of April 28, 2017.This information is current as

MeningitisStreptococcus pneumoniaePathogenesis in

DrivesγInflammasome-Dependent IFN-

Wolfgang Weninger and Nicholas H. HuntStephen L. Leib, Cheryl A. Jones, Sebastien K. Gerega,Helen J. Ball, Lay Khoon Too, Arby Abtin, Paul Hertzog, Andrew J. Mitchell, Belinda Yau, James A. McQuillan,

http://www.jimmunol.org/content/189/10/4970doi: 10.4049/jimmunol.1201687October 2012;

2012; 189:4970-4980; Prepublished online 15J Immunol 

MaterialSupplementary

7.DC1http://www.jimmunol.org/content/suppl/2012/10/15/jimmunol.120168

Referenceshttp://www.jimmunol.org/content/189/10/4970.full#ref-list-1

, 42 of which you can access for free at: cites 84 articlesThis article

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists, Inc. All rights reserved.Copyright © 2012 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on April 28, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on April 28, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on April 28, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on April 28, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on April 28, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on April 28, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

Inflammasome-Dependent IFN-g Drives Pathogenesis inStreptococcus pneumoniae Meningitis

Andrew J. Mitchell,*,†,1 Belinda Yau,*,1 James A. McQuillan,* Helen J. Ball,*

Lay Khoon Too,* Arby Abtin,† Paul Hertzog,‡ Stephen L. Leib,x Cheryl A. Jones,*,{

Sebastien K. Gerega,‖ Wolfgang Weninger,† and Nicholas H. Hunt*

The pathology associated with Streptococcus pneumoniae meningitis results largely from activation of immune-associated path-

ways. We systematically investigated the production of IFN subtypes, as well as their influence on pathology, in a mouse model of

S. pneumoniae meningitis. Despite the occurrence of a mixed IFN type I/II gene signature, no evidence for production or

involvement of type I IFNs in disease progression was found. In contrast, type II IFN (IFN-g) was strongly induced, and IFN-

g2/2 mice were significantly protected from severe disease. Using intracellular cytokine staining and targeted cell-depletion

approaches, NK cells were found to be the dominant source of IFN-g. Furthermore, production of IFN-g was found to be

dependent upon ASC and IL-18, indicating that an ASC-dependent inflammasome pathway was responsible for mediating

IFN-g induction. The influence of IFN-g gene deletion on a range of processes known to be involved in bacterial meningitis

pathogenesis was examined. Although neutrophil numbers in the brain were similar in infected wild-type and IFN-g2/2 mice, both

monocyte recruitment and CCL2 production were less in infected IFN-g2/2 mice compared with infected wild-type controls.

Additionally, gene expression of NO synthase was strongly diminished in infected IFN-g2/2 mice compared with infected controls.

Finally, bacterial clearance was enhanced in IFN-g2/2 mice, although the underlying mechanism remains unclear. Together, these

data suggest that inflammasome-dependent IFN-g contributes via multiple pathways to pathology during S. pneumoniae menin-

gitis. The Journal of Immunology, 2012, 189: 4970–4980.

Bacterial meningitis (BM) is a major cause of mortalityworldwide and is responsible for 340,000 deaths each year(1). The Gram-positive extracellular bacterium Strepto-

coccus pneumoniae (pneumococcus) is the most common patho-gen in adults and, despite antibiotic therapy, it has the highestmortality of all meningitis-causing agents: 4–16% in children andup to 60% in adults (2–4). Furthermore, pneumococcal meningitiscauses the highest rates of neurologic sequelae (5), with long-termdisabilities reported in 27–57% of survivors (6–8). The processesunderlying pathogenesis in BM are complex, with contributions

from both bacteria and host, but the immune response is believedto be a key factor (9–11).Immune targeting of S. pneumoniae is essential for clearance of

infection; however, within the brain, the attendant inflammatory

response contributes to pathogenesis. The presence of S. pneumo-

niae is initially sensed by resident cells via interaction of microbial

products with pattern recognition receptors, including TLR2 and

TLR4 (12-14), as well as NOD2 (15). More recently, it was shown

that S. pneumoniae may also be recognized by inflammasomes that

are dependent upon the adaptor molecule ASC (16–19). Together,

these pathways initiate immune activation and lead to production of

inflammatory cytokines and chemokines that subsequently mediate

the recruitment of leukocytes to the site of infection (20–26).

However, when this occurs within the brain, the ensuing inflam-

matory response is argued to ultimately cause neuronal damage and/

or death (9, 10). Indeed, administration of steroids diminishes in-

flammation in rodent models of pneumococcal meningitis (27), and

it is recommended as adjunct therapy in human disease (28). Most

studies investigating the roles of cytokines in BM have focused on

those cytokines that are traditionally regarded as mediators of acute

inflammation, such as TNF-a and IL-1b (20, 29, 30); however, the

contribution of other cytokines, including IFNs, to pathogenesis has

not been well studied. Although type I IFNs (IFN-a and IFN-b)

historically have been associated with responses to viral infection, it

is becoming increasingly clear that they may influence the outcome

of responses to other pathogens (31, 32). In contrast, type II IFN

(IFN-g) has a well-described role in protection against infection

with intracellular bacteria (33, 34).Early studies of S. pneumoniae meningitis showed the presence

of IFN bioactivity in the cerebrospinal fluid (CSF) of patients with

BM (35), but reports on the presence of IFN-g specifically are

inconsistent, with some studies reporting concentrations that are

low to undetectable (36–39), whereas others suggested that IFN-g

*Sydney Medical School, University of Sydney, Camperdown, New South Wales2050, Australia; †Centenary Institute for Cancer Medicine and Cell Biology, New-town, New South Wales 2042, Australia; ‡Centre for Innate Immunity and InfectiousDisease, Monash Institute of Medical Research, Clayton, Victoria 3168, Australia;xInstitute for Infectious Diseases, University of Bern, Bern CH-3012, Switzerland;{Children’s Hospital Westmead, University of Sydney, Westmead, New South Wales2145, Australia; and ‖Sydney Bioinformatics, University of Sydney, Sydney, NewSouth Wales 2006, Australia

1A.J.M. and B.Y. contributed equally to this work.

Received for publication June 20, 2012. Accepted for publication September 5, 2012.

This work was supported by a grant from the National Health and Medical ResearchCouncil of Australia (571024).

The array data presented in this article have been submitted to ArrayExpress(http://www.ebi.ac.uk/arrayexpress/) under accession number E-MEXP-3697.

Address correspondence and reprint requests to Nicholas H. Hunt, University of Sydney,Medical Foundation Building (K25), Sydney, NSW 2006, Australia. E-mail address:nicholas.hunt@sydney.edu.au

The online version of this article contains supplemental material.

Abbreviations used in this article: BFA, brefeldin A; BM, bacterial meningitis; CBA,cytometric bead array; COX2, cyclooxygenase-2; CSF, cerebrospinal fluid; icv, intra-cerebroventricular(ly); NOS2, NO synthase-2; p.i., postinoculation; qPCR, quantita-tive PCR; WT, wild-type.

Copyright� 2012 by TheAmericanAssociation of Immunologists, Inc. 0022-1767/12/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1201687

by guest on April 28, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

production may be present in at least some patients (40–42). Ofparticular note is the fact that many of these studies did not distin-guish between causative agents when investigating the presence ofIFN-g in CSF. This is significant, because BM caused by S. pneumo-niae was reported to be specifically associated with increased levelsof IFN-g in CSF (41, 42). Although IFN-g was reported to beproduced in the brain in a rodent model of S. pneumoniaemeningitis(43), the significance of this finding has not been investigated.The involvement of IFN-g in the development of systemic and

pulmonary pneumococcal disease has been studied in detail and hastypically been associated with enhanced clearance of bacteria. Inpatients with S. pneumoniae sepsis, plasma IFN-g levels were re-ported to be elevated, and circulating levels of the cytokine corre-lated with mortality (44). In murine models, IFN-g inhibition/ablation was reported to lead to decreased survival (25, 26, 45) orto have no effect on mortality (23, 46). Furthermore, in the absenceof IFN-g, decreased levels of chemokines involved in neutrophilrecruitment, as well as neutrophils themselves, were reported bysome groups (23, 26), and this was associated with a lower bacterialload (23). More recently, and in contrast, IFN-g–deficient micewere argued to have higher levels of neutrophils and chemokinesassociated with their recruitment, as well as increased bacterial load(25). Alternatively, IFN-g may influence pneumococcal clearancevia modulation of macrophage scavenger receptors. These patternrecognition receptors directly mediate phagocytosis of S. pneumo-niae and are critical for clearance both within the lung (47–49) andduring systemic infection (50). Expression of scavenger receptorswas shown to be modulated by IFN-g (51), and high levels of IFN-g, resulting from viral infection, for example, can lead to drasticallyreduced pneumococcal load (48).In addition to its influence on bacterial clearance, IFN-g has the

potential to exacerbate inflammation and subsequent pathology,primarily through its ability to modulate the functions of neu-trophils and monocyte/macrophages. Tight control of neutrophilfunction is particularly relevant during pneumococcal meningitis,because these cells may contribute both to protection via clearanceof bacteria (22) and, if their apoptosis is delayed, to tissue damageand pathology (52). In conjunction with stimulation through patternrecognition receptors, IFN-g activates macrophages and polarizesthem toward an M1 phenotype, which is characterized by upregu-lation of proinflammatory factors (reviewed in Ref. 53). In partic-ular, IFN-g induces NO synthase-2 (NOS2) (54), and increased NOproduction secondary to NOS2 induction was argued to contributeto pathology during pneumococcal meningitis (55).In this study, we systematically investigated the production of

both type I and type II IFNs in a murine model of S. pneumoniaeBM. We find that IFN-g, but not type I IFNs, is produced and drivespathology during pneumococcal meningitis. This IFN-g is made byNK cells and is dependent upon inflammasome activation and IL-18. Protection is associated with decreased bacterial load, but onlya limited number of inflammatory factors was IFN-g dependent.

Materials and MethodsMice

Mice of mixed sex, aged 6–14 wk at the commencement of experimen-tation, were used. Wild-type (WT; C57BL/6J) mice were sourced from theAnimal Resources Centre (Canning Vale, WA, Australia). IL-182/2 (56),IFN-g2/2 (57), CXCR32/2 (58), and CD1d2/2 (59) mice were maintainedin the Medical Foundation Building Animal House (The University ofSydney). IFNAR12/2 (60) and IFNAR22/2 (61) mice were from theMonash Institute of Medical Research. ASC2/2 mice (62) were main-tained in the Centenary Institute Animal Facility. All mice were ona C57BL/6 background. They were housed in the Medical FoundationBuilding Animal House or Centenary Institute Animal Facility in groupcages under a 12-h light–dark cycle, with food and water ad libitum. All

procedures adhered to the Australian National Health and Medical Re-search Council guidelines for animal research and were approved by theUniversity of Sydney Animal Care and Ethics Committee.

Abs

FITC–anti-CD3 (145-2C11), PE-Ly6G (1A8), PE–anti-abTCR (H57-597),PerCP–anti-CD45 (30-F11), Violet450-anti-CD3 (500-A2), FITC–anti-CD4 (H129.19), PECy7–anti-CD45 (30-F11), and anti–CD16/32 (Fc-block, 2.4G2) were purchased from Becton Dickinson. Pacific Blue–anti-NK1.1 (PK136), Pacific Blue–anti-CD19 (6D5), FITC–anti-CD19(6D5), PerCP–anti-Ly6C (HK1.4), Alexa Fluor 647–anti-Ly6G (1A8),allophycocyanin–Cy7–anti-NK1.1 (PK136), allophycocyanin–Cy7–anti-CD11b (M1/70), allophycocyanin–Cy7–anti-NK1.1 (PK136), and PE–anti-CD8 (53-6.7) were from BioLegend. FITC–anti-IFN-g (XMG 1.2),PECy5–anti-gdTCR (GL3), PECy7–anti-CD8 (53-6.7), Alexa Fluor700–anti-CD4 (GK1.5), and FITC- and Alexa Fluor 647–anti-NKp46(29A1.4) were purchased from eBiosciences. For inhibition experi-ments, anti–IFN-g (R4-6A2) and IgG1 isotype control (GL113) werepurchased from the Walter and Eliza Hall Institute Biotechnology Centre(Bundoora, VIC, Australia).

Model

S. pneumoniae, strain WU2, serotype 3 (courtesy of Prof. J. Paton, Uni-versity of Adelaide, Adelaide, SA, Australia) was stored at 280˚C. Ali-quots were thawed and cultured overnight in 5% CO2 at 37˚C in brain heartinfusion broth (Oxoid, Adelaide, SA, Australia). Bacteria were thensubcultured into brain heart infusion broth and grown to late log phase.After harvesting, bacteria were washed in PBS and adjusted to ∼5 3 105

CFU/ml based on OD at 570 nm. Exact inoculum concentration (CFU/ml)was determined by serial dilution and plating on horse blood agar plates(Oxoid), followed by overnight incubation in 5% CO2 at 37˚C. For in-duction of meningitis, animals were lightly anesthetized with isoflurane,the head was swabbed with 70% ethanol, and a 29G needle was insertedbetween the hemispheres to a depth of 3–4 mm. Ten microliters of inoc-ulum (∼5 3 103 CFU) was injected into the third ventricle. Sham-infectedmice received 10 ml PBS. In this model, infected mice begin to developsevere disease at ∼48 h postinoculation (p.i.). In some experiments, IFN-gwas neutralized by i.p. administration of anti–IFN-g (500 mg 2 h prior toinfection, then 250 mg at 48 h and 96 h postinfection), or by intra-cerebroventricular (icv) administration of 20 mg Ab mixed with bacterialinoculum. Animals were euthanized if they showed overtly hunched pos-ture, gait disturbances, fitting, or difficulty righting.

RT-quantitative PCR

Tissues were homogenized in RLT buffer (QIAGEN, Valencia, CA), using1-mm zirconium beads (Biospec Products) in a FastPrep FP120 bead mill(Thermo Fisher, Waltham, MA). After storage at 280˚C, samples werethawed, and RNAwas extracted using an RNeasy miniprep kit (QIAGEN),following the manufacturer’s instructions. RNA was reverse-transcribedusing Moloney murine leukemia virus-RT (Ambion) primed with ran-dom hexamers (GeneWorks, Hindmarsh, SA, Australia). Quantitative PCR(qPCR) was performed, using the primers listed in Table 1, on a CorbettResearch Rotor Gene 3000 using platinum SYBR Green Supermix(Invitrogen, Carlsbad, CA) with cycling conditions as follows: 2 min of95˚C denaturation and then 35 repeats of a two-step amplification cycle(95˚C for 15 s and 60˚C for 45 s). Following amplification, product puritywas assessed by melt-curve analysis. RPL13a was used as a referencegene, and the gene-expression levels in populations were normalizedagainst the brains of uninfected animals using the 2DDCt method (63).The primers used for RT-qPCR had similar amplification efficiencies to thereference gene.

Bacterial load estimation

Half brains from infected mice were removed aseptically and homoge-nized in a 1:5 (w/v) dilution in PBS using a TissueRuptor (QIAGEN)homogenizer. Bacterial load was determined by serial dilution andplating on horse blood agar plates, followed by overnight incubation in5% CO2 at 37˚C.

Array analysis

For array analysis, RNA from brains of infected and sham-infected mice(n = 4/group) was pooled and analyzed using an Affymetrix Mouse Gene1.0-ST array. Arrays were normalized using the robust multi-array averagealgorithm (64) implemented in BioConductor, and differentially expressedgenes with fold change .2.0 were identified. Array data have been sub-

The Journal of Immunology 4971

by guest on April 28, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

mitted to ArrayExpress (http://www.ebi.ac.uk/arrayexpress) under acces-sion number E-MEXP-3697.

Cytometric bead array

Half brains were collected in PBS, 5% v/v FCS containing protease inhibitormixture (Sigma) and homogenized with 1-mm high-density zirconiumbeads (Biospec Products) in a FastPrep FP120 (Thermo Fisher). Homog-enates were pelleted at 13,000 3 g for 10 min at 4˚C, and the supernatantswere stored at 280˚C until assay. Cytokine levels in homogenates werequantified using cytometric bead array (CBA; Becton Dickinson Bio-sciences). Prior to analysis with the CBA kit, total protein levels of allsamples were calculated by bicinchoninic acid protein assay (ThermoFisher). The CBA kit was used according to the manufacturer’s instruc-tions, with the modification that volumes of all reagents and samples wereeither 50% or 10% of those in the original protocol. Data were collectedusing an FC500 Flow Cytometer (Beckman Coulter, Gladesville, NSW,Australia) and analyzed with FlowJo software (Tree Star, Ashland, OR).Concentrations of cytokines in tissue homogenates were normalized tototal protein content for each sample.

Flow cytometry

Quantification of leukocyte subsets in the brains of infected animals wasperformed as described previously (65, 66), with minor modifications.Briefly, brains were mashed between frosted glass slides in RPMI 1640,digested with collagenase/DNase for 20 min at room temperature, andtriturated before a second 20-min collagenase/DNase step. Low-densitydead cells/debris/myelin were removed by resuspension in 30% Percolland centrifugation at 500 3 g for 10 min at 4˚C. Following Percollisolation, contaminating RBCs were lysed by Tris-ammonium chloridelysis. Prior to staining, cells were incubated with anti-CD16/32 (Fcblock). Ab staining for flow cytometry was performed by standardmethods, and dead cells were excluded using propidium iodide. Sampleswere analyzed on a five-laser LSR II (Becton Dickinson Biosciences),and data were analyzed using FlowJo software (Tree Star). Cells forintracellular cytokine staining were prepared using a modified isolationand staining procedure, as follows. Animals received an i.p. injection of250 mg brefeldin A (BFA; Sigma) at 42 h p.i. Following euthanasia,brains were placed into ice cold RPMI 1640 containing 10 mg/ml BFAand disaggregated immediately by pressing through a 100-mm mesh.Cells were pelleted (300 3 g, 5 min, 4˚C), resuspended in 30% Percollcontaining BFA, and centrifuged at 500 3 g for 5 min at 4˚C. FollowingPercoll isolation, contaminating RBCs were removed by a 10-s distilledwater lysis. Fc-blocked cells were stained for surface markers in PBS,1% FCS, 5 mM EDTA, 0.05% sodium azide, and BFA. Thereafter, in-tracellular IFN-g was stained using an intracellular cytokine staining kit(BioLegend), as per the manufacturer’s instructions. Staining controlsincluded cells preincubated with either an excess of unlabeled anti–IFN-g (isoclonic control) or with a molar excess of rIFN-g. Additionally,brain leukocytes isolated from S. pneumoniae-infected IFN-g2/2 miceshowed no significant IFN-g staining.

Statistical analysis

Experimental group size giving 90% statistical power was estimated basedon preliminary data using PS Power and Sample Size Calculation software(67). Because large group sizes were often required to attain this level ofpower, data from multiple independent experiments were typically pooledfor analysis, as described in the figure legends. Data were analyzed usingGraphPad Prism version 5.01 for Windows (GraphPad Software, La Jolla,CA). For two groups, statistical significance was determined using a t test,whereas one-way ANOVA with the Tukey posttest was used for multiple-group comparisons. Quantitative PCR and bacterial load data were logtransformed prior to analysis. Survival experiments were analyzed usinga log-rank test.

ResultsType II IFN, but not type I IFNs, contributes to diseaseprogression in pneumococcal meningitis

The potential for IFN subtypes to contribute to the immune re-sponse to S. pneumoniae during meningitis was initially examinedby investigating the overall patterns of gene expression by genemicroarray, followed by analysis of IFN response signatures (68).In infected mice, the majority of differentially expressed genes

during BM were identified as IFN-regulated genes that were ca-pable of being regulated by either type I or type II IFNs. Of the top200 differentially expressed genes, 132 were classified as IFNregulated. Of these 132 genes, 130 were associated with type I

FIGURE 1. IFN responses in S. pneumoniae meningitis and effect of

subtypes on development of severe disease. (A) Classification of top 200

differentially expressed (DE) genes into type I and type II IFN regulated.

Brains were collected from S. pneumoniae-infected (∼5 3 103 CFU, icv)

mice at 48 h p.i. Venn diagram shows IFN type I- and type II-dependent

genes (68). Values indicate the number of genes in each category. Data are

from a single experiment with samples pooled from n = 5 animals/group.

(B) Time course of gene expression, relative to expression in noninfected

animals, of IFN subsets. Changes in mRNA for IFN-a4, IFN-aN (IFN-a1,

IFN-a2, IFN-a7, IFN-a11, and IFN-a12), IFN-b, and IFN-g were deter-

mined by RT-qPCR in the brains of infected mice. Symbols represent in-

dividual animals, and horizontal lines represent the geometric mean of

each group. **p , 0.01, one-way ANOVA with Tukey posttest on log-

transformed data. Data are representative of two independent experiments

with similar, but nonidentical, collection times. (C) Contribution of IFN

subtypes to severe disease. Upper panels, WT and IFN-g2/2 or WT and

IFN-aR12/2 or IFN-aR22/2 mice were infected with S. pneumoniae

(∼5 3 103 CFU, icv), and survival to a predefined clinical end point was

determined (see Materials and Methods). Lower panels, WT mice were

treated with neutralizing anti–IFN-g Ab via either i.p. or icv routes at the

time of infection, and survival to a predefined clinical end point was de-

termined. Control animals received an isotype control Ab (i/c). Data sets

shown for each strain/treatment are pooled from a minimum of two in-

dependent experiments, with a minimum of n = 16/condition. **p , 0.01,

versus WT/isotype control-treated group, log-rank test.

4972 IFN-g IN PNEUMOCOCCAL MENINGITIS

by guest on April 28, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

IFN-stimulated responses, and 65 were associated with type II IFNresponses, with an intersection of 63 genes (Fig. 1A). These datasuggested that a mixed type I and type II IFN response occursfollowing infection with S. pneumoniae.Because the array data suggested that both type I and type II IFN-

regulated genes were upregulated following infection, the induc-tion of mRNA for IFN subtypes, at 24 and 48 h p.i., was examinedby RT-qPCR in Table I. In contrast to the apparent strong type IIFN gene signature suggested by array, when changes in the ex-pression of type I IFN genes (IFNa1, IFNa2, IFNa4, IFNa7,IFNa11, IFNa12, IFNb) themselves were examined, no signifi-cant induction was seen over the course of 48 h (Fig. 1B). Thiswas in marked contrast to changes seen in IFN-g mRNA levels,with IFN-g message increasing by ∼10-fold over uninfectedcontrol at 24 h and ∼100-fold by 48 h p.i. (Fig. 1B).The contribution of IFN subtypes to the development of severe

pathology was investigated by following disease progression inmice deficient either in type I IFN signaling (IFNAR12/2 orIFNAR22/2 mice) or in the gene for IFN-g. Consistent with theapparent absence of either IFN-a or IFN-b production followinginfection, mice that lacked functional type I IFN signaling bydisruption of either the IFNAR1 or IFNAR2 chain showed nosignificant differences in survival compared with infected WTmice (Fig. 1C). This was in distinct contrast to the outcome inIFN-g–deficient mice, which were significantly protected from thedevelopment of severe disease; 82% of IFN-g–deficient animalssurvived to 200 h p.i., whereas only 37% of WT animals did so(Fig. 1C). Confirmation of the involvement of IFN-g that wasproduced within the brain, in the development of pathology, wasdetermined by Ab-mediated neutralization of IFN-g. When IFN-gwas neutralized by systemic (i.p.) administration of Ab, there

was no protection; however, when anti–IFN-g was administeredicv at the time of infection, mice were significantly protected fromsevere disease (Fig. 1C).

NK cells are the major source of IFN-g in pneumococcalmeningitis

Because a wide variety of cells is capable of producing IFN-g, thecellular source of IFN-g during S. pneumoniae meningitis wasinvestigated by intracellular cytokine staining and cell-specificdepletion approaches. The inflammatory infiltrate occurring inresponse to S. pneumoniae infection was dominated by neutrophilsand inflammatory monocytes, although appreciable numbers ofCD4+ and CD8+ ab T cells, as well as gd T cells and NK cells,were also recruited to the brains of infected mice (Fig. 2A),whereas significant numbers of NKT cells were not detected (datanot shown). Although a small percentage of CD3+ T cellsexpressed IFN-g protein, a much larger number of NK1.1+,NKp46+ NK cells expressed high levels of IFN-g (Fig. 2A). Incontrast, myeloid cells, which included microglia (CD45int,CD11bhi), monocytes (CD45hi, CD11bhi, Ly6G2), and neutrophils(CD45hi, CD11bhi, Ly6Cint, Ly6G+), showed no staining for IFN-g(data not shown). The dominant contribution of NK cells to IFN-gproduction was confirmed in animals depleted of NK cells byadministration of anti–asialo-GM1 Ab. High levels of IFN-gprotein (8810 6 2890 pg/g of total protein) were detected ininfected control mice, whereas IFN-g protein was below the limitof assay detection (,400 pg/g) in the majority of infected, NK-depleted animals (Fig. 2B). Furthermore, IFN-g mRNA was de-creased .10-fold by NK depletion (Fig. 2C). The absence ofa major contribution of T cell populations and, in particular, CD8+

T cells, to IFN-g production was confirmed by infection of

Table I. Primers used for RT-qPCR

Gene Primer Sequence

RPL13a 59-CTTAGGCACTGCTCCTGTGGAT-39 (sense)59-GGTGCGCTGTCAGCTCTCTAAT-39 (antisense)

IFN-an 59-TCTGATGCAGCAGGTGGG-39 (sense)59-AGGGCTCTCCAGAYTTCTGCTCTG-39 (antisense)

IFN-a4 59-CCTGTGTGATGCAGGAACC-39 (sense)59-TCACCTCCCAGGCACTGA-39 (antisense)

IFN-g 59-CAGCAACAGCAAGGCGAAA-39 (sense)59-GCTGGATTCCGGCAACAG-39 (antisense)

IFN-b 59-TCCAAGAAAGGACGACACTTCG-39 (sense)59-TGAGGACATCTCCCACGTCAA-39 (antisense)

OAS1b 59-AAGACAGGAAGGTGCCCAAA-39 (sense)59-GGACAGAAGCCTGGAGTCTGA-39 (antisense)

MARCO 59-CGCGTCCGGATCATGGGTGG-39 (sense)59-GCATGCGACAGAAGACAGTCGCA-39 (antisense)

SRA1 59-TGCATCCCTTCCTCACAGCACT-39 (sense)59-AGAGAGCAATGAGGGCAGCCT-39 (antisense)

CCL2 59-GGCTCAGCCAGATGCAGTTAA-39 (sense)59-CCTACTCATTGGGATCATCTTGCT-39 (antisense)

CCL4 59-AGATGAAGATGTGGGCTTTGCT-39 (sense)59-GGCAGATTTCTAGCCACAAGGA-39 (antisense)

CXCL2 59-CCCTCAACGGAAGAACCAAA-39 (sense)59-AGGCACATCAGGTACGATCCA-39 (antisense)

CXCL9 59-GCCATGAAGTCCGCTGTTCT-39 (sense)59-GGGTTCCTCGAACTCCACACT-39 (antisense)

CXCL10 59-GACGGTCCGCTGCAACTG-39 (sense)59-GCTTCCCTATGGCCCTCATT-39 (antisense)

IL-1b 59-GTGGTTCGAGGCCTAATAGGCT-39 (sense)59-AGCTGCTTCAGACACCTTGCA-39 (antisense)

IL-10 59-GCCCTTTGCTATGGTGTCCTTT-39 (sense)59-TGAGCTGCTGCAGGAATGATC-39 (antisense)

TNF-a 59-AATGGCCTCCCTCTCATCAGTT-39 (sense)59-CCACTTGGTGGTTTGCTACGA-39 (antisense)

GFAP 59-GCCAAACACGAAGCTAACGACTA-39 (sense)59-GCGCATTTGCCGCTCTA-39 (antisense)

The Journal of Immunology 4973

by guest on April 28, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

RAG12/2 mice (Supplemental Fig. 1A) and mice depleted ofCD8+ T cells by administration of specific Ab (Supplemental Fig.1B). The lack of contribution to IFN-g production by invariantNKT cells was confirmed using CD1d2/2 mice, which lack in-variant NKT cells (Supplemental Fig. 1C).

IFN-g production is inflammasome and IL-18 dependent

Rapid production of IFN-g by NK cells was reported to be de-pendent upon IL-18 (56), and because cleavage of pro–IL-18 tomature IL-18 is typically driven by inflammasome formation (69),the contribution of this pathway to the production of IFN-g wasinvestigated. Initially, the presence of IL-18R on leukocytes in-filtrating the brain was examined by flow cytometry; althoughT cell populations showed variable expression of IL-18R, NKcells uniformly showed high levels of surface IL-18R (data notshown). Because MyD88 is a crucial signaling intermediary forIL-1R family members, including IL-18R, as well as TLRs, wenext investigated the production of IFN-g in MyD882/2 mice. Inthese mice, IFN-g induction was almost exclusively MyD88 de-pendent (Fig. 3A). Recently, it was shown that ASC-dependentinflammasomes may be activated within the brain during S.pneumoniae meningitis and that this contributes to IL-18–depen-dent pathological processes (17); therefore, we assessed the con-tribution of this pathway to the production of IFN-g. Infected micedeficient in ASC had significantly lower levels of IFN-g proteinand mRNA in brain homogenates compared with WT controls(Fig. 3B). Furthermore, IL-182/2 mice showed a similar inhibi-tion of IFN-g protein and mRNA induction to that seen in ASC2/2

mice following infection (Fig. 3C). In both of these gene-deficientstrains, IFN-g protein in infected animals was near or below thelimit of detection, although mRNA induction was not completelyablated, with ∼10–20% residual IFN-g induction present. Finally,the potential source of pro–IL-18 within the brain was examinedby sorting microglia, as well as splenic myeloid subsets forcomparison, followed by quantification of pro–IL-18 mRNA.

Using this approach, it was found that, under resting conditions,microglia expressed comparatively high levels of pro–IL-18mRNA (Fig. 3D).

IFN-g–dependent effects on pathology

Following infection of the CSF with S. pneumoniae, numerousprocesses may contribute to pathology and, ultimately, death, inparticular, the development of edema, enhancement of inflam-mation, as well as inhibition of clearance of bacteria; therefore,the influence of IFN-g on these processes was assessed. An initialtime point of 48 h p.i. was chosen for examination, because dif-ferences in clinical severity between WT and gene-deficient ani-mals first become apparent at this time. An additional time pointof 72 h p.i. was investigated because, although ∼10–15% of WTanimals had developed severe disease and had been euthanized bythis stage (Fig. 1C), there was a clear difference in clinical signsbetween infected WT and IFN-g2/2 mice (data not shown). Be-cause the initial investigation of the role of edema in the de-velopment of pathology found that there was no statisticallysignificant difference in brain water content between infected WTor gene-deficient animals at either 48 or 72 h p.i. (SupplementalFig. 2), alternative pathways were focused upon.

IFN-g modulates inflammation

Because inflammation is critical in the development of pathologyduring pneumococcal meningitis and because IFN-g can enhancemany inflammatory processes, modulation of inflammation mightexplain the protection of IFN-g2/2 mice from mortality. There-fore, the influence of IFN-g on a range of inflammatory parame-ters, including leukocyte recruitment, as well as cytokine, chemokine,and enzyme gene expression, was determined.At both 48 and 72 h p.i., neutrophils and inflammatory mono-

cytes were the dominant cell populations recruited to the brainsof both WT and IFN-g2/2 animals following infection, whereassmaller numbers of lymphocytes (T and NK cells) were present

FIGURE 2. NK cells are the dominant source of IFN-g in pneumococcal meningitis. (A) Intracellular cytokine staining of lymphoid populations for IFN-

g. Mice were infected with S. pneumoniae, and brain leukocytes were isolated at 48 h p.i. NK and T cell populations were identified by flow cytometry; the

percentage of IFN-g+ cells in each gated population is shown. Data are representative of n = 9 mice in three independent experiments. (B) IFN-g protein in

brain homogenates of NK-depleted mice. Animals were untreated (-) or treated with anti–asialo-GM1 (Ab) or control Ab (I/C) and then either sham

infected (PBS) or infected with S. pneumoniae (S.p.). IFN-g levels, normalized to total protein, were quantified in brain homogenates by cytometric bead

array. (C) IFN-g gene expression in brains of NK-depleted mice. Mice were treated with Ab and infected as above and then relative gene expression was

determined by RT-qPCR. Symbols represent individual animals, and horizontal lines represent the arithmetic mean (B) or geometric mean (C) of each

group. Gene-expression data were normalized to the geometric mean of the uninfected control group. Protein data (B) are from a single experiment, and

gene-expression data (C) are from four independent experiments assayed simultaneously (n = 11–12/group). *p , 0.05, **p, 0.01, one-way ANOVAwith

Tukey posttest on log-transformed data.

4974 IFN-g IN PNEUMOCOCCAL MENINGITIS

by guest on April 28, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

(Fig. 4). At 48 h p.i., similar numbers of NK cells and CD4+

T cells were seen in infected WT and IFN-g2/2 mice, whereasthere was a modest, but significant, increase in the numbers ofCD8+ T cells in infected gene-deficient animals (Fig. 4A). Al-though similar total numbers of neutrophils were seen in bothinfected groups at 48 h p.i., there were significantly fewermonocytes in the brains of infected IFN-g2/2 mice compared withinfected WT controls (1.9 3 106 6 6.9 3 105 versus 3.3 3 106 62.4 3 105 per half brain). In contrast, at 72 h p.i., although all ofthe populations that were increased in number at 48 h p.i.remained present, no effect of IFN-g gene deletion was seen onleukocyte numbers (Fig. 4B).Because differences in leukocyte recruitment were seen be-

tween infected WT and IFN-g2/2 mice, at least at 48 h, the in-fluence of IFN-g gene deletion on cytokine and chemokineproduction was examined. Both protein and mRNA of inflam-matory cytokines and key chemokines involved in myeloid cell

recruitment were determined in brain homogenates at 48 and 72 hp.i. Using these approaches, differences in inflammatory cytokineand chemokine expression between infected WT and IFN-g2/2

mice were found not to be overt. mRNA for a range of inflam-matory cytokines and chemokines was induced by infection withS. pneumoniae at both 48 and 72 h p.i. TNF-a and (pro)–IL-1bmRNA were induced to similar levels in both infected WT andIFN-g2/2 mice, whereas IL-10 mRNA abundance in infectedanimals remained similar to that seen in sham-infected animals at48 h but was induced in infected WT mice at 72 h p.i. (Fig. 5A,5B). Although CCL4, CXCL1, and CXCL2 mRNAwere inducedfollowing infection at both time points, there was no effect ofIFN-g gene deletion on their abundance at 48 or 72 h p.i. (Fig.5A, 5B). Consistent with the gene-expression data, protein levelsof TNF-a and IL-1b in brain homogenates were significantlyinduced in both WT and IFN-g2/2 mice compared with sham-infected controls (Fig. 5C, 5D), whereas IL-10 protein was notdetected in either strain following infection (data not shown). Incontrast, infection led to an induction of CCL2 mRNA in bothWT and IFN-g2/2 mice, but infected gene-deficient mice showeda diminished level of expression at 72 h p.i. compared withinfected WT controls (Fig. 5A, 5B). A similar diminution inCCL2 protein in homogenates of infected IFN-g2/2 mice, com-pared with infected controls, was seen at 48 h, but not at 72 h, p.i.(Fig. 5C, 5D). In addition to chemokines that act primarily onmyeloid cells, the influence of IFN-g on the expression of theknown IFN-g–dependent CXCR3 ligands CXCL9 and CXCL10was examined. CXCL9 mRNAwas significantly lower in infectedIFN-g2/2 mice compared with infected WT mice. However, theCXCR3 system did not make a major contribution to pathology,because CXCR32/2 mice were not protected from severe disease(data not shown).Because components of inflammation, including recruitment of

myeloid cells, appeared to be downmodulated following infectionin IFN-g2/2 mice compared with WT mice and because IFN-ginfluences the activation state of monocyte/macrophages, we in-vestigated the expression of proinflammatory enzymes that maybe produced by these cells. NOS2 and cyclooxygenase-2 (COX2)are IFN-g–inducible enzymes that were shown to contribute to thedevelopment of neuropathology in a range of diseases. NOS2, inparticular, was implicated in the development of pathology inpneumococcal meningitis (70, 71). At both 48 and 72 h p.i., NOS2and COX2 were significantly induced in the brains of infected WTanimals compared with uninfected controls. Although there wasno statistical difference in the induction of COX2 betweeninfected WT and IFN-g2/2 animals, induction of NOS2 wassignificantly decreased in infected IFN-g2/2 mice compared withinfected WT animals at both time points (p , 0.05, Fig. 5A, 5B).

IFN-g inhibits bacterial clearance

Effective bacterial clearance is necessary for the ultimate resolutionof pneumococcal meningitis. Indeed, S. pneumoniae clearance inmeningitis was reported to be IL-18 dependent (72), at least inpart, and because we previously showed that IFN-g productionwas dependent upon IL-18 (Fig. 3C), we hypothesized that IL-18might mediate this effect via induction of IFN-g. Therefore, theeffect of IFN-g gene deletion on bacterial load in infected animalswas determined. Although there was a trend toward decreasedbacterial load within the brains of infected IFN-g2/2 mice at 48 hp.i., this was not significant (Fig. 6A). However, at 72 h p.i., therewas an ∼10-fold decrease in the mean CFU in the brains ofinfected gene-deficient animals compared with infected controlmice (Fig. 6B). The mechanisms by which the immune systemclears S. pneumoniae are multiple; however, key among these are

FIGURE 3. IFN-g production is ASC and IL-18 dependent. WT or

MyD882/2 (A), ASC2/2 (B), or IL-182/2 mice (KO) (C) were inoculated

with PBS or S. pneumoniae (∼5 3 103 CFU; S.p.) icv, brains were col-

lected at 24 h p.i., and changes in IFN-g protein (left panel) and mRNA

levels (right panel) were quantified by CBA and RT-qPCR, respectively.

Symbols represent individual animals, and horizontal lines represent the

arithmetic (protein) or geometric (mRNA) mean of each group. Gene-

expression data were normalized to the geometric mean of the uninfected

control group. **p , 0.01, one-way ANOVAwith Tukey posttest on log-

transformed data. (D) Expression of pro–IL-18 mRNA by sorted spleen

and brain myeloid populations. The indicated cell populations (spleen:

red pulp macrophages, CD11c+ conventional dendritic cells, Ly6Chi and

Ly6lo monocytes; brain: microglia) from noninfected mice were sorted by

flow cytometry; the relative expression of pro–IL-18 mRNA was quan-

tified by RT-qPCR, with normalization to whole spleen expression levels.

Data in (A) are from a single experiment; data in (B) and (C) are repre-

sentative of two independent experiments for the gene-deficient strains.

Data in (D) represents median and range of three to five independent

experiments/cell type.

The Journal of Immunology 4975

by guest on April 28, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

the macrophage scavenger receptors MARCO, SRA1, and SIGNR1,which bind either to nonopsonized particles or to components ofthe polysaccharide capsule and mediate phagocytosis (47, 49,73). The influence of IFN-g on scavenger receptor expressionwas examined by quantifying mRNA for scavenger receptorswithin the brains of infected animals at 48 and 72 h p.i. (Sup-plemental Fig. 3A, 3B), as well as by determining the expressionof protein on recruited monocytes and resident microglia by flowcytometry at 72 h p.i. (Supplemental Fig. 3C). Despite the cleardifference in bacterial load at this time, and although gene ex-pression of both MARCO and SRA1 was significantly inducedfollowing infection, there were no differences in the expression

levels of any scavenger receptor between infected WT andIFN-g2/2 animals.

DiscussionKey drivers of pathology in pneumococcal meningitis are theimmune processes in response to the bacterium; however, theseimmune responses are a double-edged sword: they enhance pa-thology but are also essential for clearance of the organism. It isagainst this background that we investigated the involvement ofIFNs in both pathology and immunity in pneumococcal meningitis.Initial analysis of gene-array data revealed an apparent overall

pattern of upregulation of IFN-responsive genes. The majority of

FIGURE 4. Influence of IFN-g gene deletion on leukocyte recruitment. WT or IFN-g2/2 (KO) mice were inoculated with PBS or S. pneumoniae (∼5 3103 CFU; S.p.) icv, brains were collected at 48 h (A) or 72 h (B) p.i., and recruited leukocytes were quantified by flow cytometry. Data shown are from four

(A) or two (B) pooled experiments. Symbols represent individual animals, and horizontal lines represent the arithmetic means of groups. Total n = 12–25/

group (48 h p.i.) or n = 5–8/group (72 h p.i.). *p , 0.05, one-way ANOVA with Tukey posttest. ns, Not significant.

FIGURE 5. Effect of IFN-g gene deletion on inflammatory mediators. WT or IFN-g2/2 (KO) mice were inoculated with PBS or S. pneumoniae (∼5 3103 CFU; S.p.) icv, and mRNA levels of the indicated genes were quantified by RT-qPCR in brains from mice at 48 h (A) or 72 h (B) p.i. Levels of cytokine

proteins, normalized to total protein, were quantified in brain homogenates from mice at 48 h (C) or 72 h (D) p.i by CBA. Symbols represent individual

animals, and horizontal lines represent the geometric means of groups. Data shown are from four independent experiments with total n = 10–19/group (A),

two independent experiments with n = 6–20/group (B), or a single independent experiment with total n = 6–14/group (48 h p.i.) or n = 3–7/group (72 h p.i.)

(C, D). *p , 0.05, **p , 0.01, one-way ANOVA with Tukey posttest on log-transformed data. ns, Not significant.

4976 IFN-g IN PNEUMOCOCCAL MENINGITIS

by guest on April 28, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

these were able to be classified as responsive to type I IFNs, andapproximately one third of the 200 most differentially expressedgenes were classified as type II dependent. Although type I IFNshistorically have been associated with rapid innate protection inviral infection, it has recently become clear that they can playcritical roles in developing immune responses; thus, a potential rolefor these was investigated more fully. Despite the initial results ofthe gene-microarray analysis, a contribution of type I IFNs topathology was ruled out, because none of the type I genes testedhad upregulated expression following infection. The most likelyexplanation for this discrepancy is that IFN-regulated genes can beactivated independently of IFNs. In particular, this can occur ina primary pattern recognition response by activation of IRFs,typically IRF3 or IRF7 (reviewed in Ref. 74). Furthermore, micerendered incapable of responding to type I IFNs through geneticablation of either IFNAR1 or IFNAR2 developed severe disease,with similar kinetics to WT animals.In contrast to the absence of upregulation seen with type I

IFN genes, both IFN-g mRNA and protein were stronglyinduced following infection with S. pneumoniae. More impor-tantly, IFN-g contributed to the development of pathology, be-cause IFN-g2/2 mice were significantly protected from lethality.Evidence for the production of IFN-g in experimental models of S.pneumoniae BM is conflicting. Production of both IFN-g mRNAand protein was shown in the rat (75). In contrast, althoughZwijnenburg et al. (72) reported that IL-182/2 mice are signifi-cantly protected from the lethality of pneumococcal meningitis,these investigators were unable to detect any induction of IFN-gprotein in either WT or IL-18–deficient animals. This finding wasargued to be evidence for an IL-18–dependent, but IFN-g–inde-pendent, pathway of protection. However, in preliminary experi-ments (data not shown), we found evidence for the presence offactors in mouse brain homogenates that interfered with mea-surement of IFN-g by at least one commercially available anti–IFN-g ELISA. In contrast, the CBA that was used in the currentstudy was not affected; this may explain, in part, the discrepancybetween our results and the previously published study (72). Im-portantly, the presence of IFN-g within the CSF of patients withpneumococcal meningitis was reported. Intriguingly, IFN-g levelswere higher in the CSF of patients infected with S. pneumoniae

than in those infected with either Haemophilus influenzae orNeisseria meningiditis (41, 42). This suggests that specificpathogen-sensing and IFN-g–induction pathways may be acti-vated within the brain during S. pneumoniae infection that are nottriggered during infection with these other pathogens.When the pathways leading to the production of IFN-g were

examined in detail, inflammasome formation was found to playa dominant role. Inflammasomes are high molecular weight pro-tein platforms that sense pathogen-associated molecular patternsor danger-associated molecular patterns (reviewed in Ref. 76). Wefound that IFN-g production in the brain following S. pneumoniaeinoculation was dependent upon ASC, which is a key adaptormolecule for multiple inflammasomes. Our data are consistentwith reports that argue that the pneumococcal toxin pneumolysinis recognized via ASC-dependent inflammasomes, in particularthose containing NLRP3 and/or AIM2 (16–19). Recently, IFN-gproduction secondary to inflammasome-mediated IL-18 activationwas reported in response to flagellated bacteria (77) or to S.pneumoniae during lung infection (18). Although inflammasome/caspase-dependent cleavage is the most well-characterized path-way of IL-18 activation, a number of other enzymes, includingneutrophil proteinase-3 (78), mast cell chymase (79), and gran-zyme B (80), were argued to be capable of fulfilling this function.Moreover, a range of cytokines and costimulatory molecules hasbeen implicated in IFN-g production by NK cells (81). Of par-ticular relevance, IL-12 was reported to be involved in this processduring S. pneumoniae infection in the lung (26). However, theclear dependence of IFN-g upon ASC and IL-18 shown in thisstudy indicates that IL-18 production secondary to inflammasomeformation is the dominant pathway of IFN-g production duringpneumococcal infection in the brain. This is of particular relevance,because a recent study showed that the NLRP3 inflammasomecontributes to pathology in pneumococcal meningitis through pro-duction of IL-1b and IL-18, and this was associated with diminishedleukocyte recruitment in NLRP3-deficient animals (17). Our dataextend these findings to show that inflammasome-dependent pa-thology is, at least in large part, dependent upon IL-18–mediatedinduction of IFN-g rather than through direct effects of IL-1band IL-18.With regard to the cells producing IFN-g, our data strongly

support NK cells as the dominant source in meningitis. This iscontrast to the lung, where the cellular source of IFN-g is debated,with different groups arguing for neutrophils (82), NKT cells (83),and NK cells (18). In our model, NK cells were recruited to thebrains of infected mice in large numbers, they expressed highlevels of IL-18R, and a large proportion of these cells, but notother cell populations examined, stained for intracellular IFN-g.Moreover, depletion of NK cells greatly reduced levels of bothIFN-g protein and mRNA within the brain, whereas no such de-crease was seen in animals deficient or depleted of T cell subsets(RAG12/2, CD1d2/2, or CD8 depleted).The proposed mechanisms leading to pathology in pneumo-

coccal meningitis are multiple and include development of edema,inflammation, and the generation of reactive oxygen and nitrogenspecies (reviewed in Refs. 9, 84). A critical role for IFN-g inedema formation was discounted, because we did not find anyeffect of IFN-g on brain water content in infected animals. Per-haps more surprisingly, although some modulation of leukocyteand chemokine expression was seen, leukocyte recruitment wasnot dramatically affected by IFN-g gene deletion. Neutrophils area likely candidate target cell for IFN-g action. It was reported that,during S. pneumoniae infection in the lung, inhibition/geneticablation of IFN-g leads to increased mortality, which is typi-cally associated with decreased neutrophil influx and activation, as

FIGURE 6. Absence of IFN-g promotes clearance of S. pneumoniae.

WT or IFN-g2/2 (KO) mice were inoculated with PBS or S. pneumoniae

(S.p.), brains were collected, and bacterial load was determined by plate

count at 48 h p.i. (two pooled experiments) (A) and 72 h p.i. (three pooled

experiments) (B). Symbols represent individual animals, and horizontal

lines represent the geometric means of groups. Total n = 17–18/group (48 h

p.i.) or n = 28/group (72 h p.i.). **p , 0.01, one-way ANOVAwith Tukey

posttest on log-transformed data. ns, Not significant.

The Journal of Immunology 4977

by guest on April 28, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

well as increased bacterial load (25, 26, 45, 83). However, inmeningitis, the influence of neutrophil recruitment, activation, andsurvival appears to be more complex, perhaps because the brain islikely to be more susceptible to damage caused by inflammation.Neutrophil depletion leads to increased mortality and higherbacterial loads (22), whereas, alternatively, the prolonged presenceof activated neutrophils, secondary to inhibition of apoptosis,contributes to pathology (52). Although IFN-g was shown to havean antiapoptotic effect on neutrophils (85), it appears that in themodel used in the current study, at least, the contribution of IFN-gto neutrophil survival is minimal and this is not a dominantpathway leading to pathogenesis. The absence of an effect ofIFN-g on neutrophil recruitment contrasts with its apparent role inrecruiting monocytes. At 48 h p.i., a time when clear differenceswere noted in the clinical scores of infected WT and IFN-g2/2

mice, numbers of recruited inflammatory monocytes were sig-nificantly lower in the brains of infected IFN-g2/2 mice comparedwith infected controls. This observed decrease in monocytenumbers was consistent with a decrease in CCL2 protein in brainhomogenates at 48 h p.i., supporting a major role for this che-mokine in monocyte recruitment. Despite this clear effect onmonocyte recruitment, it is difficult to attribute the dramatic in-crease in the survival of IFN-g2/2 animals to this alone, becausethe magnitude was relatively modest, and it was reported thatblockade of CCL2-mediated inflammatory monocyte recruitmentdoes not influence disease progression (22).In addition to gross effects on the recruitment and survival of

inflammatory cells, IFN-g has the potential to drive tissue damagethrough excessive myeloid activation, which may be mediatedthrough a combination of increased oxidative burst and formationof NO and prostanoids. Indeed, IFN-g is a potent stimulator ofthese functions in both neutrophils (reviewed in Ref. 86) andmacrophages (reviewed in Ref. 53). NOS2 induction, and thesubsequent generation of NO, in particular, were argued to con-tribute to pathology in S. pneumoniae BM (55, 87). We found thatinduction of mRNA for NOS2, but not COX2, was strongly di-minished in infected IFN-g gene-deficient animals at both 48 and72 h p.i. compared with infected WT controls, suggesting thatIFN-g–dependent induction of NO production via NOS2 doesindeed contribute to pathology. Because monocytes do not appearto be critical in pathology, it is likely that CNS-resident myeloidpopulations, such as microglia or meningeal macrophages, may bethe source of the induced NOS2.A final, potentially key pathway by which IFN-g could mediate

severe meningitic disease is by inhibiting clearance of bacteria.Indeed, bacterial loads within the brains of infected IFN-g2/2 micewere decreased ∼10-fold compared with infected WT mice at 72 hp.i. Previously, evidence for a role of IFN-g in modulating clearanceof bacteria came primarily from lung infection models. In contrastto our data, in the lung, IFN-g typically has been reported to en-hance, rather than inhibit, bacterial clearance, with protection beingattributed to increased IFN-g–dependent chemokine productionleading to neutrophil recruitment (46, 83). Based on the previouslynoted lack of influence of IFN-g on chemokine expression orneutrophil recruitment, such mechanisms are unlikely to be driv-ing bacterial clearance in our model. Alternatively, in coinfectionstudies with influenza virus, it was shown that virus-dependentproduction of IFN-g by T cells inhibits the expression of themacrophage scavenger receptor MARCO, and this leads to in-creased susceptibility to secondary infection with S. pneumoniae(48). However, we were unable to demonstrate any effect of IFN-gon the expression of scavenger receptors in our model. Taken to-gether, this suggests the existence of unidentified, IFN-g–inhibitedpneumococcal clearance mechanism(s).

In conclusion, this study showed that IFN-g is a major driver ofpathology during pneumococcal meningitis. The cytokine is pro-duced following activation of inflammasome pathways and stim-ulation of NK cells by IL-18. Subsequently, IFN-g contributes topathology by modulating a range of processes, including myeloidrecruitment and activation, as well as by inhibiting bacterialclearance. These findings reinforce that dysregulated immuneprocesses contribute to morbidity and mortality in meningitisand suggest that targeting of IFN-g or related pathways may bea feasible approach for adjunctive therapy.

AcknowledgmentsWe thank Dr. Hong Duong and Yang Liu for expert technical assistance.

DisclosuresThe authors have no financial conflicts of interest.

References1. World Health Organization. 2008. The Global Burden of Disease - 2004 Update.

World Health Organization, Geneva, Switzerland.2. Gessner, B. D., J. E. Mueller, and S. Yaro. 2010. African meningitis belt pneu-

mococcal disease epidemiology indicates a need for an effective serotype 1 con-taining vaccine, including for older children and adults. BMC Infect. Dis. 10: 22.

3. Arditi, M., E. O. Mason, Jr., J. S. Bradley, T. Q. Tan, W. J. Barson, G. E. Schutze,E. R. Wald, L. B. Givner, K. S. Kim, R. Yogev, and S. L. Kaplan. 1998. Three-year multicenter surveillance of pneumococcal meningitis in children: clinicalcharacteristics, and outcome related to penicillin susceptibility and dexametha-sone use. Pediatrics 102: 1087–1097.

4. Boisier, P., H. B. Maınassara, F. Sidikou, S. Djibo, K. K. Kairo, and S. Chanteau.2007. Case-fatality ratio of bacterial meningitis in the African meningitis belt:we can do better. Vaccine 25(Suppl. 1): A24–A29.

5. Schuchat, A., K. Robinson, J. D. Wenger, L. H. Harrison, M. Farley,A. L. Reingold, L. Lefkowitz, and B. A. Perkins; Active Surveillance Team.1997. Bacterial meningitis in the United States in 1995. N. Engl. J. Med. 337:970–976.

6. Kornelisse, R. F., C. M. Westerbeek, A. B. Spoor, B. van der Heijde,L. Spanjaard, H. J. Neijens, and R. de Groot. 1995. Pneumococcal meningitis inchildren: prognostic indicators and outcome. Clin. Infect. Dis. 21: 1390–1397.

7. Merkelbach, S., H. Sittinger, I. Schweizer, and M. Muller. 2000. Cognitiveoutcome after bacterial meningitis. Acta Neurol. Scand. 102: 118–123.

8. van de Beek, D., B. Schmand, J. de Gans, M. Weisfelt, H. Vaessen, J. Dankert,and M. Vermeulen. 2002. Cognitive impairment in adults with good recoveryafter bacterial meningitis. J. Infect. Dis. 186: 1047–1052.

9. Koedel, U., M. Klein, and H. W. Pfister. 2010. New understandings on thepathophysiology of bacterial meningitis. Curr. Opin. Infect. Dis. 23: 217–223.

10. Nelson, R. P., Jr. 2006. Bacterial meningitis and inflammation. Curr. Opin.Neurol. 19: 369–373.

11. Brouwer, M. C., J. de Gans, S. G. Heckenberg, A. H. Zwinderman, T. van derPoll, and D. van de Beek. 2009. Host genetic susceptibility to pneumococcal andmeningococcal disease: a systematic review and meta-analysis. Lancet Infect.Dis. 9: 31–44.

12. Koedel, U., B. Angele, T. Rupprecht, H. Wagner, A. Roggenkamp, H. W. Pfister,and C. J. Kirschning. 2003. Toll-like receptor 2 participates in mediation ofimmune response in experimental pneumococcal meningitis. J. Immunol. 170:438–444.

13. Klein, M., B. Obermaier, B. Angele, H.-W. Pfister, H. Wagner, U. Koedel, andC. J. Kirschning. 2008. Innate immunity to pneumococcal infection of the centralnervous system depends on toll-like receptor (TLR) 2 and TLR4. J. Infect. Dis.198: 1028–1036.

14. Echchannaoui, H., K. Frei, C. Schnell, S. L. Leib, W. Zimmerli, andR. Landmann. 2002. Toll-like receptor 2-deficient mice are highly susceptible toStreptococcus pneumoniae meningitis because of reduced bacterial clearing andenhanced inflammation. J. Infect. Dis. 186: 798–806.

15. Liu, X., V. S. Chauhan, A. B. Young, and I. Marriott. 2010. NOD2 mediatesinflammatory responses of primary murine glia to Streptococcus pneumoniae.Glia 58: 839–847.

16. Fang, R., K. Tsuchiya, I. Kawamura, Y. Shen, H. Hara, S. Sakai, T. Yamamoto,T. Fernandes-Alnemri, R. Yang, E. Hernandez-Cuellar, et al. 2011. Critical rolesof ASC inflammasomes in caspase-1 activation and host innate resistance toStreptococcus pneumoniae infection. J. Immunol. 187: 4890–4899.

17. Hoegen, T., N. Tremel, M. Klein, B. Angele, H. Wagner, C. Kirschning, H.-W. Pfister,A. Fontana, S. Hammerschmidt, and U. Koedel. 2011. The NLRP3 inflamma-some contributes to brain injury in pneumococcal meningitis and is activatedthrough ATP-dependent lysosomal cathepsin B release. J. Immunol. 187: 5440–5451.

18. McNeela, E. A., A. Burke, D. R. Neill, C. Baxter, V. E. Fernandes, D. Ferreira,S. Smeaton, R. El-Rachkidy, R. M. McLoughlin, A. Mori, et al. 2010. Pneu-molysin activates the NLRP3 inflammasome and promotes proinflammatorycytokines independently of TLR4. PLoS Pathog. 6: e1001191.

4978 IFN-g IN PNEUMOCOCCAL MENINGITIS

by guest on April 28, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

19. Witzenrath, M., F. Pache, D. Lorenz, U. Koppe, B. Gutbier, C. Tabeling,K. Reppe, K. Meixenberger, A. Dorhoi, J. Ma, et al. 2011. The NLRP3inflammasome is differentially activated by pneumolysin variants and contrib-utes to host defense in pneumococcal pneumonia. J. Immunol. 187: 434–440.

20. Saukkonen, K., S. Sande, C. Cioffe, S. Wolpe, B. Sherry, A. Cerami, andE. Tuomanen. 1990. The role of cytokines in the generation of inflammation andtissue damage in experimental gram-positive meningitis. J. Exp. Med. 171: 439–448.

21. Tuomanen, E. I., K. Saukkonen, S. Sande, C. Cioffe, and S. D. Wright. 1989.Reduction of inflammation, tissue damage, and mortality in bacterial meningitisin rabbits treated with monoclonal antibodies against adhesion-promotingreceptors of leukocytes. J. Exp. Med. 170: 959–969.

22. Mildner, A., M. Djukic, D. Garbe, A. Wellmer, W. A. Kuziel, M. Mack, R. Nau,and M. Prinz. 2008. Ly-6G+CCR22myeloid cells rather than Ly-6ChighCCR2+monocytes are required for the control of bacterial infection in the central ner-vous system. J. Immunol. 181: 2713–2722.

23. Rijneveld, A. W., F. N. Lauw, M. J. Schultz, S. Florquin, A. A. Te Velde,P. Speelman, S. J. H. Van Deventer, and T. Van Der Poll. 2002. The role ofinterferon-g in murine pneumococcal pneumonia. J. Infect. Dis. 185: 91–97.

24. Standish, A. J., and J. N. Weiser. 2009. Human neutrophils kill Streptococcuspneumoniae via serine proteases. J. Immunol. 183: 2602–2609.

25. Weber, S. E., H. Tian, and L. A. Pirofski. 2011. CD8+ cells enhance resistance topulmonary serotype 3 Streptococcus pneumoniae infection in mice. J. Immunol.186: 432–442.

26. Yamamoto, N., K. Kawakami, Y. Kinjo, K. Miyagi, T. Kinjo, K. Uezu,C. Nakasone, M. Nakamatsu, and A. Saito. 2004. Essential role for the p40subunit of interleukin-12 in neutrophil-mediated early host defense againstpulmonary infection with Streptococcus pneumoniae: involvement of interferon-g. Microbes Infect. 6: 1241–1249.

27. Blaser, C., M. Wittwer, D. Grandgirard, and S. L. Leib. 2011. Adjunctivedexamethasone affects the expression of genes related to inflammation, neuro-genesis and apoptosis in infant rat pneumococcal meningitis. PLoS ONE 6:e17840.

28. van de Beek, D., J. de Gans, P. McIntyre, and K. Prasad. 2004. Steroids in adultswith acute bacterial meningitis: a systematic review. Lancet Infect. Dis. 4: 139–143.

29. Wellmer, A., J. Gerber, J. Ragheb, G. Zysk, T. Kunst, A. Smirnov, W. Bruck, andR. Nau. 2001. Effect of deficiency of tumor necrosis factor alpha or both of itsreceptors on Streptococcus pneumoniae central nervous system infection andperitonitis. Infect. Immun. 69: 6881–6886.

30. Zwijnenburg, P. J., T. van der Poll, S. Florquin, J. J. Roord, and A. M. Van Furth.2003. IL-1 receptor type 1 gene-deficient mice demonstrate an impaired hostdefense against pneumococcal meningitis. J. Immunol. 170: 4724–4730.

31. Vigario, A. M., E. Belnoue, A. C. Gruner, M. Mauduit, M. Kayibanda,J. C. Deschemin, M. Marussig, G. Snounou, D. Mazier, I. Gresser, and L. Renia.2007. Recombinant human IFN-alpha inhibits cerebral malaria and reducesparasite burden in mice. J. Immunol. 178: 6416–6425.

32. O’Connell, R. M., S. K. Saha, S. A. Vaidya, K. W. Bruhn, G. A. Miranda,B. Zarnegar, A. K. Perry, B. O. Nguyen, T. F. Lane, T. Taniguchi, et al. 2004.Type I interferon production enhances susceptibility to Listeria monocytogenesinfection. J. Exp. Med. 200: 437–445.

33. Cooper, A. M., D. K. Dalton, T. A. Stewart, J. P. Griffin, D. G. Russell, andI. M. Orme. 1993. Disseminated tuberculosis in interferon gamma gene-disrupted mice. J. Exp. Med. 178: 2243–2247.

34. Buchmeier, N. A., and R. D. Schreiber. 1985. Requirement of endogenousinterferon-gamma production for resolution of Listeria monocytogenes infection.Proc. Natl. Acad. Sci. USA 82: 7404–7408.

35. Haahr, S. 1968. The occurrence of interferon in the cerebrospinal fluid in patientswith bacterial meningitis. Acta Pathol. Microbiol. Scand. 73: 264–274.

36. Fassbender, K., O. Mielke, T. Bertsch, F. Muehlhauser, M. Hennerici,M. Kurimoto, and S. Rossol. 1999. Interferon-gamma-inducing factor (IL-18)and interferon-gamma in inflammatory CNS diseases. Neurology 53: 1104–1106.

37. Larke, R. P. 1967. Interferon in the cerebrospinal fluid of children with centralnervous system disorders. Can. Med. Assoc. J. 96: 21–32.

38. Ohga, S., T. Aoki, K. Okada, H. Akeda, K. Fujioka, A. Ohshima, T. Mori,I. Minamishima, and K. Ueda. 1994. Cerebrospinal fluid concentrations ofinterleukin-1 beta, tumour necrosis factor-alpha, and interferon gamma in bac-terial meningitis. Arch. Dis. Child. 70: 123–125.

39. Chonmaitree, T., and S. Baron. 1991. Bacteria and viruses induce production ofinterferon in the cerebrospinal fluid of children with acute meningitis: a study of57 cases and review. Rev. Infect. Dis. 13: 1061–1065.

40. Abbott, R. J., I. Bolderson, P. J. Gruer, and R. C. Peatfield. 1987. Immunore-active IFN-gamma in CSF in neurological disorders. J. Neurol. Neurosurg.Psychiatry 50: 882–885.

41. Glimaker, M., P. Olcen, and B. Andersson. 1994. Interferon-gamma in cere-brospinal fluid from patients with viral and bacterial meningitis. Scand. J. Infect.Dis. 26: 141–147.

42. Kornelisse, R. F., C. E. Hack, H. F. Savelkoul, T. C. van der Pouw Kraan,W. C. Hop, G. van Mierlo, M. H. Suur, H. J. Neijens, and R. de Groot. 1997.Intrathecal production of interleukin-12 and gamma interferon in patients withbacterial meningitis. Infect. Immun. 65: 877–881.

43. Diab, A., J. Zhu, L. Lindquist, B. Wretlind, M. Bakhiet, and H. Link. 1997.Haemophilus influenzae and Streptococcus pneumoniae induce different intra-cerebral mRNA cytokine patterns during the course of experimental bacterialmeningitis. Clin. Exp. Immunol. 109: 233–241.

44. Bjerre, A., B. Brusletto, E. A. Høiby, P. Kierulf, and P. Brandtzaeg. 2004. Plasmainterferon-gamma and interleukin-10 concentrations in systemic meningococcaldisease compared with severe systemic Gram-positive septic shock. Crit. CareMed. 32: 433–438.

45. Rubins, J. B., and C. Pomeroy. 1997. Role of gamma interferon in the patho-genesis of bacteremic pneumococcal pneumonia. Infect. Immun. 65: 2975–2977.

46. Sun, K., S. L. Salmon, S. A. Lotz, and D. W. Metzger. 2007. Interleukin-12promotes gamma interferon-dependent neutrophil recruitment in the lung andimproves protection against respiratory Streptococcus pneumoniae infection.Infect. Immun. 75: 1196–1202.

47. Arredouani, M., Z. Yang, Y. Ning, G. Qin, R. Soininen, K. Tryggvason, andL. Kobzik. 2004. The scavenger receptor MARCO is required for lung defenseagainst pneumococcal pneumonia and inhaled particles. J. Exp. Med. 200: 267–272.

48. Sun, K., and D. W. Metzger. 2008. Inhibition of pulmonary antibacterial defenseby interferon-g during recovery from influenza infection. Nat. Med. 14: 558–564.

49. Arredouani, M. S., Z. Yang, A. Imrich, Y. Ning, G. Qin, and L. Kobzik. 2006.The macrophage scavenger receptor SR-AI/II and lung defense against pneu-mococci and particles. Am. J. Respir. Cell Mol. Biol. 35: 474–478.

50. Lanoue, A., M. R. Clatworthy, P. Smith, S. Green, M. J. Townsend, H. E. Jolin,K. G. Smith, P. G. Fallon, and A. N. McKenzie. 2004. SIGN-R1 contributes toprotection against lethal pneumococcal infection in mice. J. Exp. Med. 200:1383–1393.

51. Grewal, T., E. Priceputu, J. Davignon, and L. Bernier. 2001. Identification ofa gamma-interferon-responsive element in the promoter of the human macro-phage scavenger receptor A gene. Arterioscler. Thromb. Vasc. Biol. 21: 825–831.

52. Koedel, U., T. Frankenberg, S. Kirschnek, B. Obermaier, H. Hacker, R. Paul, andG. Hacker. 2009. Apoptosis is essential for neutrophil functional shutdown anddetermines tissue damage in experimental pneumococcal meningitis. PLoSPathog. 5: e1000461.

53. Murray, P. J., and T. A. Wynn. 2011. Protective and pathogenic functions ofmacrophage subsets. Nat. Rev. Immunol. 11: 723–737.

54. Xie, Q. W., R. Whisnant, and C. Nathan. 1993. Promoter of the mouse geneencoding calcium-independent nitric oxide synthase confers inducibility by in-terferon gamma and bacterial lipopolysaccharide. J. Exp. Med. 177: 1779–1784.

55. Kornelisse, R. F., K. Hoekman, J. J. Visser, W. C. Hop, J. G. Huijmans, P. J. vander Straaten, A. J. van der Heijden, R. N. Sukhai, H. J. Neijens, and R. de Groot.1996. The role of nitric oxide in bacterial meningitis in children. J. Infect. Dis.174: 120–126.

56. Takeda, K., H. Tsutsui, T. Yoshimoto, O. Adachi, N. Yoshida, T. Kishimoto,H. Okamura, K. Nakanishi, and S. Akira. 1998. Defective NK cell activity andTh1 response in IL-18-deficient mice. Immunity 8: 383–390.

57. Dalton, D. K., S. Pitts-Meek, S. Keshav, I. S. Figari, A. Bradley, andT. A. Stewart. 1993. Multiple defects of immune cell function in mice withdisrupted interferon-gamma genes. Science 259: 1739–1742.

58. Hancock, W. W., B. Lu, W. Gao, V. Csizmadia, K. Faia, J. A. King, S. T. Smiley,M. Ling, N. P. Gerard, and C. Gerard. 2000. Requirement of the chemokinereceptor CXCR3 for acute allograft rejection. J. Exp. Med. 192: 1515–1520.

59. Mendiratta, S. K., W. D. Martin, S. Hong, A. Boesteanu, S. Joyce, and L. VanKaer. 1997. CD1d1 mutant mice are deficient in natural T cells that promptlyproduce IL-4. Immunity 6: 469–477.

60. Hwang, S. Y., P. J. Hertzog, K. A. Holland, S. H. Sumarsono, M. J. Tymms,J. A. Hamilton, G. Whitty, I. Bertoncello, and I. Kola. 1995. A null mutation inthe gene encoding a type I interferon receptor component eliminates anti-proliferative and antiviral responses to interferons alpha and beta and altersmacrophage responses. Proc. Natl. Acad. Sci. USA 92: 11284–11288.

61. Fenner, J. E., R. Starr, A. L. Cornish, J. G. Zhang, D. Metcalf, R. D. Schreiber,K. Sheehan, D. J. Hilton, W. S. Alexander, and P. J. Hertzog. 2006. Suppressor ofcytokine signaling 1 regulates the immune response to infection by a uniqueinhibition of type I interferon activity. Nat. Immunol. 7: 33–39.

62. Mariathasan, S., K. Newton, D. M. Monack, D. Vucic, D. M. French, W. P. Lee,M. Roose-Girma, S. Erickson, and V. M. Dixit. 2004. Differential activation ofthe inflammasome by caspase-1 adaptors ASC and Ipaf. Nature 430: 213–218.

63. Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expressiondata using real-time quantitative PCR and the 2(2Delta Delta C(T)) Method.Methods 25: 402–408.

64. Irizarry, R. A., B. M. Bolstad, F. Collin, L. M. Cope, B. Hobbs, and T. P. Speed.2003. Summaries of Affymetrix GeneChip probe level data. Nucleic Acids Res.31: e15.

65. Mitchell, A. J., L. C. Pradel, L. Chasson, N. Van Rooijen, G. E. Grau,N. H. Hunt, and G. Chimini. 2010. Technical advance: autofluorescence as a toolfor myeloid cell analysis. J. Leukoc. Biol. 88: 597–603.

66. Miu, J., A. J. Mitchell, M. Muller, S. L. Carter, P. M. Manders, J. A. McQuillan,B. M. Saunders, H. J. Ball, B. Lu, I. L. Campbell, and N. H. Hunt. 2008.Chemokine gene expression during fatal murine cerebral malaria and protectiondue to CXCR3 deficiency. J. Immunol. 180: 1217–1230.

67. Dupont, W. D., and W. D. Plummer, Jr. 1990. Power and sample size calcu-lations. A review and computer program. Control. Clin. Trials 11: 116–128.

68. Samarajiwa, S. A., S. Forster, K. Auchettl, and P. J. Hertzog. 2008. INTER-FEROME: the database of interferon regulated genes. Nucleic Acids Res. 37:D852–D857.

69. Martinon, F., V. Petrilli, A. Mayor, A. Tardivel, and J. Tschopp. 2006. Gout-associated uric acid crystals activate the NALP3 inflammasome. Nature 440:237–241.

70. Kastenbauer, S., M. Klein, U. Koedel, and H. W. Pfister. 2001. Reactive nitrogenspecies contribute to blood-labyrinth barrier disruption in suppurative laby-rinthitis complicating experimental pneumococcal meningitis in the rat. BrainRes. 904: 208–217.

71. Braun, J. S. 2009. Inducible nitric oxide synthase mediates hippocampalcaspase-3 activation in pneumococcal meningitis. Int. J. Neurosci. 119: 455–459.

72. Zwijnenburg, P. J., T. van der Poll, S. Florquin, S. Akira, K. Takeda, J. J. Roord,and A. M. van Furth. 2003. Interleukin-18 gene-deficient mice show enhanced

The Journal of Immunology 4979

by guest on April 28, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

defense and reduced inflammation during pneumococcal meningitis. J. Neuro-immunol. 138: 31–37.

73. Park, J.-Y., H. J. Choi, M. G. Prabagar, W.-S. Choi, S.-J. Kim, C. Cheong,C. G. Park, C. Y. Chin, and Y.-S. Kang. 2009. The C-type lectin CD209b isexpressed on microglia and it mediates the uptake of capsular polysaccharides ofStreptococcus pneumoniae. Neurosci. Lett. 450: 246–251.

74. Tamura, T., H. Yanai, D. Savitsky, and T. Taniguchi. 2008. The IRF familytranscription factors in immunity and oncogenesis. Annu. Rev. Immunol. 26:535–584.

75. Diab, A., J. Zhu, L. Lindquist, B. Wretlind, H. Link, and M. Bakhiet. 1997.Cytokine mRNA profiles during the course of experimental Haemophilusinfluenzae bacterial meningitis. Clin. Immunol. Immunopathol. 85: 236–245.

76. Schroder, K., and J. Tschopp. 2010. The inflammasomes. Cell 140: 821–832.77. Kupz, A., G. Guarda, T. Gebhardt, L. E. Sander, K. R. Short,

D. A. Diavatopoulos, O. L. Wijburg, H. Cao, J. C. Waithman, W. Chen, et al.2012. NLRC4 inflammasomes in dendritic cells regulate noncognate effectorfunction by memory CD8+ T cells. Nat. Immunol. 13: 162–169.

78. Sugawara, S., A. Uehara, T. Nochi, T. Yamaguchi, H. Ueda, A. Sugiyama,K. Hanzawa, K. Kumagai, H. Okamura, and H. Takada. 2001. Neutrophil pro-teinase 3-mediated induction of bioactive IL-18 secretion by human oral epi-thelial cells. J. Immunol. 167: 6568–6575.

79. Omoto, Y., K. Tokime, K. Yamanaka, K. Habe, T. Morioka, I. Kurokawa,H. Tsutsui, K. Yamanishi, K. Nakanishi, and H. Mizutani. 2006. Human mastcell chymase cleaves pro-IL-18 and generates a novel and biologically active IL-18 fragment. J. Immunol. 177: 8315–8319.

80. Omoto, Y., K. Yamanaka, K. Tokime, S. Kitano, M. Kakeda, T. Akeda,I. Kurokawa, E. C. Gabazza, H. Tsutsui, N. Katayama, et al. 2010. Granzyme Bis a novel interleukin-18 converting enzyme. J. Dermatol. Sci. 59: 129–135.

81. Newman, K. C., and E. M. Riley. 2007. Whatever turns you on: accessory-cell-dependent activation of NK cells by pathogens. Nat. Rev. Immunol. 7: 279–291.

82. Yamada, M., J. C. Gomez, P. E. Chugh, C. A. Lowell, M. C. Dinauer,D. P. Dittmer, and C. M. Doerschuk. 2011. Interferon-g production by neu-trophils during bacterial pneumonia in mice. Am. J. Respir. Crit. Care Med. 183:1391–1401.

83. Nakamatsu, M., N. Yamamoto, M. Hatta, C. Nakasone, T. Kinjo, K. Miyagi,K. Uezu, K. Nakamura, T. Nakayama, M. Taniguchi, et al. 2007. Role of inter-feron-g in Valpha14+ natural killer T cell-mediated host defense against Strepto-coccus pneumoniae infection in murine lungs. Microbes Infect. 9: 364–374.

84. Mook-Kanamori, B. B., M. Geldhoff, T. van der Poll, and D. van de Beek. 2011.Pathogenesis and pathophysiology of pneumococcal meningitis. Clin. Microbiol.Rev. 24: 557–591.

85. Colotta, F., F. Re, N. Polentarutti, S. Sozzani, and A. Mantovani. 1992. Modu-lation of granulocyte survival and programmed cell death by cytokines andbacterial products. Blood 80: 2012–2020.

86. Ellis, T. N., and B. L. Beaman. 2004. Interferon-gamma activation of poly-morphonuclear neutrophil function. Immunology 112: 2–12.

87. Boje, K. M., D. Jaworowicz, Jr., and J. J. Raybon. 2003. Neuroinflammatory roleof prostaglandins during experimental meningitis: evidence suggestive of anin vivo relationship between nitric oxide and prostaglandins. J. Pharmacol. Exp.Ther. 304: 319–325.

4980 IFN-g IN PNEUMOCOCCAL MENINGITIS

by guest on April 28, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

Supplementary Figure legends

Supplementary Figure S1. T cells and NKT cells are not major contributors to production of IFN� in pneumococcal meningitis. (A) Wild-type (WT) or RAG1-/- (KO) mice were either sham infected (PBS) or infected with S. pneumoniae (S.p.), and brains collected at 24 hours p.i. Interferon-� protein, normalized to total protein (left panel), and IFN� mRNA (right panel) were quantified by cytometric bead array and RT-qPCR respectively. (B) Wild-type mice were untreated (-), treated with anti-CD8 antibody (Ab) or control antibody (I/C) then either sham infected (PBS) or infected with S. pneumoniae (S.p.). Depletion of CD8+ T cells was confirmed by flow cytometry using an anti-CD8 antibody recognizing a epitope non-overlapping with the depletion antibody (data not shown). Relative mRNA levels in brain homogenates were quantified by RT-qPCR. (C) Wild-type (WT) or CD1d-/- (CD1dKO) mice were either sham infected (PBS) or infected with S. pneumoniae (S.p.), and IFN� mRNA in brain homogenates was quantified by RT-qPCR. For all panels, symbols represent individual animals and horizontal lines the arithmetic (protein data) or geometric (mRNA data) mean of each group. Gene expression data were normalized to the geometric mean of the uninfected control group. Data are representative of two independent experiments (panels A and B) or a single experiment (panel C). Statistical significance was determined by one-way ANOVA with Tukey post test on log transformed data. ns=not significant.

Supplementary Figure S2. IFN� gene deletion does not affect brain oedema. Percent water content was determined in sham (PBS) and S. pneumoniae (~5x103 cfu -S.p.) infected wild type (WT) and IFN�-/- (KO) mice at 48 hours p.i. (A) and 72 hours p.i. (B). Pooled data from 2 independent experiments per time-point are shown. Symbols represent individual animals and lines group means.

Supplementary Figure S3. IFN� does not modulate macrophage scavenger receptor expression. Wild type (WT) or IFN�-/- (KO) mice were inoculated with PBS or S. pneumoniae (~5x103 cfu -S.p.) icv, and brains were collected at 48 (A) and 72 (B) hours p.i. for analysis of the indicated genes by RT-qPCR. Symbols represent individual animals and horizontal lines the geometric mean of each group. Gene expression data were normalized to the geometric mean of the uninfected control group. Data represent two independent experiments processed and assayed simultaneously. Statistical significance was determined by one-way ANOVA with Tukey post test on log transformed data. ns=not significant. Gene expression for both MARCO and SRA1, but not CD209b, were significantly increased (p<0.05) compared to the corresponding sham-infected controls at both 48 and 72 hours p.i. (C) The surface expression levels of the indicated macrophage scavenger receptors on recruited monocytes (mono) and resident microglia (M/G) was determined at 72 hours p.i. by flow cytometry. Blue histograms show antibody staining and red histograms are staining controls. Data are representative of 3 mice per infected group.

A

B C

Supplementary Figure S1

- Ab - I/C Ab0.01

0.11

10100

100010000

ns

PBS S.p.

ns

Rel

ativ

e ge

neex

pres

sion

WT KO WT KO0

1000200030004000 ns

PBS S.p.

IFN�

prot

ein

(pg/

g)

WT KO WT KO0.1

110

1001000

PBS S.p.

ns

Rel

ativ

e ge

neex

pres

sion

WT

CD1dKO WT

CD1dKO

0.11

10100

1000 ns

PBS S.p.

Rel

ativ

e ge

neex

pres

sion

A

Supplementary Figure S2

WT KO WT KO767778798081 ns

PBS S.p.

% w

et w

eigh

t

WT KO WT KO767778798081

ns

PBS S.p.

% w

et w

eigh

t

B

Mono

M/G

WT KO WT KOPBS S.p.

WT KO WT KOPBS S.p.

WT KO WT KOPBS S.p.

Supplementary Figure S3

MARCO

0.1

1

10

100 ns

PBS S.p.

Rel

ativ

e ge

neex

pres

sion

SRA1

1

10

100 ns

PBS S.p.

Rel

ativ

e ge

neex

pres

sion

CD209b

0.1

1

10

100 ns

PBS S.p.

Rel

ativ

e ge

neex

pres

sion

MARCO

0.01

0.1

1

10

100ns

PBS S.p.

Rel

ativ

e ge

neex

pres

sion

CD209b

1

10

100 ns

PBS S.p.

Rel

ativ

e ge

neex

pres

sion

SRA1

1

10

100 ns

PBS S.p.

Rel

ativ

e ge

neex

pres

sion

A

B

C

WT KO WT KOWT KO WT KO WT KO WT KO

WT KO WT KOWT KO WT KO WT KO WT KO

MARCO CD209bSRA1