complement activation instead (48). Colm Cunningham for helpful discussions and Richard Reynolds for ex- In contrast to studies in peripheral tissues, the differential ex- cellent technical assistance. pression of activating versus inhibitory FcgR has not been studied in the brain. Furthermore, to our knowledge, this is the first time Disclosures FcgRIV has been directly studied in a mouse model of neurologic The authors have no financial conflicts of interest. disease. The role of FcgR and complement in the latency of prion disease has been previously studied (49). In this study, Fcg-chain– deficient mice showed an increased survival time after in- References 1. Dheen, S. T., C. Kaur, and E. A. Ling. 2007. Microglial activation and its tracerebral inoculation of the prion agent. Notably, survival times implications in the brain diseases. Curr. Med. Chem. 14: 1189–1197. of FcgRII- and FcgRIII-deficient mice were not different from 2. ADAPT Research Group, C. L Meinert, L. D. McCaffrey, and J. C. Breitner. those of WT mice (49), suggesting a possible role for FcgRIV in 2009. Alzheimer’s Disease Anti-inflammatory Prevention Trial: design, methods, and baseline results. Alzheimers Dement. 5: 93–104. disease onset in this mouse model. Using a unique panel of anti- 3. Neumann, H., M. R. Kotter, and R. J. Franklin. 2009. Debris clearance by FcgR Abs, we confirmed our microarray and quantitative PCR microglia: an essential link between degeneration and regeneration. Brain 132: results and showed strong increase in FcgRIII and FcgRIV protein 288–295. 4. Betmouni, S., V. H. Perry, and J. L. Gordon. 1996. Evidence for an early in- levels in the thalamus of LPS-challenged ME7 animals. No flammatory response in the central nervous system of mice with scrapie. Neu- changes in protein levels of the inhibitory FcgRII were observed, roscience 74: 1–5. 5. Cunningham, C., R. M. Deacon, K. Chan, D. Boche, J. N. Rawlins, and although the Ab used for histological analysis showed clear V. H. Perry. 2005. Neuropathologically distinct prion strains give rise to similar staining of immune cells in the spleen. This was surprising, as temporal profiles of behavioral deficits. Neurobiol. Dis. 18: 258–269. 7224 SYSTEMIC INFLAMMATION MODULATES FcgR EXPRESSION IN BRAIN

6. Cunningham, C., D. C. Wilcockson, D. Boche, and V. H. Perry. 2005. Com- 30. Combrinck, M. I., V. H. Perry, and C. Cunningham. 2002. Peripheral infection parison of inflammatory and acute-phase responses in the brain and peripheral evokes exaggerated sickness behaviour in pre-clinical murine prion disease. organs of the ME7 model of prion disease. J. Virol. 79: 5174–5184. Neuroscience 112: 7–11. 7. Boche, D., C. Cunningham, J. Gauldie, and V. H. Perry. 2003. Transforming 31. Hughes, M. M., R. H. Field, V. H. Perry, C. L. Murray, and C. Cunningham. growth factor-beta 1-mediated neuroprotection against excitotoxic injury in vivo. 2010. Microglia in the degenerating brain are capable of phagocytosis of beads J. Cereb. Blood Flow Metab. 23: 1174–1182. and of apoptotic cells, but do not efficiently remove PrPSc, even upon LPS 8. Minghetti, L., A. Greco, F. Cardone, M. Puopolo, A. Ladogana, S. Almonti, stimulation. Glia 58: 2017–2030. C. Cunningham, V. H. Perry, M. Pocchiari, and G. Levi. 2000. Increased brain 32. Billadeau, D. D., and P. J. Leibson. 2002. ITAMs versus ITIMs: striking a bal- synthesis of prostaglandin E2 and F2-isoprostane in human and experimental ance during cell regulation. J. Clin. Invest. 109: 161–168. transmissible spongiform encephalopathies. J. Neuropathol. Exp. Neurol. 59: 33. Clynes, R., J. S. Maizes, R. Guinamard, M. Ono, T. Takai, and J. V. Ravetch. 866–871. 1999. Modulation of immune complex-induced inflammation in vivo by the 9. Walsh, D. T., V. H. Perry, and L. Minghetti. 2000. Cyclooxygenase-2 is highly coordinate expression of activation and inhibitory Fc receptors. J. Exp. Med. expressed in microglial-like cells in a murine model of prion disease. Glia 29: 189: 179–185. 392–396. 34. Peress, N. S., H. B. Fleit, E. Perillo, R. Kuljis, and C. Pezzullo. 1993. Identifi- 10. Cunningham, C., R. Deacon, H. Wells, D. Boche, S. Waters, C. P. Diniz, cation of Fc gamma RI, II and III on normal human brain ramified microglia and H. Scott, J. N. Rawlins, and V. H. Perry. 2003. Synaptic changes characterize on microglia in senile plaques in Alzheimer’s disease. J. Neuroimmunol. 48: 71– early behavioural signs in the ME7 model of murine prion disease. Eur. J. 79. Neurosci. 17: 2147–2155. 35. Piccio, L., C. Buonsanti, M. Mariani, M. Cella, S. Gilfillan, A. H. Cross, 11. Fadok, V. A., D. L. Bratton, A. Konowal, P. W. Freed, J. Y. Westcott, and M. Colonna, and P. Panina-Bordignon. 2007. Blockade of TREM-2 exacerbates P. M. Henson. 1998. Macrophages that have ingested apoptotic cells in vitro experimental autoimmune encephalomyelitis. Eur. J. Immunol. 37: 1290–1301. inhibit proinflammatory cytokine production through autocrine/paracrine 36. Walker, D. G., J. E. Dalsing-Hernandez, N. A. Campbell, and L. F. Lue. 2009. mechanisms involving TGF-beta, PGE2, and PAF. J. Clin. Invest. 101: 890–898. Decreased expression of CD200 and CD200 receptor in Alzheimer’s disease: 12. Savill, J., I. Dransfield, C. Gregory, and C. Haslett. 2002. A blast from the past: a potential mechanism leading to chronic inflammation. Exp. Neurol. 215: 5–19. clearance of apoptotic cells regulates immune responses. Nat. Rev. Immunol. 2: 37. Frank, S., G. J. Burbach, M. Bonin, M. Walter, W. Streit, I. Bechmann, and 965–975. T. Deller. 2008. TREM2 is upregulated in amyloid plaque-associated microglia 13. Dantzer, R., and K. W. Kelley. 2007. Twenty years of research on cytokine- in aged APP23 transgenic mice. Glia 56: 1438–1447. Downloaded from induced sickness behavior. Brain Behav. Immun. 21: 153–160. 38. Castells, M. C. 1994. Surface makers for mast cell subtypes: low affinity IgG 14. Cunningham, C., S. Campion, K. Lunnon, C. L. Murray, J. F. Woods, receptors and gp49 family. Allerg. Immunol. (Paris) 26: 127–131. R. M. Deacon, J. N. Rawlins, and V. H. Perry. 2009. Systemic inflammation 39. Nimmerjahn, F., and J. V. Ravetch. 2006. Fcgamma receptors: old friends and induces acute behavioral and cognitive changes and accelerates neurodegener- new family members. Immunity 24: 19–28. ative disease. Biol. Psychiatry 65: 304–312. 40. Ravetch, J. V., and S. Bolland. 2001. IgG Fc receptors. Annu. Rev. Immunol. 19: 15. Cunningham, C., D. C. Wilcockson, S. Campion, K. Lunnon, and V. H. Perry. 275–290. 2005. Central and systemic endotoxin challenges exacerbate the local in- 41. Ravetch, J. V., and L. L. Lanier. 2000. Immune inhibitory receptors. Science 290:

flammatory response and increase neuronal death during chronic neuro- 84–89. http://www.jimmunol.org/ degeneration. J. Neurosci. 25: 9275–9284. 42. Vedeler, C., E. Ulvestad, I. Grundt, G. Conti, H. Nyland, R. Matre, and 16. Godbout, J. P., J. Chen, J. Abraham, A. F. Richwine, B. M. Berg, K. W. Kelley, D. Pleasure. 1994. Fc receptor for IgG (FcR) on rat microglia. J. Neuroimmunol. and R. W. Johnson. 2005. Exaggerated neuroinflammation and sickness behavior 49: 19–24. in aged mice following activation of the peripheral innate immune system. 43. Loughlin, A. J., M. N. Woodroofe, and M. L. Cuzner. 1992. Regulation of Fc FASEB J. 19: 1329–1331. receptor and major histocompatibility complex antigen expression on isolated rat 17. Palin, K., C. Cunningham, P. Forse, V. H. Perry, and N. Platt. 2008. Systemic microglia by tumour necrosis factor, interleukin-1 and lipopolysaccharide: inflammation switches the inflammatory cytokine profile in CNS Wallerian de- effects on interferon-gamma induced activation. Immunology 75: 170–175. generation. Neurobiol. Dis. 30: 19–29. 44. Herber, D. L., J. L. Maloney, L. M. Roth, M. J. Freeman, D. Morgan, and 18. Nguyen, M. D., T. D’Aigle, G. Gowing, J. P. Julien, and S. Rivest. 2004. Ex- M. N. Gordon. 2006. Diverse microglial responses after intrahippocampal ad- acerbation of motor neuron disease by chronic stimulation of innate immunity in ministration of lipopolysaccharide. Glia 53: 382–391. a mouse model of amyotrophic lateral sclerosis. J. Neurosci. 24: 1340–1349. 45. Li, Y. N., X. J. Qin, F. Kuang, R. Wu, X. L. Duan, G. Ju, and B. R. Wang. 2008.

19. McColl, B. W., N. J. Rothwell, and S. M. Allan. 2007. Systemic inflammatory Alterations of Fc gamma receptor I and Toll-like receptor 4 mediate the anti- by guest on September 27, 2021 stimulus potentiates the acute phase and CXC chemokine responses to experi- inflammatory actions of microglia and astrocytes after adrenaline-induced blood- mental stroke and exacerbates brain damage via interleukin-1- and neutrophil- brain barrier opening in rats. J. Neurosci. Res. 86: 3556–3565. dependent mechanisms. J. Neurosci. 27: 4403–4412. 46. Breij, E. C., B. P. Brink, R. Veerhuis, C. van den Berg, R. Vloet, R. Yan, 20. McColl, B. W., N. J. Rothwell, and S. M. Allan. 2008. Systemic inflammation C. D. Dijkstra, P. van der Valk, and L. Bo¨. 2008. Homogeneity of active de- alters the kinetics of cerebrovascular tight junction disruption after experimental myelinating lesions in established multiple sclerosis. Ann. Neurol. 63: 16–25. stroke in mice. J. Neurosci. 28: 9451–9462. 47. Ulvestad, E., K. Williams, C. Vedeler, J. Antel, H. Nyland, S. Mørk, and R. Matre. 21. Buljevac, D., H. Z. Flach, W. C. Hop, D. Hijdra, J. D. Laman, H. F. Savelkoul, 1994. Reactive microglia in multiple sclerosis lesions have an increased expres- F. G. van Der Meche´, P. A. van Doorn, and R. Q. Hintzen. 2002. Prospective sion of receptors for the Fc part of IgG. J. Neurol. Sci. 121: 125–131. study on the relationship between infections and multiple sclerosis exacer- 48. Urich, E., I. Gutcher, M. Prinz, and B. Becher. 2006. Autoantibody-mediated bations. Brain 125: 952–960. demyelination depends on complement activation but not activatory Fc-recep- 22. Dunn, N., M. Mullee, V. H. Perry, and C. Holmes. 2005. Association between tors. Proc. Natl. Acad. Sci. USA 103: 18697–18702. dementia and infectious disease: evidence from a case-control study. Alzheimer 49. Klein, M. A., P. S. Kaeser, P. Schwarz, H. Weyd, I. Xenarios, R. M. Zinkernagel, Dis. Assoc. Disord. 19: 91–94. M. C. Carroll, J. S. Verbeek, M. Botto, M. J. Walport, et al. 2001. Complement 23. Johnston, K. C., J. Y. Li, P. D. Lyden, S. K. Hanson, T. E. Feasby, R. J. Adams, facilitates early prion pathogenesis. Nat. Med. 7: 488–492. R. E. Faught, Jr., E. C. Haley Jr.; RANTTAS Investigators. 1998. Medical and 50. Chicheportiche, Y., and P. Vassalli. 1994. Tumor necrosis factor induces a block neurological complications of ischemic stroke: experience from the RANTTAS in the cotranslation of Fc gamma RIIb mRNA in mouse peritoneal macrophages. trial. Stroke 29: 447–453. J. Biol. Chem. 269: 20134–20138. 24. Teeling, J. L., and V. H. Perry. 2009. Systemic infection and inflammation in 51. Le, W., D. Rowe, W. Xie, I. Ortiz, Y. He, and S. H. Appel. 2001. Microglial acute CNS injury and chronic neurodegeneration: underlying mechanisms. activation and dopaminergic cell injury: an in vitro model relevant to Parkinson’s Neuroscience 158: 1062–1073. disease. J. Neurosci. 21: 8447–8455. 25. Beers, S. A., R. R. French, H. T. Chan, S. H. Lim, T. C. Jarrett, R. M. Vidal, 52. Boche, D., E. Zotova, R. O. Weller, S. Love, J. W. Neal, R. M. Pickering, S. S. Wijayaweera, S. V. Dixon, H. J. Kim, K. L. Cox, et al. 2010. Antigenic D. Wilkinson, C. Holmes, and J. A. Nicoll. 2008. Consequence of Abeta im- modulation limits the efficacy of anti-CD20 antibodies: implications for antibody munization on the vasculature of human Alzheimer’s disease brain. Brain 131: selection. Blood 115: 5191–5201. 3299–3310. 26. Jain, N., J. Thatte, T. Braciale, K. Ley, M. O’Connell, and J. K. Lee. 2003. Local- 53. Wilcock, D. M., P. T. Jantzen, Q. Li, D. Morgan, and M. N. Gordon. 2007. pooled-error test for identifying differentially expressed genes with a small Amyloid-beta vaccination, but not nitro-nonsteroidal anti-inflammatory drug number of replicated microarrays. Bioinformatics 19: 1945–1951. treatment, increases vascular amyloid and microhemorrhage while both reduce 27. Teeling, J. L., C. Cunningham, T. A. Newman, and V. H. Perry. 2010. The effect parenchymal amyloid. Neuroscience 144: 950–960. of non-steroidal anti-inflammatory agents on behavioural changes and cytokine 54. Pfeifer, M., S. Boncristiano, L. Bondolfi, A. Stalder, T. Deller, M. Staufenbiel, production following systemic inflammation: Implications for a role of COX-1. P. M. Mathews, and M. Jucker. 2002. Cerebral hemorrhage after passive anti- Brain Behav. Immun. 24: 409–419. Abeta immunotherapy. Science 298: 1379. 28. Wisniewski, H. M., A. S. Lossinsky, R. C. Moretz, A. W. Vorbrodt, H. Lassmann, 55. Racke, M. M., L. I. Boone, D. L. Hepburn, M. Parsadainian, M. T. Bryan, and R. I. Carp. 1983. Increased blood-brain barrier permeability in scrapie- D. K. Ness, K. S. Piroozi, W. H. Jordan, D. D. Brown, W. P. Hoffman, et al. infected mice. J. Neuropathol. Exp. Neurol. 42: 615–626. 2005. Exacerbation of cerebral amyloid angiopathy-associated microhemorrhage 29. Guenther, K., R. M. Deacon, V. H. Perry, and J. N. Rawlins. 2001. Early in amyloid precursor protein transgenic mice by immunotherapy is dependent on behavioural changes in scrapie-affected mice and the influence of dapsone. Eur. antibody recognition of deposited forms of amyloid beta. J. Neurosci. 25: 629– J. Neurosci. 14: 401–409. 636.