1 Contributes to Host Defenses against Mycobacterium tuberculosis Kazunori Matsumura, Hiroki Iwai, Masako Kato-Miyazawa, Fumiko Kirikae, Jizi Zhao, Toru Yanagawa, Tetsuro Ishii, This information is current as Tohru Miyoshi-Akiyama, Keiji Funatogawa and Teruo of September 25, 2021. Kirikae J Immunol published online 7 September 2016 http://www.jimmunol.org/content/early/2016/09/07/jimmun

ol.1601010 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2016/09/07/jimmunol.160101 Material 0.DCSupplemental 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 25, 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 © 2016 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published September 7, 2016, doi:10.4049/jimmunol.1601010 The Journal of Immunology

Peroxiredoxin 1 Contributes to Host Defenses against Mycobacterium tuberculosis

Kazunori Matsumura,* Hiroki Iwai,* Masako Kato-Miyazawa,* Fumiko Kirikae,* Jizi Zhao,* Toru Yanagawa,† Tetsuro Ishii,† Tohru Miyoshi-Akiyama,* Keiji Funatogawa,‡ and Teruo Kirikae*

Peroxiredoxin (PRDX)1 is an that detoxifies hydrogen peroxide and peroxinitrite. Compared with wild-type (WT) mice, Prdx1-deficient (Prdx12/2) mice showed increased susceptibility to Mycobacterium tuberculosis and lower levels of IFN-g and IFN- g–producing CD4+ T cells in the lungs after M. tuberculosis infection. IL-12 production, c-Rel induction, and p38 MAPK activation levels were lower in Prdx12/2 than in WT bone marrow–derived macrophages (BMDMs). IFN-g–activated Prdx12/2 BMDMs did not kill M. tubercuosis effectively. NO production levels were lower, and arginase activity and arginase 1 (Arg1) expression levels were higher, in IFN-g–activated Prdx12/2 than in WT BMDMs after M. tuberculosis infection. An arginase Downloaded from inhibitor, Nv-hydroxy-nor-arginine, restored antimicrobial activity and NO production in IFN-g–activated Prdx12/2 BMDMs after M. tuberculosis infection. These results suggest that PRDX1 contributes to host defenses against M. tuberculosis. PRDX1 positively regulates IL-12 production by inducing c-Rel and activating p38 MAPK, and it positively regulates NO production by suppressing Arg1 expression in macrophages infected with M. tuberculosis. The Journal of Immunology, 2016, 197: 000–000.

ycobacterium tuberculosis, a causative agent of tuber- cannot inhibit the intracellular growth of M. tuberculosis as ef- http://www.jimmunol.org/ culosis (TB), is responsible for ∼9 million new cases of fectively as do WT Mfs (7). In contrast, the roles of ROS in host M active TB and 1.5 million deaths per year (1). Th1 re- defenses against M. tuberculosis are unclear. NADPH oxidase is sponses play a central role in host defenses against M. tuberculosis composed of five polypeptide subunits, gp91-phox, p22-phox, (2). Following M. tuberculosis infection, macrophages (Mfs) and p40-phox, p47-phox, and p67-phox, encoded by Cybb, Cyba, dendritic cells (DCs) produce inflammatory such as Ncf4, Nrf1, and Ncf2, respectively, which bind to the Rac1 and IL-1, IL-6, IL-12, and TNF-a, with IL-12 being essential for the Rac2 GTPases, encoded by Rac1 and Rac2, respectively (8). One 2/2 differentiation of CD4+ T cells into Th1 cells (3). Th1 cells release study reported that Cybb mice were more susceptible to

IFN-g, which activates the antimicrobial activities of Mfs. These M. tuberculosis than WT mice (5), whereas another study reported by guest on September 25, 2021 2/2 activated Mfs produce reactive nitrogen species (RNS) and re- no difference in susceptibility (9). Nrf1 mice showed increased active oxygen species (ROS) via the NO synthase (NOS) M. tuberculosis growth during the early period of infection (10). Moreover, compared with WT Mfs, IFN-g–activated Cybb2/2 2 and reduced NADPH oxidase, respectively (4). 2/2 RNS play a critical role in host defenses against M. tuberculosis. and Nrf1 Mfs similarly inhibited the intracellular growth of NOS2-deficient (Nos22/2) mice are more susceptible to M. tuberculosis M. tuberculosis (5, 7). These findings were likely due to the ability 2/2 M. tuberculosis than are wild-type (WT) mice (5, 6), and MfsfromNos2 mice of to produce ROS scavenging enzymes (11). Alteration of RNS metabolism affects susceptibility to M. tuberculosis (12). L-Arginine is converted by NOS2 to NO and *Department of Infectious Diseases, Research Institute, National Center for Global Health and Medicine, Shinjuku, Tokyo 162-8655, Japan; †Faculty of Medicine, Uni- L-citrulline or by ARG1 to urea and L-ornithine (13). ARG1, versity of Tsukuba, Tsukuba, Ibaraki 305-8575, Japan; and ‡Department of Micro- therefore, competes with NOS2 for the same substrate, reducing NO biology, Tochigi Prefectural Institute of Public Health and Environmental Science, production by activated Mfs (14). Compared with WT mice, mice Utsunomiya, Tochigi 329-1196, Japan lacking Mf ARG1 showed decreased numbers of M. tuberculosis ORCID: 0000-0001-8807-3957 (H.I.). and increased NO in the lungs after M. tuberculosis infection (12). Received for publication June 10, 2016. Accepted for publication August 16, 2016. It is unclear whether alteration of ROS metabolism affects This work was supported by National Center for Global Health and Medicine Grants susceptibility to M. tuberculosis. Peroxiredoxin (PRDX)1, a major 21-A-105, 24-A-103, and 27-A-1102, Japan Society for the Promotion of Science Grant-in-Aid for Scientific Research 22590411, and by Japan Agency for Medical cytosolic PRDX, is expressed ubiquitously in tissues and affects Research and Development Grant 16fk0108302h0003. ROS metabolism (15, 16). PRDXs are a mammalian family of Address correspondence and reprint requests to Dr. Teruo Kirikae, Department of that reduce hydrogen peroxide, organic hydroperox- Infectious Diseases, Research Institute, National Center for Global Health and Medicine, 1-21-1 Toyama, Shinjuku, Tokyo 162-8655, Japan. E-mail address: ides, and peroxinitrite, protecting cells and tissues from oxidative [email protected] and nitrosative stresses (15, 16). PRDX1 has protective effects on The online version of this article contains supplemental material. oxidative injury induced by Helicobacter pylori (17), kidney Abbreviations used in this article: ASK1, apoptosis signal–regulating kinase 1; damage by administration of ferric nitrilotriacetate (18), and BMDM, bone marrow–derived Mf; 9-cRA, 9-cis retinoic acid; DC, dendritic cell; pulmonary inflammation induced by bleomycin administration Mf, macrophage; MKK, MAPK kinase; MOI, multiplicity of infection; nor-NOHA, v (19), suggesting that protective effects of PRDX1 depend on its N -hydroxy-nor-L-arginine; NOS, NO synthase; PFA, paraformaldehyde; p.i., post- infection; PRDX, peroxiredoxin; PTEN, phosphatase and tensin homolog; qRT-PCR, radical scavenger activity. quantitative RT-PCR; RNS, reactive nitrogen species; ROS, reactive oxygen species; It is known that PRDX1 has functions other than the radical SOCS, suppressor of signaling; TB, tuberculosis; WT, wild-type. scavenger activity. PRDX1 serves as chaperone in the form of Copyright Ó 2016 by The American Association of Immunologists, Inc. 0022-1767/16/$30.00 oligomer (16, 20). PRDX1 regulates innate immune responses

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1601010 2 PRDX1 CONTRIBUTES TO HOST DEFENSES AGAINST TB necessary for restoration of damaged tissue. PRDX1 increases bright-field images using Photoshop CS 11 (Adobe Systems, San Jose, CA). acute pulmonary inflammation due to O3 exposure (21). In the case of brain stroke, which causes ischemia–reperfusion injury, Cytokine/chemokine measurements PRDX1 released from necrotic cells worsened the brain damage via initiating inflammation (22). Recombinant PRDX1 binds Cytokine/chemokine levels in organs were measured using Milliplex MAP mouse cytokine/chemokine premixed 22-plex kits (Merck, Kenilworth, NJ) TLR4 and positively regulates inflammation (22–24). and a Luminex 200 (Luminex, Austin, TX), according to the manufacturers’ We hypothesized that loss of PRDX1 would increase resistance instructions. The kits detect G-CSF, GM-CSF, IFN-g, IL-1a, IL-1b, sIL-2, to M. tuberculosis. Prdx12/2 Mfs will contain excess RNS and IL-4, IL-5, IL-6, IL-7, IL-9, IL-10, IL-12 p70, IL-13, IL-15, IL-17, TNF-a, ROS, inhibiting the intracellular growth of M. tuberculosis.ROS CCL2, CCL3, CCL5, CXCL1, and CXCL10. Cytokine concentrations in cell culture supernatants were measured using ELISA kits (R&D Systems, activate NF-kB (25), leading to increased production of inflammatory Minneapolis, MN), according to the manufacturer’s instructions. cytokines by Prdx12/2 relative to WT MfsduringM. tuberculosis infection. Unexpectedly, we found that Prdx12/2 mice were more Flow cytometric analysis susceptible to M. tuberculosis than were WT mice. Following Lungs were removed from euthanized mice, cut into small pieces (∼1mm3), 2 2 M. tuberculosis infection, Prdx1 / Mfs produced lower amounts digested in Dulbecco’s PBS (Nacalai Tesque) containing 1 mg/ml colla- of IL-12 and NO than did WT Mfs, and they negatively regulated genase type II (Worthington Biochemical, Lakewood, NJ) and 0.5 mg/ml c-Rel induction (26) and p38 MAPK activation (27), both positive DNase I (Roche, Basel, Switzerland) and incubated at 37˚C for 1 h. The digests were passed through a 0.40-mm strainer (BD Biosciences) using regulators of IL-12, and upregulated Arg1 expression. These the plunger of a 1-ml syringe. After exclusion of RBCs with ammonium findings indicate that PRDX1 contributes to host defenses against chloride solution, the lung cell suspensions were washed with Dulbecco’s

M. tuberculosis by regulating Th1 responses and the antimicrobial PBS and the numbers counted by trypan blue exclusion. To detecting IFN- Downloaded from + + activity of Mfs. g–producing CD4 T cells, lung CD4 T cells were negatively selected from lung cell suspensions using mouse a CD4 T lymphocyte enrichment set–DM (BD Biosciences), fixed with 4% paraformaldehyde (PFA), and permeabilized Materials and Methods with BD Perm/Wash buffer (BD Biosciences). After blocking with anti-mouse Bacterial strains and culture conditions CD16/CD32 (BioLegend, San Diego, CA), total lung cells were incubated with the following fluorescent dye–conjugated Abs (BioLegend): CD3ε-FITC M. tuberculosis Erdman strain (TMC107), originally from the Trudeau (clone145-2C11), CD4-PE/Cy7 (RM4-5), CD8a-allophycocyanin (53-6.7), Mycobacterium Culture Collection, was obtained from the National CD19-PE (6D5), CD11b-FITC (M1/70), CD11c-PE (N418), and Ly-6G/Ly-6C http://www.jimmunol.org/ Institute of Infectious Diseases (Tokyo, Japan) and grown at 37˚C in (Gr-1)–allophycocyanin (RB6-8C5). Lung CD4+ cells were incubated with Middlebrook 7H9 medium (BD Biosciences, Franklin Lakes, NJ) con- IFN-g–FITC (XMG1.2) and CD4-PE/Cy7. Stained cells were fixed with taining 10% oleic acid/albumin/dextrose/ (BD Biosciences), 5% 4% PFA for 1 h and analyzed by flow cytometry (Cytomics FC500; glycerol (Nacalai Tesque, Kyoto, Japan), and 0.05% Tween 80 (Sigma- Beckman Coulter, Brea, CA), according to the manufacturer’s instructions. Aldrich, St. Louis, MO) or in MycoBroth (Kyokuto Pharmaceutical Industrial, Tokyo, Japan). The GFP was amplified using the primers Bone marrow–derived Mf infection and stimulation 59-AAAAGAATTCATGGTGAGCAAGGGCGCCGA-39 (forward, contain- ing an EcoRI site) and 59-AAAAGTCGACCAGATCCTCTTCTGAGATGA- Bone marrow–derived Mfs (BMDMs) were cultured in RPMI 1640 39 (reverse, containing an SalI site) and cloned in pMV261 vector to yield the (Nakarai Tesgue) supplemented with 5% FBS (HyClone; Thermo Fisher plasmid pMV261-GFP. The plasmid was electroporated into WT Erdman Scientific, Waltham, MA) and 10% L929 culture supernatant for 48 h. by guest on September 25, 2021 strain, and GFP-expressing strains were selected on kanamycin-containing For activation, 10 ng/ml IFN-g (PeproTech, Rocky Hill, NJ) was added to cell cultures the day before infection. For inhibiting arginase activity, 7H10 agar plates (BD Biosciences). v N -hydroxy-nor-L-arginine (nor-NOHA; Sigma-Aldrich) was added to cell Mice cultures. were grown to OD530 0.5–0.8, washed twice with RPMI 1640, and resuspended in culture medium. Bacterial cell suspensions 2/2 Eight-week-old C57BL/6 mice were purchased from CREA Japan. Prdx1 were passed through 5-mm filters (Pall) and their concentrations adjusted mice were generated as described on a C57BL/6 background (18) to 6 3 105 CFU/ml. BMDMs were infected with bacteria at a multiplicity (OmniBank no. OST422296; Lexicon Pharmaceuticals). All mice were of infection (MOI) of 5 unless otherwise indicated. After 4 h, the BMDMs maintained in pathogen-free conditions in the animal facility at the Na- were washed twice with RPMI 1640 to remove nonadherent bacilli and tional Center for Global Medicine. fresh culture medium was added. Culture media were changed every 48 h. Culture supernatants were collected at indicated time points and passed Mice infection through Spin-X for measurements of cytokine/chemokine and NO con- centrations. To measure arginase activity, cells were lysed at indicated time Bacteria were grown to OD530 0.5–0.8 in MycoBroth, collected by cen- trifugation, washed twice with PBS (Nacalai Tesque), passed through points with 10 mM Tris-HCl containing 0.4% Triton X-100 and protease 5-mm filters (Pall, Port Washington, NY) to remove aggregates, and adjusted inhibitor mixture tablets (Roche). To evaluate bacterial growth in BMDMs, to 107 CFU/ml. Mice were i.v. injected with 0.2 ml of the bacterial sus- the cells were lysed with PBS containing 0.025% SDS. Serial dilutions 2 pension. Infected mice were euthanized with diethylether. Their spleen, were plated on 7H10 agar plates, which were incubated for 3 4wkat37˚C, liver, and lungs were dissected and homogenized immediately in 9-fold (v/w) and CFU of bacteria were counted. Cells were stimulated with 100 ng/ml vol of PBS. The homogenates were plated on 7H10 agar plates and LPS (Sigma-Aldrich) unless otherwise indicated. For investigating the incubated for 3–4 wk at 37˚C, followed by counting of bacterial CFU. To effect of cytokines, BMDMs were treated with G-CSF, IL-1b, IL-4, IL-6, measure cytokine/chemokine concentrations in organs, the homogenates and IL-17 (10 ng/ml each, all from PeproTech). After 24 h treatment, were centrifuged for 5 min at 13,000 rpm, and the supernatants were BMDMs were stimulated with 100 ng/ml LPS for another 24 h. Culture passed through Spin-X centrifuge tube filters (Corning, Corning, NY) and supernatants were collected for determination of NO production. For in- stored at 280˚C. All animal experiments were reviewed and approved by vestigating the effect of inhibitors, BMDMs were treated with apocynin the National Center for Global Medicine Animal Care and Use Committee (Sigma-Aldrich), N-acetyl-L-cysteine (Sigma-Aldrich), SB202190 (Wako), and the Ethical Committee of Tochigi Prefectural Institute of Public Health SB239063 (Enzo Life Sciences, Farmingdale, NY), and IT-603 (Merck) and Environmental Science. (10 mM each) before 1 h of stimulation. Histology NO production and arginase activity assay Lungs were infused with 10% formaldehyde neutral buffer solution NO concentrations were determined by Griess reagent. Arginase activity (Wako, Osaka, Japan). Sections made from paraffin blocks were stained was determined with QuantiChrom arginase assay kits (BioAssay Systems, with H&E (Wako). To detect bacilli in lungs, sections were washed with Hayward, CA), according to the manufacturer’s instructions. xylene, ethanol, and PBS (all from Nacalai Tesque) and stained with a RNA analysis TB fluorescent stain kit T (BD Biosciences) according to the manufac- turer’s instructions. All bright-field and fluorescent images were acquired Total RNA was isolated with RNeasy mini kits (Qiagen, Venlo, the using an IX-70 microscope (Olympus, Tokyo, Japan) equipped with an Netherlands). First-strand cDNA was synthesized with iScript (Bio-Rad Olympus PEN digital camera. Fluorescent images were overlaid on Laboratories, Hercules, CA), and quantitative RT-PCR (qRT-PCR) assays The Journal of Immunology 3

2 2 were performed using SYBR Green real-time PCR master mix (Toyobo, Low IFN-g levels in M. tuberculosis–infected Prdx1 / mice Osaka, Japan) and LightCycler (Roche). All procedures were performed according to the manufacturers’ instructions. Primers used in this study are To investigate the immunological responses of mice to M. tuberculosis listed in Table I and were purchased from Greiner Japan. infection, we measured the levels of 22 cytokines and chemokines levels were normalized to GAPDH expression levels and reported relative in their spleens, livers, and lungs from 1 to 9 wk p.i. IFN-g is to normalized expression levels in uninfected cells, unstimulated cells, and critical in host defenses against M. tuberculosis by activating the naive mice, as appropriate. Relative gene expression levels were calculated by the DDCt method. antimicrobial activities of phagocytes (28, 29) and reduces tissue inflammation by limiting neutrophil accumulation in lesions (30). Western blot analysis Extremely high levels of IFN-g were detected in the spleens and 2/2 Cells were lysed with RIPA buffer (25 mM Tris-HCl [pH 7.6], 150 mM livers of both Prdx1 and WT mice at 1 wk p.i., followed by NaCl, 1% Nonidet P-40; 1% sodium deoxycholate, 0.1% SDS) containing dramatic reductions in subsequent weeks (Fig. 2A). The level of protease inhibitor mixture tablets (Roche) and phosphatase inhibitor IFN-g was significantly lower in the spleens of Prdx12/2 than WT mixture set II (Merck). were separated by SDS-PAGE and mice at 1 wk p.i. At 1 wk, IFN-g levels were lower in the lungs transferred to polyvinylidene difluoride membranes (Bio-Rad Laborato- ries). The membranes were blocked with 5% nonfat dry milk or 5% BSA than in the spleen and liver of their respective strains, gradually (Nacalai Tesque) in TBS containing 0.05% Tween 20, according to the increasing at 3 and 9 wk p.i., but they were significantly lower in manufacturer’s instructions. The membranes were incubated with primary the lungs of Prdx12/2 than WT mice (Fig. 2A). These results Ab, followed by incubation with HRP-linked secondary Ab (GE Health- indicate that PRDX1 positively regulates IFN-g production after care, Little Chalfont, U.K.). Binding was detected using SuperSignal West Femto maximum sensitivity substrate (Thermo Fisher Scientific). Images M. tuberculosis infection and that reduced levels of IFN-g lead to 2/2 were acquired with ChemiDoc XRS (Bio-Rad Laboratories). The primary higher susceptibility of Prdx1 mice to M. tuberculosis. Abs used in this study were Abs against c-Rel (sc-71; Santa Cruz Biotech- The cytokine and chemokine profiles in the lungs of the two Downloaded from nology, Dallas, TX), ERK (sc-94; Santa Cruz Biotechnology), phospho-Erk1/2 mouse strains differed markedly at 9 wk p.i. (Fig. 2B). The levels (4370; Cell Signaling Technology, Danvers, MA), SAPK/JNK (9258; Cell of G-CSF, IFN-g, IL-1a, IL-6, IL-17, TNF-a, CCL2, CCL3, and Signaling Technology), phospho-SAPK/JNK (9255; Cell Signaling Technol- 2/2 ogy), p38 (9212; Cell Signaling Technology), phospho–p38 MAPK (9211; Cell CXCL1 were significantly higher in the lungs of Prdx1 than Signaling Technology), and GAPDH (6C5; Thermo Fisher Scientific). Bands WT mice, suggesting that PRDX1 negatively regulates the ex- were analyzed densitometrically with ImageJ software (National Institutes of pression of these selected cytokines and chemokines. No signifi- Health, Bethesda, MD). The expression of each was normalized to that cant differences in cytokine and chemokine levels were observed http://www.jimmunol.org/ of GAPDH and represented as fold increase relative to its expression in 2/2 unstimulated cells. Phosphorylated proteins were normalized relative to total in the spleens and livers of Prdx1 and WT mice or in the lungs protein expression. of these two strains at 1 and 3 wk p.i. (data not shown). + Detection of M. tuberculosis colocalized with lysosomes Low CD4 cell infiltration and high neutrophil/monocyte 2/2 2 2 accumulation in lungs of M. tuberculosis–infected Prdx1 mice Untreated and 100 ng/ml IFN-g–activated WT and Prdx1 / BMDMs were infected with GFP-expressing M. tuberculosis at an MOI of 1. After 4 h, Immune cells that had infiltrated into mouse lungs after M. tu- BMDMs were stained with LysoTracker (Molecular Probes, Eugene, OR) berculosis infection were examined (Fig. 3). Representative flow for 30 min, according to the manufacturer’s instructions, and fixed with cytometry plots of the lung cells are shown in Fig. 3A–E. Although

4% PFA for 1 h. Quantification (n = 100 per group) of colocalization of by guest on September 25, 2021 M. tuberculosis Prdx12/2 the total numbers of lung cells gradually increased in both WT and and lysosomes in WT and BMDMs was achieved 2 2 by confocal imaging. Coverslips were mounted onto microscope slides Prdx1 / mice during infection, the numbers of lung cells at 9 wk using Vectashield mounting medium (Vector Laboratories, Burlingame, p.i. were significantly higher in Prdx12/2 than in WT mice (Fig. 3F). CA). Confocal images were acquired with IX1000 (Olympus). During M. tuberculosis infection, the numbers of CD3+ and CD4+ Statistical analysis T cells that had infiltrated into the lungs increased over time in both WT and Prdx12/2 mice, but the numbers of both at 1 and 3 wk p.i. All statistical analyses were performed using GraphPad Prism 5 software 2/2 (GraphPad Software, La Jolla, CA). Survival curves were compared using were significantly lower in Prdx1 than in WT lungs (Fig. 3G, + the generalized Wilcoxon test. Multiple groups were compared by one-way 3H). Similarly, the numbers of IFN-g–producing CD4 T cells at 1, 2 2 ANOVA. Statistical significance was defined as p , 0.05. 3, and 9 wk p.i. were significantly lower in Prdx1 / than in WT lungs (Fig. 3I), with the latter correlated with the low IFN-g levels in 2/2 Results Prdx1 lungs. These results suggest that PRDX1 contributes to 2 2 Th1 responses against M. tuberculosis. Increased susceptibility of Prdx1 / mice to M. tuberculosis The numbers of neutrophils (CD11bhighCD11clowGr-1high)in Infection with M. tuberculosis led to significantly shorter mean both WT and Prdx12/2 lungs were lower at 1 and 3 wk p.i. than at 2/2 postinfection (p.i.) survival in Prdx1 (80.8 6 10.0 d; range, baseline, although the numbers of these cells were significantly 70–100 d) than in WT mice (143.4 6 17.5 d; range, 120–170 d) higher in Prdx12/2 than in WT lungs at 9 wk p.i. (Fig. 3J). The (Fig. 1A). The numbers of bacteria in the spleens and livers of numbers of monocytes (CD11bhighCD11clowGr-1middle)inWT these mice were not altered during infection and did not differ in lungs were not altered during M. tuberculosis infection, whereas 2/2 WT and Prdx1 mice (Fig. 1B). The numbers of bacteria in the the numbers in Prdx12/2 lungs significantly increased at 3 and 9 wk lungs increased from 1 to 9 wk p.i. in both strains, but the numbers p.i. (Fig. 3K). These results indicate that PRDX1 negatively reg- 2/2 at 9 wk were 10-fold higher in Prdx1 than in WT lungs (Fig. 1B). ulates the accumulation of neutrophils and monocytes in lungs Histological examinations showed high accumulation of inflam- during M. tuberculosis infection. 2/2 matory cells in the lungs of both WT and Prdx1 mice at 9 wk The numbers of CD8+ T, CD19+ B, and CD11c+ cells in lungs 2/2 (Fig. 1C). Prdx1 lungs were characterized by alveolar sacs were similar in M. tuberculosis–infected WT and Prdx12/2 mice filled with necrotic cells and separated by thickened septa (Fig. 1D). (Fig. 3L–N), suggesting that PRDX1 contributes little to the Auramine-rhodamine–stained M. tuberculosis was present pre- numbers of these cells in lungs during M. tuberculosis infection. 2/2 dominantly in the alveolar sacs of Prdx1 lungs (Fig. 1E), but 2/2 few lesions were observed in WT lungs (Fig. 1D). Collectively, Low IL-12 production by Prdx1 BMDMs after these results indicate that PRDX1 contributes to host defenses M. tuberculosis infection and LPS stimulation against M. tuberculosis by inhibiting M. tuberculosis growth in the Because we found that PRDX1 positively regulated Th1 responses lungs and the progression of lung pathology. against M. tuberculosis in vivo (Figs. 2, 3), we assessed whether 4 PRDX1 CONTRIBUTES TO HOST DEFENSES AGAINST TB Downloaded from http://www.jimmunol.org/

FIGURE 1. Prdx12/2 mice are more susceptible to M. tuberculosis than are WT mice. (A) WT (C57BL/6) and Prdx12/2 mice (n = 8 each) were infected i.v. with M. tuberculosis (2 3 106 CFU). Survival curves were constructed by the Kaplan–Meier method and compared using a Wilcoxon test (p , 0.001). (B) Numbers of bacteria in spleens (left), livers (middle), and lungs (right) of WT and Prdx12/2 mice 1, 3, and 9 wk p.i. with M. tuberculosis. The mean 6 SD of six mice is shown. *p , 0.005 compared with WT mice using one-way ANOVA. (C–E) Lung histopathology of WT (upper panel) and Prdx12/2 (lower panel) mice at 9 wk p.i. Formalin-fixed, paraffin-embedded lung sections were stained with H&E. Stained lung sections are shown at 34(C, scale m 3 D m M. tuberculosis 3 bar, 500 m) or 40 ( , scale bar, 50 m) original magnification. Images of auramine-rhodamine–stained at original magnification 40 by guest on September 25, 2021 (E) are overlaid on the images (D), representing the same field of view. (B–E) Mice were infected as in (A). Data are representative of three (A) and two (B–E) independent experiments.

PRDX1 regulates IL-12 production by M. tuberculosis–infected Low c-Rel induction and p38 MAPK activation in Mfs in vitro by measuring the amount of IL-12 produced by LPS-stimulated Prdx12/2 BMDMs M. tuberculosis–infected BMDMs. IL-12 is composed of two Although the levels of IL-12p40 produced by both WTand Prdx12/2 subunits, p40 and p35, which are encoded by Il12b and Il12a, 2/2 BMDMs increased beginning 4 h after LPS stimulation, the levels respectively (3). Both WT and Prdx1 BMDMs produced IL- in Prdx12/2 BMDMs were significantly lower than those in WT 12p40 at day 3 p.i., with the amounts dependent on the MOI BMDMs (Fig. 5A). The levels of Il12b mRNA in both WT and 2/2 (Fig. 4A). At an MOI of 1, the levels of p40 in Prdx1 BMDMs Prdx12/2 BMDMs peaked 2 h after LPS stimulation and gradu- 2/2 were 46.3% of those in WT BMDMs. WT and Prdx1 BMDMs ally decreased thereafter, but they were significantly lower in 2/2 produced the same levels of TNF-a (Fig. 4B), whereas Prdx1 Prdx12/2 than in WT BMDMs (Fig. 5B). The level of expression BMDMs produced slightly, but significantly, higher levels of IL-6 of c-Rel, a member of the NF-kB family encoded by Rel and a than did WT BMDMs. Similar results were observed at mRNA critical regulator of Il12b transcription (26), was increased in both levels analyzed by qRT-PCR using primers listed in Table I WT and Prdx12/2 BMDMs 2 h after LPS stimulation, but the (Fig. 4C). The level of Il12b mRNA was significantly higher in levels of c-Rel expression were significantly lower in Prdx12/2 2/2 WT than in Prdx1 BMDMs, Tnf mRNA levels were similar in than in WT BMDMs (Fig. 5C, 5D). Levels of Rel mRNA in both 2/2 WT and Prdx1 BMDMs, and Il6 mRNA levels were signifi- WT and Prdx12/2 BMDMs gradually increased after stimulation, 2/2 cantly higher in Prdx1 than in WT BMDMs. and both levels showed no difference (Fig. 5E). These results 2/2 LPS-stimulated WT and Prdx1 BMDMs produced IL-12p40 suggest that PRDX1 positively regulate c-Rel translation. As c-Rel in a dose-dependent manner (Fig. 4D). Although both BMDMs also regulates the expression of Il12a and Il23a mRNAs (31, 32), produced the same levels of IL-12p40 following stimulation the levels of both in WT and Prdx12/2 BMDMs after LPS 2 2 with 0.1 ng/ml LPS, Prdx1 / BMDMs produced significantly stimulation were assayed. The levels of both Il12a and Il23a mRNAs lower levels of IL-12p40 than did WT BMDMs when stimulated were significantly lower in Prdx12/2 than in WT BMDMs (Fig. 5F), with 1–1000 ng/ml. Both BMDMs produced the same levels of indicating that PRDX1 positively regulates c-Rel induction. TNF-a andIL-6whenreactedwith100ng/mlLPS(Fig.4E). LPS has been shown to induce the phosphorylation of three Similar results were observed at mRNA levels (Fig. 4F). These MAPKs, ERK, JNK, and p38, within minutes to hours (33), with results suggest that PRDX1 positively regulates IL-12 produc- the p38 MAPK pathway being critical for IL-12 production (27). tion by BMDMs after both M. tuberculosis infection and LPS ERK and JNK in both WT and Prdx12/2 BMDMs were phos- stimulation. phorylated 15 min after LPS stimulation (Fig. 5G) with similar The Journal of Immunology 5 Downloaded from

FIGURE 2. IFN-g levels are low and selected cytokines and chemokines are high in the lungs of Prdx12/2 mice after M. tuberculosis infection. (A) IFN-g levels in spleen (left), liver (middle), and lung (right) homogenates of WT and Prdx12/2 mice at 1, 3, and 9 wk p.i. (B) Cytokine and chemokine levels in lung homogenates of WT and Prdx12/2 mice at 9 wk p.i. (A and B) All mice were infected as described in the legend to Fig. 1A. Organs were isolated and homogenates were prepared at indicated time points. Cytokine and chemokine concentrations in each homogenate were analyzed using multiplex ELISA kits. Data are representative of two independent experiments. The mean 6 SD of three mice is shown. *p , 0.05 compared with WT using one-way ANOVA. http://www.jimmunol.org/

kinetics and to similar levels in both cell types (data not shown). after the addition of apocynin and N-acetyl-L-cysteine (H2O2 2/2 LPS stimulation of both WT and Prdx1 BMDMs also resulted scavengers), indicating that H2O2 scavenger function of PRDX1 in p38 phosphorylation within 15 min, with phosphorylation levels plays little role in Il12b induction. declining rapidly thereafter. Unlike findings with ERK and JNK, The anti-inflammatory cytokines IL-10 (34) and TGF-b (35) and the levels of phosphorylated p38 from 15 to 60 min were signif- suppressor of cytokine signaling (SOCS)1 (36) have been shown icantly lower in Prdx12/2 than in WT BMDMs (Fig. 5G). These to inhibit the expression of IL-12, as well as TNF-a, IL-6, and results indicate that PRDX1 positively regulates the p38 MAPK NOS2, in Mfs. LPS stimulation of WT and Prdx12/2 BMDMs by guest on September 25, 2021 pathway. induced the expression of similar levels of Il10, Tgfb1, and Socs1 When the induction levels of Il12b expression were compared mRNAs (Supplemental Fig. 1A), as well similar levels of Tnf, Il6, after LPS stimulation between those in WT and Prdx12/2 and Nos2 mRNAs (Fig. 4F, Supplemental Fig. 1A) and similar levels BMDMs, the levels in WT BMDMs were significantly higher than of TNF-a and IL-6 proteins and NO (Fig. 4E, Supplemental Fig. 1B). those in Prdx12/2 BMDMs (Fig. 5H). The effects of five inhibi- These results indicate that IL-10, TGF-b, and SOCS1 are unrelated tors on the induction levels of Il12b expression were examined to the decreased production of IL-12 by Prdx12/2 BMDMs.

(Fig. 5H). IT-603, a selective inhibitor of c-Rel, suppressed higher 2/2 levels of Il12b expression in WT BMDMs, resulting in levels IFN-g–activated Prdx1 BMDMs do not kill M. tuberculosis identical to those in Prdx12/2 BMDMs (Fig. 5H). This effect was effectively absent after addition of SB202190 and SB239063 (p38 MAPK Assays of the antimicrobial activities of Prdx12/2 BMDMs against inhibitors), indicating that p38 MAPK signaling contributes less in M. tuberculosis showed that, in the absence of IFN-g treatment, Il12b induction. The effect introduced by IT-603 was also absent the numbers of M. tuberculosis at day 4 p.i. increased to the same

Table I. List of primers used for qRT-PCR

Gene Forward (59→39) Reverse (59→39) Arg1 CTCCAAGCCAAAGTCCTTAGAG AGGAGCTGTCATTAGGGACATC Arg2 TCCTCCACGGGCAAATTCC GCTGGACCATATTCCACTCCTA Clec7a GACTTCAGCACTCAAGACATCC TTGTGTCGCCAAAATGCTAGG Gapdh AGGTCGGTGTGAACGGATTTG TGTAGACCATGTAGTTGAGGTCA Il6 TAGTCCTTCCTACCCCAATTTCC TTGGTCCTTAGCCACTCCTTC Il10 GCTCTTACTGACTGGCATGAG CGCAGCTCTAGGAGCATGTG Il12a CTGTGCCTTGGTAGCATCTATG GCAGAGTCTCGCCATTATGATTC Il12b CAGAAGCTAACCATCTCCTGGTTTG TCCGGAGTAATTTGGTGCTTCACAC Il23a ATGCTGGATTGCAGAGCAGTA ACGGGGCACATTATTTTTAGTCT Nos2 GTTCTCAGCCCAACAATACAAGA GTGGACGGGTCGATGTCAC Rel AGAGGGGAATGCGGTTTAGAT TTCTGGTCCAAATTCTGCTTCAT Retnla CCAATCCAGCTAACTATCCCTCC ACCCAGTAGCAGTCATCCCA Socs1 CTGCGGCTTCTATTGGGGAC AAAAGGCAGTCGAAGGTCTCG Tgfb1 CTCCCGTGGCTTCTAGTGC GCCTTAGTTTGGACAGGATCTG Tnf TCTTCTCATTCCTGCTTGTGG GGTCTGGGCCATAGAACTGA 6 PRDX1 CONTRIBUTES TO HOST DEFENSES AGAINST TB Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021

FIGURE 3. Infiltrations of total and IFN-g+CD4+ T cells are low and accumulations of neutrophils/monocytes are high in lungs of Prdx12/2 mice after M. tuberculosis infection. (A–E) Representative flow cytometry plots of lung cells in WT and Prdx12/2 mice after M. tuberculosis infection and percentages of gated cells. (A) CD3+ T cells and CD19+ B cells at 3 wk p.i., (B) CD4+ T cells and CD8+ T cells at 3 wk p.i., (C) CD11b+ cells and CD11c+ cells at 9 wk p.i., (D) CD11c+ cells and Gr-1+ cells at 9 wk p.i., and (E) IFN-g–producing CD4+ T cells at 1 wk p.i. (F–N) Numbers or fold increases of cells in lungs of WT and Prdx12/2 mice at 0, 1, 3, and 9 wk p.i. were analyzed by flow cytometry. (F) Total cells in lungs, (G) CD3+ T cells, (H) CD4+ T cells, (I) IFN-g– producing CD4+ T cells, (J) neutrophils (CD11bhighCD11clowGr-1high), (K) monocytes (CD11bhighCD11clowGr-1middle), (L) CD8+ T cells, (M) CD19+ B cells, and (N) CD11c+ cells. The cells in the lungs of naive mice are represented as those of mice at 0 wk p.i. (A–N) All mice (Figure legend continues) The Journal of Immunology 7 Downloaded from FIGURE 4. IL-12 production levels are low in Prdx12/2 BMDMs after M. tuberculosis infection and LPS stimulation. (A) IL-12p40 production by WT and Prdx12/2 BMDMs infected with M. tuberculosis at MOIs of 1, 5, and 10. IL-12 levels in culture supernatants collected 3 d p.i. were determined by ELISA. (B) Cytokine production of WT and Prdx12/2 BMDMs infected with M. tuberculosis at an MOI of 1. Culture supernatants were collected after 3 d, and concentrations of IL-12p40, TNF-a, and IL-6 were determined by ELISA. (C) Cytokine mRNA expression by WT and Prdx12/2 BMDMs infected with M. tuberculosis at an MOI of 1. Total RNAs were isolated 24 h p.i., and levels of Il12b, Tnf (insert), and Il6 mRNAs were analyzed by qRT-PCR. The level of each mRNA was normalized to that of GAPDH mRNA and is expressed relative to expression in uninfected cells. (D) IL-12p40 production by WT and 2/2 Prdx1 BMDMs stimulated with 0, 0.1, 1, 10, 100, and 1000 ng/ml LPS. After 24 h, culture supernatants were collected and IL-12p40 concentrations http://www.jimmunol.org/ analyzed by ELISA. (E) Cytokine production by WT and Prdx12/2 BMDMs stimulated with 100 ng/ml LPS. After 24 h, culture supernatants were collected and IL-12p40, TNF-a, and IL-6 levels were analyzed by ELISA. (F) Cytokine mRNA expression by WT and Prdx12/2 BMDMs stimulated with 100 ng/ml LPS. Total RNAs were isolated 12 h after stimulation, and levels of Ill2b, Tnf (insert), and IL6 mRNAs were analyzed by qRT-PCR. The level of each mRNA was normalized to that of GAPDH mRNA and is expressed relative to expression in unstimulated cells. (A–F) Mean 6 SD of three replicates is shown. Data are representative of three independent experiments. *p , 0.005 compared with WT using one-way ANOVA. levels in WT and Prdx12/2 BMDMs (Fig. 6A). Following treat- was significantly higher in IFN-g–activated Prdx12/2 than WT ment with IFN-g, the numbers of M. tuberculosis at day 4 p.i. BMDMs (Fig. 6C). were reduced in both BMDMs, but the numbers were significantly In the absence of IFN-g, both WT and Prdx12/2 BMDMs by guest on September 25, 2021 lower in WT than in Prdx12/2 BMDMs (Fig. 6A). IFN-g enhances expressed little Nos2 and Arg1 mRNA before or after M. tuberculosis autophagy of Mfs to promote fusion of mycobacterial phagosomes infection (Fig. 6D). IFN-g–activated WT and Prdx12/2 BMDMs with lysosomes, leading to inhibition of M. tuberculosis survival (37). expressedrelativelyhighbutsimilarlevelsofNos2 mRNA before In the absence and presence of IFN-g, percentages of M. tuberculosis and after M. tuberculosis infection. Both types of IFN-g–activated colocalizing with lysosomes in WT and Prdx12/2 BMDMs showed BMDMs expressed little Arg1 mRNA before M. tuberculosis infec- no significant difference (Supplemental Fig. 2). These results suggest tion. After M. tuberculosis infection, IFN-g–activated WT BMDMs that PRDX1 positively regulates antimicrobial activity of IFN-g– expressed low levels of Arg1,whereasIFN-g–activated Prdx12/2 activated BMDMs, but plays little role in autophagy. BMDMs expressed significantly higher levels of Arg1 (Fig. 6D). In both the presence and absence of IFN-g, both types of BMDMs Low NO production and high arginase activity and Arg1 expressed little Arg2 (39) mRNA after M. tuberculosis infection 2/2 expression in IFN-g–activated Prdx1 BMDMs infected with (data not shown). These results indicate that PRDX1 suppresses M. tuberculosis Arg1 induction in IFN-g–activated BMDMs infected with IFN-g activation of BMDMs has been found to enhance NO M. tuberculosis and the induced ARG1 reduces NO production 2 2 production (38). In the absence of IFN-g treatment, both WT and of IFN-g–activated Prdx1 / BMDMs, leading to more survival Prdx12/2 BMDMs produced little NO after M. tuberculosis in- of M. tuberculosis. 2 2 fection (Fig. 6B). IFN-g activation of both BMDMs led to the Cytokines whose levels were high in lungs of Prdx1 / mice production of relatively high amounts of NO after M. tuberculosis infected with M. tuberculosis (Fig. 2B), anti-inflammatory cyto- infection, although IFN-g–activated Prdx12/2 BMDMs produced kines (IL-10 and TGF-b), and culture supernatants of IFN-g– 2 2 significantly lower amounts of NO than did IFN-g–activated activated Prdx1 / BMDMs infected with M. tuberculosis did not WT BMDMs (Fig. 6B). In the absence of IFN-g treatment, both affect NO production of LPS-stimulated BMDMs (Supplemental types of BMDMs showed low arginase activity (Fig. 6C). In Fig. 3 and data not shown). IL-4/IL-13 (40) and 9-cis retinoic acid contrast, IFN-g–activated BMDMs showed relatively high argi- (9-cRA) (41) are known to induce Arg1 expression in Mfs. nase activity after M. tuberculosis infection, but the activity Indeed, IL-4–treated WT and Prdx12/2 BMDMs reduced NO

were infected as described in the legend to Fig. 1A. Lungs were isolated and single-cell suspensions were prepared at indicated time points. Total cell numbers were counted with a hemocytometer. Lung single-cell suspensions were fixed with 4% PFA and stained with Abs for flow cytometry. (I) CD4+ T cells enriched from single-cell suspension of lung cells were cultured with GolgiStop for 5 h, fixed, permeabilized, and stained with anti–IFN-g and anti- CD4 Abs. Data are representative of two independent experiments. The mean 6 SD of three mice is shown. *p , 0.05 compared with WT using one-way ANOVA. 8 PRDX1 CONTRIBUTES TO HOST DEFENSES AGAINST TB Downloaded from http://www.jimmunol.org/

FIGURE 5. c-Rel induction and p38 MAPK activation levels are low in LPS-stimulated Prdx12/2 BMDMs. (A and B) Time course of IL-12p40 production by WT and Prdx12/2 BMDMs stimulated with 100 ng/ml LPS. Culture supernatants were collected and total RNAs were isolated at indicated time points. (A) IL-12p40 production levels were analyzed by ELISA. (B) Il12b expression levels were analyzed by qRT-PCR. The level of each mRNA was normalized to that of GAPDH mRNA and is expressed relative to expression in unstimulated cells. Expression levels in unstimulated BMDMs are rep- resented as 1. (C–E) c-Rel mRNA and protein expression levels by WT and Prdx12/2 BMDMs stimulated with 100 ng/ml LPS. Whole-cell lysates were prepared and total RNAs isolated at indicated time points. (C) c-Rel expressions were analyzed by Western blotting and normalized relative to GAPDH expression. (D) Quantification of c-Rel expression, with each normalized to that of GAPDH. Expression in unstimulated BMDMs was represented as 1. (E) by guest on September 25, 2021 Rel expression levels analyzed by qRT-PCR. The level of each mRNA was normalized to relative to that of GAPDH mRNA and is expressed relative to mRNA levels in unstimulated cells. (F) Levels of Il12a and Il23a expression in LPS-stimulated WT and Prdx12/2 BMDMs. (G) Western blot analysis of phospho-ERK, JNK, and p38 in WT and Prdx12/2 BMDMs stimulated with 100 ng/ml LPS. Cell lysates were prepared at indicated time points. Total amounts of ERK, JNK, p38 and GAPDH were also presented as loading controls. (H) Effect of inhibitors on Il12b expression in LPS-stimulated WT and 2/2 Prdx1 BMDMs. H2O2 scavengers (apocynin and N-acetyl-L-cysteine [NAC]), p38 MAPK inhibitors (SB202190 and SB239063), and a c-Rel inhibitor (IT-603) were added to culture medium (10 mM each) before 1 h of LPS stimulation. Total RNAs were isolated at 2 h after LPS stimulation. (A–H) Data are representative of three independent experiments. (A, B, D–F,andH) The mean 6 SD of three replicates is shown. *p , 0.005 compared with WT using one-way ANOVA. production after LPS stimulation (Supplemental Fig. 3). WT and being significantly lower than the numbers in nor-NOHA–untreated Prdx12/2 BMDMs expressed the same levels of IL-4/IL-13– and IFN-g–activated Prdx12/2 BMDMs (Fig. 6E). inducible (40) and 9-cRA–inducible genes (42) after treatment Although treatment with nor-NOHA did not affect NO production of these ARG1 inducers (Supplemental Fig. 4A–C). IFN-g–activated by IFN-g–activated WT BMDMs infected with M. tuberculosis WT and Prdx12/2 BMDMs infected with M. tuberculosis did not (Fig. 6F), nor-NOHA treatment enhanced NO production by IFN- 2 2 express IL-4/IL-13–inducible genes (40) and expressed the same g–activated Prdx1 / BMDMs infected with M. tuberculosis. The levels of 9-cRA–inducible genes (42) (Supplemental Fig. 4D). level of NO produced on day 4 p.i. by nor-NOHA–treated, IFN-g– 2 2 These results suggest that cytokines, extracellular factors, IL-4/IL- activated Prdx1 / BMDMs was significantly higher than those 2 2 13, and retinoic acid are unrelated to Arg1 induction in IFN-g– produced by nor-NOHA–untreated, IFN-g–activated Prdx1 / activated Prdx12/2 BMDMs infected with M. tuberculosis. BMDMs and was equal to the level produced by IFN-g–activated WT BMDMs (Fig. 6F). These results suggest that ARG1 reduces An arginase inhibitor restores antimicrobial activities and NO 2/2 2 2 NO production of IFN-g–activated Prdx1 BMDMs infected production of IFN-g–activated Prdx1 / BMDMs infected with M. tuberculosis. with M. tuberculosis 2/2 We tested the effects of the arginase inhibitor nor-NOHA (43) on Arg1 induction in lungs of Prdx1 mice infected with antimicrobial activities and NO production of IFN-g–activated M. tuberculosis 2 2 Prdx1 / BMDMs infected with M. tuberculosis. Although We tested the levels of Nos2 and Arg1 mRNA expression in the treatment of IFN-g–activated WT BMDMs with nor-NOHA did lungs of WT and Prdx12/2 mice infected with M. tuberculosis. not affect their antimicrobial activity, treatment markedly enhanced Nos2 mRNA levels in the lungs of both WT and Prdx12/2 mice the antimicrobial activity of IFN-g–activated Prdx12/2 BMDMs, increased from 1 to 9 wk p.i. (Fig. 6G) and were significantly with the numbers of M. tuberculosis decreasing markedly at day 4 higher in the lungs of Prdx12/2 than those of WT mice at 9 wk p.i. p.i.,tothelevelobservedinIFN-g–activated WT BMDMs, and Arg1 mRNA levels were low in the lungs of both types of mice The Journal of Immunology 9 Downloaded from http://www.jimmunol.org/

FIGURE 6. Arg1 expression is induced in IFN-g–activated Prdx12/2 BMDMs and mice after M. tuberculosis infection. (A) M. tuberculosis survival in WT and Prdx12/2 BMDMs. Untreated and 10 ng/ml IFN-g–activated BMDMs were infected with M. tuberculosis at an MOI of 5, and the numbers of bacteria were counted on days 0 and 4 p.i. (B) NO production by WT and Prdx12/2 BMDMs infected with M. tuberculosis. Culture supernatants were collected at day 4 p.i. and NO production levels were determined by Griess assay. (C) Arginase activity of WT and Prdx12/2 BMDMs after M. tuberculosis infection. Cell lysates were prepared at day 4 p.i. and arginase activity levels were determined using a colorimetric assay. (D) Levels of Nos2 and Arg1 mRNAs in uninfected (UI) and M. tuberculosis–infected WT and Prdx12/2 BMDMs in the presence or absence of IFN-g. Total RNAs were isolated at day

4 p.i. Levels of each mRNA were analyzed by qRT-PCR and normalized relative to that of GAPDH mRNA in the same sample and are expressed relative to by guest on September 25, 2021 the level of mRNA in uninfected cells. (E) M. tuberculosis survival in arginase inhibitor–pretreated WT and Prdx12/2 BMDMs. BMDMs pretreated with 100 nM nor-NOHA and activated with 10 ng/ml IFN-g were infected with M. tuberculosis, and the numbers of bacteria were counted on days 0 and 4 p.i. (F) NO production by arginase inhibitor–pretreated WT and Prdx12/2 BMDMs after M. tuberculosis infection. Culture supernatants were collected at day 4 p.i. and NO production levels were determined by Griess assay. Nos2 (G) and Arg1 (H) mRNA expression levels in lungs of WT and Prdx12/2 mice after M. tuberculosis infection. Total RNAs were isolated from lungs of mice at 0, 1, 3, and 9 wk p.i. The mRNA levels in the lungs of naive mice are represented as those of mice at 0 wk p.i. The levels of each mRNA were normalized to that of GAPDH mRNA in the same sample and are expressed relative to the level of mRNA in the lungs of uninfected mice. (B–F) BMDMs were infected as in (A). (G and H) Mice were infected as described in the legend to Fig. 1A. Data are representative of three (A–F) and two (G and H) independent experiments. The mean 6 SD of three replicates (A–F) or three mice (G and H) is shown. *p , 0.005 compared with WT using one-way ANOVA. before infection and at 1 and 3 wk p.i. (Fig. 6H). At 9 wk p.i., NO plays a central role in host defenses against M. tuberculosis in however, Arg1 mRNA levels were significantly higher in the lungs mice (5–7). Nos22/2 mice showed increased susceptibility to 2 2 of Prdx1 / than WT mice. These results indicate that PRDX1 M. tuberculosis (5, 6), and Nos22/2 Mfs were unable to inhibit negatively regulates Arg1 induction in IFN-g–activated Mfsin intracellular growth of M. tuberculosis as effectively as WT Mfs(7). the lungs of mice infected with M. tuberculosis. In inducing IL-12 production, PRDX1 positively regulates translation of c-Rel. However, the mechanisms by which PRDX1 Discussion regulates c-Rel translation remain unclear. PRDX1 can bind RNA This study showed that in regulating host defenses against and acts as a RNA chaperone (47, 48). UV crosslinking studies M. tuberculosis in mice, PRDX1 functions primarily in Mfs. suggest that PRDX1 binds RNA in cells (49–51). These reports PRDX1 was found to positively regulate IL-12 and NO production indicate that PRDX1 binds the 59 untranslated region of c-Rel in M. tuberculosis–infected Mfs. IL-12 activates CD4+ T cells to mRNA and helps translation. c-Rel binds to the promoters of produce IFN-g (3), and IFN-g–activated Mfs produce NO to kill Il12a (31), Il12b (26), and Il23a (32), in agreement with our M. tuberculosis (38). IL-12 plays a critical role in host defenses finding that the expression of Il12a, Il12b, and Il23a mRNAs was 2 2 against M. tuberculosis in mice and humans (44), with Il12b / lower in Prdx12/2 than in WT BMDMs. c-Rel is uninvolved in mice showing increased susceptibility to M. tuberculosis (45). regulating the transcription of TNF-a, IL-6, and NOS2 (26), in Severe mycobacterial infections were reported in IL-12R– agreement with our finding that the levels of Tnf, Il6, and Nos2 deficient patients and their cells were deficient in IL-12R were similar in Prdx12/2 and WT BMDMs stimulated with LPS. signaling and IFN-g production (46). PRDX1 positively reg- The pathway by which PRDX1 induces IL-12 production involves ulates NO production by suppressing the induction of ARG1, a the positive regulation of p38 MAPK activation, a step crucial for competitor of NOS2, for their common substrate of L-arginine. IL-12 expression (27). SB203580, a specific inhibitor of p38, was 10 PRDX1 CONTRIBUTES TO HOST DEFENSES AGAINST TB found to block IL-12 production in DCs, whereas wortmannin, an M. Kato-Miyazawa, F. Kirikae, J. Zhao, T. Yanagawa, T. Ishii, activator of p38, was found to enhance IL-12 production (52). T. Miyoshi-Akiyama, K. Funatogawa, and T. Kirikae, unpublished data). PRDX1-mediated p38 MAPK activation may be regulated by ac- Decreased numbers of CD4+ T cells (Fig. 3H) and increased 2 2 tivation of apoptosis signal–regulating kinase 1 (ASK1) and mole- numbers of neutrophils (Fig. 3J) in lungs of Prdx1 / mice induce cules downstream of MAPK kinase (MKK)3/6 (53). PRDX1 pathological lesions characterized by alveolar sacs filled with interacts with ASK1 to regulate multimerization and subsequent necrotic cells separated by thickened septa (Fig. 1D). Similar phosphorylation of ASK1 (54). Activated ASK1 phosphorylates characteristic lesions were observed in lungs of CD4+ T cell– MKK3/6, which, in turn, phosphorylates p38 MAPK (53). However, depleted mice infected with M. tuberculosis, but not in WT mice PRDX1 overexpression has also been reported to inhibit the phos- (67). Neutrophils accumulate in situations of high pathogen load phorylation of ASK1, MKK3/6, and p38 MAPK, although PRDX1 and immunological dysfunction, and they are likely to contribute interacts with ASK1 via the thioredoxin-binding domain of the to the pathology of TB (68). G-CSF, GM-CSF, IL-6, IL-8, IL-17, latter (55). PRDX1-mediated IL-12 production may be regulated by and other CXC chemokines recruit neutrophils to sites of infection a pathway involving phosphatase and tensin homolog (PTEN) and (69). Indeed, we found that the concentrations of G-CSF, IL-6, 2 2 PI3K, independent of c-Rel induction and p38 MAPK activation. IL-17, and CXCL1 were significantly higher in Prdx1 / than in PRDX1 protects PTEN from oxidation-induced inactivation (56). WT lungs at 9 wk after M. tuberculosis infection (Fig. 2B). 2/2 PTEN negatively regulates PI3K signaling (57). PI3K is an en- It would be better to investigate TB susceptibility of Prdx1 dogenous suppressor of IL-12 production triggered by TLR sig- mice by using an aerosol M. tuberculosis infection model. There naling (52). In response to various TLR ligands, PTEN-deficient are multiple routes of M. tuberculosis inoculation in mice, and the Mfs produce less IL-12 than do WT cells (58). two of the most common approaches are the i.v. and aerosol routes Downloaded from Our present results indicate that PRDX1 does not directly mod- (70). The aerosol infection approach was not available in the ulate TLR4 and MyD88 molecules to the positive regulation of IL-12 study. The i.v. infection model showed that IL-6–deficient mice production, although these molecules are required for IL-12 pro- were more susceptible than WT mice (71), whereas the aerosol duction (59). Low levels of IL-12 production in Prdx12/2 compared infection model showed that IL-6–deficient mice were slightly with WT BMDMs were observed, but not TNF-a or IL-6 produc- more susceptible than WT mice (72). Both i.v. and aerosol infection tion (Fig. 4E). A number of studies, however, demonstrated that models showed that IFN-g–deficient mice were more significantly http://www.jimmunol.org/ recombinant human PRDX1 bound to TLR4 and stimulated Mfsor susceptible than WT mice (28). DCs to produce inflammatory cytokines (22, 23). The production There is a discrepancy in results of IL-12 production between was dependent on the TLR4-MyD88 signaling pathway, which re- in vivo and in vitro infection models, that is, the levels of IL-12 were 2/2 sults in activation of NF-kB (23). PRDX1 will induce IL-12 pro- higher in the lungs of Prdx1 mice than in WT mice at 9 wk p.i. duction in a TLR4 molecule–independent manner. (Fig. 2B), whereas IL-12 production by M. tuberculosis–infected 2/2 PRDX1 negatively regulates ARG1 induction during M. tuber- Prdx1 BMDMs were lower than that by WT BMDMs (Fig. 4B, culosis infection. Prdx12/2 mice, therefore, may show enhanced 4C). The discrepancy could be explained by the numbers of IL-12– producing cells and M. tuberculosis in each infection model. In

Th2 responses against helminth infection because induced ARG1 by guest on September 25, 2021 plays protective roles against helminth infection (60). However, the the in vivo infection model, the numbers of monocytes and Prdx12/2 f viable M. tuberculosis were 3- and 10-fold higher in the lungs of mechanisms of ARG1 induction in M s have not yet 2/2 been determined. At least nine transcription factors are known to Prdx1 mice than those of WT mice at 9 wk p.i., respectively (Figs. 1B, 3K). In the in vitro infection model, the numbers of regulate ARG1 expression in Mfs (61), indicating that PRDX1 2/2 must activate or regulate some of these transcription factors. WT and Prdx1 BMDMs were the same, and the numbers of PRDX1 may interact with intracellular proteins that suppress ARG1 M. tuberculosis infected into the both BMDMs were also the induction, such as SHIP (62) and histone deacetylase 3 (63). ARG1 same (Fig. 4). induction by IL-10 and TGF-b was suppressed in WT BMDMs but not in Ship2/2 BMDMs (62), whereas ARG1 induction by IL-4 was Acknowledgments markedly enhanced in Hdac32/2 BMDMs (63). We thank Noriko Toyama-Sorimachi (National Center for Global Medicine) Although PRDX1, PRDX2, and PRDX3 belong to the same and Satoshi Takaki (National Center for Global Medicine) for technical ad- vice and support. We thank Miwa Tamura-Nakano (National Center for protein family, PRDX1 plays different roles in immune responses Global Medicine) and Kayo Okumura (Obihiro University of Agriculture than do PRDX2 and PRDX3. PRDX2 negatively regulates NF-kB and Veterinary Medicine) for technical support. activation and phosphorylation of all three MAPKs in response to 2/2 LPS (64). Moreover, following LPS stimulation, Prdx2 Disclosures BMDMs expressed higher levels of TNF-a and IL-6 than do WT The authors have no financial conflicts of interest. BMDMs (64). Similar to PRDX2, PRDX3 may negatively reg- ulate NF-kB and MAPKs, because Prdx32/2 BMDMs expressed higher levels of TNF-a in response to LPS (65). In contrast, References PRDX1 positively regulates c-Rel induction and p38 MAPK 1. World Health Organization. 2015. Global Tuberculosis Report 2015. Available at: http://www.who.int/tb/publications/global_report/en/. phosphorylation in response to LPS. 2. Flynn, J. L., and J. Chan. 2001. Immunology of tuberculosis. Annu. Rev. + PRDX1 will not contribute to the balance of CD4 T cell dif- Immunol. 19: 93–129. ferentiation into Th1 or Th2 cells. PRDX1 acts as a negative 3. Trinchieri, G. 2003. Interleukin-12 and the regulation of innate resistance and adaptive immunity. Nat. Rev. Immunol. 3: 133–146. regulator of Th2-dominant allergic asthma (66). Th2 differen- 4. Ehrt, S., and D. Schnappinger. 2009. Mycobacterial survival strategies in the 2 2 tiation, however, was not observed in Prdx1 / mice during phagosome: defence against host stresses. Cell. Microbiol. 11: 1170–1178. M. tuberculosis infection because cytokine/chemokine mea- 5. Adams, L. B., M. C. Dinauer, D. E. Morgenstern, and J. L. Krahenbuhl. 1997. Comparison of the roles of reactive oxygen and nitrogen intermediates in the surements showed undetectable levels of IL-4 in lungs of host response to Mycobacterium tuberculosis using transgenic mice. Tuber. Lung M. tuberculosis–infected Prdx12/2 mice and the same levels of Dis. 78: 237–246. 2/2 6. MacMicking, J. D., R. J. North, R. LaCourse, J. S. Mudgett, S. K. Shah, and IL-5, IL-9, and IL-13 in lungs of Prdx1 and WT mice during C. F. Nathan. 1997. Identification of nitric oxide synthase as a protective locus M. tuberculosis infection (Fig. 2B and K. Matsumura, H. Iwai, against tuberculosis. Proc. Natl. Acad. Sci. USA 94: 5243–5248. The Journal of Immunology 11

7. Ehrt, S., D. Schnappinger, S. Bekiranov, J. Drenkow, S. Shi, T. R. Gingeras, 35. Li, M. O., Y. Y. Wan, S. Sanjabi, A. K. Robertson, and R. A. Flavell. 2006. T. Gaasterland, G. Schoolnik and C. Nathan. 2001. Reprogramming of the Transforming growth factor-b regulation of immune responses. Annu. Rev. macrophage transcriptome in response to interferon-g and Mycobacterium Immunol. 24: 99–146. tuberculosis: signaling roles of nitric oxide synthase-2 and phagocyte oxi- 36. Kubo, M., T. Hanada, and A. Yoshimura. 2003. Suppressors of cytokine sig- dase. J. Exp. Med. 194: 1123–1140. naling and immunity. Nat. Immunol. 4: 1169–1176. 8. Babior, B. M. 2004. NADPH oxidase. Curr. Opin. Immunol. 16: 42–47. 37. Gutierrez, M. G., S. S. Master, S. B. Singh, G. A. Taylor, M. I. Colombo, and 9. Jung, Y. J., R. LaCourse, L. Ryan, and R. J. North. 2002. Virulent but not V. Deretic. 2004. Autophagy is a defense mechanism inhibiting BCG and My- avirulent Mycobacterium tuberculosis can evade the growth inhibitory action of cobacterium tuberculosis survival in infected macrophages. Cell 119: 753–766. a T helper 1-dependent, nitric oxide synthase 2-independent defense in mice. 38. Chan, J., Y. Xing, R. S. Magliozzo, and B. R. Bloom. 1992. Killing of virulent J. Exp. Med. 196: 991–998. Mycobacterium tuberculosis by reactive nitrogen intermediates produced by 10. Cooper, A. M., B. H. Segal, A. A. Frank, S. M. Holland, and I. M. Orme. 2000. activated murine macrophages. J. Exp. Med. 175: 1111–1122. Transient loss of resistance to pulmonary tuberculosis in p47phox2/2 mice. Infect. 39. Wu, G., and S. M. Morris, Jr. 1998. Arginine metabolism: nitric oxide and be- Immun. 68: 1231–1234. yond. Biochem. J. 336: 1–17. 11. Voskuil, M. I., I. L. Bartek, K. Visconti, and G. K. Schoolnik. 2011. The re- 40. Van Dyken, S. J., and R. M. Locksley. 2013. Interleukin-4- and interleukin-13- sponse of Mycobacterium tuberculosis to reactive oxygen and nitrogen species. mediated alternatively activated macrophages: roles in homeostasis and disease. Front. Microbiol. 2: 105. Annu. Rev. Immunol. 31: 317–343. 12. El Kasmi, K. C., J. E. Qualls, J. T. Pesce, A. M. Smith, R. W. Thompson, 41. Gallardo-Soler, A., C. Go´mez-Nieto, M. L. Campo, C. Marathe, P. Tontonoz, M. Henao-Tamayo, R. J. Basaraba, T. Ko¨nig, U. Schleicher, M. S. Koo, et al. A. Castrillo, and I. Corraliza. 2008. Arginase I induction by modified lipopro- 2008. Toll-like receptor-induced arginase 1 in macrophages thwarts effective teins in macrophages: a peroxisome proliferator-activated receptor-g/d-mediated immunity against intracellular pathogens. Nat. Immunol. 9: 1399–1406. effect that links lipid metabolism and immunity. Mol. Endocrinol. 22: 1394–1402. 13. Boucher, J. L., C. Moali, and J. P. Tenu. 1999. Nitric oxide biosynthesis, nitric 42. Sze´les, L., S. Po´liska, G. Nagy, I. Szatmari, A. Szanto, A. Pap, M. Lindstedt, oxide synthase inhibitors and arginase competition for L-arginine utilization. S. J. Santegoets, R. Ruhl,€ B. Dezso¨, and L. Nagy. 2010. Research resource: Cell. Mol. Life Sci. 55: 1015–1028. transcriptome profiling of genes regulated by RXR and its permissive and 14. Rutschman, R., R. Lang, M. Hesse, J. N. Ihle, T. A. Wynn, and P. J. Murray. nonpermissive partners in differentiating monocyte-derived dendritic cells. Mol. 2001. Cutting edge: Stat6-dependent substrate depletion regulates nitric oxide Endocrinol. 24: 2218–2231. production. J. Immunol. 166: 2173–2177. 43. Tenu, J. P., M. Lepoivre, C. Moali, M. Brollo, D. Mansuy, and J. L. Boucher. Downloaded from v 15. Wood, Z. A., E. Schro¨der, J. Robin Harris, and L. B. Poole. 2003. Structure, 1999. Effects of the new arginase inhibitor N -hydroxy-nor-L-arginine on NO mechanism and regulation of . Trends Biochem. Sci. 28: 32–40. synthase activity in murine macrophages. Nitric Oxide 3: 427–438. 16. Rhee, S. G., and H. A. Woo. 2011. Multiple functions of peroxiredoxins: per- 44. Cooper, A. M., A. Solache, and S. A. Khader. 2007. Interleukin-12 and tuber- oxidases, sensors and regulators of the intracellular messenger H2O2, and protein culosis: an old story revisited. Curr. Opin. Immunol. 19: 441–447. chaperones. Antioxid. Redox Signal. 15: 781–794. 45. Cooper, A. M., J. Magram, J. Ferrante, and I. M. Orme. 1997. Interleukin 12 17. Sato, D., A. Yanaka, T. Shibahara, H. Matsui, A. Nakahara, T. Yanagawa, (IL-12) is crucial to the development of protective immunity in mice intrave- E. Warabi, T. Ishii, and I. Hyodo. 2008. Peroxiredoxin I protects gastric mucosa nously infected with Mycobacterium tuberculosis. J. Exp. Med. 186: 39–45.

from oxidative injury induced by H. pylori infection. J. Gastroenterol. Hepatol. 46. de Jong, R., F. Altare, I. A. Haagen, D. G. Elferink, T. Boer, P. J. van Breda http://www.jimmunol.org/ 23: 652–659. Vriesman, P. J. Kabel, J. M. Draaisma, J. T. van Dissel, F. P. Kroon, et al. 1998. 18. Uwayama, J., A. Hirayama, T. Yanagawa, E. Warabi, R. Sugimoto, K. Itoh, Severe mycobacterial and Salmonella infections in interleukin-12 receptor- M. Yamamoto, H. Yoshida, A. Koyama, and T. Ishii. 2006. Tissue Prx I in the deficient patients. Science 280: 1435–1438. protection against Fe-NTA and the reduction of nitroxyl radicals. Biochem. 47. Kim, J. H., J. M. Lee, H. N. Lee, E. K. Kim, B. Ha, S. M. Ahn, H. H. Jang, and Biophys. Res. Commun. 339: 226–231. S. Y. Lee. 2012. RNA-binding properties and RNA chaperone activity of human 19. Kikuchi, N., Y. Ishii, Y. Morishima, Y. Yageta, N. Haraguchi, T. Yamadori, . Biochem. Biophys. Res. Commun. 425: 730–734. H. Masuko, T. Sakamoto, T. Yanagawa, E. Warabi, et al. 2011. Aggravation of 48. Hakimi, H., K. Suganuma, M. Usui, H. Masuda-Suganuma, J. M. Angeles, bleomycin-induced pulmonary inflammation and fibrosis in mice lacking per- M. Asada, S. Kawai, N. Inoue, and S. Kawazu. 2014. Plasmodium knowlesi oxiredoxin I. Am. J. Respir. Cell Mol. Biol. 45: 600–609. thioredoxin 1 binds to nucleic acids and has RNA chaperone activity. 20. Jang, H. H., K. O. Lee, Y. H. Chi, B. G. Jung, S. K. Park, J. H. Park, J. R. Lee, Parasitol. Res. 113: 3957–3962. S. S. Lee, J. C. Moon, J. W. Yun, et al. 2004. Two enzymes in one; two yeast 49. Baltz, A. G., M. Munschauer, B. Schwanha¨usser, A. Vasile, Y. Murakawa, peroxiredoxins display -dependent switching from a peroxidase M. Schueler, N. Youngs, D. Penfold-Brown, K. Drew, M. Milek, et al. 2012. The by guest on September 25, 2021 to a molecular chaperone function. Cell 117: 625–635. mRNA-bound proteome and its global occupancy profile on protein-coding 21. Yanagisawa, R., E. Warabi, K. Inoue, T. Yanagawa, E. Koike, T. Ichinose, transcripts. Mol. Cell 46: 674–690. H. Takano, and T. Ishii. 2012. Peroxiredoxin I null mice exhibits reduced acute 50. Castello, A., B. Fischer, K. Eichelbaum, R. Horos, B. M. Beckmann, C. Strein, lung inflammation following ozone exposure. J. Biochem. 152: 595–601. N. E. Davey, D. T. Humphreys, T. Preiss, L. M. Steinmetz, et al. 2012. Insights 22. Shichita, T., E. Hasegawa, A. Kimura, R. Morita, R. Sakaguchi, I. Takada, into RNA biology from an atlas of mammalian mRNA-binding proteins. Cell T. Sekiya, H. Ooboshi, T. Kitazono, T. Yanagawa, et al. 2012. Peroxiredoxin 149: 1393–1406. family proteins are key initiators of post-ischemic inflammation in the brain. Nat. 51. Kwon, S. C., H. Yi, K. Eichelbaum, S. Fo¨hr, B. Fischer, K. T. You, A. Castello, Med. 18: 911–917. J. Krijgsveld, M. W. Hentze, and V. N. Kim. 2013. The RNA-binding protein 23.Riddell,J.R.,X.Y.Wang,H.Minderman,andS.O.Gollnick.2010. repertoire of embryonic stem cells. Nat. Struct. Mol. Biol. 20: 1122–1130. Peroxiredoxin 1 stimulates secretion of proinflammatory cytokines by binding 52. Fukao, T., M. Tanabe, Y. Terauchi, T. Ota, S. Matsuda, T. Asano, T. Kadowaki, to TLR4. J. Immunol. 184: 1022–1030. T. Takeuchi, and S. Koyasu. 2002. PI3K-mediated negative feedback regulation 24. Ishii, T., E. Warabi, and T. Yanagawa. 2012. Novel roles of peroxiredoxins in of IL-12 production in DCs. Nat. Immunol. 3: 875–881. inflammation, and innate immunity. J. Clin. Biochem. Nutr. 50: 91–105. 53. Ichijo, H., E. Nishida, K. Irie, P. ten Dijke, M. Saitoh, T. Moriguchi, M. Takagi, 25. Janssen-Heininger, Y. M., M. E. Poynter, and P. A. Baeuerle. 2000. Recent ad- K. Matsumoto, K. Miyazono, and Y. Gotoh. 1997. Induction of apoptosis by vances towards understanding redox mechanisms in the activation of nuclear ASK1, a mammalian MAPKKK that activates SAPK/JNK and p38 signaling factor kB. Free Radic. Biol. Med. 28: 1317–1327. pathways. Science 275: 90–94. 26. Sanjabi, S., A. Hoffmann, H. C. Liou, D. Baltimore, and S. T. Smale. 2000. 54. Jarvis, R. M., S. M. Hughes, and E. C. Ledgerwood. 2012. Peroxiredoxin 1 Selective requirement for c-Rel during IL-12 p40 gene induction in macro- functions as a signal peroxidase to receive, transduce, and transmit peroxide phages. Proc. Natl. Acad. Sci. USA 97: 12705–12710. signals in mammalian cells. Free Radic. Biol. Med. 53: 1522–1530. 27. Lu, H. T., D. D. Yang, M. Wysk, E. Gatti, I. Mellman, R. J. Davis, and 55. Kim, S. Y., T. J. Kim, and K. Y. Lee. 2008. A novel function of peroxiredoxin 1 R. A. Flavell. 1999. Defective IL-12 production in mitogen-activated protein (Prx-1) in apoptosis signal-regulating kinase 1 (ASK1)-mediated signaling (MAP) kinase kinase 3 (Mkk3)-deficient mice. EMBO J. 18: 1845–1857. pathway. FEBS Lett. 582: 1913–1918. 28. Cooper, A. M., D. K. Dalton, T. A. Stewart, J. P. Griffin, D. G. Russell, and 56. Cao, J., J. Schulte, A. Knight, N. R. Leslie, A. Zagozdzon, R. Bronson, I. M. Orme. 1993. Disseminated tuberculosis in interferon g gene-disrupted Y. Manevich, C. Beeson, and C. A. Neumann. 2009. Prdx1 inhibits tumorigen- mice. J. Exp. Med. 178: 2243–2247. esis via regulating PTEN/AKT activity. EMBO J. 28: 1505–1517. 29. Flynn, J. L., J. Chan, K. J. Triebold, D. K. Dalton, T. A. Stewart, and 57. Carracedo, A., and P. P. Pandolfi. 2008. The PTEN-PI3K pathway: of feedbacks B. R. Bloom. 1993. An essential role for interferon g in resistance to Myco- and cross-talks. Oncogene 27: 5527–5541. bacterium tuberculosis infection. J. Exp. Med. 178: 2249–2254. 58. Kuroda, S., M. Nishio, T. Sasaki, Y. Horie, K. Kawahara, M. Sasaki, M. Natsui, 30. Nandi, B., and S. M. Behar. 2011. Regulation of neutrophils by interferon-g T. Matozaki, H. Tezuka, T. Ohteki, et al. 2008. Effective clearance of intracel- limits lung inflammation during tuberculosis infection. J. Exp. Med. 208: 2251–2262. lular Leishmania major in vivo requires Pten in macrophages. Eur. J. Immunol. 31. Grumont, R., H. Hochrein, M. O’Keeffe, R. Gugasyan, C. White, I. Caminschi, 38: 1331–1340. W. Cook, and S. Gerondakis. 2001. c-Rel regulates interleukin 12 p70 expression 59. Scanga, C. A., J. Aliberti, D. Jankovic, F. Tilloy, S. Bennouna, E. Y. Denkers, in CD8+ dendritic cells by specifically inducing p35 gene transcription. J. Exp. R. Medzhitov, and A. Sher. 2002. Cutting edge: MyD88 is required for resistance Med. 194: 1021–1032. to Toxoplasma gondii infection and regulates parasite-induced IL-12 production 32. Carmody, R. J., Q. Ruan, H. C. Liou, and Y. H. Chen. 2007. Essential roles of by dendritic cells. J. Immunol. 168: 5997–6001. c-Rel in TLR-induced IL-23 p19 gene expression in dendritic cells. J. Immunol. 60. Allen, J. E., and T. E. Sutherland. 2014. Host protective roles of type 2 immu- 178: 186–191. nity: parasite killing and tissue repair, flip sides of the same coin. Semin. 33. Sweet, M. J., and D. A. Hume. 1996. Endotoxin signal transduction in macro- Immunol. 26: 329–340. phages. J. Leukoc. Biol. 60: 8–26. 61. Pourcet, B., and I. Pineda-Torra. 2013. Transcriptional regulation of macrophage 34. Moore, K. W., R. de Waal Malefyt, R. L. Coffman, and A. O’Garra. 2001. arginase 1 expression and its role in atherosclerosis. Trends Cardiovasc. Med. 23: Interleukin-10 and the interleukin-10 receptor. Annu. Rev. Immunol. 19: 683–765. 143–152. 12 PRDX1 CONTRIBUTES TO HOST DEFENSES AGAINST TB

62. Rauh, M. J., V. Ho, C. Pereira, A. Sham, L. M. Sly, V. Lam, L. Huxham, 67. Dunn, P. L., and R. J. North. 1995. Virulence ranking of some Mycobacterium A. I. Minchinton, A. Mui, and G. Krystal. 2005. SHIP represses the generation of tuberculosis and Mycobacterium bovis strains according to their ability to alternatively activated macrophages. Immunity 23: 361–374. multiply in the lungs, induce lung pathology, and cause mortality in mice. Infect. 63. Mullican, S. E., C. A. Gaddis, T. Alenghat, M. G. Nair, P. R. Giacomin, Immun. 63: 3428–3437. L. J. Everett, D. Feng, D. J. Steger, J. Schug, D. Artis, and M. A. Lazar. 2011. 68. Lowe, D. M., P. S. Redford, R. J. Wilkinson, A. O’Garra, and A. R. Martineau. Histone deacetylase 3 is an epigenomic brake in macrophage alternative acti- 2012. Neutrophils in tuberculosis: friend or foe? Trends Immunol. 33: 14–25. ´ vation. Genes Dev. 25: 2480–2488. 69. Linden, A., M. Laan, and G. P. Anderson. 2005. Neutrophils, interleukin-17A 64. Yang, C. S., D. S. Lee, C. H. Song, S. J. An, S. Li, J. M. Kim, C. S. Kim, and lung disease. Eur. Respir. J. 25: 159–172. 70. Orme, I. M., and F. M. Collins. 1994. Mouse model of tuberculosis. In Tuber- D. G. Yoo, B. H. Jeon, H. Y. Yang, et al. 2007. Roles of peroxiredoxin II in the culosis: Pathogenesis, Protection, and Control, 1st Ed. B. R. Bloom, ed. regulation of proinflammatory responses to LPS and protection against American Society for Microbiology, Washington, D.C., p. 113–134. endotoxin-induced lethal shock. [Published erratum appears in 2007 J. Exp. Med. 71. Ladel, C. H., C. Blum, A. Dreher, K. Reifenberg, M. Kopf, and S. H. Kaufmann. 204: 1237.] J. Exp. Med. 204: 583–594. 1997. Lethal tuberculosis in interleukin-6-deficient mutant mice. Infect. Immun. 65. Li, L., T. Kaifu, M. Obinata, and T. Takai. 2009. Peroxiredoxin III-deficiency 65: 4843–4849. sensitizes macrophages to oxidative stress. J. Biochem. 145: 425–427. 72. Saunders, B. M., A. A. Frank, I. M. Orme, and A. M. Cooper. 2000. Interleukin-6 66. Inoue, K., H. Takano, E. Koike, E. Warabi, T. Yanagawa, R. Yanagisawa, and induces early g interferon production in the infected lung but is not required for T. Ishii. 2009. Peroxiredoxin I is a negative regulator of Th2-dominant allergic generation of specific immunity to Mycobacterium tuberculosis infection. Infect. asthma. Int. Immunopharmacol. 9: 1281–1288. Immun. 68: 3322–3326. Downloaded from http://www.jimmunol.org/ by guest on September 25, 2021