Leishmania donovani Exploits Tollip, a Multitasking Protein, To Impair TLR/IL-1R Signaling for Its Survival in the Host

This information is current as Naveen Parmar, Pragya Chandrakar, Preeti Vishwakarma, of September 30, 2021. Kavita Singh, Kalyan Mitra and Susanta Kar J Immunol 2018; 201:957-970; Prepublished online 15 June 2018; doi: 10.4049/jimmunol.1800062

http://www.jimmunol.org/content/201/3/957 Downloaded from

References This article cites 46 articles, 12 of which you can access for free at: http://www.jimmunol.org/content/201/3/957.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication by guest on September 30, 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 © 2018 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Leishmania donovani Exploits Tollip, a Multitasking Protein, To Impair TLR/IL-1R Signaling for Its Survival in the Host

Naveen Parmar,*,† Pragya Chandrakar,*,† Preeti Vishwakarma,*,† Kavita Singh,‡ Kalyan Mitra,†,‡ and Susanta Kar*,†

IL-1R/TLR signaling plays a significant role in sensing harmful foreign pathogens and mounting effective innate and adaptive immune responses. However, the precise mechanism by which Leishmania donovani, an obligate intramacrophagic pathogen, breaches IL-1R/TLR signaling and host-protective immunity remains obscure. In this study, we report the novel biphasic role of Toll-interacting protein (Tollip), a negative regulator of the IL-1R/TLR pathway, in the disease progression of experimental visceral leishmaniasis. We observed that during early hours of infection, L. donovani induced phosphorylation of IRAK-1, resulting in the release of Tollip from the IL-1R–associated kinase (IRAK)-1 complex in J774 macrophages, which then acted as an endocytic adaptor on cell surface IL-1R1 and promoted its lysosomal degradation. In the later stage, Tollip shuttled back to Downloaded from IRAK-1, thereby inhibiting IRAK-1 phosphorylation in association with IRAK-M to neutralize downstream TLR signaling in infected macrophages. Moreover, during late infection, L. donovani enhanced nuclear translocation and recruitment of transcrip- tion factors early growth response protein 2, NF erythroid 2–related factor 2, and Ahr on Tollip promoter for its induction. Small interfering RNA–mediated silencing of Tollip in infected macrophages significantly enhanced NF-kB activation and induced host- defensive IL-12 and TNF-a synthesis, thereby reducing amastigote multiplication. Likewise, abrogation of Tollip in L. donovani– infected BALB/c mice resulted in STAT-1–, IRF-1–, and NF-kB–mediated upregulation of host-protective cytokines and re- http://www.jimmunol.org/ duced organ parasite burden, thereby implicating its role in disease aggravation. Taken together, we conclude that L. donovani exploited the multitasking function of Tollip for its own establishment through downregulating IL-1R1/TLR signaling in macrophages. The Journal of Immunology, 2018, 201: 957–970.

athogenicity of Leishmania donovani has been attributed members played an influential role in initiating host innate im- to the exploitation of host cellular machinery, which fa- mune responses and directing adaptive immune responses against cilitates favorable adaptation of parasites to the intra- invading foreign pathogens (1). However, L. donovani employed

P by guest on September 30, 2021 cellular milieu of the innate immune sentinel cells such as numerous strategies to evade the first-line defense of the host by macrophages. Armed with a variety of pattern recognition re- manipulation of TLR and its downstream signaling components. ceptors, macrophages sense parasite surface molecules to trigger L. donovani hindered the TLR-mediated immune response at the appropriate host immune response. IL-1R/TLR superfamily multiple levels by exploiting host cellular proteins such as IL-1R– associated kinase (IRAK)-M (2), A20 (3), TRAF-3 (4), and phosphatases such as MKP-1, MKP-3, and PP2A (5). However, *Division of Parasitology, Council of Scientific and Industrial Research–Central the role of another Toll/IL-1R (TIR) superfamily member recep- Drug Research Institute, Lucknow 226031, India; †Academy of Scientific and Innovative Research, Anusandhan Bhawan, New Delhi 110001, India; and tor, IL-1R1, and its downstream signaling are poorly understood ‡Electron Microscopy Unit, Sophisticated Analytical Instrument Facility, Council during fatal visceral leishmaniasis (VL). IL-1 signaling plays an of Scientific and Industrial Research–Central Drug Research Institute, Lucknow important role in inflammation and early activation of host innate 226031, India immune response following foreign pathogens. Signal transduc- Received for publication January 16, 2018. Accepted for publication May 21, 2018. tion via IL-1R1 has a fundamental role in host defense against a This work was supported by Department of Science and Technology Grant SB/FT/ LS-310/2012, Council of Scientific and Industrial Research Grant CSIR NWP wide variety of pathogens, including Mycobacterium (6), Strep- BSC0114, and by Indian National Science Academy Grant SP/YSP/115/2015. N.P. tococcus (7), and Listeria monocytogenes (8). However, to prevent received a fellowship from the University Grants Commission (New Delhi), and P.C. irrelevant and adverse inflammatory responses, the IL-1R/TLR and P.V. are the recipients of fellowships from the Council of Scientific and Industrial Research (New Delhi). The funders had no role in study design, data collection and signaling pathway is tightly controlled by endogenous negative analysis, decision to publish, or preparation of the manuscript. regulators at multiple levels (9). This manuscript has Central Drug Research Institute Communication number 9670. Earlier reports suggested that a relationship exist between en- Address correspondence and reprint requests to Dr. Susanta Kar, Council of Scientific docytosis of ligand-activated receptors and further interaction with and Industrial Research–Central Drug Research Institute, Sector-10, Jankipuram downstream signaling molecules to continue signaling (10, 11). Extension, Sitapur Road, Lucknow 226031, Uttar Pradesh, India. E-mail address: [email protected] However, internalization followed by degradation in lysosomes is another fate of receptors that leads to attenuation of signaling and Abbreviations used in this article: BMDC, bone marrow–derived DC; ChIP, chroma- tin immunoprecipitation; CUE, coupling of ubiquitin to endoplasmic reticulum deg- is an extremely regulated process that involves recognition of radation; DC, dendritic cell; EEA1, early endosome Ag 1; Egr2, early growth ubiquitinated receptors by cargo proteins involved in their traf- response protein 2; IRAK, IL-1R–associated kinase; LAMP-1, lysosomal- associated membrane protein 1; LDU, Leishman–Donovan unit; LPG, lipophospho- ficking (12). Internalization of IL-1R1 from the plasma membrane glycan; Nrf2, NF erythroid 2–related factor 2; siRNA, small interfering RNA; TIR, played an important role in IL-1 signaling (13), and IL-1R1 also Toll/IL-1R; Tollip, Toll-interacting protein; TSS, transcription start site; VL, visceral undergoes monoubiquitnation before its degradation in lysosomes. leishmaniasis. The endosomal adaptor protein, Toll-interacting protein (Tollip), Copyright Ó 2018 by The American Association of Immunologists, Inc. 0022-1767/18/$35.00 is a multitasking regulator, as it not only participates in trafficking www.jimmunol.org/cgi/doi/10.4049/jimmunol.1800062 958 TOLLIP INHIBITS TLR/IL-1R1 SIGNALING IN VL and endosomal sorting of receptors such as TLR2, TLR4, and Parasite, cell culture conditions, and infection IL-1R1, but also in inhibiting IL/Toll/NF-kB signaling by L. donovani strain MHOM/IN/80/Dd8 was maintained as promastigotes in restricting phosphorylation of IRAK (14). Tollip is a multi- medium 199 (Sigma-Aldrich) supplemented with 10% heat-inactivated functional protein that contains the Tom1 binding domain at the FBS (Life Technologies) at 24 6 2˚C. J774 murine macrophage were N-terminal, a C2 domain that enables Tollip interaction with obtained from the National Centre for Cell Sciences (Pune, India) and were phosphoinositides, and the coupling of ubiquitin to endoplasmic grown as monolayers in RPMI 1640 medium (Sigma-Aldrich), supple- mented with 10% FBS, 100 U/ml penicillin, and 100 mg/ml streptomycin reticulum degradation (CUE) domain at the C-terminal. The (Invitrogen) in 5% CO2 atmosphere at 37˚C. Amastigotes of the MHOM/ CUE domain is responsible for associationofTolliptodifferent IN/80/Dd8 strain of L. donovani were freshly isolated from the spleen of cellular proteins, including IRAK-1, IL-1R1, TLR2, and TLR4. infected BALB/c mice. For in vitro studies, J774 macrophages were cul- Reports suggested that Tollip and its common variants negatively tured in petri dishes (BD Biosciences) overnight and were infected with stationary phase Dd8 promastigotes or freshly isolated amastigotes at a regulate TLR signaling in human monocytes and are associated parasite/macrophage ratio of 10:1 for different time intervals. Intracellular with susceptibility to tuberculosis (15). Tollip has an anti- parasite numbers were determined by Giemsa staining. For in vivo in- inflammatory effect on TLR signaling, as it suppressed TNF-a fection, 2 3 107 stationary phase L. donovani promastigotes were injected and IL-6 (16) production after stimulation with TLR2 and TLR4 in the tail vein of BALB/c mice, and potency of infection was ascertained ligands and induced the anti-inflammatory cytokine IL-10 (17). at the second week postinfection in Giemsa-stained impression smears of spleen after autopsy of three to four randomly selected mice. At each time Reports have suggested that Tollip-deficient cells showed in- point, the number of amastigotes per 1000 cell nuclei was counted in creased activation and binding of NF-kB to its regulated genes Giemsa-stained impressions of liver and spleen tissue, and organ parasite (14, 18). Increasing emerging evidence has also suggested that burden was expressed in Leishman–Donovan units (LDU) as described 3 several variants of the Tollip gene are associated with immu- previously (22), where 1 LDU = amastigote per nucleated cell organ Downloaded from weight in milligrams. nopathogenesis of cutaneous leishmaniasis (19), leprosy (20), and malaria (21), but in context of experimental VL, the role of In vivo translation inhibition of Tollip Tollip is still obscure. In this study, we observed that Tollip To inhibit Tollip translation in vivo, translation blocking vivo-morpholino played a bifunctional role in the negative regulation of IL-1R1 oligonucleotides targeting Tollip (59-TCTGCCTCGCCAACTGCACAGA- and the TLR2/TLR4 signaling pathway following L. donovani GAC-39) or control morpholino (59-CCTCTTACCTCAGTTACAATTTATA-39) were synthesized and purchased from Gene Tools. Both vivo-morpholinos infection and might have played a decisive task in survival of the http://www.jimmunol.org/ were dissolved in sterile PBS at 0.5 mM concentration and were injected into parasite, followed by disease progression, during experimental BALB/c mice (n = 5) through the tail vein twice a week at a dose of 10 nmol VL via dual means. (∼5mg/kg),startingat1wkpriortoinfectionanduptothesixthweek postinfection (see schematic diagram, Fig 6). The efficiency of knockdown was evaluated at the second, fourth, and sixth weeks postinfection in the Materials and Methods splenocytes of control and Tollip morpholino-treated animals by immuno- Reagents, Abs, and kits blotting. Spleen and liver parasite burdens were determined on a weekly basis and expressed as LDU (23). Splenocytes from different experimental RPMI 1640, medium 199, DMSO, BAPTA-AM, and DAPI were com- groups of BALB/c mice were also isolated at the second, fourth, and sixth mercially acquired from Sigma-Aldrich. Heat-inactivated FBS was weeks postinfection and processed for cytokine estimation and nuclear obtained from Life Technologies. Abs for Tollip, IL-1R1, a-tubulin, protein isolation (see schematic diagram, Fig 6). by guest on September 30, 2021 p–IkB-a,IkB-a, TRAF-6, p–TAK-1, TAK-1, early endosome Ag 1 (EEA1), Rab7, lysosomal-associated membrane protein 1 (LAMP-1), RNA extraction, reverse transcription, and quantitative rabbit anti-goat IgG-PE, goat anti-rabbit IgG-PE, goat anti-mouse real-time PCR IgG-PE and chromatin immunoprecipitation (ChIP)–grade Abs for Egr2, Ahr, and NF erythroid 2–related factor 2 (Nrf2) were purchased Total RNA was extracted from murine J774 macrophages using TRIzol from Santa Cruz Biotechnology (Dallas, TX). Abs for ERK1/2, p–ERK1/2, reagent and reverse transcribed using a RevertAid first-strand cDNA JNK1/2, p–JNK1/2, p38, p–p38, p–IRAK-1, IRAK-1, NF-kB–p65, his- synthesis kit (Fermentas) following the manufacturer’s instructions. tone H3, GAPDH, and b-actin were purchased from Cell Signaling Quantitative real-time PCR was carried out using SYBR Select Master Technology (Danvers, MA). Small interfering RNA (siRNA) for Tollip, Mix and transcript-specific primers with optimized thermal cycling con- IRAK-M, Egr2, Ahr, Nrf2, scrambled siRNA, transfection medium, and ditions. Expression of these transcripts was normalized to the expression transfection reagent were purchased from Santa Cruz Biotechnology. level of mouse GAPDH and relative quantitation was performed using the IL-12, TNF-a, TGF-b,IL-10,andIFN-g cytokines BD OptEIA kits 22DDCT method (24). All real-time RT-PCR reactions were conducted in were purchased from BD Biosciences. Cy3 oligonucleotides for NF-kB, duplicate on the StepOnePlus (Applied Biosystems). The primer pairs for IRF-1, and STAT-1 were synthesized and HPLC purified by Inte- the real-time RT-PCR assay for Tollip, TLR2, TLR4, IL-1R1, and GAPDH grated DNA Technologies. A ChIP assay was performed based on the are as follows: Tollip, 59-CCCCAGGACTTCCTCCGTAT-39 (forward), 59- SimpleChIP enzymatic ChIP kit (Cell Signaling Technology) protocol. AGCATAGCCCAGACGCAGAC-39 (reverse); TLR2, 59-GACGACTG- Tollip and control vivo-morpholinos were purchased from Gene Tools TACCCTCAATGG-39 (forward), 59-TTAAATGCTGGGAGAACGAG-39 (Philomath, OR). (reverse); TLR4, 59-AGACCTCAGCTTCAATGGTG-39 (forward), 59-GAGAC- TGGTCAAGCCAAGAA-39 (reverse); IL-1R1, 59-TTGGAGGGACAGTTTGG- Ethics statement ATA-39 (forward), 59-TTCTCCAACTCAAGCAGGAC-39 (reverse); GAPDH, 9 9 9 Animal care and experimental procedures were performed in strict ac- 5 -TTGCAGTGGCAAAGTGGAGA-3 (forward), 5 -GGGTCTCGCTCCTG- GAAGAT-39 (reverse). cordance with the guidelines by the Committee for the Purpose of Control and Supervision of Experiments on Animals (New Delhi, India). The Western blot and immunoprecipitation protocol was approved by the Institutional Animal Ethics Committee (registration no. 34/GO/Re-SL/Bi-S/99/CPCSEA) of the Council for Sci- Control and L. donovani–infected macrophages or splenocytes (2 3 106) entific and Industrial Research–Central Drug Research Institute (Lucknow, seeded in petri dishes were harvested and washed with cold 13 PBS. For India) (Institutional Animal Ethics Committee approval no. IAEC/2013/ the preparation of whole-cell lysates, RIPA buffer (150 mM NaCl, 50 mM 69/Renew01 [119/14]). Age-matched female BALB/c mice (20–25 g) of Tris, 1.0% IGEPAL, 0.5% sodium deoxycholate, 0.1% SDS, supplemented both sexes were bred in the National Animal Laboratory Centre in the with a final concentration of 1 mM PMSF and 13 protease inhibitor Council for Scientific and Industrial Research–Central Drug Research mixture) was used. Lysates in RIPA buffer were kept on ice for 15 min Institute. For experimental studies, female BALB/c mice were housed in followed by centrifugation at 10,000 rpm for 10 min at 4˚C to obtain climate-controlled (23 6 2˚C; relative humidity, 60%) and photoperiod- supernatant, which was stored at 220˚C for further use. Protein concen- controlled (12-h light/dark cycles) animal quarters with standard rodent trations in the supernatants were estimated using Bradford reagent (Sigma- pellets supplemented with grain and drinking water ad libitum. Anesthesia Aldrich). Furthermore, equal amounts of proteins were separated in 10% was performed by i.p. injection of ketamine combined with xylazine. acrylamide SDS-PAGE and transferred to nitrocellulose membrane using a Euthanasia of mice was performed by CO2 inhalation at the end of the semidry blotting system (Amersham Biosciences). This was followed by experiments. blocking with 5% (w/v) BSA, incubation with the primary Abs and The Journal of Immunology 959 secondary Abs, and detection using SuperSignal West Pico chemilumi- PMSF, 0.5 mM DTT, and a protease mixture inhibitor mix. Nuclei were nescent substrate (Thermo Scientific). The blot image was acquired in pelleted and resuspended in buffer supplemented with DTT, digested by ChemiDOC imagers (Bio-Rad Laboratories). micrococcal nuclease, and further homogenized on ice. After centrifu- For immunoprecipitation, cell lysates containing 500 mg of proteins gation, sheared chromatin was incubated with anti–early growth re- were first precleared by incubating with protein A/G plus agarose (Santa sponse protein 2 (Egr2; sc-20690X), anti-Ahr (sc-7474X), anti-Nrf2 Cruz Biotechnology) followed by incubation with appropriate Ab (sc-722X), anti-Rfx1 (sc-10652X), and anti-IgG as a negative control overnight with rotation at 4˚C. Immunoprecipitates were collected by overnight at 4˚C and precipitated using ChIP-grade protein G–agarose centrifugation at 3000 rpm, washed with 13 PBS, dissociated by beads. An aliquot of chromatin fraction not incubated with any Ab was boiling in electrophoresis sample buffer, and analyzed by Western used as the input control sample (2%). Immunoprecipitated chromatin blotting. complexes were then washed sequentially in low- and high-salt wash buffers at 4˚C and protein/DNA cross-links were reversed in the presence Nuclear protein isolation and EMSA of proteinase K at 65˚C for 2 h. PCR was performed with denaturation at 94˚C for 30 s, annealing at 49˚C for 30 s, and extension at 72˚C for 30 s Nuclear proteins were isolated as described by Haglund and Rothblum (25) for 40 cycles followed by 10 min at 72˚C using primers designed to L. donovani– with slight modifications. Briefly, control, infected macro- amplify the region encompassing putative transcription factor binding phages or splenocytes were homogenized in ice-cold hypotonic buffer sites for Egr2, Ahr, and Nrf2 on Tollip promoter. The sequences for (10 mM HEPES, 10 mM KCl, 1 mM DTT, 0.5 mM PMSF, 0.3 mM MgCl2, 3 primers are as follows: forward, 59-CATTTTAACGGGCTGCTGA-39 0.1 mM EDTA, and 1 protease inhibitor mixture) using a syringe and reverse, 59-AGGCAATCGGCTCTTCAGT-39. equipped with a 21-gauge needle. Homogenized mixtures were incubated for 15 min on ice with a 0.5% final concentration of Nonidet P-40 and then centrifuged at 10,000 3 g for 20 min. Supernatant containing cytosolic Immunofluoroscence and image analysis protein was collected and nuclear protein in the pellet was further extracted Macrophages (1 3 105 cells) were seeded overnight on 18-mm2 glass using high-salt lysis buffer (10 mM HEPES, 40 mM NaCl, 0.1 mM EDTA, coverslips kept in 35-mm petri dishes. Cells were then infected with Downloaded from 1 mM DTT, 0.5 mM PMSF, 25% glycerol, and 13 protease inhibitor L. donovani for different time periods, washed twice with 13 PBS, and mixture) after incubation for 30 min on ice. Finally, mixtures were fixed in ice-chilled methanol for 10 min. Fixed macrophages were per- centrifuged at 16,000 3 g for 25 min at 4˚C to obtain nuclear protein. For meabilized with 0.1% Triton X-100 and blocked with PBS containing EMSA, nuclear protein (5 mg) was incubated with 20 pmol of a 59 Cy3- 1% BSA for 30 min. For fluorescent labeling, cells were incubated with labeled double-stranded oligonucleotides probe in appropriate binding primary Abs overnight at 4˚C followed by washing and incubation with buffer (250 mM HEPES, 10 mM KCl, 1 mM EDTA, 10 mM DTT, 1 mM appropriate fluorochrome-conjugated secondary Abs for 1 h at 4˚C. PMSF, and 100 mM NaCl), 1 mg of poly(deoxyinosinic-deoxycytidylic) After fluorescent labeling, cell nuclei were counterstained with DAPI acid, 1 mg of tRNA, and 10% glycerol. To analyze specificity of the in- (1 mg/ml) in PBS plus 10 mg/ml RNase A. Finally, coverslips were http://www.jimmunol.org/ teraction, a 100-fold molar excess of unlabeled competitor oligonucleotide mounted in antifade reagent (Thermo Scientific) on slides and were was added for 30 min at 25˚C and analyzed by EMSA in the presence of all examined using a confocal laser-scanning microscope (LSM510 META; components of the binding reaction. After incubation, the DNA/protein Carl Zeiss). complex was electrophoresed on 6% nondenaturing polyacrylamide gels 3 in 0.5 TBE buffer. Shifted complexes on gel were scanned directly on an NF-kB reporter assay ImageQuant LAS 4000 (GE Healthcare Life Sciences). Consensus oligo- nucleotide sequences used for preparation of Cy3-labeled double-stranded NF-kB promoter activation in control, L. donovani–infected, and siRNA- oligonucleotide probes for different transcription factors are as follows: treated macrophages were determined using a luciferase-based reporter STAT1, 59-Cy3-CATGTTATGCATATTCCTGTAAGTG-39,39-GTACAA- assay system (Promega). Briefly, J774 macrophages were cotransfected TACGTATAAGGACATTCAC-59; NF-kB, 59-Cy3-AGTTGAGGGGA- with pGL4.32 (luc2P/NF-kB-RE/Hygro) and Renilla luciferase (pRL-TK– by guest on September 30, 2021 CTTTCCCAGGC-39,39-TCAACTCCCCTGAAAGGGTCCG-59; IRF1, Renilla luciferase) as an internal control vector (in a 10:1 ratio) and 59-Cy3- GGAAGCGAAAATGAAATTGACT-39,39- CCTTCGCTTTAC- cultured for 24 h in serum-free medium using Lipofectamine reagent TTTAACTGA-59. (Invitrogen) according to the manufacturer’s instructions. The luciferase activity was determined using the Dual-Luciferase assay kit (Promega) ELISA according to the manufacturer’s instructions, and luminescence was de- tected on a plate reader luminometer (Berthold, Orion). For siRNA- The levels of various cytokines in the culture supernatants of J774 mac- mediated silencing, J774 macrophages were transfected with Tollip, rophages or splenocytes were measured using selective kit as per the manufacturer’s instructions (BD Biosciences). Isolated splenocytes from IRAK-M, Tollip plus IRAK-M, or control siRNA according to the man- mice were plated in triplicate (105 cells per well) in 96-well plates and ufacturer’s instructions (Santa Cruz Biotechnology) as mentioned above, L. donovani allowed proliferated for 72 h in a 5% CO incubator at 37˚C either in followed by infection with for 24 h and then transfection with 2 k Renilla presence or absence of soluble leishmanial Ag. Soluble leishmanial Ag luc2P/NF- B and pRL-TK– luciferase to determine luciferase ac- was prepared as described elsewhere (26) and used at 5 mg/ml for ex vivo tivity. Transfections were performed in duplicate, and experiments were repeated at least three times. Luciferase activity data were normalized for stimulation of splenocytes. transfection efficiency by dividing firefly luciferase activity with that of siRNA-mediated knockdown in macrophages Renilla luciferase. siRNA targeted against Tollip, IRAK-M, Egr2, Ahr, and Nrf2 (Santa Cruz Purification of lipophosphoglycan Biotechnology) or scrambled siRNA was transfected into J774 macro- phages according to the manufacturer’s instructions. Briefly, 50 nM Lipophosphoglycan (LPG) was purified from log-phase culture of siRNA diluted in transfection medium was coincubated with siRNA L. donovani promastigotes as described previously with minor modifi- 9 transfectionreagentandthenaddedtomacrophagesalongwithaddi- cations (27, 28). Briefly, L. donovani promastigotes (1 3 10 cells) were tional siRNA transfection reagent followed by incubation for 5 h. After harvested by centrifugation and extracted with methanol/chloroform/ transfection, macrophages were incubated with normal growth medium water (2:1:0.5 v/v) for 2 h at room temperature. The insoluble fraction containing twice the normal serum and antibiotics concentration for an containing LPG was used for LPG extraction with 9% 1-butanol in water additional 24 h. After replacing media with fresh 13 growth media, cells and the supernatants were vacuum dried. A further fraction was resus- were infected with L. donovani for different time periods and processed pended in 0.1 M ammonium acetate and purified using Octyl-Sepharose for cytokine estimation by ELISA, lysate preparation for immunoblot, chromatography using a propan-1-ol gradient (5–60%) in 0.1 M am- and coimmunoprecipitation analysis. Intracellular parasite numbers were monium acetate. Purified LPG was resuspended in sterile PBS, and also determined for Tollip siRNA-treated L. donovani–infected macro- hexose concentration was estimated in fractions by a modified phenol– phages by Giemsa staining. sulphuric acid assay. LPG was sonicated in RPMI 1640 before treating macrophages. ChIP assay Statistical analysis ChIP assay was performed using the SimpleChIP enzymatic ChIP kit from Cell Signaling Technology as per the manufacturer’s instructions. In brief, Results are presented as mean 6 SD of at least three independent transcription factors were cross-linked to DNA with 1% formaldehyde for experiments performed in replicates, and differences between two or 10 min. The cells were washed with ice-cold PBS, pooled, pelleted, and more groups were analyzed for statistical significance by a one-way incubated on ice for 10 min in lysis buffer supplemented with 1 mM ANOVA followed by a Tukey posttest using GraphPad Prism (version 5) 960 TOLLIP INHIBITS TLR/IL-1R1 SIGNALING IN VL software. Differences with p , 0.05 were considered statistically (725 6 65 pg/ml). In accordance with previous findings (3), secre- significant. tions of IL-6 (51 6 4.2 pg/ml, 66 6 4.9 pg/ml, Fig. 1A) and TNF-a (158 6 12 pg/ml, 326 6 26 pg/ml, Fig. 1B) were attenuated Results from macrophages incubated with stationary phase L. donovani L. donovani exploits Tollip to manipulate proximal events of promastigotes along with TLR2 and TLR4 respective ligands LPG IL-1R1/TLR signaling in macrophages and LPS. L. donovani multiplies and thrives as an intracellular L. donovani inhibited TLR-mediated immune responses in macro- pathogen in macrophages by destabilization of signaling intermedi- phages by different adopted strategies to create a favorable niche for ates and stabilization of negative regulators (2) of the TLR signaling its multiplication at different stages (29). However, the precise pathway to dampen the host innate immune response, and Tollip has mechanism by which L. donovani neutralized IL-1R1 signaling– been implicated as a crucial negative regulatory protein of the mediated host immunity during infection is still obscured. To IL-1R1, TLR2, and TLR4 signaling pathway (18, 30). To determine understand the effect of L. donovani on IL-1R1 signaling, macro- whether Tollip was induced in L. donovani–infected macrophages, phages were cocultured with L. donovani promastigotes and IL-1b, we checked for its mRNA and protein expression levels by using the cognate ligand for IL-1R1, and the level of host defensive real-time PCR and Western blot, respectively. Infection of J774 proinflammatory cytokines was measured. As shown in Fig. 1A, the macrophages with stationary phase L. donovani promastigotes led level of IL-6 was significantly reduced (76.2%) after cotreatment of to the upregulation of Tollip, both at the mRNA and protein levels macrophages with L. donovani plus IL-1b as compared with IL-1b in a time-dependent fashion. The mRNA level of Tollip in macro- treatment alone. Similarly, we found that the level of TNF-a in phages was induced at 4 h postinfection (2.6-fold) with maximum cultured macrophage supernatant was significantly decreased upregulation observed at 16 h (9.3-fold, as compared with nonin- Downloaded from (71.4%) when coincubated with L. donovani and IL-1b (Fig. 1B). fected control), and sustained transcript expression was detected up Alternatively, stimulation of macrophages with IL-1b resulted in to 48 h (6.1-fold) postinfection (Fig. 1C). A significant and consis- significant upregulation of IL-6 (892 6 81 pg/ml) and TNF-a tent upregulation of Tollip was also observed at the protein level http://www.jimmunol.org/ by guest on September 30, 2021

FIGURE 1. Tollip modulates the TLR/IL-1R1 signaling pathway in L. donovani–infected J774 macrophages. (A and B) J774 macrophages were in- cubated with IL-1b (10 ng/ml), LPS (100 ng/ml), LPG (6 mg/ml), IL-1b plus L. donovani, LPS plus L. donovani, LPG plus L. donovani, and L. donovani alone for 24 h. Levels of IL-6 and TNF-a were evaluated by sandwich ELISA. (C) mRNA expression of Tollip in infected macrophages at different time points was evaluated by SYBR Green–based real-time PCR. The expression changes were calculated relative to noninfected control macrophages after normalizing with GAPDH as endogenous control. (D) J774 macrophages were infected with stationary phase L. donovani promastigotes for the indicated time intervals and expression of Tollip was analyzed by Western blotting. b-Actin served as the loading control. (E) J774 macrophages were transfected with Tollip or scrambled control siRNA for 24 h, followed by treatment as mentioned in (A) and (B). Concentrations of IL-6 and TNF-a in the cell-free culture supernatants were detected by ELISA. (F) The efficacy of Tollip siRNA was determined in whole-cell extracts from macrophages expressing Tollip- targeting or control siRNAs by immunoblotting using a specific Ab against Tollip. (G) The TLR2 mRNA level was determined by real-time PCR in normal, L. donovani, LPS (100 ng/ml), and LPS plus L. donovani incubated J774 macrophages for the indicated time periods, as described in (C). (H) Levels of TLR4 mRNA expression were assessed by real-time PCR in normal, L. donovani, PGN plus ionomycin (10 mg/ml, 25 ng/ml), and PGN plus ionomycin plus L. donovani incubated J774 macrophages for the indicated different time periods. (I) The level of IL-1R1 mRNA expression was assessed by real-time PCR as in (G). (J) IL-1R1 expression was detected by Western blot analysis in normal, L. donovani, LPS (100 ng/ml), and LPS plus L. donovani incubated J774 macrophages after 16 h of treatment. Anti–a-tubulin Ab was used as a loading control. (K) J774 macrophages were infected with stationary phase L. donovani promastigotes for different time periods, and whole-cell lysates were subjected to immunoblot analysis to analyze the expression of phos- phorylated IRAK-1 and total IRAK-1. (L and M) J774 macrophages were infected with stationary phase L. donovani promastigotes for the indicated time periods. The cells were lysed and subjected to immunoprecipitation and Western blotting. Results are representative of three individual experiments. The significance between different experimental groups was calculated by a one-way ANOVA followed by a Tukey posttest using GraphPad Prism 5. Signi- ficance was determined for infected versus noninfected groups. *p , 0.05, **p , 0.01, *** p , 0.001. ns, not significant. IP, immunoprecipitation using the indicated Ab; WB, Western blot analysis using the indicated Abs. The Journal of Immunology 961 in infected macrophages, which was maximal at 16 h postinfection might interact with cell surface IL-1R1 after dissociating from and significant up to 48 h (Fig. 1D). Therefore, we next performed IRAK-1 and facilitate IL-1R1 endosomal sorting in infected mac- siRNA-mediated knockdown of Tollip prior to infection to investi- rophages. As shown in Fig. 1M, coimmunoprecipitation studies gate its functional role in modulating IL-1R1/TLR signaling in revealed that a strong association occurs between IL-1R1 and Tollip L. donovani–infected macrophages. Functional knockdown of Tollip during early phases of infection (2–8 h). However, in the same bysiRNAmarkedlyrestoredbothIL-6andTNF-a levels (3.1-fold coimmunoprecipitation study, a very low level of IL-1R1 was de- increase for IL-6 and 3.3-fold increase in TNF-a as compared with tected in association with Tollip at late hours (12 and 24 h) of in- control siRNA-treated macrophages) in macrophages coincubated fection, which might be due to degradation of IL-1R1 in the with L. donovani plus IL-1b, the cognate ligand for IL-1R1 lysosomes. (Fig. 1E). Similarly, pretreatment of macrophages with Tollip siRNA significant restored the level of IL-6 and TNF-a (6.1-fold L. donovani–induced host Tollip promotes trafficking and and 3.2-fold increase, respectively, as compared with control lysosomal degradation of IL-1R1 as an early stage siRNA-treated cells) in macrophages incubated with L. donovani event in macrophages plus LPS or LPG (cognate ligand for TLR4 and TLR2, respectively; Endocytosis-mediated lysosomal degradation of receptors associ- Fig. 1E). The efficacy of siRNA on Tollip expression was assessed ated with cell plasma membrane is an alternative mechanism for by Western blotting. Tollip expression at the protein level was re- negative regulation of TLR/IL-1R signaling. Being a multitasking duced significantly in cells expressing Tollip-specific siRNA com- protein, Tollip also functions as an endocytic adaptor protein and pared with cells expressing control siRNA (Fig. 1F). Collectively, thus modulates intracellular trafficking and degradation of recep- these observations suggested an immune-suppressive role of Tollip tors (31). Strong interaction between Tollip and IL-1R1 during Downloaded from during experimental VL and prompted us to evaluate its effect in early hours of infection followed by specific downregulation of modulating proximal signaling events of IL-1R1/TLR signaling. IL-1R1 at both the mRNA and protein levels led us to speculate The CUE domain present at the C-terminal region of Tollip is re- that being an endocytic adaptor protein, Tollip might be involved sponsible for its interaction with the TIR domain of various re- in intracellular trafficking and degradation of IL-1R1 in infected ceptors, including TLR2, TLR4, and IL-1R1 (30, 31). Generalizing macrophages. To test this hypothesis, we further used confocal

these observations, we hypothesized next that L. donovani might microscopy analysis to ensure interaction of Tollip with IL-1R1 http://www.jimmunol.org/ use Tollip for degradation of receptors belonging to the IL-1R1/ that is required for internalization and lysosomal degradation TLR superfamily to evade host immunity. To test this hypothesis, during the early course of L. donovani infection. To this end, we we first evaluated the effect of L. donovani infection on mRNA first monitored the localization of internalized IL-1RI after dif- levels of TLR2, TLR4, and IL-1R1 in J774 macrophages. Stimu- ferent intervals of L. donovani infection by costaining it with lation of J774 macrophages with LPS for different time intervals EEA1, Rab7, and LAMP-1, markers specific for early endosomes, resulted in increased mRNA levels of TLR2 (5.4-fold increase as late endosomes, and lysosomes, respectively. In control cells, IL- compared with control at 8 h posttreatment) and IL-1R1 (9.2-fold 1R1 is predominantly localized on the cell surface, and it barely increase as compared with control at 8 h posttreatment) (Fig. 1G, had any colocalization with EEA1, Rab7, or LAMP-1 as observed 1I), whereas TLR4 mRNA levels were upregulated (4.2-fold in- upto 12 h (Fig. 2A, 2C, 2E). On the contrary, L. donovani infection by guest on September 30, 2021 crease as compared with control at 8 h posttreatment) after exposure stimulated rapid internalization of IL-1R1 in J774 macrophages, of J774 macrophages to peptidoglycan plus ionomycin (Fig. 1H). as an increased proportion of IL-1R1 was colocalized with the We observed that LPS-induced TLR2 and peptidoglycan plus ion- early endosomal marker EEA1 (Fig. 2B) as observed at 6 h omycin–induced TLR4 mRNA levels were not changed postinfec- postinfection, whereas such colocalization was absent in unin- tion of J774 macrophages with L. donovani.Similarly,L. donovani fected cells. Further immunolabeling of J774 macrophages after infection could not change LPS-induced IL-1R1 expression up to longer incubation with L. donovani (9 h) confirmed that inter- 8 h of infection. However, L. donovani infection significantly re- nalized IL-1R1 was colocalized with the late endosomal marker duced LPS-induced IL-1R1 mRNA expression at 16 h (27.1% re- Rab7 at later time points, thereby suggesting trafficking of IL-1R1 duction as compared with LPS-stimulated macrophages, Fig. 1I) in from early to late endosomes (Fig. 2D). However, at late hours of J774 macrophages coincubated with LPS and L. donovani. Inter- infection (12 h) we did not observed any costaining of IL-1R1 estingly, in similar experimental conditions when we observed the with the lysosomal marker LAMP-1, which might be due to ly- protein level of TLR2, TLR4, and IL-1R1, we found that sosomal degradation of IL-1R1 (Fig. 2F). In fact, after 12 h of L. donovani infection could only reduce IL-1R1 expression at 16 h infection, we hardly detected any IL-1R1 expression at the cellular (Fig. 1J) but not TLR2 and TLR4 expression at the protein level level as observed by immunostaining. To further confirm the role (data not shown). In unstimulated cells, Tollip sequesters IRAK-1, of Tollip in IL-1R1 trafficking, we employed two complementary an upstream kinase of the IL-1R1 and TLR2/TLR4 signaling approaches. First, we checked Tollip trafficking in early endo- pathways (30) in the cytosol, keeping it in an inactive form. Inter- somes, late endosomes, and lysosomes at similar time points estingly, we observed that during early hours, L. donovani induced postinfection as we employed for IL-1R1. Costaining analysis via transient phosphorylation of IRAK-1 in J774 macrophages, which confocal microscopy revealed that similar to IL-1R1, Tollip was was observed up to 4 h postinfection, followed by a sharp decrease, also co localized with EEA1, Rab7, and LAMP-1 at 6, 9, and 12 h and remained dephosphorylated throughout the later stage of in- postinfection, respectively (Fig. 2H–J). In uninfected macrophages, fection (8–24 h, Fig. 1K). Interestingly, at early hours, IRAK-1 we did not observe any trafficking of Tollip in the early endosome phosphorylation coincided with release of Tollip from the IRAK-1 as observed by the absence of costaining of Tollip with EEA1, thus complex as observed by coimmunoprecipitation studies. As shown indicating that the subcellular distribution of Tollip was induced in Fig. 1L, we observed a strong association of Tollip with IRAK-1 following L. donovani infection in the host (Fig. 2G). Second, in control macrophages that was gradually decreased following siRNA-mediated silencing of Tollip was performed to further cor- L. donovani infection up to 8 h and then further restored back at late roborate its role in IL-1R1 trafficking and lysosomal degradation in hours (16–24 h). As IL-1R1 and Tollip are known to physically infected macrophages. As shown in Fig. 2K, knockdown of Tollip interact with each other via their specific TIR-CUE domains, we in infected macrophages restored the normal distribution of IL-1R1 further hypothesized that being an endocytic adaptor protein, Tollip on the cell surface in comparison with control siRNA-treated 962 TOLLIP INHIBITS TLR/IL-1R1 SIGNALING IN VL Downloaded from http://www.jimmunol.org/ by guest on September 30, 2021

FIGURE 2. Trafficking and localization of IL-1R1 and Tollip following L. donovani infection in J774 macrophages. (A–F) Time course confocal image analysis of lysosomal degradation of IL-1R1 in L. donovani–infected J774 macrophages. Confocal fluoroscence microscopy shows colocalization of IL-1R1 with anti-EEA1, anti-Rab7, and anti–LAMP-1 in early endosomes (B), late endosomes (D), and lysosomes (F), respectively, in infected macrophages after immunostaining as indicated by arrows (merged panel). Similarly, (A), (C), and (E) represent localization of IL-1R1 in control macrophages when costained with the markers for early endosomes (A), late endosomes (C), and lysosomes (E) in uninfected macrophages. Intracellular localization of IL-1R1 is shown in green (Alexa Fluor 488) and localization of organelles markers are shown in red. DAPI as a counterstain is shown in blue. (G–J) J774 macrophages were infected with stationary phase L. donovani promastigotes for the indicated time period. Fixed cells were immunostained to detect trafficking of Tollip in early endosome (H), late endosome (I), and lysosome (J) by using anti-EEA1, anti-Rab7, and anti–LAMP-1 Abs, respectively. For uninfected cells Tollip trafficking was observed in early endosomes (G). Secondary Abs were anti- rabbit FITC (visualized in green for Tollip) and anti-goat PE (visualized in red for all organelle markers). Arrows indicate colocalization of Tollip in early endosomes, late endosomes, and lysosomes. (K) J774 macrophages were transfected with Tollip siRNA or control scrambled siRNA for 24 h. After transfection, cells were infected with L. donovani for 12 h, permeabilized, and incubated with secondary Ab against IL-1R and immunostained for LAMP-1 to evaluate the IL-1R1 surface expression and LAMP-1 localization via confocal microscopy. IL- 1RI is shown in green, LAMP-1 is shown in red, and nuclear staining by DAPI is shown in blue. (L) J774 macrophages were transfected in the presence of Tollip siRNA or control scrambled siRNA for 24 h. After transfection, cells were coincubated with LPS and L. donovani for 12 h and then harvested for immunobloting with Abs against IL-1R1 and a-tubulin. Results are representative of three separate individual experiments (n =3). infected macrophages, where we observed negligible detection of LPS and L. donovani. Tollip knockdown markedly restored the IL-1R1. Although Tollip knockdown restored IL-1R1 expression at expression level of IL-1R1 in LPS plus L. donovani–stimulated the cell surface of infected macrophages, we did not observe any cells, which were attenuated in control siRNA-treated macrophages colocalization of IL-1R1 with LAMP-1, which suggested that IL- coincubated with LPS and L. donovani (Fig. 2L). These observa- 1R1 trafficking and lysosomal degradation was hampered in the tions clearly depicted that L. donovani induced downregulation of absence of Tollip (Fig. 2K). To further reconfirm our observation, IL-1R1 from the cell surface of infected hosts, which was pre- immunoblot analysis was performed in macrophages that underwent dominantly facilitated by its interaction with Tollip followed by siRNA-mediated knockdown of Tollip followed by coincubation with sorting and degradation in lysosomes. The Journal of Immunology 963

L. donovani diminished the magnitude of IL-1R1/TLR partially restored IRAK-1 phosphorylation in infected phagocytic signaling by inhibiting IRAK-1 phosphorylation through cells, dual knockdown of Tollip and IRAK-M markedly restored upregulation of Tollip in macrophages phosphorylation of IRAK-1 in infected macrophages (Fig. 3A). Tollip controls the magnitude of inflammatory cytokine production IRAK-1 phosphorylation facilitated its association with TRAF-6 in response to IL-1b and LPS (32). Therefore, the effect of Tollip followed by phosphorylation of downstream kinase, TAK-1 (33). on IL-1R1/TLR signaling was further determined by assessing Coimmunoprecipitation studies revealed that Tollip knockdown activation of downstream signaling components in infected mac- partially restored the association of IRAK-1 with TRAF-6 in in- rophages after silencing Tollip. Because ligand binding and fected macrophages; however, this association was greatly in- agonist-induced stimulation of both IL-1R1/TLR signaling resul- creased in infected cells that underwent dual knockdown of Tollip ted in phosphorylation of IRAK-1, which was also negatively and IRAK-M (Fig. 3B). This was further reflected in TAK-1 regulated by Tollip, we first checked the p–IRAK-1 level in Tollip phosphorylation that was absent in L. donovani–infected cells, siRNA-treated infected cells. Our earlier observation suggested partially restored in Tollip knockdown-infected cells, and mark- that IRAK-1 phosphorylation was diminished in L. donovani– edly induced in Tollip plus IRAK-M knockdown cells (Fig. 3C). infected macrophages at late hours, so we checked IRAK-1 Phosphorylated and activated TAK-1 acts as an upstream kinase phosphorylation at 16 h postinfection in Tollip knockdown mac- for MAPK family members as well as IKK/NF-kB signaling, and rophages. In this context, Tollip resembles the negative regulator L. donovani could downregulate both of these pathway. Tollip IRAK-M, which also binds to IRAK-1 and halts signaling by knockdown could not restore phosphorylation-mediated activation blocking recruitment of IRAK-4 and thereby the phosphorylation of any of the MAPKs (p38, ERK1/2, and JNK1/2) in infected Downloaded from and/or activation of IRAK-1. Moreover, IRAK-M is already macrophages, and a similar finding was observed following known for attenuating phosphorylation of IRAK-1 and its down- IRAK-M knockdown (Fig. 3D). However, as shown in Fig. 3E, stream signaling in L. donovani–infected macrophages (2). Our Tollip knockdown partially restored IkB phosphorylation followed observation clearly demonstrated that although Tollip knockdown by its degradation (a prerequisite for NF-kB activation) in http://www.jimmunol.org/ by guest on September 30, 2021

FIGURE 3. Effect of Tollip knockdown on modulation of early TLR/IL-1R1 signaling during infection. (A and C) J774 macrophages were transfected with control siRNA, Tollip siRNA, IRAK-M siRNA, or Tollip plus IRAK-M siRNA (24 h) and then infected with L. donovani promastigotes for 16 h. Lysates were subjected to immunoblot to detect expression level of kinases using anti–p-IRAK1 and anti-IRAK1 Abs (A) and anti–p-TAK1 and anti-TAK1(C)Abs.(B)J774 macrophages were transfected with different siRNA followed by infection with L. donovani. The cell lysates were immunoprecipitated with anti-IRAK1 Ab and the immunoprecipitates were analyzed for TRAF-6 expression by Western blotting. (D and E) Cell lysates were also subjected to immunoblot to detect expression level of phosphorylated and nonphosphorylated anti-ERK1/2, anti-JNK, anti-p38, and IkB-a using the respective Abs. (F) J774 macrophages were treated with different siRNA as mentioned above and then infected with L. donovani promastigotes for 24 h. Nuclear and cytosolic extracts were prepared and lysates were subjected to immunoblot to detect nuclear and cytosolic expression levels of NF-kB p65. Histone H3 and GAPDH were used as nuclear and cytosolic markers, respectively. (G) J774 macrophages were transiently transfected with control siRNA, Tollip siRNA, IRAK-M siRNA, or Tollip plus IRAK-M siRNA along with pGL4.32 (luc2P/NF-kB-RE/Hygro) and Renilla luciferase (pRL-TK–Renilla luciferase) vector. After 24 h of transfection, cells were infected with L. donovani promastigotes for 24 h. Cells were lysed and luciferase levels were measured. The error bars represent the 6 SD of the mean from at least three independent experiments. IRAK-M siRNA– or Tollip plus IRAK-M siRNA–treated macrophages were compared with control siRNA-treated macrophages. (H and I) J774 macrophages were transfected with control siRNA, Tollip siRNA, IRAK-M siRNA, and Tollip plus IRAK-M siRNA followed by infection for 24 h. Concen- trations of IL-12, TNF-a (H), IL-10, and TGF-b (I) in the cell-free culture supernatants were measured by ELISA. (J) J774 macrophages were transfected with control siRNA, Tollip siRNA, IRAK-M siRNA, and Tollip plus IRAK-M siRNA followed by infection for 48 h. Intracellular amastigotes number was assessedby Giemsa staining. The significance between different experimental groups was calculated by a one-way ANOVA followed by a Tukey posttest using GraphPad Prism 5. Significance was determined for control siRNA versus Tollip siRNA, IRAK-M siRNA, and Tollip plus IRAK-M siRNA. **p , 0.01, ***p , 0.001. 964 TOLLIP INHIBITS TLR/IL-1R1 SIGNALING IN VL

L. donovani–infected cells, which was more prominent in the case L. donovani–infected macrophages at different time points, but not of Tollip plus IRAK-M siRNA–treated cells. These observations in the control IgG Ab immunoprecipitates. ChIP analysis showed suggested that Tollip might have some role in restoring IKK/NF- maximum binding of Egr2 (Fig. 4C) and Nrf2 (Fig. 4E) at 12 h kB signaling in infected macrophages. To further confirm this, we postinfection on Tollip, whereas for Ahr maximum binding was observed nuclear translocation of the NF-kB p65 subunit (Fig. 3F) observed at 24 h postinfection (Fig. 4D). Additionally, we ob- followed by assessing NF-kB activity by reporter-based luciferase served sustained recruitment and binding of Egr2, Nrf2, and Ahr assay (Fig. 3G). As depicted in Fig. 3F, the p65 subunit of NF-kB on the Tollip promoter up to 24 h of L. donovani infection in is predominantly localized in the cytosol of infected macrophages, macrophages (Fig. 4B). On the contrary, we did not observe any whereas Tollip knockdown partially induced its translocation to detectable binding of transcription factors Rfx1 on the Tollip the nucleus, as evident by increased expression of p65 in the promoter in L. donovani–infected macrophages (data not shown). nuclear fraction. Cytosolic-to-nuclear translocation of p65 was Chromatin fractions from uninfected macrophages exhibited no greatly restored in infected macrophages pretreated with Tollip binding of Egr2, Nrf2, and Ahr to the Tollip promoter, suggesting plus IRAK-M siRNA. Similarly, a luciferase-based reported assay that the basal expression level of Tollip is not modulated by these revealed significant NF-kB activation in infected cells subjected to transcription factors (Fig. 4B). Relative enrichment of transcrip- Tollip knockdown that was greatly enhanced in Tollip plus IRAK- tion factor Egr2, Nrf2, and Ahr on the Tollip promoter was also M siRNA–treated macrophages (Fig. 3G). This was further re- confirmed by quantitative real-time PCR (Fig. 4C–E). Because flected in downstream host defensive cytokine synthesis such as binding of Egr2, Ahr, and Nrf2 on Tollip promoters required their IL-12 and TNF-a, which was regulated by transcription factor NF- nuclear translocation, we next performed confocal analysis to kB. As shown in Fig. 3H and 3I, siRNA-mediated silencing of further confirm our observation by studying nuclear localization of Downloaded from Tollip in infected macrophages indeed resulted in increased IL-12 all of these transcription factors. To ensure this, macrophages were (3.6-fold) as compared with control siRNA-treated infected cells infected with L. donovani for 12 h and analyzed for nuclear and TNF-a (3.1-fold) secretion, which was further enhanced in translocation of Egr2 and Nrf2, whereas for Ahr nuclear locali- Tollip plus IRAK-M siRNA–pretreated infected cells (Fig. 3H). In zation was observed at 24 h postinfection. Immunofluoroscence the similar experimental systems, we observed a decrease in the analysis clearly showed that L. donovani infection promoted nu-

level of Th2 cytokines (IL-10 and TGF-b levels) in Tollip clear translocation of Egr2, Ahr, and Nrf2, as evident by cos- http://www.jimmunol.org/ siRNA– and Tollip plus IRAK-M siRNA–pretreated infected cells taining of the dye Alexa Fluor 594 conjugated with the secondary (Fig. 3I). Finally, the effect of functional knockdown of Tollip in Ab used for these transcription factors with nuclear staining dye, infected macrophages was also reflected at the level of intracel- that is, DAPI (Fig. 4F). On the contrary, Egr2, Ahr, and Nrf2 were lular parasite multiplication. As shown in Fig. 3J, in comparison predominantly present in cytosol in uninfected macrophages (Fig. with control siRNA-treated infected macrophages there were 4F), as we did not observe any costaining of nuclear dye DAPI considerable reductions (61.4% decrease) in parasite numbers in with the dye Alexa Fluor 594 conjugated with the secondary Ab macrophages treated with Tollip siRNA at 48 h postinfection. used for these transcription factors. Therefore, the binding of Intracellular amastigote multiplication was observed to decrease Nrf2, Egr2, and Ahr to the promoter region was also corroborated by 83.2% in infected macrophages pretreated with Tollip plus with their increased nuclear translocation after L. donovani in- by guest on September 30, 2021 IRAK-M siRNA. Collectively, these results indicated that fection. To further reconfirm the fact that transcriptional activation L. donovani induced Tollip along with IRAK-M to attenuate IL-1R1/ of Tollip was under the control of Egr2, Nrf2, and Ahr in L. TLR signaling in infected phagocytic cells and that Tollip does not donovani–infected macrophages, we used siRNA-mediated si- act interchangeably with IRAK-M but may function as a suppressor lencing of all of these transcription factors. We observed that in a co-operative manner. knockdown of Egr2, Ahr, and Nrf2 significantly reduced the ex- pression of Tollip in infected macrophages in comparison with Transcriptional activation of Tollip is regulated by Egr2, Ahr, infected macrophages treated with control siRNA (Fig. 4G). Our and Nrf2 in L. donovani–infected macrophages studies, therefore, clearly suggest that transcription factors such as Because L. donovani infection markedly induced Tollip at both the Egr2, Ahr, and Nrf2 are specifically recruited on the Tollip pro- mRNA and protein levels, we therefore sought to determine the moter and thus exploited by L. donovani for induction of Tollip transcriptional machinery behind these in infected macrophages. expression. To our knowledge, there is no report to date that depicts tran- scriptional regulation of Tollip and identifies the putative tran- L. donovani amastigotes induce Tollip for its survival by scription factors that could bind to the Tollip promoter. To this attenuating TLR/IL-1R1 signaling in macrophages end, we used the Champion ChIP Transcription Factor Search The amastigote stage of the Leishmania parasite represents the Portal based on SABiosciences’ proprietary database known as mammalian form that can reside long term within the hostile DECODE (for more information, see http://saweb2.sabiosciences. environment of macrophages and is responsible for causing com/chipqpcrsearch.php?app=TFBS). The program predicted the chronic infections. Because amastigotes are the disease-causing hypothetical binding sites for the most relevant transcription fac- intramacrophage form of the parasites and also more physiologi- tors such as Egr2, Nrf2, and Ahr to the Tollip promoter, which was cally relevant in context of long-term infection, we next sought to found between 300 bases upstream of the transcription start site identify whether L. donovani amastigotes could also modulate (TSS; chr7:141902407–14190270) and downstream of TSS for TLR/IL-1R1 signaling events in macrophages by exploiting Rfx1 (Fig. 4A). To determine whether the putative Egr2, Nrf2, Tollip. We observed that Tollip transcripts and protein were sig- Rfx1, and Ahr binding sites of the Tollip promoter can actually nificantly enhanced in amastigote-infected macrophages in a time- bind with all of these proteins, we performed ChIP analysis. dependent fashion similar to L. donovani promastigote–infected As shown in Fig. 4B, the infection of J774 macrophages with macrophages (Fig. 5A, 5B). However, higher mRNA induction of L. donovani increased the binding of Egr2, Ahr, and Nrf2 on the Tollip at late time points (12.6-, 11.8-, and 10.6-fold at 16, 24, and Tollip promoter. We observed that the PCR-amplified DNA 48 h postinfection, respectively) was observed when amastigotes bands in the endogenous Tollip promoter were detected in the were used as compared with promastigote-infected macrophages anti-Egr2, anti-Ahr, and anti-Nrf2 Ab immunoprecipitates of (Fig. 5A). We next assessed whether Tollip deficiency in hosts The Journal of Immunology 965 Downloaded from http://www.jimmunol.org/ by guest on September 30, 2021

FIGURE 4. Analysis of in vivo interaction of transcription factors Egr2, Ahr, and Nrf2 with the Tollip promoter in J774 macrophages. (A) Schematic representation of the murine Tollip promoter region and binding sites for Egr2, Ahr, Rfx1, and Nrf2 near the TSS of the Tollip promoter were identified by SABiosciences’ database known as DECODE. Binding sites for Egr2 (2154 to 2165), Ahr (2110 to 2127), and Nrf2 (293 to 2103) are located ∼200 bp upstream from the TSS, and the Rfx1 binding site is located downstream from the TSS (+91 to +108). (B–E) A ChIP assay was performed to observe the binding of Egr2, Nrf2, and Ahr to the Tollip promoter in control and L. donovani–infected macrophages at the indicated time points. Binding of Egr2 (C), Ahr (D), and Nrf2 (E) on the Tollip promoter was measured and quantified by both semiquantitative PCR (B) and real-time (Figure legend continues) 966 TOLLIP INHIBITS TLR/IL-1R1 SIGNALING IN VL affected the secretion of defensive proinflammatory cytokines in Role of Tollip in controlling in vivo parasitemia and host amastigote-infected macrophages. As compared with the unin- defensive transcription factors and cytokine induction fected control, infection with amastigotes slightly downregulated The above in vitro observations suggesting a crucial role of Tollip IL-6 and TNF-a, although the variation was not significant in L. donovani–induced inhibition of macrophage IL-1R1/TLR (Fig. 5C). However, siRNA-mediated knockdown of Tollip signaling and led us to investigate its role in the disease pro- markedly restored both IL-6 and TNF-a levels in macrophages gression of VL in the BALB/c mouse model (Fig. 6A). To this coincubated with amastigotes and cognate ligands for TLR2, end, we first checked expression of Tollip at the protein level by TLR4, and IL-1R1 (LPG, LPS, and IL-1b, respectively) immunoblot analysis in splenocytes isolated from control and (Fig. 5C).This was also similar to what we observed in the case L. donovani–infected BALB/c mice at different time points. In of promastigote infection. Next, we sought to determine the agreement with our in vitro data, splenocytes from L. donovani– nuclear translocation and activation of NF-kB, and likewise infected mice showed augmented expression levels of Tollip in promastigotes. In this study, we also observed restoration of infected BALB/c mice (maximum induction at fourth week cytosolic-to-nuclear translocation of the NF-kB p65 subunit in postinfection), whereas we observed basal expression levels of Tollip-siRNA–treated amastigote-infected macrophages, which Tollip in uninfected controls without any induction (Fig. 6B). To was even higher in Tollip plus IRAK-M siRNA–treated infected further validate the role of Tollip in disease progression, vivo- macrophages (Fig. 5D). The induction of NF-kB transcriptional morpholino–mediated knockdown of Tollip in infected BALB/c activity was further observed in a luciferase-based reporter assay. mice was carried out. To this end, BALB/c mice were in- After knockdown of Tollip and Tollip plus IRAK-M, we ob-

jected with control or Tollip-specific vivo-morpholino at a dose of Downloaded from served a 7.24- and 13.62-fold increase in NF-kB luciferase ac- 5 mg/kg per week up to 7 wk starting at 1 wk prior to infection tivity, respectively, in amastigote-infected macrophages when (Fig. 6A). To confirm the knockdown efficiency, the expression of compared with control siRNA-treated amastigote-infected mac- Tollip at the protein level was observed in control and Tollip- rophages (Fig. 5E). This was even higher than its promastigote morpholino–treated BALB/c mice at the second, fourth, and sixth counterpart (5.25- and 9.58-fold induction in Tollip- and Tollip weeks postinfection. Western blot analysis confirmed reduced plus IRAK-M siRNA–treated infected macrophages, respec- expression of Tollip at the protein level in Tollip-morpholino– http://www.jimmunol.org/ tively, when compared with control siRNA-treated promastigote- treated mice as compared with infected mice treated with control infected macrophages, Fig. 3G). We next addressed the impact vivo-morpholino at all time points (Fig. 6C). Additionally, we also of L. donovani amastigote infection in inducing transcrip- observed that administration of vivo-morpholino showed no ad- tional machinery behind Tollip expression. In comparison with verse effects on health of infected mice. We next determined the L. donovani promastigote-infected macrophages, L. donovani effect of Tollip knockdown in controlling spleen and liver parasite amastigote infection induced much enhanced recruitment of burden in infected BALB/c mice and observed a maximum re- transcription factors such as Egr2, Ahr, and Nrf2 on the Tollip duction in hepatic (65.4%, Fig. 6D) and splenic (68.6%, Fig. 6E) promoter at 6 h, and further binding of these transcription factors parasitic burden at the sixth week postinfection in infected mice reached a maximum level between 12 and 24 h in amastigote- treated with Tollip vivo-morpholino as compared with infected by guest on September 30, 2021 infected macrophages (Fig. 5F). Furthermore, knockdown of mice treated with control vivo-morpholino. These observations Egr2, Ahr, and Nrf2 using specific siRNA revealed significantly strongly argue that Tollip induction in vivo is essential for suc- reduced Tollip expression at the protein level in amastigote- cessful manipulation of host immune response and favors survival infected macrophages (Fig. 5G). Collectively, these results sug- of L. donovani. During the course of VL, the fate of infection is gested that both L. donovani promastigotes and amastigotes used determined by interplay between Th1/Th2 cytokine balance, and similar transcriptional induction machinery behind Tollip ex- inhibition of Th1 cytokines is a characteristic mark of disease pression, but with different time kinetics. Finally, knockdown of progression. In this context, we observed that treatment of infected Tollip markedly reduced parasitemia in amastigote-infected mice with Tollip vivo-morpholino resulted in an increased level of macrophages. We observed significantly reduced amastigote Th1 polarizing cytokine IL-12 (4.5-fold, induction as compared burden in macrophages pretreated with Tollip siRNA (66.2% with infected mice treated with control vivo-morpholino) and decrease) or Tollip plus IRAK-M siRNA (87.6% decrease) TNF-a (6.4-fold induction as compared with infected mice treated (Fig. 5H), and these decreases in parasitemia are slightly greater with control vivo-morpholino) at the fourth week postinfection than those in promastigote-infected macrophages (Fig. 3J) that (Fig. 6F, 6G). The expression of proinflammatory cytokines such underwent similar siRNA treatment. Taken together, our findings as IL-12 and TNF-a is tightly regulated by the transcription factor indicate that L. donovani amastigotes, the disease-causing NF-kB (34, 35). Therefore, we next examined whether knock- intramacrophage form of the parasite, neutralize TLR/IL-1R1– down of Tollip could enhance DNA binding activity of NF-kB mediated host defense by exploiting Tollip to establish long-term in vivo that in turn upregulates proinflammatory cytokines in chronic infection. Tollip-morpholino–treated infected BALB/c mice. Interestingly,

PCR (C–E). (B) Lane 1 represents the PCR amplification of input DNA in each sample for corresponding treatment. Lane 2 is the PCR results for immunoprecipitated samples with the IgG Ab. Lane 3, 4, and 5 are the PCR amplifications of target sequences for the Tollip promoter in immunopre- cipitated chromatin fractions with anti-Egr2, anti-Ahr, and anti-Nrf2 Abs. Data were normalized to input DNA, and nonimmune IgG was used as the negative control. (F) J774 macrophages were infected with L. donovani promastigotes for 12 or 24 h. Cells were stained with anti-Egr2, anti-Nrf2, and anti- Ahr primary Ab followed by incubation with secondary Alexa Fluor 594–conjugated Ab. Nuclei were stained with DAPI, and cells were analyzed under a confocal microscope. (G) J774 macrophages were transfected with scrambled siRNA oligonucleotides or Egr2-, Ahr-, and Nrf2-specific siRNA (24 h) and then infected with L. donovani promastigotes for 16 h. Cell lysates were analyzed for Tollip expression by immunoblotting. b-Actin served as the loading control. The results shown are representative of at least three individual experiments. The significance between different experimental groups was cal- culated by a one-way ANOVA followed by a Tukey posttest using GraphPad Prism 5. Significance was determined for infected versus noninfected groups. *p , 0.05, ***p , 0.001. The Journal of Immunology 967 Downloaded from http://www.jimmunol.org/

FIGURE 5. Intracellular amastigotes attenuate IL-1R1/TLR signaling by exploiting Tollip. (A and B) J774 macrophages were infected with L. donovani amastigotes isolated form spleen of BALB/c mice for the indicated time periods and the expression level of Tollip at the mRNA and protein levels was quantified using real-time PCR (A) and immunoblot analysis (B), respectively. (C) J774 macrophages were transfected with Tollip or scrambled control by guest on September 30, 2021 siRNA for 24 h, followed by treatment with L. donovani amastigotes plus IL-1b, L. donovani amastigotes plus LPS, and L. donovani amastigotes plus LPG. Concentrations of IL-6 and TNF-a in the cell-free culture supernatants were detected by ELISA. (D) J774 macrophages were transfected with control siRNA, Tollip siRNA, IRAK-M siRNA, or Tollip plus IRAK-M siRNA (24 h) and then infected with L. donovani amastigotes for 24 h. Nuclear and cytosolic extracts were prepared and lysates were subjected to immunoblot to detect nuclear and cytosolic expression levels of NF-kB p65. Histone H3 and GAPDH were used as nuclear and cytosolic loading controls, respectively. (E) J774 macrophages were transfected with control siRNA, Tollip siRNA, IRAK-M siRNA, or Tollip plus IRAK-M siRNA along with pGL4.32 (luc2P/NF-kB-RE/Hygro) and Renilla luciferase (pRL-TK–Renilla luciferase) vector. After 24 h of transfection, cells were infected with L. donovani amastigotes for 24 h. Cells were lysed and luciferase activity was measured as described in Materials and Methods.(F) A ChIP assay was performed to observe the binding of Egr2, Nrf2, and Ahr to the Tollip promoter in control and L. donovani amastigote–infected macrophages at indicated time points. The binding of Egr2, Ahr, and Nrf2 on the Tollip promoter was measured and quantified by semiquantitative PCR. (G) J774 macrophages were transfected with control siRNA oligonucleotides or Egr2-, Ahr-, and Nrf2-specific siRNA (24 h) and then infected with L. donovani amastigotes for 16 h. Cell lysates were analyzed for Tollip expression by immunoblotting. b-Actin served as the loading control. (H) J774 macrophages were transfected with control siRNA, Tollip siRNA, IRAK-M siRNA, and Tollip plus IRAK-M siRNA followed by infection with amastigotes for 48 h. Intracellular amastigotes number was assessed by Giemsa staining. The significance between different experimental groups was calculated by one-way ANOVA followed by a Tukey posttest using GraphPad Prism 5. Significance was determined for control siRNA versus Tollip siRNA, IRAK-M siRNA, and Tollip plus IRAK-M siRNA. **p , 0.01, ***p , 0.001. ns, not significant. silencing of Tollip restored the DNA binding activity of NF-kBin vivo-morpholino whereas we did not observed any binding of both infected BALB/c mice, as much higher DNA binding was ob- STAT-1 and IRF-1 in control vivo-morpholino–treated infected served at the second, fourth, and sixth weeks postinfection in mice. These results suggested that strong upregulation of Tollip the splenocytes of Tollip-morpholino–treated BALB/c mice as in vivo might be associated with inhibition of NF-kB, STAT-1, compared with control vivo-morpholino–treated BALB/c mice IRF-1, and host defensive proinflammatory cytokine synthesis that (Fig. 6I). Moreover, in comparison with mice treated with control in turn facilitated parasite survival in the BALB/c mice model of vivo-morpholino, splenocytes isolated from mice treated with experimental VL. Tollip-morpholino were able to produce significant amounts of host-protective IFN-g (9.1-fold) (Fig. 6H). As transcription factors Discussion STAT-1 and IRF-1 were involved in IFN-g–mediated induction of L. donovani, an intracellular obligate pathogen and causative host immunity, we next observed their activation in the spleno- agent of VL, has evolved several sophisticated strategies to create cytes of control and Tollip-morpholino–treated mice by EMSA. its niche by manipulation of host cell functions and immune de- As shown in Fig. 6J and 6K, much higher DNA binding of STAT-1 fense mechanisms, and to achieve this, the parasite neutralizes (Fig. 6J) and IRF-1 (Fig. 6K) was observed in nuclear extracts robust macrophage activation machinery and dampens the host- isolated from splenocytes of infected mice treated with Tollip protective proinflammatory response. TLR/IL-1R1 superfamily 968 TOLLIP INHIBITS TLR/IL-1R1 SIGNALING IN VL Downloaded from http://www.jimmunol.org/

FIGURE 6. Effect of in vivo knockdown of Tollip on parasite survival and cytokine response. (A) Schematic representation of the experimental

protocol for vivo-morpholino–mediated translation blocking of Tollip in BALB/c mice to determine parasite burden and cellular immune response. (B) by guest on September 30, 2021 BALB/c mice were infected with stationary phase promastigotes of L. donovani through the tail vein, and splenocyte lysates were analyzed for the expression of Tollip by immunoblotting at the time periods indicated. Equal protein loading was verified by probing with b-actin Ab. (C) In vivo knockdown of Tollip was achieved by administering vivo-morpholino against Tollip as indicated in Materials and Methods. Expression of Tollip in splenocytes of control or Tollip-morpholino–treated infected BALB/c mice was evaluated by immunoblotting at the indicated time periods. (D and E) Hepatic (D)andsplenic(E) parasite burden was determined weekly from different groups of infected mice and expressed as the LDU 6 SD for five mice per group. (F–H) Levels of IL-12 (F), TNF-a (G), and IFN-g (H) in culture supernatants isolated from splenocytes of different experimental group of mice at the second, fourth, and sixth week postinfection were determined by ELISA. (I–K) Cy3-labeled double-stranded oligonucleotide probes for NF-kB(I), STAT-1 (J), and IRF-1 (K) were incubated with nuclear extracts prepared from splenocytes of different groups of infected mice as indicated above, and EMSA was performed to analyze the DNA binding activity. To confirm specificity of the interaction, unlabeled specific probes were added as competitors to EMSA binding reactions before adding the Cy3-labeled probe. The significance between two or different experimental groups was calculated by a one-way ANOVA followed by a Tukey posttest using GraphPad Prism 5. Significance was determined for control vivo-morpholino versus Tollip-vivo morpholino. **p , 0.01, ***p , 0.001. members play a fundamental role in defending against pathogenic yeast two-hybrid screen using the IL-1R accessory protein (14). microbial infection through the induction of inflammatory cyto- Subsequent studies showed that Tollip was found to be associated kines and type I IFNs in macrophages. TLR/IL-1R signaling is with the TIR domain of the IL-1R accessory protein, TLR2 and required for regulation of innate inflammatory and immune re- TLR4. Overexpression of Tollip impairs IL-1RI–, TLR2–, and sponses to infections, injury, stress, and allergies (23); however, it TLR4-stimulated MAPK and NF-kB pathways crucial for inflam- should be tightly controlled, as aberrant activation of these path- mation and host defense (15). However, the role of Tollip in the ways promotes the onset of inflammatory diseases, autoimmune context of immunosuppression associated with parasitic diseases is diseases, and cancers (36). Several intracellular negative regulators poorly understood. In our present findings, we provided substantial of different TLR signaling have been well reported, such as IRAK- evidence that L. donovani exploited the multitasking function of M, short MyD88, SOCS-1, SIGIRR, ST-2, A20, and Triad3A (37), Tollip to inhibit TLR/IL-1R signaling–mediated protective immu- that are very often induced by TLR ligands in a negative feedback nity during experimental VL. Our results clearly demonstrated that manner to maintain the magnitude and duration of inflammatory functional knockdown of Tollip could significantly restore protein– response. However, these negative regulators are very often protein integrity of TLR/IL-1R signaling that ultimately induced exploited by successful intracellular pathogens such as Leishmania, host-protective cytokine responses and reduced parasitemia in both Toxoplasma gondii,andTrypanosoma cruzi to manipulate host in vitro and in vivo models of VL. immune responses and to maintain a long-term immunosuppression Tollip’s role as an endocytic adaptor protein is already well (38). One such crucial intracellular negative regulator of TLR/IL-1R documented, and structural analysis revealed that it has three signaling is Tollip, which was originally identified through a distinct domains, that is, the N-terminal Tom1-binding domain The Journal of Immunology 969

(TBD), the conserved 2 domain, and the CUE domain, which is markedly elevated the expression of Tollip and inhibited proin- involved in protein sorting by association with Tom1, ubiquitin, flammatory cytokine synthesis in bone marrow–derived DCs and clathrin (39). The conserved 2 domain of Tollip enables it to (BMDCs) cultured in the presence of LPS (44). Tolerance in- bind specifically to phospholipids and shows a broad preference duction in mouse BMDCs with Staphylococcus aureus lipo- for phosphoinositides, thereby promoting Tollip localization with teichoic acid significantly suppressed proinflammatory cytokine cellular membranes rich in phospholipid such as cell membrane, (TNF-a and IL-6) secretion in BMDCs along with sustained ex- endosome, and lysosome. Evidences suggested that Tollip may pression of Tollip (45). These observations, therefore, suggested a aggregate at cellular and/or lysosome membranes with IL-1R1 and proinflammatory phenotype of ex vivo–cultured DCs that are TLR4 following high doses of LPS stimulation, contributing to the devoid of Tollip. Moreover, the mechanism of CD4+ T activation inhibition of the TLR4-mediated immune response via the CUE in the presence of various types of DCs such as BMDCs, domain (40). Being a multitasking regulator, Tollip not only monocyte-derived human DCs, or skin-derived mouse Langerhans participated in trafficking and endosomal sorting of receptors, but cells (46) is already well established. In our case, therefore, it is also in modulation of IL-1R/TLR/NF-kB/MAPK signaling, quite possible that in vivo IL-12 or IFN-g synthesis in Tollip- autophagic clearance of protein aggregates, bacterial entry, and morpholino–treated infected mice might be due to crosstalk be- regulation of sumoylation (41). In the current study, we therefore tween DCs and T cells, as they are the main producers of IL-12 hypothesized that innate immune receptors belonging to the and IFN-g, and this warrants further detailed investigation. IL-1R/TLR superfamily might be internalized into lysosomal Moreover, a wealth of evidences suggested that Leishmania has compartments by Tollip for degradation in L. donovani–infected developed a range of strategies to efficiently neutralize or subvert macrophages. In our case, we clearly observed specific down- macrophage microbicidal machinery, and immunosuppression is Downloaded from regulation of IL-1R1 but not TLR2 or TLR4 in infected macro- accomplished by a range of Leishmania-derived virulence factors. phages, and Tollip played a crucial role in endocytic trafficking of Therefore, in the future more detailed biochemical approaches IL-1R1 for its degradation. In our study, active participation of will hopefully provide insights into the putative role of specific Tollip in the recognition and trafficking of cell surface IL-1R1 Leishmania-derived virulence factors. Overall, our results pro- within endocytic pathways during experimental VL was con- vided additional findings and prompted us to conclude that suc-

firmed by two complementary approaches. First, we observed cessful survival and propagation of L. donovani inside the host http://www.jimmunol.org/ strong association of Tollip with cell surface IL-1R1 during early might be mediated by exploiting multiple subversion strategies hours of infection by coimmunoprecipitation studies, and silenc- where Tollip acts as a crucial negative regulator of early IL-1R1 ing of Tollip by siRNA markedly restored IL-1R1 protein levels in and late TLR signaling. LPS-stimulated infected macrophages. Second, confocal analysis clearly showed that trafficking of IL-1R1 and Tollip to early and Acknowledgments late endosomal compartments occurred at the same time points We thank the Director of the Central Drug Research Institute (Lucknow, (6 and 9 h postinfection, respectively) and Tollip deficiency India) for providing research facilities and encouragement. markedly impaired IL-1R1 trafficking and degradation, as we observed that IL-1R1 was widely distributed on the cell surface of Disclosures by guest on September 30, 2021 L. donovani–infected macrophages treated with Tollip siRNA. The authors have no financial conflicts of interest. However, different negative regulatory protein such as A20 (3) and IRAK-M (2) of the TLR signaling pathway have been im- plicated in promoting Leishmania parasite survival and disease References progression. Nonetheless, we cannot exclude the possibility that 1. Mogensen, T. H. 2009. Pathogen recognition and inflammatory signaling in innate immune defenses. Clin. Microbiol. Rev. 22: 240–273. infection with successful invading parasites such as L. donovani 2. Srivastav, S., A. Saha, J. Barua, A. Ukil, and P. K. Das. 2015. IRAK-M regulates might also exploit other host negative regulatory proteins of IL- the inhibition of TLR-mediated macrophage immune response during late 1R1/TLR signaling. This is because starting from the cell surface in vitro Leishmania donovani infection. Eur. J. Immunol. 45: 2787–2797. 3. Srivastav, S., S. Kar, A. G. Chande, R. Mukhopadhyaya, and P. K. Das. 2012. receptor to the nuclear transcription factor, there are arrays of Leishmania donovani exploits host deubiquitinating enzyme A20, a negative second messengers, adaptor proteins, kinases, and transcription regulator of TLR signaling, to subvert host immune response. J. Immunol. 189: factors that are tightly regulated by several negative regulators that 924–934. 4. Gupta, P., J. Giri, S. Srivastav, A. G. Chande, R. Mukhopadhyaya, P. K. Das, and act at a different level to control the magnitude of innate immune A. Ukil. 2014. Leishmania donovani targets tumor necrosis factor receptor- responses downstream of the IL-1R/TLR pathway (42). We fur- associated factor (TRAF) 3 for impairing TLR4-mediated host response. ther validated the functional role of Tollip in an in vivo model of FASEB J. 28: 1756–1768. 5. Kar, S., A. Ukil, G. Sharma, and P. K. Das. 2010. MAPK-directed phosphatases experimental VL by morpholino-mediated knockdown in infected preferentially regulate pro- and anti-inflammatory cytokines in experimental BALB/c mice. Morpholino-mediated silencing of Tollip signifi- visceral leishmaniasis: involvement of distinct protein kinase C isoforms. J. cantly suppressed organ parasite burden along with strong in- Leukoc. Biol. 88: 9–20. 6. Juffermans, N. P., S. Florquin, L. Camoglio, A. Verbon, A. H. Kolk, duction of host-protective IL-12 and IFN-g synthesis in the P. Speelman, and S. J. H. Van Deventer. 2000. Interleukin-1 signaling is es- splenocytes of infected BALB/c mice. However, these cytokine sential for host defense during murine pulmonary tuberculosis. J Infect Dis. 182: 902–908. responses (IL-12 and IFN-g) in vivo might be due to the absence 7. Biondo, C., G. Mancuso, A. Midiri, G. Signorino, M. Domina, V. Lanza of Tollip in surrounding immune cells, as splenocytes contain Cariccio, M. Venza, I. Venza, G. Teti, and C. Beninati. 2014. Essential role of heterogeneous populations of dendritic cells (DCs), T cells, interleukin-1 signaling in host defenses against group B streptococcus. MBio 5: e01428–14. B cells, and other cells apart from macrophages. In this context, 8. Labow, M., D. Shuster, M. Zetterstrom, P. Nunes, R. Terry, E. B. Cullinan, note that Tollip also acts as a negative regulator of TLR-mediated T. Bartfai, C. Solorzano, L. L. Moldawer, R. Chizzonite, and K. W. McIntyre. inflammatory responses in DCs in both ex vivo cultured conditions 1997. Absence of IL-1 signaling and reduced inflammatory response in IL-1 type I receptor-deficient mice. J. Immunol. 159: 2452–2461. as well as in vivo. For instances, in vivo LPS exposure upregulated 9. Lang, T., and A. Mansell. 2007. The negative regulation of Toll-like receptor and the expression of Tollip and other TLR negative regulatory pro- associated pathways. Immunol. Cell Biol. 85: 425–434. teins in Peyer’s patch DCs that is unable to secrete IL-12p70, 10. Sorkin, A., and M. von Zastrow. 2009. Endocytosis and signalling: intertwining molecular networks. Nat. Rev. Mol. Cell Biol. 10: 609–622. TNF-a, and IL-1b when compared with spleen DCs (43). 11. Platta, H. W., and H. Stenmark. 2011. Endocytosis and signaling. Curr. Opin. Epigallocatechin-3-gallate, a major active polyphenol of green tea, Cell Biol. 23: 393–403. 970 TOLLIP INHIBITS TLR/IL-1R1 SIGNALING IN VL

12. Bonifacino, J. S., and A. M. Weissman. 1998. Ubiquitin and the control of 28. Piedrafita, D., L. Proudfoot, A. V. Nikolaev, D. Xu, W. Sands, G. J. Feng, protein fate in the secretory and endocytic pathways. Annu. Rev. Cell Dev. Biol. E. Thomas, J. Brewer, M. A. Ferguson, J. Alexander, and F. Y. Liew. 1999. 14: 19–57. Regulation of macrophage IL-12 synthesis by Leishmania phosphoglycans. Eur. 13. Windheim, M., and B. Hansen. 2014. Interleukin-1-induced activation of the J. Immunol. 29: 235–244. small GTPase Rac1 depends on receptor internalization and regulates gene ex- 29. Chandra, D., and S. Naik. 2008. Leishmania donovani infection down-regulates pression. Cell. Signal. 26: 49–55. TLR2-stimulated IL-12p40 and activates IL-10 in cells of macrophage/ 14. Burns, K., J. Clatworthy, L. Martin, F. Martinon, C. Plumpton, B. Maschera, monocytic lineage by modulating MAPK pathways through a contact- A. Lewis, K. Ray, J. Tschopp, and F. Volpe. 2000. Tollip, a new component of dependent mechanism. Clin. Exp. Immunol. 154: 224–234. the IL-1RI pathway, links IRAK to the IL-1 receptor. Nat. Cell Biol. 2: 346–351. 30. Zhang, G., and S. Ghosh. 2002. Negative regulation of Toll-like receptor- 15. Capelluto, D. G. S. 2012. Tollip: a multitasking protein in innate immunity and mediated signaling by Tollip. J. Biol. Chem. 277: 7059–7065. protein trafficking. Microbes Infect. 14: 140–147. 31. Brissoni, B., L. Agostini, M. Kropf, F. Martinon, V. Swoboda, S. Lippens, 16. Humbert-Claude, M., D. Duc, D. Dwir, L. Thieren, J. Sandstrom von Tobel, ¨ H. Everett, N. Aebi, S. Janssens, E. Meylan, et al. 2006. Intracellular trafficking C. Begka, F. Legueux, D. Velin, M. H. Maillard, K. Q. Do, et al. 2016. Tollip, an of interleukin-1 receptor I requires Tollip. Curr. Biol. 16: 2265–2270. early regulator of the acute inflammatory response in the substantia nigra. J. 32. Didierlaurent, A., B. Brissoni, D. Velin, N. Aebi, A. Tardivel, J. C. Sirard, Neuroinflammation 13: 303. 17. Shah, J. A., J. C. Vary, T. T. Chau, N. D. Bang, N. T. Yen, J. J. Farrar, G. Angelov, and K. Burns. 2006. Tollip regulates proinflammatory responses to S. J. Dunstan, and T. R. Hawn. 2012. Human TOLLIP regulates TLR2 and TLR4 interleukin-1 and lipopolysaccharide. Mol. Cell Biol. 26: 735–742. signaling and its polymorphisms are associated with susceptibility to tubercu- 33. Chen, Z. J. 2005. Ubiquitin signalling in the NF-kappaB pathway. Nat. Cell Biol. losis. J. Immunol. 189: 1737–1746. 7: 758–765. 18. Li, T., J. Hu, and L. Li. 2004. Characterization of Tollip protein upon lipo- 34. Trede, N. S., A. V. Tsytsykova, T. Chatila, A. E. Goldfeld, and R. S. Geha. 1995. polysaccharide challenge. Mol. Immunol. 41: 85–92. Transcriptional activation of the human TNF-alpha promoter by superantigen in 19. de Araujo, F. J., L. D. da Silva, T. G. Mesquita, S. K. Pinheiro, S. Vital Wde, human monocytic cells: role of NF-kappa B. J. Immunol. 155: 902–908. A. Chrusciak-Talhari, J. A. Guerra, S. Talhari, and R. Ramasawmy. 2015. 35. Murphy, T. L., M. G. Cleveland, P. Kulesza, J. Magram, and K. M. Murphy. Polymorphisms in the TOLLIP gene influence susceptibility to cutaneous 1995. Regulation of interleukin 12 p40 expression through an NF-kappa B half- leishmaniasis caused by Leishmania guyanensis in the Amazonas state of Brazil. site. Mol. Cell. Biol. 15: 5258–5267. PLoS Negl. Trop. Dis. 9: e0003875. 36. Luddy, K. A., M. Robertson-Tessi, N. K. Tafreshi, H. Soliman, and D. L. Morse. Downloaded from 20. Shah, J. A., W. R. Berrington, J. C. Vary, Jr., R. D. Wells, G. J. Peterson, 2014. The role of Toll-like receptors in colorectal cancer progression: evidence C. B. Kunwar, S. Khadge, D. A. Hagge, and T. R. Hawn. 2016. Genetic variation for epithelial to leucocytic transition. Front. Immunol. 5: 429. in toll-interacting protein is associated with leprosy susceptibility and cutaneous 37. Liew, F. Y., D. Xu, E. K. Brint, and L. A. J. O’Neill. 2005. Negative regulation of expression of interleukin 1 receptor antagonist. J. Infect. Dis. 213: 1189–1197. Toll-like receptor-mediated immune responses. Nat. Rev. Immunol. 5: 446–458. 21. Ojurongbe, O., R. I. Funwei, T. J. Snyder, I. Farid, N. Aziz, Y. Li, C. O. Falade, 38. Gazzinelli, R. T., and E. Y. Denkers. 2006. Protozoan encounters with Toll-like and B. N. Thomas. 2018. Genetic variants of tumor necrosis factor-a -308G/A receptor signalling pathways: implications for host parasitism. Nat. Rev. (rs1800629) but not Toll-interacting proteins or vitamin D receptor genes en- Immunol. 6: 895–906.

hances susceptibility and severity of malaria infection. Immunogenetics 70: 135– 39. Katoh, Y., Y. Shiba, H. Mitsuhashi, Y. Yanagida, H. Takatsu, and K. Nakayama. http://www.jimmunol.org/ 140. 2004. Tollip and Tom1 form a complex and recruit ubiquitin-conjugated proteins 22. Vishwakarma, P., N. Parmar, P. Chandrakar, T. Sharma, M. Kathuria, onto early endosomes. J. Biol. Chem. 279: 24435–24443. P. K. Agnihotri, M. I. Siddiqi, K. Mitra, and S. Kar. 2018. Ammonium trichloro 40. Kowalski, E. J. A., and L. Li. 2017. Toll-interacting protein in resolving and non- [1,2-ethanediolato-O,O9]-tellurate cures experimental visceral leishmaniasis by resolving inflammation. Front. Immunol. 8: 511. redox modulation of Leishmania donovani trypanothione reductase and inhib- 41. Torun, A., E. Szyma, I. Castanon, and L. Woli. 2015. Endocytic adaptor protein Cell. Mol. Life Sci. iting host integrin linked PI3K/Akt pathway. 75: 563–588. Tollip inhibits canonical Wnt signaling. PLoS One. 10: e0130818. Leishmania 23. Bradley, D. J., and J. Kirkley. 1977. Regulation of populations 42. Newton, K., and V. M. Dixit. 2012. Signaling in innate immunity and inflam- within the host. I. the variable course of Leishmania donovani infections in mice. mation. Cold Spring Harb. Perspect. Biol. 4: a006049. Clin. Exp. Immunol. 30: 119–129. 43. Davies, J. M., J. MacSharry, and F. Shanahan. 2010. Differential regulation of 24. Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expression Toll-like receptor signalling in spleen and Peyer’s patch dendritic cells. Immu- data using real-time quantitative PCR and the 2(2Delta Delta C(T)) method. nology 131: 438–448.

Methods 25: 402–408. by guest on September 30, 2021 25. Haglund, R. E., and L. I. Rothblum. 1987. Isolation, fractionation and recon- 44. Byun, E.-B., H.-G. Choi, N.-Y. Sung, and E.-H. Byun. 2012. Green tea poly- stitution of a nuclear extract capable of transcribing ribosomal DNA. Mol. Cell. phenol epigallocatechin-3-gallate inhibits TLR4 signaling through the 67-kDa Biochem. 73: 11–20. laminin receptor on lipopolysaccharide-stimulated dendritic cells. Biochem. 26. Shivahare, R., P. Vishwakarma, N. Parmar, P. K. Yadav, W. Haq, M. Srivastava, Biophys. Res. Commun. 426: 480–485. S. Gupta, and S. Kar. 2014. Combination of liposomal CpG oligodeox- 45. Yoon, H. S., G. Kim, Y. J. Ju, I. S. Cheon, S. W. Hong, D. W. Kim, B. C. Park, ynucleotide 2006 and miltefosine induces strong cell-mediated immunity during S. H. Han, and C. H. Yun. 2017. Distinct pattern of immune tolerance in den- experimental visceral leishmaniasis. PLoS One 9: e94596. dritic cells treated with lipopolysaccharide or lipoteichoic acid. Mol. Immunol. 27. Descoteaux, A., S. J. Turco, D. L. Sacks, and G. Matlashewski. 1991. Leish- 91: 57–64. mania donovani lipophosphoglycan selectively inhibits signal transduction in 46. Soong, L. 2008. Modulation of dendritic cell function by Leishmania parasites. macrophages. J. Immunol. 146: 2747–2753. J. Immunol. 180: 4355–4360.