Discovery of the DIGIRR Gene from Teleost Fish: A Novel Toll−IL-1 Receptor Family Member Serving as a Negative Regulator of IL-1 Signaling This information is current as of October 1, 2021. Yi-feng Gu, Yu Fang, Yang Jin, Wei-ren Dong, Li-xin Xiang and Jian-zhong Shao J Immunol 2011; 187:2514-2530; Prepublished online 29 July 2011;

doi: 10.4049/jimmunol.1003457 Downloaded from http://www.jimmunol.org/content/187/5/2514

Supplementary http://www.jimmunol.org/content/suppl/2011/07/29/jimmunol.100345 Material 7.DC1 http://www.jimmunol.org/ References This article cites 87 articles, 26 of which you can access for free at: http://www.jimmunol.org/content/187/5/2514.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on October 1, 2021

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

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

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

Discovery of the DIGIRR Gene from Teleost Fish: A Novel Toll–IL-1 Receptor Family Member Serving as a Negative Regulator of IL-1 Signaling

Yi-feng Gu, Yu Fang, Yang Jin, Wei-ren Dong, Li-xin Xiang, and Jian-zhong Shao

Toll–IL-1R (TIR) family members play crucial roles in a variety of defense, inflammatory, injury, and stress responses. Although they have been widely investigated in mammals, little is known about TIRs in ancient vertebrates. In this study, we report a novel double Ig IL-1R related molecule (DIGIRR) from three model fish (Tetraodon nigroviridis, Gasterosteus aculeatus, and Takifugu rubripes), adding a previously unknown homolog to the TIR family. This DIGIRR molecule contains two Ig-like domains in the extracellular region, one Arg-Tyr–mutated TIR domain in the intracellular region, and a unique subcellular distribution within the Golgi apparatus. These characteristics distinguish DIGIRR from other known family members. In vitro injection of DIGIRR into zebrafish embryos dramatically inhibited LPS-induced and IL-1b–induced NF-kB activation. Moreover, in vivo knockdown Downloaded from of DIGIRR by small interfering RNA significantly promoted the expression of IL-1b–stimulated proinflammatory cytokines (IL-6 and IL-1b) in DIGIRR-silenced liver and kidney tissues and in leukocytes. These results strongly suggest that DIGIRR is an important negative regulator of LPS-mediated and IL-1b–mediated signaling pathways and inflammatory responses. The Arg- Tyr–mutated site disrupted the signal transduction ability of DIGIRR TIR. Evolutionally, we propose a hypothesis that DIGIRR and single Ig IL-1R related molecule (SIGIRR) might originate from a common ancient IL-1R–like molecule that lost one (in

DIGIRR) or two (in SIGIRR) extracellular Ig-like domains and intracellular Ser and Arg-Tyr amino acids. DIGIRR might be an http://www.jimmunol.org/ evolutionary “transitional molecule” between IL-1R and SIGIRR, representing a shift from a potent receptor to a negative regulator. These results help define the evolutionary history of TIR family members and their associated signaling pathways and mechanisms. The Journal of Immunology, 2011, 187: 2514–2530.

he Toll–IL-1R (TIR) superfamily encompasses a group of presence of microorganisms and activate a complex, multifaceted structurally homologous proteins that are essential for the cellular defense response (9–14). The Ig domain-containing sub- T initiation of innate host defenses against infection, aller- group includes the IL-1R type I (IL-1RI) (15) and its accessory gic and nonallergic inflammation, injury, and stress (1–4). Mem- protein (IL-1RAcP) (16), the IL-18Rs (IL-18Ra and IL-18Rb) by guest on October 1, 2021 bers are distinguished by the presence of a conserved intracellular (17, 18), the regulatory receptors T1/ST2 (19–21), and a number TIR domain that signals to the NF-kB, MAPK, and AP-1 path- of orphan receptors, including three Ig IL-1R related molecule-1 ways (5–7). This superfamily can be further divided into two and molecule-2 (TIGIRR-1/2) (22), the IL-1R related protein subgroups on the basis of its extracellular domains: the leucine- 2 (IL-1Rrp2) (23), and the single Ig IL-1R related molecule rich repeats receptors, such as TLRs (1), and the Ig domain- (SIGIRR) (24–27). The pleiotropic cytokine IL-1 is known to be containing receptors (8). The TLRs function as sensors for the involved in a variety of immune activities, such as T cell and B cell proliferation and differentiation, as well as IL-2 and IL-6 College of Life Sciences, Zhejiang University, Hangzhou 310058, People’s Republic production (28, 29). When agonist IL-1 binds to the IL-1R com- of China; Key Laboratory for Cell and Gene Engineering of Zhejiang Province, plex, it initiates an amplification cascade of innate resistance, Hangzhou 310058, People’s Republic of China; and Key Laboratory of Animal contributes to the activation of adaptive immunity, and modulates Epidemic Etiology and Immunology Prevention of the Ministry of Agriculture, Hangzhou 310058, People’s Republic of China inflammatory reactions (29, 30). Received for publication October 20, 2010. Accepted for publication June 20, 2011. A series of negative regulators of IL-1 signaling pathways has been identified, including IL-1R–associated kinase (IRAK)-M This work was supported by grants from the National Basic Research Program of China (973) (2006CB101805), the Hi-Tech Research and Development Program of (31), MyD88s (32), SOCS-1 (33), type II IL-1R (IL-1RII) (34, China (863) (2008AA09Z409), the National Natural Science Foundation of China 35), T1/ST2 (36), and SIGIRR (24–27), which suppress IL-1 (30871936, 31072234), and the Science and Technology Foundation of Zhejiang Province (2007C12011). signaling pathways through different mechanisms. Among these inhibitors, IL-1RII, T1/ST2, and SIGIRR are members of the IL- The data presented in this article have been submitted to GenBank under accession numbers EF095151, EU305619, and EU360722. 1R family. In addition, both IL-1RII and T1/ST2 receptors contain Address correspondence and reprint requests to Prof. Jian-zhong Shao and Dr. Li-xin three Ig-like domains, whereas SIGIRR contains a single Ig-like Xiang, Zhejiang University, Hangzhou 310058, Zhejiang, People’s Republic of domain. An intriguing issue is whether double Ig domain recep- China. E-mail addresses: [email protected] and [email protected] tors exist and whether they act as positive or negative regulators. The online version of this article contains supplemental material. Although the TIR superfamily has been widely studied in Abbreviations used in this article: DIG, digoxigenin; DIGIRR, double Ig IL-1R re- mammals, little is known about its existence, occurrence, and lated molecule; EST, expressed sequence tag; IL-1RAcP, IL-1R accessory protein; IL-1RI, type I IL-1R; IL-1RII, type II IL-1R; IL-1Rrp2, IL-1R related protein 2; function in lower vertebrates like fish. In recent years, a number of IRAK, IL-1R–associated kinase; ORF, open reading frame; RNAi, RNA interference; IL-1 family members and TIR-containing receptors, such as IL- SIGIRR, single Ig IL-1R related molecule; siRNA, small interfering RNA; TIGIRR, 1b, IL-18, IL-1RI, IL-1RII, IL-1RAcP, ST2L, and TLR1–TLR23, three Ig IL-1R related molecule; TIR, Toll–IL-1R; UTR, untranslated region. have been identified in fish (37–47), indicating the presence of IL- Copyright Ó 2011 by The American Association of Immunologists, Inc. 0022-1767/11/$16.00 1 and TIR signaling. However, few reports regarding the negative www.jimmunol.org/cgi/doi/10.4049/jimmunol.1003457 The Journal of Immunology 2515 regulators in TIR signaling are available, except for a Tollip ho- boundaries) were elucidated by comparing DIGIRR cDNAs with genome molog recently identified from Oncorhynchus mykiss (48). In- sequences, and illustrations were drawn using GeneMaper 2.5 (http:// vestigating the extent of the negative feedback mechanisms in genemaper.googlepages.com). Selected species included Gallus gallus, T. nigroviridis, G. aculeatus, T. rubripes, Homo sapiens, Mus musculus, ancient vertebrates such as fishes and the evolutionary changes and Rattus norvegicus. The potential functional motifs in the DIGIRR that have occurred in this cytokine regulatory network during protein were analyzed by PROSITE (50), SMART database (51), and the vertebrate evolution would be intriguing. TMHMM (http://www.cbs.dtu.dk/services/TMHMM-2.0/) program (52, In the current study, the TIRs were identified in three fish species, 53). The signal peptidase cleavage sites were predicted with SignalP 3.0 (http://www.cbs.dtu.dk/services) (54). BLAST was used to examine the Tetraodon nigroviridis, Gasterosteus aculeatus,andTakifugu deduced amino acid sequence against the SwissProt protein database rubripes, and a novel family member, named double Ig IL-1R (http://www.expasy.org/tools/blast/) (55). Percentage amino acid sequence related molecule (DIGIRR), was successfully characterized. Phy- identity was calculated using the MEGALIGN program from DNASTAR, logenetic analysis revealed that fish DIGIRR was homologous and a multiple alignment was generated using the ClustalW program to SIGIRR, but distinct in both protein structure and subcellu- (version 1.83) (56). lar localization. In vivo and in vitro functional characterization Southern blot analysis demonstrated that DIGIRR served as a negative regulator of IL-1– mediated signaling. Importantly, our results also suggest that Genomic DNA was prepared from the Tetraodon head kidney according to a previously reported method (57). A digoxigenin (DIG)-labeled probe was DIGIRR appears to be an evolutionary “transitional molecule” prepared according to the intracellular domain of the DIGIRR gene. The between IL-1R and SIGIRR and represents a link between the DNA fragment was amplified using primers DIGIRR-P-F and DIGIRR-P- potent receptors and the negative regulators. Our aim is that the R (Table I) from genomic DNA and then purified using a Gel Extraction current study will contribute to the understanding of the evolu- Kit (Qiagen). The probe was randomly labeled with digoxigenin-11- deoxyuridine triphosphate according to the protocol of DIG High Prime Downloaded from tionary history of the TIR superfamily and its associated signaling (Roche). Southern blot analysis was performed following a protocol pre- pathways in both lower and higher vertebrates as a whole. viously described (58). Briefly, the genomic DNA (10–15 mg) was digested overnight at 37˚C using 100 U of the restriction enzymes HindIII, BamHI, Materials and Methods and EcoRI (TaKaRa). Digested DNA was separated by electrophoresis Experimental fish overnight on 0.8% agarose gels, transferred to positively charged nylon membranes (Amersham), and fixed by baking at 120˚C for 1.5 h. After One-year-old pufferfish, T. nigroviridis, of both sexes and weighing ∼4–6 prehybridization in 10 ml DIG Easy Hyb solution at 42˚C for 2–3 h, hy- g, were kept in running water at 25˚C and fed commercial food pellets bridization was carried out at 47˚C overnight in a hybridization oven. http://www.jimmunol.org/ twice a day. The fish were held in the laboratory for at least 2 wk prior to The membrane was washed twice in 23 SSC containing 0.1% SDS at experimental use to allow for acclimatization and evaluation of overall room temperature for 5 min each, and then twice in 0.53 SSC containing health. Only healthy fish, as determined by general appearance and level of 0.1% SDS at 68˚C for 15–20 min each. Genomic DNA, hybridized with activity, were used for these studies. probe, was immunologically detected by anti-DIG Ab conjugated with alkaline phosphatase. The alkaline phosphatase substrate CSPD was used Sequence retrieval in the chemiluminescence assay, and signals were exposed on X-ray film (Kodak) for 5–30 min. Marker Wide Range (500–15,000 bp; TaKaRa) The sequence of DIGIRR was initially searched on the Tetraodon genome was added to the same gel and stained in 13 TBE (0.1 mg/ml; EB) after database (http://www.genoscope.cns.fr) using the human IL-1R amino acid electrophoresis. Band sizes were calculated by 1D Image Analysis Soft- sequence (M27492) as a query. When a Tetraodon homolog was found

ware (Kodak). by guest on October 1, 2021 within a DNA contig (Scaffold 14769) in the database, it was further used as a probe to search homologous sequences on Takifugu and Gasterosteus Phylogenetic analysis EST/genome databases (http://genome.jgi-psf.org/ fugu6/fugu6.home.html and http://www.ensembl.org/Gasterosteus_aculeatus). Aided by various The genomic and expressed sequence tag (EST) databases maintained at the software programs, such as Genscan and BLAST (49), a possible fish National Center for Biotechnology Information (http://www.ncbi.nlm.nih. homolog of human SIGIRR was compiled, and it was used for the design gov/), Ensembl (http://www.ensembl.org/), UCSC Genome Bioinformatics of primers for the cloning of full-length Tetraodon DIGIRR cDNA. (http://www.genome.ucsc.edu/), and TIGR Gene Indices (http://www.tigr. org/tdb/tgi/) were used to search for IL-1R homologs in other species of Molecular cloning primate, rodent, bird, and teleost fish. A phylogenetic tree was constructed Fish were sacrificed after anesthesia, and total RNA was isolated from the by the neighbor-joining method with MEGA 3.0 for 1000 replicates whole fish using TRIzol reagent (Invitrogen) and then treated with RNase- (59, 60). free DNase I (Qiagen). The first-strand cDNA was transcribed using an RNA PCR kit (AMV version 3.0; TaKaRa) according to the manu- Tissue distribution and expression of DIGIRR facturer’s instructions. The full-length Tetraodon DIGIRR cDNA was To determine the distribution of Tetraodon DIGIRR in various tissues and cloned by RT-PCR and RACE-PCR using the specific primers shown in organs, fish were i.m. injected with 0.1 ml LPS (LPS from E. coli serotypes Table I. The PCR amplification was performed in 10-ml reaction mixtures m m m 055:B5; 0.01 mg/ml; Sigma). Healthy unstimulated fish served as controls. containing 0.8 l10 M forward and reverse primers, 0.2 lDNAtem- Brain, heart, gill, intestine, skin, liver, muscle, spleen, and head kidney plate, 1 ml PCR buffer (TaKaRa), 0.2 ml deoxy-ribonucleoside triphos- 2 m were carefully removed, flash frozen in liquid nitrogen, and stored at 80˚C phate mixture (for a final concentration of 2.5 mM each), 7.75 l dis- until processed. In another experiment, fish were i.m. injected with 0.1 ml tilled water, and 0.05 ml Taq polymerase (5 U/ml; TaKaRa). The cycling LPS for 0 (control), 3, 6, 12, 24, and 48 h, and the head kidney tissue protocol was an initial denaturing cycle at 94˚C for 4 min followed by 35 was collected and stored in liquid nitrogen. Total RNA was extracted as cycles of 94˚C for 30 s, 55˚C for 30 s, and 72˚C for 30 s, and a final 9 9 described earlier, then reversed transcribed into cDNA using an RNA PCR elongation at 72˚C for 10 min. The 5 and 3 RACE-PCRs were performed kit (AMV version 3.0; TaKaRa) following the manufacturer’s instructions. according to the protocols provided by the manufacturer (TaKaRa). The Two pairs of specific PCR primers, b-actin-F/b-actin-R and DIGIRR-EF/ Tetraodon b open reading frame (ORF) of zebrafish and IL-1 cDNA were DIGIRR-ER (Table I), were designed for the amplification of Tetraodon also amplified by RT-PCR using specific primers listed in Table I. PCR b-actin (as an internal control) and Tetraodon DIGIRR. The real-time PCR products were purified by using a gel extraction kit (Qiagen) and were then amplification was carried out on the Mastercyclerep realplex real-time ligated into the pGEM-T EASY vector (Promega) and transformed into PCR system (Eppendorf) using an SYBR Premix Ex Taq kit (TaKaRa) competent Escherichia coli top10 cells (Invitrogen). Plasmid DNA was following the manufacturer’s instructions. Briefly, all real-time PCR re- purified by using a plasmid Miniprep method and sequenced on a Mega- actions were performed in a 10-ml total reaction volume. The experi- BACE 1000 Sequencer (GE) using the DYEnamic ET Dye Terminator ment protocol consisted of 1) amplification repeated 40 times, using 30 s at Cycle Sequencing Kit (Pharmacia). 95˚C, then 20 s at 60˚C, and 2) melting curve analysis using 5 s at 95˚C, Structural characterization of DIGIRR 15 s at 65˚C, then 95˚C, and 3) cooling at 40˚C for 30 s. Relative gene expression was calculated using the 22DDCT method with initial normali- The comparative gene map positions were determined using Map viewer zation of DIGIRR against b-actin. In all cases, each PCR trial was per- (http://www.ncbi.nlm.nih.gov/mapview/). Gene organizations (intron/exon formed with triplicate samples and repeated at least three times. 2516 DISCOVERY OF A DIGIRR GENE FROM FISH

Plasmids construction 5 h posttransfection, the medium was changed to DMEM with 10% FBS. At 24 h posttransfection, cells were lysed for reporter gene assays or fixed The ORF of Tetraodon DIGIRR was inserted into pcDNA6/myc-HisB for laser confocal imaging. (Invitrogen) between EcoRI and XhoI sites to construct a eukaryotic ex- pression vector named pc6-DIGIRR, in which the C terminus of DIGIRR Subcellular distribution was fused with a myc-tag. Subcellular localization plasmid was conducted by fusing DIGIRR with GFP by placing cDNA between EcoRI and XhoI The HEK293T, HepG2, and HeLa cells were subcultured and seeded onto sites of pEGFP-N1 (Clontech); the plasmid was referred to as pEGFPN1- coverslips in 24-well plates before transfection. The next day, the cells were DIGIRR. The cDNA of zebrafish and Tetraodon IL-1b was digested with cotransfected with the overexpression plasmids pcDNA6-myc-DIGIRR BamHI and XhoI and subcloned into pCMV-Tag2B (Stratagene) to pro- and/or pEGFPN1-DIGIRR with or without pDsRed2-Golgi, the latter of duce the overexpression vectors pCMV-ZeIL-1 and pCMV-TnIL-1 (Ze, which contains the human b-1,4-galactosyltransferase sequence contain- zebrafish; Tn, T. nigroviridis). For preparation of recombinant Tetraodon ing a membrane-anchoring signal peptide that targets the protein to the IL-1b protein, prokaryotic expression vector pET28a-TnIL-1b was con- transmedial region of the Golgi apparatus. At 24 h posttransfection, cells structed by inserting full-length Tetraodon IL-1b cDNA into pET28a were washed three times with PBS and fixed for 10 min in 3% (v/v) (Novagan) between BamHI and XhoI sites. All the primers used for in- formaldehyde in PBS. Fluorescence images of pEGFPN1-DIGIRR and troducing enzyme sites are shown in Table I. Overexpression plasmid pDsRed2-Golgi in cells were obtained using a laser scanning confo- pFLAG-CMV-MouseSIGIRR was a gift from Prof. Alberto Mantovani cal microscope (Zeiss LSM510). In addition, colocalization images of (Istituto Clinico Humanitas, Milan, Italy) (61), and Golgi apparatus sub- pcDNA6-myc-DIGIRR and pDsRed2-Golgi were obtained using immu- cellular localization plasmid pDsRed2-Golgi was kindly provided by Prof. nofluorescence staining. For this, the fixed cells were permeabilized with Thomas Kietzmann (University of Kaiserslautern, Kaiserslautern, Ger- 0.2% Triton X-100 in PBS for 15 min. The cells were then blocked with many) (62). All the plasmids for transfection and microinjection were 5% (v/v) goat serum at 37˚C for 1 h, incubated with rabbit anti-myc Abs prepared free of endotoxin by using the Spin Miniprep Kit (Qiagen). (Abcam) at 37˚C for 1 h and then with FITC-labeled goat anti-rabbit IgG secondary Abs at 37˚C for 1 h. The cells transfected with pEGFPN1-

Chimeras and mutant constructions DIGIRR were successively incubated with BODIPY TR C5-ceramide (a Downloaded from Golgi indicator; Molecular Probes) and DiI (a cell membrane probe; A chimera molecule was constructed by fusing the extracellular and Beyotime). For C -ceramide staining, vital cells were incubated without transmembrane domains of human IL-1R (1–359 aa) with the cytoplasmic 5 fixation with 5 mMC5-ceramide at 4˚C for 30 min, rinsed several times DIGIRR region (255–524 aa) to detect whether the intracellular TIR do- with an ice-cold medium, and incubated in fresh medium at 37˚C for main in DIGIRR is capable of signal transduction. Two pairs of primers another 30 min. For DiI staining, the cells were fixed with formaldehyde, (ChimF1 and ChimR1, ChimF2 and ChimR2) were designed to amplify as described above, and stained with 10 mM DiI at 37˚C for 10 min. The the extracellular and transmembrane domains of human IL-1R and the fluorescence images of these stained cells were visualized under a laser cytoplasmic region of DIGIRR, respectively. Both amplified fragments scanning confocal microscope at 3600 magnification. http://www.jimmunol.org/ contain 25 overlapping bp. The chimeric cDNA was obtained using an overlapping PCR protocol with ChimF1 and ChimR2. Thereafter, it was NF-kB luciferase assay in mammalian cells inserted into pcDNA6/myc-HisB between the EcoRI and XhoI sites, and the resulting plasmid was designated as pc6-Chimera-1. Four chimera The NF-kB–dependent luciferase assay was performed to examine NF- mutants (chimera-2, -3, -4, and -5) were constructed by site mutation based kB activity and signaling. For this purpose, HeLa or HEK293T cells on chimera-1 using overlap PCR to investigate the potential role of the were cotransfected with 400 ng of either pc6-DIGIRR, pFLAG-CMV- amino acids Ser335 and Arg-Tyr420 in the TIR domain of DIGIRR in MouseSIGIRR, pc6-Chimeras1–5, pc6-HumanIL-1R, or empty control signal transduction. Chimera-2 was obtained by replacing the Leu420 in vector pcDNA6 along with the reporter plasmid pNFkB-TA-Luc (100 ng) chimera-1 with Tyr. For this, two pairs of primers (ChimF1 and ChimR3, (Clontech) and pRL-TK (10 ng) (Promega). At 24 h posttransfection, the ChimF3 and ChimR2) were used to amplify the two fragments (A and B) HeLa or HEK293T cells were stimulated for 6 h with human recombinant of chimera-1 before and after the Leu420 site. The mutational fragment C IL-1b (Invitrogen). Cells were lysed using Reporter Lysis Buffer (Prom- by guest on October 1, 2021 was then generated by introducing the mutation to the 59 terminal of ega), and luciferase activity was assessed using Dual-Luciferase Assay fragment B with the primers ChimF4 and ChimR2. Finally, fragments C Reagent (Promega). All results reported represent duplicate experiments and A were fused by overlap PCR using primers ChimF1 and ChimR2 to from at least three independent transfections. generate chimera-2, which was subcloned into pcDNA6/myc-HisB be- tween the EcoRI and XhoI sites. Similar protocols were used for the NF-kB luciferase assay in zebrafish embryos construction of chimera-3, -4, and -5. In chimera-3, the TIR domain Zebrafish embryo model was also used to examine NF-kB activity contained two substitutions, with Ala419 replaced with Arg and Leu420 according to the method previously described (63, 64). In brief, expres- with Tyr. For chimera-4, the TIR domain was generated by replacing sion plasmids (0.1–1.0 ng/egg), including pc6-DIGIRR, pCMV-ZeIL-1 or Ala419 with Arg. For chimera-5, the Ser335 was mutated into Cys in the pCMV-TnIL-1, pcDNA6/pCMV-Tag2B (used to ensure all samples re- TIR domain. In parallel, a human IL-1RI–encoding plasmid was con- ceived equal amounts of DNA), firefly and Renilla reporter plasmids (100 structed by inserting the IL-1RI cDNA into pcDNA6/myc-HisB between and 10 pg/egg), with or without 4 ng/egg E. coli LPS (E. coli serotypes the EcoRI and XhoI sites. The human IL-1RI cDNA was kindly provided 055:B5), were combined with microinjection buffer (0.5% phenol red, 240 by Dr. Justin V. McCarthy (University College Cork, Cork, Ireland). A mM KCl, 40 mM HEPES, pH 7.4) and injected (2–3 nl) into one-cell-stage schematic illustration of the chimeras is shown in Fig. 11. All the primers embryos using a microinjector (ASI MPPI-3). The microinjected embryos used for chimera and mutant construction are shown in Table I. were cultured at 28˚C in E3 medium (5 mM NaCl, 0.17 mM KCl, 0.33 mM Production of Tetraodon IL-1b protein CaCl2, 0.33 mM MgSO4). The firefly and Renilla luciferase activities were assayed at 24 h postmicroinjection as described above with 5–10 replicates The pET28a-TnIL-1b was transformed into E. coli BL21 (DE3). One (each replicate containing the tissue extracts from 30 embryos). single colony was inoculated into 100 ml Luria–Bertani medium with kanamycin (25 mg/l). The inoculated Luria–Bertani medium was shaken Screening of small interfering RNA against DIGIRR and incubated at 37˚C until the OD value reached 0.6, isopropyl b-D- 600 The small interfering RNA (siRNA) against DIGIRR was initially designed thio-galactoside was added to a final concentration of 1 mM, and the in- by an siRNA Template Design Tool (Applied Biosystem) following the cubation was continued for 6 h. The expression level of the protein was method previously reported (65). From the identified 46 regions of DIGIRR assessed by 10% SDS-PAGE gel followed by Coomassie brilliant blue targeted for siRNA, four regions (Table I) were selected to screen the R250 staining. The recombinant IL-1b was purified using Ni-NTA agarose activity. DNA oligonucleotides for hairpin RNA expression were synthe- affinity chromatography according to the QIA Expressionist manual sized by Invitrogen and dissolved in nuclease-free H O at a concentration (Qiagen). 2 of 3 mg/ml and then mixed with equimolar amounts of complementary Cell culture and transient transfection sense and antisense strands in annealing buffer (103 buffer stock, 100 mM Tris-HCl, pH 7.5, 10 mM EDTA, 1 M NaCl). In a 50-ml reaction, the The cultured cell lines HEK293T, HepG2, and HeLa were maintained in mixture was incubated at 90˚C for 4 min, then at 70˚C for 10 min, and DMEM supplemented with 10% FBS, penicillin (100 U/ml), and strep- slowly cooled to 10˚C. The pSUPER vector (pSUPER.retro.puro; Oligo- tomycin (100 mg/ml) at 37˚C in 5% CO2. Cells were transiently transfected engine) was digested with HindIII and BglII, and the annealed oligonu- by Lipofectamine 2000 (Invitrogen) according to the manufacturer’s pro- cleotides were ligated into pSUPER downstream of the H1 promoter to tocol. Cells (1 3 105/ml) were seeded into 24-well plates and transfected generate four constructs with different siRNAs named pSUPER-DIGIRR- with 1 mg DNA in Opti-MEM I medium without serum and antibiotics. At 1, -2, -3, and -4. The pSUPER-DIGIRR constructs or control pSUPER The Journal of Immunology 2517

(1 mg) along with overexpression plasmid DIGIRR (1 mg) were cotrans- in the same protocols. The nucleotide and deduced amino acid fected into HEK293T cells in 12-well plates. The specificity of siRNAs for sequences of DIGIRR cDNA from Tetraodon, Gasterosteus, and DIGIRR was determined by real-time PCR. Takifugu are shown in Fig. 1 and Supplemental Fig. 1. Production of lentivirus The full-length Tetraodon DIGIRR cDNA (Fig. 1, GenBank accession number EF095151) is 1916 bp long (excluding the Based on data (see Results) showing that pSUPER-DIGIRR-3 was the most poly-A tail) and consists of a 112-bp 59 untranslated region effective siRNA construct to suppress the DIGIRR expression, the U6 promoter cassette in lentiviral vector PLB was replaced by the H1-siRNA (UTR), a 1575-bp ORF encoding a predicted polypeptide of 524 cassette excised from pSUPER-DIGIRR-3 construct to produce the PLB- aa, and a 229-bp 39UTR that contains one polyadenylation signal DIGIRR lentiviral vector. Lentiviruses were produced by transient trans- (AATAAA) 15 nucleotides upstream of the poly-A tail. Similarly, fection of packaging cell line HEK293T cells according to standard pro- the Takifugu DIGIRR cDNA (Supplemental Fig. 1A, GenBank tocols (66). In brief, confluent cells were subcultured, seeded into 10-cm dishes, and cotransfected with PLB (20 mg), pCMV-VSVG (packaging accession number EU305619) consists of a 1569-bp ORF that plasmid for HIV-1–based vectors) (6 mg), and pCMV-dR8.2 (the vesicular encodes a 522-aa polypeptide, and the Gasterosteus DIGIRR stomatitis virus G protein expression plasmid) (15 mg) by Lipofectamine cDNA (Supplemental Fig. 1B, GenBank accession number 2000. The lentiviral supernatant was harvested at 48–96 h posttransfection, EU360722) is composed of a 1638-bp ORF that encodes a 545-aa concentrated via ultracentrifugation at 25,000 rpm for 90 min at 4˚C (SW- polypeptide. In comparison with the counterparts of DIGIRR in 41Ti rotor), and dissolved in 100 ml PBS and stored at 280˚C. Lentivirus titers were determined by transduction and flow cytometric analysis of other species, fish DIGIRR genes are 112–135 bp longer than GFP expression in HEK293T cells. The ability of lentiviral supernatant to their mammalian homolog SIGIRR. Significantly, SIGIRR was silence the DIGIRR expression was also evaluated. The HEK293T cells not found in Tetraodon, Gasterosteus,orTakifugu;rather,we m were transfected with DIGIRR, then at 24 h posttransfection, 10 l con- found DIGIRR, which we considered a possible new immune gene centrated supernatant was added into culture medium, and the silencing Downloaded from efficiency of lentiviral supernatant was detected by real-time PCR. in the TIR superfamily.

In vivo knockdown of DIGIRR Characterization of DIGIRR structures The lentivirus delivery system developed as described earlier was used for Using Genscan and BLAST programs, genes adjacent to the in vivo knockdown of DIGIRR expression in Tetraodon. For this purpose, DIGIRR locus were retrieved from Tetraodon chromosome 13. the lentiviruses (5 3 106 TU/ml) concentrated from HEK293T supernatant Similar to SIGIRR, plakophilin 3, transmembrane protein 16J, was repeatedly delivered into fish by muscle injection (20 ml/fish) once phosphatidylserine synthase 2, v-Ha-ras, and Harvey rat sarcoma http://www.jimmunol.org/ a day for 3 d. In parallel, the control group was injected with PBS. Total viral oncogene homolog are all found in the identical order on RNA from liver, kidney, spleen, and leukocytes from head kidney were isolated and reversed transcribed into cDNA. Real-time PCR was con- Tetraodon chromosome 13, human chromosome 11, and mouse ducted to evaluate the efficiency of in vivo suppression of DIGIRR. chromosome 7. Clearly, there is a high degree of conservation of genome synteny between Tetraodon DIGIRR and mouse and hu- Evaluation of DIGIRR in IL-1–induced inflammation man SIGIRR (Fig. 2A). By comparing the full-length Tetraodon, Lentiviral supernatant (20 ml/fish) was repeatedly injected into Tetraodon Takifugu, and Gasterosteus DIGIRR cDNAs with the corre- as described earlier. At the time of the third injection, some fish were also sponding genomic sequences, the organization of DIGIRR genes m b stimulated with 6 g of recombinant IL-1 (the control group received was elucidated (Fig. 2D). The Tetraodon DIGIRR gene is located PBS). At 24 h after IL-1b challenge, total RNA from liver, kidney, spleen, by guest on October 1, 2021 and head kidney leukocytes was isolated and transcribed into cDNA. The within a 3094-bp genomic fragment on chromosome 13, the head kidney leukocytes of Tetraodon were isolated as described previously Takifugu DIGIRR gene is located within a 3163-bp genomic (67). Briefly, the head kidney was removed from fish under sterile con- fragment of scaffold 198, and the Gasterosteus DIGIRR gene is ditions, washed twice with L-15 medium, and gently passed through a 100- located within a 4313-bp genomic fragment of group XIX. The m m nylon mesh. The cell suspension was layered onto a 34/51% discon- intron-splicing consensus (GT/AG) is constant at the 59 and 39 tinuous Percoll (Sigma) gradient and centrifuged at 400 3 g for 30 min at 4˚C. Cells at the 34/51% interface were collected for RNA isolation. Real- ends of the introns in all three fish DIGIRR genes. time PCR was performed to detect the IL-1b–induced inflammatory Unlike the vertebrate homolog SIGIRR genes with nine cytokines (IL-1b and IL-6) using the primers listed in Table I. encoding-exons, the fish DIGIRR genes have 10 encoding-exons. Particularly, the first exon is much longer (67–70 bp) than that (7 bp) Statistical analysis in all higher vertebrate SIGIRR genes. This exon is predicted Statistical evaluation of differences between experimental group means to encode an unexpected signal peptide sequence, which is not , was done by ANOVA and multiple Student t tests. A p value 0.05 was found in SIGIRR, suggesting that the final subcellular DIGIRR considered to be statistically significant. The sample numbers of each group were always .10 fish of about equal body weight. Data points were localization is different (Figs. 1, 2B). Exon 2, which encodes the from at least three independent experiments. first extracellular Ig-like domain (named D1) of the DIGIRR molecule, is unique to the DIGIRR gene. Except for the unique features of the first two encoding exons, however, the size and Results organization of other exons in DIGIRR genes are well matched Identification of DIGIRR cDNA with SIGIRR genes. Exons 3–10 in DIGIRR correspond with Through computational searching of the Tetraodon EST/genome exons 2–9 in SIGIRR. Exon 3, together with exon 4 of DIGIRR databases using the human IL-1R sequence, we discovered one IL- genes, encode the second extracellular Ig-like domain, which 1R–related sequence (named as DIGIRR) homologous to SIGIRR, confers DIGIRR with a unique two-Ig-like domain structure dis- which had never been characterized. Based on the sequences tinct from SIGIRR (with one) and IL-1R (with three). Exon 5 searched, specific primers (Table I) were designed for the mo- encodes the transmembrane domain, and the remaining exons lecular cloning of Tetraodon DIGIRR gene. Two ORF fragments (exons 6–10) encode the intracellular regions, including one TIR of DIGIRR obtained by different primers combined with the domain (Fig. 2B). The similar organization between DIGIRR fragments obtained from 59 and 39 RACE-PCR were compiled (exons 3–10) and SIGIRR (exons 2–9) genes, together with the together to form a full-length Tetraodon DIGIRR cDNA. Simi- conserved genomic synteny, largely provide evolutionary evidence larly, the full-length Gasterosteus and Takifugu DIGIRR sequen- that these two genes probably originated from a common ancestor. ces were also predicted by an in silico screen of the EST/genome The copy number of the Tetraodon DIGIRR gene is analyzed databases using the Tetraodon DIGIRR cDNA sequence as a probe by Southern blotting under stringent hybridization and washing 2518 DISCOVERY OF A DIGIRR GENE FROM FISH

Table I. Oligonucleotide primers used for amplifying cDNAs and gene expression analysis

Primer Name Sequence (59–39)Use DIGIRR-F ATCCTCAAACCGTACTTGG Tetraodon DIGIRR ORF DIGIRR-R CATCGGGTCCTTGTCG Tetraodon DIGIRR ORF DIGIRR-3F GCAGGACGACAAGGACCCGATG Tetraodon DIGIRR 39RACE 39Adapter CTGATCTAGAGGTACCGGATCC Tetraodon DIGIRR 39RACE DIGIRR-5F GGTCCGCTGATGGTAACTAACGC Tetraodon DIGIRR 59UTR DIGIRR-5R CAGTCAGGCGGAGGGTAAACG Tetraodon DIGIRR 59UTR TnIL-1b-F ATGGAATCACAGATGTCCAACAACG Tetraodon IL-1b ORF TnIL-1b-R TTACATCTCTCCCTCGCAACTGG Tetraodon IL-1b ORF ZeIL-1b-F ATGGCATGCGGGCAATATGAAGTC Zebrafish IL-1b ORF ZeIL-1b-R CTAGATGCGCACTTTATCCTGCAGC Zebrafish IL-1b ORF b-actin-F ACACCTTCTACAATGAGCTG Gene expression b-actin-R CTGCTTGCTGATCCACATCT Gene expression DIGIRR-EF GCAACCACACGCTTCGCT Gene expression DIGIRR-ER GCGACAGCAAGACGATGAGAC Gene expression DIGIRR-RT-F GACGACAAGGACCCGATGCT Real-time PCR DIGIRR-RT-R GGTGACCAGGCGGTAGAAGT Real-time PCR IL-1-RT-F ACGGTGGTGGAAGAGCAAAT Real-time PCR IL-1-RT-R GTCGTGCTTGTAGAACAGAAATCG Real-time PCR IL-6-RT-F GTCTCCTCGTCTACTCGGTTCT Real-time PCR Downloaded from IL-6-RT-R CCTTGTTCTCCACCTCTTTGAT Real-time PCR b-actin-RT-F AGGTCATCACCATCGGCAAT Real-time PCR b-actin-RT-R GATGTCCACGTCGCACTTCA Real-time PCR DIGIRR-P-F GACGACAAGGACCCGATGCT Southern blot probe preparation DIGIRR-P-R GGTGACCAGGCGGTAGAAGT Southern blot probe preparation pET28a-TnIL-1b-F TTTGGATCCATGGAATCACAG Prokaryotic expression b TTTCTCGAGCATCTCTCCCTC pET28a-TnIL-1 -R Prokaryotic expression http://www.jimmunol.org/ Pc6-DIGIRR-F TTTGAATTCGCCACCATGGCTCTAATG Eukaryotic expression Pc6-DIGIRR-R AAACTCGAGCGGACGTCTGCCTC Eukaryotic expression pCMV-TnIL-1b-F TTTGGATCCATGGAATCACAG Eukaryotic expression pCMV-TnIL-1b-R TTTCTCGAGCATCTCTCCCTC Eukaryotic expression pCMV-ZeIL-1b-F AAAGGATCCATGGCATGCGGG Eukaryotic expression pCMV-ZeIL-1b-R AAACTCGAGGATGCGCACTTT Eukaryotic expression DIGIRR-siRNA1-sense GATCCCCCGAGCCCTCAGCAACCACA RNAi TTCAAGAGATGTGGTTGCTGAGGGCT CGTTTTTA DIGIRR-siRNA1-antisense AGCTTAAAAACGAGCCCTCAGCAAC RNAi

CACATCTCTTGAATGTGGTTGCTGAG by guest on October 1, 2021 GGCTCGGGG DIGIRR-siRNA2-sense GATCCCCACAATGACCAGGACAGAA RNAi ATTCAAGAGATTTCTGTCCTGGTCAT TGTTTTTTA DIGIRR-siRNA2-antisense AGCTTAAAAAACAATGACCAGGAC RNAi AGAAATCTCTTGAATTTCTGTCCTG GTCATTGTGGG DIGIRR-siRNA3-sense GATCCCCTATCAAACTCTGGTACAA RNAi GTTCAAGAGACTTGTACCAGAGTTT GATATTTTTA DIGIRR-siRNA3-antisense AGCTTAAAAATATCAAACTCTGGTA RNAi CAAGTCTCTTGAACTTGTACCAGAG TTTGATAGGG DIGIRR-siRNA4-sense GATCCCCGAATGCTTACAAGGTGCAT RNAi TTCAAGAGAATGCACCTTGTAAGCAT TCTTTTTA DIGIRR-siRNA4-antisense AGCTTAAAAAGAATGCTTACAAGGTG RNAi CATTCTCTTGAAATGCACCTTGTAAGC ATTCGGG ChimF1 TTTGAATTCGCCACCATGAAAGTGTT Chimera construction ChimF2 TTCTGTTTTCATCTATAAAATCTTC Chimera construction AGGTGTCACC ChimF3 GCCATGCCTCGCAGGGTG Chimera construction ChimF4 CTCCGTCTTCTGGAAGGAGCTAGC Chimera construction TTACGCCATGCCTCG ChimF5 CTCCGTCTTCTGGAAGGAGCTAAG Chimera construction GTACGCCATGCCTCG ChimF6 CTCCGTCTTCTGGAAGGAGCTAAG Chimera construction GTTAGCCATGCCTCG ChimF7 CGGCGTCTCATCGTCTTG Chimera construction ChimF8 AGAGCTGCTGATGAACATGAGTCGC Chimera construction TGCCGGCGTCTCAT ChimR1 TTTGAATTCGCCACCATGAAAGTGTT Chimera construction ChimR2 AAACTCGAGCGGACGTCTGCCTC Chimera construction ChimR3 GCCATGCCTCGCAGGGTG Chimera construction ChimR4 GCGACTCATGTTCATCAG Chimera construction The Journal of Immunology 2519 Downloaded from

FIGURE 1. The nucleotide and deduced amino acid sequences of DIGIRR cDNA from Tetraodon (Gen- Bank accession number EF095151). The putative sig-

nal peptide is underlined, and the transmembrane re- http://www.jimmunol.org/ gion is shown in gray. The start and stop codons are in boldface. The three conserved motifs (boxes 1–3) of the TIR domain are underlined and noted. The poly- adenylation signal “AATAAA” in the 39UTR is in bold italic. by guest on October 1, 2021

conditions. As shown in Fig. 2C, the digests obtained with three 174–177; NTTS, 181–184; NMSR, 355–358; NQSA, 454–457), different enzymes displayed one band. As the genomic sequence five N-myristoylation sites (GNytCV, 94–99; GNgtAS, 102–107; covered by the probe did not contain any cleavage sites specific to GNvaTF, 129–134; GSevTL, 139–144; GGeqSA, 345–350), six any of the restriction enzymes used here, the results suggest that protein kinase C phosphorylation sites (TlR, 109–111; SrR, 151– Tetraodon has only one copy of the DIGIRR gene. 153; TlR, 176–178; SyR, 249–251; SrR, 359–361; TwR, 426– The molecular masses of the fish DIGIRR proteins are predicted 428), eight casein kinase II phosphorylation sites (TcvD, 25–28; to be 58.05–61.08 kDa with theoretical isoelectric points of 5.85– TvdE, 113–116; SygD, 291–294; SagD, 456–459; TlqD, 463–466; 6.54. The mature DIGIRR contains a number of potential func- SdtD, 483–486; SdiD, 521–524; TeaD, 544–547), and two tyro- tional motifs including seven N-glycosylation sites (NYTC, amino sine kinase phosphorylation sites (Rcr. DrykY, 213–220; Rcr. acid positions 95–98; NGTA, 103–106; NCTA, 145–148; NHTL, DgykY, 233–240). 2520 DISCOVERY OF A DIGIRR GENE FROM FISH Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021

FIGURE 2. A, Order and orientation of genes syntenic to DIGIRR and SIGIRR in sequenced genomes. Because genomes are not completely sequenced, a small possibility exists that any gene portrayed as absent is actually present. Genes are linked with lines and aligned in columns to facilitate visualization of synteny. Such alignments help guide the search for conserved regulatory elements. B, Schematic representation of domain organization of DIGIRR and SIGIRR proteins. Each domain was predicted by the SMART and PROSITE programs. Some transmembrane regions were predicted by the TMHMM program. Unlike all reported mammalian SIGIRR molecules, all fish DIGIRR molecules have a signal peptide in the N-terminal region, followed by two Ig- like domains in the extracellular region. However, the TIR domain at the C-terminal end is conserved in both DIGIRR and SIGIRR. C, Southern blot analysis of Tetraodon DIGIRR. Genomic DNA samples extracted from head kidney were completely digested with HindIII, BamHI, and EcoRI. D, Genomic structures of fish DIGIRRs genes compared with SIGIRRs of human, mouse, rat, and chicken. The rectangles represent the exons, and the lines between them indicate the introns. The sizes of exons are indicated above the exons, and the sizes of introns are indicated below the introns. The red boxes indicated the unique exons found in fish, and the silvery boxes indicate the conserved exons in both SIGIRR and DIGIRR. The cysteine residues forming the disulfide bonds of Ig domains are indicated with “C” and turning curve. The two Ig-like domains of DIGIRR (D1, D2) and the single Ig domain of SIGIRR (S1) are boxed out with dashed lines.

Multiple sequence alignment and phylogenetic analysis and DIGIRRs show 78.3% identity between Tetraodon and Taki- An amino acid sequence multiple alignment of the DIGIRR/ fugu, 66.8% between Tetraodon and Gasterosteus, and 68.3% SIGIRR family encompassing representatives of various species, between Gasterosteus and Takifugu. The Tetraodon DIGIRR is including primate, rodent, bird, and teleost fish, was prepared using clearly more similar to that of Takifugu, indicating the close re- ClustalW. The result shows that the fish DIGIRRs share an overall lationship between Tetraodon and Takifugu. The phylogenetic tree sequence similarity with other species (Fig. 3). As shown in Table was constructed by using the neighbor-joining method and included II, the teleost fish DIGIRRs show 31.0–41.2% sequence identity DIGIRRs, SIGIRRs, and other IL-1R family members (Fig. 4). The with mammalian SIGIRRs and 35.1–37.9% homology with avian result shows that fish DIGIRRs are not only clustered together to (chicken) SIGIRR. In general, homology is much higher between form an exclusive group but also, along with mammal SIGIRRs, fish species than between fishes and mammals or between fishes merge into a larger group with high bootstrap probability. These and birds. Indeed, SIGIRRs demonstrate 81.2–89.8% identity findings indicate that DIGIRR and SIGIRR share sequence ho- among mammals and 61.8–66.2% between mammals and avians, mology and might originate from a common ancestor molecule. The Journal of Immunology 2521 Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021

FIGURE 3. Alignment of the predicted amino acid sequence of fish DIGIRR with those of corresponding SIGIRR proteins of other species. Residues shaded in black are completely conserved across all species, and residues shaded in gray are similar with respect to side chains. The dashes in the amino acid sequences indicate gaps introduced to maximize alignment. Three conserved motifs of TIR are boxed out in solid lines. Black triangles indicate cysteines predicted to form the disulfide bonds of an Ig-like domain fold. Gene accession numbers are noted in the legend of Fig. 4.

Functional domain analysis compared in DIGIRR, both newly identified Ig-like domains We then performed a more detailed analysis of the sequence ho- (named as D1 and D2) are highly conserved among the three fish mology between functional domains among DIGIRR, SIGIRR, and species (D1, 65.8–73.9%; D2, 54.3–66.4%). However, when D1 other IL-1R family members. When the Ig-like domains are and D2 are compared with each other, the identity is lower (13.9–

Table II. Percentages (%) of amino acid sequence identity for SIGIRR and DIGIRR sequences

Human Mouse Rat Bovine Chicken Takifugua Tetraodona Gasterosteusa Human 81.3 87.1 88.2 66.2 40.1 39.3 35.4 Mouse 89.8 81.2 61.8 35.7 35.2 31 Rat 87.1 65.8 38.9 38.6 34.6 Bovine 66 41.2 39.8 36 Chicken 36.8 37.9 35.1 Takifugua 78.3 68.3 Tetraodona 66.8 Gasterosteusa aDIGIRR sequence. 2522 DISCOVERY OF A DIGIRR GENE FROM FISH

18.4%), indicating that D1 and D2 are two independently con- served structures in fish. When both D1 and D2 are compared with the SIGIRR Ig-like domain S1, D1 shows lower sequence iden- tity to S1 (16.8–24.3%) compared with D2 (27.0–39.6%). These results, together with the gene organization analysis (Fig. 2D), strongly suggest that an evolutionary relationship exists between D2 and S1. In addition, there is conservation in the intracellular region between DIGIRR and SIGIRR, with an overall 37.8–56.7% sequence identity in the intracellular region and a 51.7–64.4% identity in the TIR domains. However, the extracellular regions between DIGIRR and SIGIRR are less conserved. Two sites of conserved amino acids, Ser and Arg-Tyr, in the TIR domains of the IL-1R family are believed to be crucial for receptor signal transduction (25, 27, 68). However, both sites are substituted in the SIGIRR molecules. As shown in Fig. 5, the conserved sites of Ser447 and Arg-Tyr536 in the TIR domains of IL-1R family members are replaced by Cys222 and Gln-Leu305 in SIGIRR molecules. However, the DIGIRR is more homologous to IL-1Rs, as only one site mutation, Arg-Tyr536, is replaced by Ala- Leu420. Accordingly, the same substitution is also found in Downloaded from mammalian IL-1RAcP, in which the Arg-Tyr is replaced by Gln- Val. Thus, the mutation of a crucial conserved amino acids site (Arg-Tyr) in the TIR domain might functionally transform DIG- IRR. To know for certain, however, it is necessary to assess the function of DIGIRR in IL-1–mediated immunity. http://www.jimmunol.org/ Tissue distribution and expression analyses RT-PCR and real-time PCR were used to evaluate the expression patterns of DIGIRR in different tissues in both healthy and LPS- stimulated fish. The result showed that in healthy fish, the DIGIRR mRNA could be amplified in gill, brain, liver, skin, intestine, spleen, head kidney, and leukocytes from blood, spleen, and head kidney, but was totally undetectable in heart and muscle (Fig. 6A, FIGURE 4. An unrooted phylogenetic tree of protein sequence of 6B). After LPS stimulation for 12 h, however, a significant ele- by guest on October 1, 2021 DIGIRR, SIGIRR, and IL-1R family members was constructed with the vation of DIGIRR expression was observed in head kidney, neighbor-joining method by Mega 3.0. Node values represent percent spleen, liver, brain, skin, leukocytes from blood, spleen, and head bootstrap confidence derived from 1000 replicates. GenBank accession kidney (all p , 0.01), and intestine (p , 0.05) compared with that numbers are as follows: Human IL-1RI, M27492 (http://www.ncbi.nlm. in untreated control fish (Fig. 6C,6D). This result indicated that nih.gov/nuccore/M27492); Mouse IL-1RI, M20658 (http://www.ncbi.nlm. DIGIRR may play a role in inflammatory responses. The ex- nih.gov/nuccore/M20658); Rat IL-1RI, Q05KR1 (http://www.ncbi.nlm. pression pattern of DIGIRR appeared different from that pre- nih.gov/protein/Q05KR1); Chicken IL-1RI, M81846 (http://www.ncbi. viously observed in SIGIRR, which was expressed in most tissues nlm.nih.gov/nuccore/M81846); Salmon IL-1RLP, AJ293335 (http://www. examined in mouse (26), but was significantly downregulated after ncbi.nlm.nih.gov/nuccore/AJ293335); Trout IL-1RI, CAC82935 (http:// www.ncbi.nlm.nih.gov/protein/CAC82935); Human IL-1Rrp2, Q9HB29 stimulation with LPS (24). This converse regulation suggests there (http://www.ncbi.nlm.nih.gov/protein/Q9HB29); Mouse IL-1Rrp2, Q9ERS7 might be a different mechanism involved in the inflammatory (http://www.ncbi.nlm.nih.gov/protein/Q9ERS7); Human ST2, Q01638 response between DIGIRR and SIGIRR. Furthermore, kinetic ex- (http://www.ncbi.nlm.nih.gov/protein/Q01638); Mouse ST2, P14719 pression analysis demonstrated that an upregulation of DIGIRR (http://www.ncbi.nlm.nih.gov/protein/P14719); Rat ST2, U04319 (http:// could be observed after 3-h stimulation with LPS in head kidney. www.ncbi.nlm.nih.gov/nuccore/U04319); Chicken ST2, Q9DEE5 (http:// At 6 h and 12 h after LPS challenge, there was a significant (3.3- www.ncbi.nlm.nih.gov/protein/Q9DEE5); Salmon ST2, Q8AXT5 (http:// fold and 4.2-fold) increase in the abundance of DIGIRR mRNA. www.ncbi.nlm.nih.gov/protein/Q8AXT5); Human IL-1RAcP O14915 At 24 h postchallenge, the expression of DIGIRR reached its peak (http://www.ncbi.nlm.nih.gov/protein/34222652); Mouse IL-1RAcP, Q61730 of 6.6-fold higher than that measured in the control group (0 h). (http://www.ncbi.nlm.nih.gov/protein/Q61730); Rat IL-1RAcP, Q63621 After 48 h, however, the expression level decreased (Fig. 6E). This (http://www.ncbi.nlm.nih.gov/protein/Q63621); Salmon IL-1RAcP, AJ505008 (http://www.ncbi.nlm.nih.gov/nuccore/AJ505008); Human IL-18R, Q13478 LPS-induced increase in DIGIRR expression mirrors several other (http://www.ncbi.nlm.nih.gov/protein/Q13478); Mouse IL-18R, Q61098 IL-1R family members, such as IL-1R, IL-1RAcP, and ST2, which (http://www.ncbi.nlm.nih.gov/protein/Q61098); Sus scrofa IL-18R, are crucial in the initiation of inflammatory signaling (36, 40, 43, BAC19817 (http://www.ncbi.nlm.nih.gov/protein/BAC19817); Human IL- 69, 70). 18RAcP, O95256 (http://www.ncbi.nlm.nih.gov/protein/O95256); Mouse IL-18RAcP, Q9Z2B1 (http://www.ncbi.nlm.nih.gov/protein/Q9Z2B1); Chicken IL-18RAcP-like, XP_001234110 (http://www.ncbi.nlm.nih.gov/ protein/XP_001234110); Human SIGIRR, Q9H733 (http://www.ncbi. Chicken SIGIRR, XP_420927 (http://www.ncbi.nlm.nih.gov/protein/NP_ nlm.nih.gov/protein/Q9H733?report=genpept); Mouse SIGIRR, Q9JLZ8 001186471); Tetraodon DIGIRR, EF095151 (http://www.ncbi.nlm.nih.gov/ (http://www.ncbi.nlm.nih.gov/protein/Q9JLZ8); Rat SIGIRR, XP_219479 nuccore/EF095151); Takifugu DIGIRR, EU305619 (http://www.ncbi.nlm. (http://www.ncbi.nlm.nih.gov/protein/NP_001020058); Bovine SIGIRR, nih.gov/nuccore/EU305619); Gasterosteus DIGIRR, EU360722 (http:// NP_001075912.1 (http://www.ncbi.nlm.nih.gov/protein/NP_001075912.1); www.ncbi.nlm.nih.gov/nuccore/EU360722). The Journal of Immunology 2523

thermore, the cells transfected with pEGFPN1-DIGIRR were stained with the Golgi tracker C5-ceramide and the cell membrane indicator DiI. The results show that DIGIRR was clearly distrib- uted in the cytoplasm in all three cell lines examined (Fig. 7). Furthermore, the fluorescence images of both enhanced GFP- (Fig. 7B,7E,7F) and the FITC–anti-myc–labeled (Fig. 7C) DIGIRR molecules merged well with the DsRed protein or C5- ceramide signals targeting the Golgi apparatus (Fig. 7D). How- ever, no colocalized signals of the DIGIRR were detected with DiI (Fig. 7A). In contrast, the Tetraodon IL-1RI, a known cell membrane-localized receptor, was clearly demonstrated to be colocalized with the DiI, illustrating that IL-1RI is distributed on the cell surface (data not shown). These observations preclude the presence of DIGIRR on the cell surface and suggest that DIGIRR is a Golgi-distributed protein. k FIGURE 5. Alignment of the peptide sequences of the two conserved DIGIRR downregulates NF- B signaling in fish amino acids sites in the cytoplasmic TIR regions of representative IL-1R The TIR domain-containing family members were usually found to family members. Important differences within the two conserved amino be positive or negative regulators of two classical (IL-1–mediated Downloaded from acids sites of the DIGIRR sequence compared with other IL-1R family and LPS-mediated) NF-kB signaling pathways. To investigate members are boldface and underlined. The corresponding residues of IL- whether DIGIRR participates in these signaling pathways, the 1RI are essential for IL-1RI signaling. Gene accession numbers are noted involvement of this factor in IL-1–mediated and LPS-mediated in the legend of Fig. 4. NF-kB signaling pathways was examined. For this purpose, zebrafish and Tetraodon IL-1b expression plasmids were con- Subcellular distribution of DIGIRR structed, and the typical NF-kB dual-luciferase reporter assay To investigate the subcellular distribution of DIGIRR, EGFP- system and the newly developed zebrafish embryo signaling de- http://www.jimmunol.org/ labeled or myc-tagged DIGIRR constructs (pEGFPN1-DIGIRR tection model were used. The result showed that injection of and pcDNA6-myc-DIGIRR) together with the pDsRed-Golgi zebrafish embryos with both zebrafish and Tetraodon IL-1– were cotransfected into HEK293T, HepG2, and HeLa cells. Fur- expression plasmids induced significant (p , 0.01) activation of by guest on October 1, 2021 FIGURE 6. Tissue distribution and expression analyses of Tetraodon DIGIRR. A and B, RT-PCR data il- lustrating the DIGIRR distribution in control fish and in fish stimulated with LPS (1 mg/fish) for 12 h. RT- PCR was performed for 25 cycles using specific primers for DIGIRR and b-actin genes, and products from brain, heart, head kidney, spleen, li- ver, gill, intestine, skin, and muscle (A) and leukocytes from the blood, spleen, and head kidney (B) were loaded from left to right. C and D, The gene expression of DIGIRR af- ter LPS challenge was measured by quantitative PCR. E, Expression pro- file of DIGIRR in Tetraodon head kidney at different time points after LPS injection. DIGIRR mRNA ex- pression was measured using real- time quantitative PCR and is shown relative to the gene expression of b-actin (DIGIRR/b-actin). Values are mean 6 SEM. The quantitative PCR value was averaged from three du- plicates, each of which contained .10 fish. *p , 0.05, **p , 0.01. 2524 DISCOVERY OF A DIGIRR GENE FROM FISH

DIGIRR acts as a crucial negative regulator in both IL-1–mediated and LPS-mediated signaling. In addition, we also examined whether DIGIRR shared a conserved function across species. For this purpose, the inhibitory effect of DIGIRR on IL-1–mediated signaling was examined in HeLa cells. The result showed that ectopic expression of DIGIRR in human cells had no significant effect on IL-1–mediated activation of NF-kB. However, in the positive control cells in which mouse SIGIRR was overexpressed, IL-1–stimulated NF-kB activation was dramatically reduced (Fig. 8D). These results indicated that DIGIRR failed to regulate the IL-1–mediated NF-kB signaling in mammalian cells, suggesting functional diversity exists between DIGIRR and other family members, including SIGIRR, in different species. In vivo knockdown of DIGIRR by siRNA In an attempt to silence the expression of DIGIRR for in vivo functional studies, we developed a lentivirus-based siRNA delivery system for fish. Forty-six potential siRNAs, targeted to different regions of DIGIRR, were initially predicted by a template design program, from which four siRNAs (Table I) were selected for Downloaded from further evaluation. Of the four generated constructs (pSUPER- DIGIRR-1 to -4), the pSUPER-DIGIRR-3 was identified to be the most effective (75%) in inducing DIGIRR mRNA degradation (Fig. 9A). This siRNA encoding sequence was inserted into a lentiviral vector (PLB), and the construct was transfected into HEK293T cells to produce lentiviruses together with pCMV- http://www.jimmunol.org/ VSVG and pCMV-dr8.2. The titer of the concentrated viruses reached ∼5 3 106 TU/ml. The siRNA expression efficiency of the lentiviruses detected in HEK293T cells reached above 90% using the expression of GFP as an indicator (Fig. 9C), and the in- terference efficiency of this system reached above 74% (Fig. 9B). These results demonstrated that the recombinant lentiviruses could silence the DIGIRR expression in vitro. To investigate further whether this lentiviral system could be used for in vivo knock- by guest on October 1, 2021 down purpose, we administered the recombinant viruses into Tetraodon once a day for 3 d. Real-time PCR was performed to test the silencing of DIGIRR in different tissues. The result FIGURE 7. Subcellular distribution analysis of DIGIRR in cultured showed that DIGIRR expression could be dramatically downreg- cells. A, HeLa cells were transfected with plasmid pEGFPN1-DIGIRR, ulated (46–78%) in liver, kidney, and in leukocytes from head fixed, and stained with the cell membrane probe DiI, which demonstrates kidney, although it seems not obviously reduced in spleen (Fig. that DIGIRR was distributed in the cytoplasm, and no colocalized signals 10A). The development of this in vivo knockdown technology of DIGIRR were detected with DiI. B, HeLa cells were cotransfected with might represent a new method to deliver siRNAs into fish and pEGFPN1-DIGIRR and pDsRed-Golgi, which shows that the enhanced greatly benefit functional studies of DIGIRR in vivo. GFP-labeled DIGIRR merged well with the DsRed protein. C, HeLa cells were cotransfected with pcDNA6-myc-DIGIRR and pDsRed-Golgi, fixed, DIGIRR downregulates IL-1–induced inflammation and stained with FITC–anti-myc Abs, showing that DIGIRR merged well b with the DsRed protein. D, HeLa cells were transfected with pEGFPN1- The observation that DIGIRR suppressed IL-1 –induced and LPS- induced NF-kB signaling strongly suggested that DIGIRR may DIGIRR and stained with the Golgi tracer C5-ceramide, which demon- strates that DIGIRR merged well with the C5-ceramide signal. E and F, modulate inflammatory responses. To clarify this effect further, we HEK293T and HepG2 cells were cotransfected with pEGFPN1-DIGIRR explored the function of DIGIRR in IL-1–mediated inflammatory and pDsRed2-Golgi, showing DIGIRR and DsRed proteins were colocalized responses. We delivered the lentiviruses into Tetraodon once a day in the Golgi apparatus. Scale bars, 10 mm (original magnification 3600). for 3 d, followed by coadministration with recombinant IL-1b protein (see Supplemental Fig. 2) on the last day. Quantitative NF-kB signaling (Fig. 8A,8B). In contrast, coinjection of zebra- PCR data showed that, after induction by IL-1b stimulation, the fish embryos with IL-1–expression and DIGIRR-expression plas- expression of the proinflammatory cytokines IL-6 and IL-1b were mids severely reduced (p , 0.01) the activation of NF-kB sig- dramatically upregulated in the liver, kidney, and head kidney naling (Fig. 8A,8B). In parallel, little NF-kB activation could be leukocytes from DIGIRR-silenced fish relative to IL-1b–stimu- detected after injection of the mock plasmid (pcDNA6) in the lated control fish (Fig. 10B,10C). We also found that the DIGIRR- control group (Fig. 8A,8B). These results suggested that DIGIRR silenced fish with no external IL-1b stimulation had higher basal negatively regulated the IL-1–mediated NF-kB signaling pathway. expression of IL-6 and IL-1b compared with control fish (Fig. It has previously been shown that LPS can activate NF-kB sig- 10B,10C). These observations strongly suggested that DIGIRR- naling in zebrafish embryos (63), so we investigated whether silenced fish were hyperresponsive to either exogenous or en- DIGIRR regulated this LPS-mediated NF-kB pathway. As expected, dogenous IL-1b–stimulated inflammatory responses and further DIGIRR decreased (p , 0.01) the activation of NF-kB signaling confirmed that DIGIRR served as a negative regulator in IL-1– induced by LPS (Fig. 8C). Based on these data, we suggest that mediated signaling. The Journal of Immunology 2525 Downloaded from

FIGURE 8. Overexpression of DIGIRR downregulated IL-1–mediated and LPS-mediated NF-kB signaling. A and B, In zebrafish embryos, both Tet- k k raodon and zebrafish IL-1 constructs could activate NF- B signaling, and DIGIRR was able to inhibit IL-1–mediated NF- B signaling. Zebrafish embryos http://www.jimmunol.org/ were developed as a model to study NF-kB signaling by microinjecting overexpression plasmids (pCMV-ZeIL-1b and pCMV-TnIL-1b) along with firefly (pNFkB-TA-Luc) and Renilla (pRL-TK) reporter plasmids (10:1). At 24 h postmicroinjection, activation of NF-kB was assayed as described in Materials and Methods. C, DIGIRR was also demonstrated to downregulate LPS-mediated NF-kB signaling. D, Ectopic expression of DIGIRR in HeLa cells had no significant effect on IL-1b–mediated activation of NF-kB. Each bar represents the mean 6 SEM of at least five samples. The asterisks denote statistically significant differences between the indicated samples. **p , 0.01

NF-kB signaling of chimeras with amino acid replacements were unresponsive to IL-1 stimulation, suggesting that any mu- Two sites of conserved amino acids (Ser and Arg-Tyr) in the TIR tations in the Arg-Tyr amino acids might result in the inter- by guest on October 1, 2021 domains of the IL-1R family are suggested to be extremely im- ruption of receptor signaling. The chimera resembling SIGIRR portant for receptor signal transduction. Substitutions at these two (chimera-5), in which the Ser335 was replaced with Cys, also k sites by any other amino acid in molecules such as SIGIRR and IL- exhibited no effect on the activation of NF- B signaling, which 1RAcP results in the interruption of IL-1R–mediated NF-kB ac- is in agreement with the nonresponsiveness of the SIGIRR TIR tivation, and they become negative regulators in the signaling domain to IL-1 stimulation (25). Thus, crucial amino acid changes pathways. A series of chimeras with functional domains or amino (Ala419 and Leu420) are largely responsible for the loss of the acid replacements were constructed to investigate whether one sub- signal transduction ability of the DIGIRR TIR domain. stitution at the Arg-Tyr site may functionally transform DIGIRR. An IL-1–mediated NF-kB reporter assay was used for this pur- Discussion pose. As shown in Fig. 11, the chimera with the extracellular and The IL-1R family is a subgroup of the TIR superfamily. The IL-1Rs transmembrane portions of human IL-1RI and the cytoplasmic are characterized by extracellular Ig-like domains and a conserved region of DIGIRR (chimera-1) exhibited no response to IL-1 TIR domain in the intracellular region. In most cases, the extra- stimulation in the HEK293T cells, in which the NF-kB activi- cellular portion of IL-1R family members is highly conserved, with ties were kept at the basal level, similar to that of the control group three Ig-like domains, except for SIGIRR, which consists of only (empty vector pcDNA6). In contrast, a significant, ∼4-fold in- one Ig-like domain. In the current study, we identified a novel crease (p , 0.01) in NF-kB activation was observed in the pos- member of the IL-1R family that displayed a high degree of se- itive group, in which the human IL-1RI dramatically responded to quence homology in the cytoplasmic TIR domain but expressed IL-1. This result indicates that the DIGIRR TIR domain with an two Ig-like domains in the extracellular region. This novel mol- Arg-Tyr site mutation (Arg-Tyr was replaced by Ala-Leu) may ecule was referred to as DIGIRR based on this unique extracellular have lost its IL-1 signal transduction ability. To provide evidence structure. The novel DIGIRR protein also displayed another unique for this, another chimera with a “reverse” mutation at the Arg-Tyr characteristic not observed in other known IL-1R family members: site (chimera-3) was created, in which the Ala-Leu420 in the subcellular distribution within the Golgi apparatus. Thus, our study DIGIRR TIR domain was replaced with the two conserved amino has added a new member to this family (DIGIRR, which could also acids Arg and Tyr (equal to the consensus sequence in IL-1RI). be named TIR9 after SIGIRR/TIR8). The reporter assays clearly show that this chimera significantly Most characterized members of the TIR superfamily serve as (p , 0.01) restored its ability to activate IL-1–induced NF-kB positive regulators in various TLRs-mediated and IL-1–mediated signaling, in which a 1.8-fold increase in NF-kB activation was signaling pathways (1–3, 8), whereas negative regulators in these detected. However, both chimera-2 and chimera-4, with only one signaling pathways are rare. In recent years, several negative conserved amino acid replacement at the Arg-Tyr site (Leu420 regulators, such as IL-1RII, ST2, and SIGIRR, have been identi- replaced with Tyr, and Ala419 replaced with Arg, respectively), fied in humans and other mammals, which provide preliminary 2526 DISCOVERY OF A DIGIRR GENE FROM FISH Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021

FIGURE 10. DIGIRR-silenced fish showed enhanced inflammatory re- sponses to IL-1b. A, The DIGIRR expression could be dramatically downregulated in liver, kidney, and leukocytes from head kidney. B and C, DIGIRR-silenced fish showed greater induction of the proinflammatory cytokines IL-6 and IL-1b after administration of recombinant IL-1b. The lentivirus with siRNAs was repeatedly delivered into fish by muscle in- jection once a day for 3 d and followed by coadministration with recombinant IL-1b on the last day. Twenty-four hours after IL-1b chal- lenge, relative gene expression of DIGIRR, IL-1b, and IL-6 was calculated FIGURE 9. Screening of siRNAs targeting Tetraodon DIGIRR expres- and normalized to b-actin expression. Each sample was quantified by PCR, sion and detection of siRNAs produced by lentivirus. A, Screening of ef- and results are expressed as the mean of three independent experiments. fective siRNAs. Four different siRNAs against DIGIRR were designed to Expression level is normalized to the control group (“No siRNA + PBS”). target distinct parts of the DIGIRR mRNA and inserted into siRNA- Bars represent SEMs. Statistical significance is denoted with an asterisk as expressing vectors to construct four recombinant plasmids (pSUPER- compared with controls: *p , 0.05, **p , 0.01. DIGIRR-1 to -4). The HEK293T cells were cotransfected with pSUPER- DIGIRRs or control plasmid (pSUPER) together with overexpression plasmid pc6-DIGIRR. The inhibitory efficiency of siRNAs was measured insight into the origin, occurrence, and evolution of negative by real-time PCR. The pSUPER-DIGIRR-3 was found to be the most ef- regulatory mechanisms in TIR signaling pathways. Structurally, fective in silencing DIGIRR, so this siRNAs encoding sequence was these negative regulators share many similarities with their posi- then subcloned into the lentiviral vector PLB. Lentiviruses were pro- tive counterparts. However, distinct alterations or mutations in key duced by cotransfection with PLB, pCMV-VSVG, and pCMV-dr8.2 into HEK293T cells. B, The efficacy of DIGIRR knockdown by lentiviruses was also evaluated by real-time PCR. C, The siRNA expression efficiency of the lentiviruses detected in HEK293T cells exceeded 90%, as revealed by GFP Scale bars, 200 mm (original magnification 350). Means 6 SEM of results fluorescence under laser scanning confocal microscopy (Zeiss LSM 510). from three independent experiments are shown. **p , 0.01. The Journal of Immunology 2527

FIGURE 11. Evaluation of the functional amino acids site in the TIR domain of DIGIRR by construc- tion of chimeras with amino acid replacements. The NF-kB signaling of the chimeras was assayed in HEK293T cells. A, Schematic illustrations of the chi- mera human IL-1R plus DIGIRR constructs transfected for the NF-kB activation assay. B, Different chime- ras and the human IL-1RI were transfected into HEK293T cells along with reporter plasmids. At 24 h posttransfection, the cells were stimulated for 6 h with 5 ng/ml human recombinant IL-1b. NF-kB activation was assayed as described in Materials and Methods. The reporter assays show that chimera-1, with extra- cellular and transmembrane portions of human IL-1RI and the cytoplasmic region of DIGIRR, exhibited no response to IL-1 stimulation; both chimera-2 and chimera-4, with only one conserved amino acid re- placement each (Leu419→Tyr and Ala419→Arg, re- spectively), were unresponsive to IL-1 stimulation; Downloaded from chimera-3 (replaced with conserved amino acids, Ala- Leu420→Arg-Tyr) regained its ability to activate the IL-1–induced NF-kB signaling; chimera-5 (Ser335→ Cys) exhibited no effect on NF-kB signaling activation; human IL-1RI, as the positive control, was significantly responsive to IL-1 stimulation. **p , 0.01. http://www.jimmunol.org/

functional domains are observed in these molecules. Therefore, it SIGIRR were in turn three, two, and one. The IL-1Rs with three seems reasonable to suggest that these negative and positive reg- Ig-like domains were shown to be potent receptors that have ulators might be derived from a common ancestor but that nega- the ability to bind IL-1a, IL-1b, or IL-1R antagonist, whereas tive regulators have lost distinct functional elements. The SIGIRR SIGIRR, with only one Ig-like domain, was unable to bind these molecule may be an example. It appears to be derived from its cytokines. Thus, the decreased number of Ig-like domains among by guest on October 1, 2021 potent counterpart, the IL-1R. IL-1R possesses three Ig-like IL-1R, DIGIRR, and SIGIRR suggested that the DIGIRR may domains in its extracellular region and one intact TIR domain represent a link between the potent receptor (IL-1R) and the with two conserved amino acids sites (Ser and Arg-Tyr) in its negative regulator (SIGIRR) that lacks two extracellular Ig-like intracellular region, which are required for signal transduction in domains necessary to recognize ILs. It is generally believed that all TIRs (25, 27, 68). The SIGIRR, however, contains only one Ig- all three Ig-like domains in mammalian IL-1Rs are necessary for like domain in its extracellular portion, whereas the TIR domain IL-la or IL-1b binding (72), but an exception was found in fish. has two substitutions at the conserved sites of amino acids, For example, the trout IL-1RII, with only two Ig-like domains, Cys222 and Gln-Leu305. As a result, the SIGIRR lost the abi- had the ability to bind IL-1b (73), indicating that the fish DIGIRR lity directly to interact with IL-1a, IL-1b, or IL-1R antagonist with two Ig-like domains may still bind ILs. Functional studies are and to transduce signals from IL-agonist binding (24, 25, 71). It needed to evaluate this possibility. Such studies could also define is therefore possible that SIGIRR became a negative regulator which one of these two Ig-like domains is the more crucial for the through loss of two extracellular Ig-like domains and by two recognition of ligand. mutations in the TIR domain. This is supported by the isolation of Third, two conserved amino acids sites, Ser and Arg-Tyr, within DIGIRR from fishes, ancient lower vertebrates that are an in- the TIR domains of the IL-1R family important for the signaling dispensable model organism for evolutionary studies. of TIRs (25, 27, 68), were changed by two in SIGIRR and by one In this study, we suggest that DIGIRR bears the hallmarks of in DIGIRR, underscoring the possibility that DIGIRR is a tran- a “transitional molecule” between IL-1R and SIGIRR; in other sitional molecule between IL-1R and SIGIRR. In SIGIRR, the words, DIGIRR may be an early example of the transition from Ser and Arg-Tyr were changed to Cys and Gln-Leu, leading to potent receptor to negative regulator during vertebrate evolu- a receptor that cannot mediate signal transduction (24, 25, 27). tion. A model for the evolutionary relationships among IL-1R, However, only one amino acid site substitution occurred in the DIGIRR, and SIGIRR is proposed (Fig. 12). This conclusion is DIGIRR intracellular domain (Ala-Leu420 replaces the Arg- supported by several observations as discussed below. Tyr536), whereas Ser was conserved in IL-1R and other family First, DIGIRR shares sequence identity and functional domains members. Notably, the Tyr in the Arg-Tyr site was replaced with with SIGIRR and IL-1Rs, indicating the existence of evolutionary an aliphatic residue, such as Leu (in SIGIRR and DIGIRR) and relationships among these TIR superfamily members. Second, the Val (in IL-1RAcP). cysteines involved in the first Ig-like domain in DIGIRR, which Fourth, phylogenetic analysis revealed the genetic relationship were absent in SIGIRR, were well conserved in fish. This con- between DIGIRR and SIGIRR, and these two molecules were firmed that the two Ig-like domains were common structures in clustered to one branch, suggesting that DIGIRR and SIGIRR most fish DIGIRR proteins. Remarkably, the numbers of the Ig-like likely share a common ancestor molecule in addition to their domains in the extracellular region of IL-1R, DIGIRR, and genome synteny and gene organization. Fifth, the fish DIGIRRs 2528 DISCOVERY OF A DIGIRR GENE FROM FISH

flammatory signaling and inflammation. The results also indicate that once the ancient IL-1R molecule lost one Ig-like domain in its extracellular region or one conserved amino acid site in its in- tracellular TIR domain, it probably became a negative regulator, emphasizing the importance of any one Ig-like domain and the Ser/Arg-Tyr amino acids in the signaling pathway. To provide evidence for this, a series of human IL-1RI/DIGIRR chimeras with different amino acid substitutions at the Ser/Arg-Tyr sites in the TIR domain of DIGIRR were constructed for the signaling assay. The results demonstrate that the TIR domain of DIGIRR with the Arg-Tyr site mutation (Arg-Tyr replaced with Ala-Leu) lost its IL-1 signal transduction capacity. However, the TIR do- main of DIGIRR with the “reverse” replacement (Ala-Leu re- placed with Arg-Tyr) significantly regained its ability for IL-1 signal transduction. Furthermore, chimeras with the DIGIRR TIR domains in which only one amino acid was reversely replaced (Leu420 replaced with Tyr, or Ala419 replaced with Arg) at the Arg-Tyr site did not regain their signaling activity, implying that the alteration in either Arg or Tyr may result in signal interruption, Downloaded from FIGURE 12. A hypothetical model for the interpretation of DIGIRR as which is in agreement with previous reports that the sequence a “transitional molecule” between SIGIRR and IL-1R. A hypothesized “FWKXXRY” in the TIR domain may be a conserved motif ancient molecule, indicated as ancient IL-1R–like molecule containing among mammalian and avian IL-1RIs and is important for re- three Ig-like domains and one TIR domain, plays the role of a cytokine ceptor signal transduction (68). receptor. During gene duplication events, ancient IL-1R–like molecule The detailed mechanisms of SIGIRR in the negative regulation probably evolved into two classes: positive receptors and negative regu- of IL-1 signaling were addressed in several previous studies (24, lators. Positive receptors are represented by IL-1RI and IL-18R. These 71). The results demonstrated that SIGIRR exerted its inhibitory http://www.jimmunol.org/ molecules not only retained the three Ig-like domains and intact TIR do- role by disturbing the heterodimerization of IL-1R and IL-1RAcP main with no mutation among the two conserved amino acids sites (Ser and Arg-Tyr) but also inherited cytokine binding and signal transduction by binding to these molecules through the extracellular Ig domain. from the ancient IL-1R–like molecule. Another class of molecules was Also, the altered intracellular TIR domain inhibited signaling by found to have lost Ig-like domains extracellularly or TIR domain in- attenuating the recruitment of receptor-proximal signaling com- tracellularly, and the molecules have mutations at the conserved amino ponents, such as MyD88, TNFR-associated factor 6, and IRAK, to acids sites in the TIR domain. The appearance of DIGIRR in lower ver- the receptor (71). Most members of the IL-1R family, including tebrate fish is an evolutionary event resulting from the loss of one Ig-like SIGIRR and Tetraodon IL-1RI (data not shown), are distributed on domain and one site mutation (Arg-Tyr→Ala-Leu) during the genome the cellular membrane (74). In contrast, DIGIRR is localized in by guest on October 1, 2021 duplication of the ancient IL-1R–like molecule. With further evolution, the cytoplasm and may be a Golgi-distributed protein. Therefore, most probably in higher vertebrates, DIGIRR disappeared after losing the we presumed that there might be different posttranslational pro- second Ig-like domain and through another site mutation (Ser→Cys). It cessing and subcellular targeting for DIGIRR and SIGIRR. Most was thus transformed into the negative regulator SIGIRR. In addition, the ancient IL-1R–like molecule retained three Ig-like domains but lost the likely, DIGIRR suppresses IL-1R–mediated signaling pathways entire TIR domain to become the IL-1RII “decoy” receptor. by interacting with adapter proteins like MyD88 and TNFR- associated factor 6 or directly with the IL-1R complex through its Arg-Tyr–mutated TIR domain. Further investigation is needed have a putative 16–19 aa signal secretion peptide at the N-terminal to clarify the detailed molecular mechanisms of negative regula- region, whereas there is no predictable signal peptide in all tion of IL-1R signaling by DIGIRR. SIGIRRs. However, similar signal peptides were also found in There are still very few protocols available for functional studies other IL-1R subfamily members, such as IL-1RI, IL-1RII, IL- of genes in fish. As suggested in preceding reports, morpholino 1RAPL, TIGIRR, ST2, and IL-18R, providing further evidence for demonstrated effective loss of function of embryonic genes in an evolutionary link between IL-1R and DIGIRR and suggesting several species, such as zebrafish (75). However, technologies for that fish DIGIRR is translocated into the membrane through the suppression of genes in adult fish tissues are undeveloped. The a conserved pathway that is similar to IL-1R subfamily members. RNA interference (RNAi) method has emerged as a powerful Finally, IL-1R possesses three Ig-like domains and an entire TIR technique to inhibit the expression of specific genes in living cells. domain, with no amino acids substitutions, implying that IL-1 It is mediated by siRNAs produced from long dsRNAs of exog- binding and signal transduction requires these elements. It is the enous or endogenous origin by the endonuclease Dicer (76–79). loss of the Ig-like domains and crucial amino acid changes that Generally, siRNAs can be obtained by chemical synthesis, in- transformed SIGIRR into a negative regulator (24, 27, 71). Thus, terference plasmids, or viral plasmids (80, 81). It has also been the “transitional position” of DIGIRR between IL-1R and SIGIRR reported that short hairpin RNAs, precursors to siRNA, can be raises an intriguing question of whether DIGIRR still retains the expressed using lentivirus, allowing for RNAi in a variety of positive activity like IL-1R or has evolved into a negative regu- cell types (82). The lentiviruses, such as the HIV, are capable of lator similar to SIGIRR. infecting both dividing and nondividing cells and integrating into To address this question, the role of DIGIRR in IL-1b–mediated the host cell chromosome. Indeed, siRNAs produced by lentivirus and LPS-mediated NF-kB signaling, as well as IL-1b–induced are widely applied. In vitro delivery of siRNAs into cultured cells inflammation, was examined. We found that DIGIRR inhibited IL- can be achieved by several protocols, including transfection 1b–mediated and LPS-mediated NF-kB activation, and DIGIRR- with liposome (78). However, in vivo delivery appears far more silenced animals were hyperresponsive to IL-1b challenge. This difficult and therefore receives much attention. Recently, protocols suggested that DIGIRR served as a negative regulator in proin- for in vivo siRNA delivery have been developed or improved, The Journal of Immunology 2529 including i.v. or orally delivery synthetic siRNA or siRNA- recognition of pathogens by the innate immune system is defined by cooperation between toll-like receptors. Proc. Natl. Acad. Sci. USA 97: 13766–13771. expressing plasmids with various modifications (83). These pro- 13. Hemmi, H., O. Takeuchi, T. Kawai, T. Kaisho, S. Sato, H. Sanjo, M. Matsumoto, tocols have been used in mouse models and may prove beneficial K. Hoshino, H. Wagner, K. Takeda, and S. Akira. 2000. A Toll-like receptor for drug delivery in clinical application (80, 84–87). However, the recognizes bacterial DNA. Nature 408: 740–745. 14. Hoshino, K., O. Takeuchi, T. Kawai, H. Sanjo, T. Ogawa, Y. Takeda, K. Takeda, delivery of siRNAs into specific adult fish tissues is still elusive. In and S. Akira. 1999. Cutting edge: Toll-like receptor 4 (TLR4)-deficient mice are the current study, we have successfully used a lentivirus vector to hyporesponsive to lipopolysaccharide: evidence for TLR4 as the Lps gene deliver the DIGIRR-silencing siRNA into fish by i.m. adminis- product. J. Immunol. 162: 3749–3752. 15. Sims, J. E., M. A. Gayle, J. L. Slack, M. R. Alderson, T. A. Bird, J. G. Giri, tration. The results demonstrated that expression of DIGIRR could F. Colotta, F. Re, A. Mantovani, K. Shanebeck, et al. 1993. Interleukin 1 sig- be effectively reduced by 46–78% in immune-relevant tissues naling occurs exclusively via the type I receptor. Proc. Natl. Acad. Sci. USA 90: 6155–6159. (liver and kidney) and cells (leukocytes). This was effective in 16. Greenfeder, S. A., P. Nunes, L. Kwee, M. Labow, R. A. Chizzonite, and G. Ju. silencing DIGIRR, and suppression resulted in a hyperresponse 1995. Molecular cloning and characterization of a second subunit of the in- to IL-1b, strongly supporting the idea that DIGIRR functioned terleukin 1 receptor complex. J. Biol. Chem. 270: 13757–13765. 17. Parnet, P., K. E. Garka, T. P. Bonnert, S. K. Dower, and J. E. Sims. 1996. IL-1Rrp as a negative regulator in IL-1 signaling and inflammation. The is a novel receptor-like molecule similar to the type I interleukin-1 receptor and successful application of this lentivirus-based siRNAs delivery its homologues T1/ST2 and IL-1R AcP. J. Biol. Chem. 271: 3967–3970. in fish immune system could greatly benefit further functional 18. Torigoe, K., S. Ushio, T. Okura, S. Kobayashi, M. Taniai, T. Kunikata, T. Murakami, O. Sanou, H. Kojima, M. Fujii, et al. 1997. Purification and studies of both DIGIRR and other genes in fish immunity. characterization of the human interleukin-18 receptor. J. Biol. Chem. 272: In conclusion, we report in this study a novel DIGIRR gene from 25737–25742. 19. Werenskiold, A. K., U. Ro¨ssler, M. Lo¨wel, J. Schmidt, K. Heermeier, a model teleost fish, T. nigroviridis. Structural and functional M. T. Spanner, and P. G. Strauss. 1995. Bone matrix deposition of T1, a homo- characterizations showed it represents a previously unknown TIR logue of interleukin 1 receptors. Cell Growth Differ. 6: 171–177. family member that serves as a negative regulator in IL-1 sig- 20. Klemenz, R., S. Hoffmann, and A. K. Werenskiold. 1989. Serum- and onco- Downloaded from protein-mediated induction of a gene with sequence similarity to the gene naling. We propose that DIGIRR and SIGIRR might originate encoding carcinoembryonic antigen. Proc. Natl. Acad. Sci. USA 86: 5708–5712. from an ancient IL-1R–like molecule through the loss of one and 21. Thomassen, E., G. Kothny, S. Haas, J. Danescu, L. Hu¨ltner, P. Do¨rmer, and two extracellular Ig-like domains and mutation of intracellular Ser A. K. Werenskiold. 1995. Role of cell type-specific promoters in the de- velopmental regulation of T1, an interleukin 1 receptor homologue. Cell Growth and Arg-Tyr amino acids. DIGIRR might be an evolutionary Differ. 6: 179–184. “transitional molecule” between IL-1R and SIGIRR, representing 22. Born, T. L., D. E. Smith, K. E. Garka, B. R. Renshaw, J. S. Bertles, and

J. E. Sims. 2000. Identification and characterization of two members of a novel http://www.jimmunol.org/ the connector between the potent receptor and the negative reg- class of the interleukin-1 receptor (IL-1R) family. Delineation Of a new class of ulator. This study also demonstrates the importance of all Ig-like IL-1R-related proteins based on signaling. J. Biol. Chem. 275: 41528. domains and Ser and Arg-Tyr amino acids for IL-1R–mediated 23. Lovenberg, T. W., P. D. Crowe, C. Liu, D. T. Chalmers, X. J. Liu, C. Liaw, W. Clevenger, T. Oltersdorf, E. B. De Souza, and R. A. Maki. 1996. Cloning of signaling transduction. Our results not only add a new member to a cDNA encoding a novel interleukin-1 receptor related protein (IL 1R-rp2). J. the TIR superfamily but also contribute to a better understanding Neuroimmunol. 70: 113–122. of the evolutionary history of this family and its associated sig- 24. Wald, D., J. Qin, Z. Zhao, Y. Qian, M. Naramura, L. Tian, J. Towne, J. E. Sims, G. R. Stark, and X. Li. 2003. SIGIRR, a negative regulator of Toll-like receptor- naling network. interleukin 1 receptor signaling. Nat. Immunol. 4: 920–927. 25. Thomassen, E., B. R. Renshaw, and J. E. Sims. 1999. Identification and char- acterization of SIGIRR, a molecule representing a novel subtype of the IL-1R

Disclosures superfamily. Cytokine 11: 389–399. by guest on October 1, 2021 The authors have no financial conflicts of interest. 26. Garlanda, C., F. Riva, N. Polentarutti, C. Buracchi, M. Sironi, M. De Bortoli, M. Muzio, R. Bergottini, E. Scanziani, A. Vecchi, et al. 2004. Intestinal in- flammation in mice deficient in Tir8, an inhibitory member of the IL-1 receptor family. Proc. Natl. Acad. Sci. USA 101: 3522–3526. References 27. Garlanda, C., H. J. Anders, and A. Mantovani. 2009. TIR8/SIGIRR: an IL-1R/ 1. Takeda, K., T. Kaisho, and S. Akira. 2003. Toll-like receptors. Annu. Rev. TLR family member with regulatory functions in inflammation and T cell po- Immunol. 21: 335–376. larization. Trends Immunol. 30: 439–446. 2. Dinarello, C. A. 1996. Biologic basis for interleukin-1 in disease. Blood 87: 28. Dinarello, C. A. 1992. Role of interleukin-1 in infectious diseases. Immunol. Rev. 2095–2147. 127: 119–146. 3. Medzhitov, R. 2001. Toll-like receptors and innate immunity. Nat. Rev. Immunol. 29. Sims, J. E., and D. E. Smith. 2010. The IL-1 family: regulators of immunity. Nat. 1: 135–145. Rev. Immunol. 10: 89–102. 4. O’Neill, L. A. 2002. Toll-like receptor signal transduction and the tailoring of 30. O’Neill, L. A., and C. A. Dinarello. 2000. The IL-1 receptor/toll-like receptor innate immunity: a role for Mal? Trends Immunol. 23: 296–300. superfamily: crucial receptors for inflammation and host defense. Immunol. 5. Kyriakis, J. M., P. Banerjee, E. Nikolakaki, T. Dai, E. A. Rubie, M. F. Ahmad, Today 21: 206–209. J. Avruch, and J. R. Woodgett. 1994. The stress-activated protein kinase sub- 31. Kobayashi, K., L. D. Hernandez, J. E. Gala´n, C. A. Janeway, Jr., R. Medzhitov, family of c-Jun kinases. Nature 369: 156–160. and R. A. Flavell. 2002. IRAK-M is a negative regulator of Toll-like receptor 6. Ridley, S. H., S. J. Sarsfield, J. C. Lee, H. F. Bigg, T. E. Cawston, D. J. Taylor, signaling. Cell 110: 191–202. D. L. DeWitt, and J. Saklatvala. 1997. Actions of IL-1 are selectively controlled 32. Burns, K., S. Janssens, B. Brissoni, N. Olivos, R. Beyaert, and J. Tschopp. 2003. by p38 mitogen-activated protein kinase: regulation of prostaglandin H synthase- Inhibition of interleukin 1 receptor/Toll-like receptor signaling through the al- 2, metalloproteinases, and IL-6 at different levels. J. Immunol. 158: 3165–3173. ternatively spliced, short form of MyD88 is due to its failure to recruit IRAK-4. 7. Bird, T. A., J. M. Kyriakis, L. Tyshler, M. Gayle, A. Milne, and G. D. Virca. J. Exp. Med. 197: 263–268. 1994. Interleukin-1 activates p54 mitogen-activated protein (MAP) kinase/stress- 33. Kinjyo, I., T. Hanada, K. Inagaki-Ohara, H. Mori, D. Aki, M. Ohishi, H. Yoshida, activated protein kinase by a pathway that is independent of p21ras, Raf-1, and M. Kubo, and A. Yoshimura. 2002. SOCS1/JAB is a negative regulator of LPS- MAP kinase kinase. J. Biol. Chem. 269: 31836–31844. induced macrophage activation. Immunity 17: 583–591. 8. Boraschi, D., and A. Tagliabue. 2006. The interleukin-1 receptor family. Vitam. 34. Colotta, F., S. K. Dower, J. E. Sims, and A. Mantovani. 1994. The type II ‘decoy’ Horm. 74: 229–254. receptor: a novel regulatory pathway for interleukin 1. Immunol. Today 15: 562– 9. Alexopoulou, L., A. C. Holt, R. Medzhitov, and R. A. Flavell. 2001. Recognition 566. of double-stranded RNA and activation of NF-kappaB by Toll-like receptor 3. 35. Colotta, F., F. Re, M. Muzio, R. Bertini, N. Polentarutti, M. Sironi, J. G. Giri, Nature 413: 732–738. S. K. Dower, J. E. Sims, and A. Mantovani. 1993. Interleukin-1 type II receptor: 10. Hayashi, F., K. D. Smith, A. Ozinsky, T. R. Hawn, E. C. Yi, D. R. Goodlett, a decoy target for IL-1 that is regulated by IL-4. Science 261: 472–475. J. K. Eng, S. Akira, D. M. Underhill, and A. Aderem. 2001. The innate immune 36. Brint, E. K., D. Xu, H. Liu, A. Dunne, A. N. McKenzie, L. A. O’Neill, and response to bacterial flagellin is mediated by Toll-like receptor 5. Nature 410: F. Y. Liew. 2004. ST2 is an inhibitor of interleukin 1 receptor and Toll-like re- 1099–1103. ceptor 4 signaling and maintains endotoxin tolerance. Nat. Immunol. 5: 373–379. 11. Takeuchi, O., K. Hoshino, T. Kawai, H. Sanjo, H. Takada, T. Ogawa, K. Takeda, 37. Wu, X. Y., L. X. Xiang, L. Huang, Y. Jin, and J. Z. Shao. 2008. Characterization, and S. Akira. 1999. Differential roles of TLR2 and TLR4 in recognition of gram- expression and evolution analysis of Toll-like receptor 1 gene in pufferfish negative and gram-positive bacterial cell wall components. Immunity 11: 443– (Tetraodon nigroviridis). Int. J. Immunogenet. 35: 215–225. 451. 38. Roach, J. C., G. Glusman, L. Rowen, A. Kaur, M. K. Purcell, K. D. Smith, 12. Ozinsky, A., D. M. Underhill, J. D. Fontenot, A. M. Hajjar, K. D. Smith, L. E. Hood, and A. Aderem. 2005. The evolution of vertebrate Toll-like C. B. Wilson, L. Schroeder, and A. Aderem. 2000. The repertoire for pattern receptors. Proc. Natl. Acad. Sci. USA 102: 9577–9582. 2530 DISCOVERY OF A DIGIRR GENE FROM FISH

39. Meijer, A. H., S. F. Gabby Krens, I. A. Medina Rodriguez, S. He, W. Bitter, B. Ewa 63. Sepulcre, M. P., F. Alcaraz-Pe´rez, A. Lo´pez-Mun˜oz, F. J. Roca, J. Meseguer, Snaar-Jagalska, and H. P. Spaink. 2004. Expression analysis of the Toll-like re- M. L. Cayuela, and V. Mulero. 2009. Evolution of lipopolysaccharide (LPS) ceptor and TIR domain adaptor families of zebrafish. Mol. Immunol. 40: 773–783. recognition and signaling: fish TLR4 does not recognize LPS and negatively 40. Subramaniam, S., C. Stansberg, L. Olsen, J. Zou, C. J. Secombes, and regulates NF-kappaB activation. J. Immunol. 182: 1836–1845. C. Cunningham. 2002. Cloning of a Salmo salar interleukin-1 receptor-like 64. Alcaraz-Pe´rez, F., V. Mulero, and M. L. Cayuela. 2008. Application of the dual- cDNA. Dev. Comp. Immunol. 26: 415–431. luciferase reporter assay to the analysis of promoter activity in Zebrafish em- 41. Zou, J., P. S. Grabowski, C. Cunningham, and C. J. Secombes. 1999. Molecular bryos. BMC Biotechnol. 8: 81. cloning of interleukin 1beta from rainbow trout Oncorhynchus mykiss reveals no 65. Elbashir, S. M., J. Harborth, K. Weber, and T. Tuschl. 2002. Analysis of gene evidence of an ice cut site. Cytokine 11: 552–560. function in somatic mammalian cells using small interfering RNAs. Methods 26: 42. Huising, M. O., R. J. M. Stet, H. F. J. Savelkoul, and B. M. L. Verburg-van 199–213. Kemenade. 2004. The molecular evolution of the interleukin-1 family of cyto- 66. Kissler, S., P. Stern, K. Takahashi, K. Hunter, L. B. Peterson, and L. S. Wicker. kines; IL-18 in teleost fish. Dev. Comp. Immunol. 28: 395–413. 2006. In vivo RNA interference demonstrates a role for Nramp1 in modifying 43. Sangrador-Vegas, A., S. A. Martin, P. G. O’Dea, and T. J. Smith. 2000. Cloning susceptibility to type 1 diabetes. Nat. Genet. 38: 479–483. and characterization of the rainbow trout (Oncorhynchus mykiss) type II 67. Peddie, S., J. Zou, C. Cunningham, and C. J. Secombes. 2001. Rainbow trout interleukin-1 receptor cDNA. Eur. J. Biochem. 267: 7031–7037. (Oncorhynchus mykiss) recombinant IL-1beta and derived peptides induce mi- 44. Stansberg, C., S. Subramaniam, B. Collet, C. J. Secombes, and C. Cunningham. gration of head-kidney leucocytes in vitro. Fish Shellfish Immunol. 11: 697–709. 2005. Cloning of the Atlantic salmon (Salmo salar) IL-1 receptor associated 68. Heguy, A., C. T. Baldari, G. Macchia, J. L. Telford, and M. Melli. 1992. Amino protein. Fish Shellfish Immunol. 19: 53–65. acids conserved in interleukin-1 receptors (IL-1Rs) and the Drosophila toll 45. Stansberg, C., S. Subramaniam, L. Olsen, C. J. Secombes, and C. Cunningham. protein are essential for IL-1R signal transduction. J. Biol. Chem. 267: 2605– 2003. Cloning and characterisation of a putative ST2L homologue from Atlantic 2609. salmon (Salmo salar). Fish Shellfish Immunol. 15: 211–224. 69. Ito, A., T. Takii, T. Soji, and K. Onozaki. 1997. Endotoxin-induced upregulation 46. Matsuo, A., H. Oshiumi, T. Tsujita, H. Mitani, H. Kasai, M. Yoshimizu, of type I interleukin-1 receptor mRNA expression in hepatocytes of mice: role of M. Matsumoto, and T. Seya. 2008. Teleost TLR22 recognizes RNA duplex to cytokines. J. Interferon Cytokine Res. 17: 55–61. induce IFN and protect cells from birnaviruses. J. Immunol. 181: 3474–3485. 70. Penton-Rol, G., S. Orlando, N. Polentarutti, S. Bernasconi, M. Muzio, 47. Ribeiro, C. M., T. Hermsen, A. J. Taverne-Thiele, H. F. Savelkoul, and M. Introna, and A. Mantovani. 1999. Bacterial lipopolysaccharide causes rapid G. F. Wiegertjes. 2010. Evolution of recognition of ligands from Gram-positive shedding, followed by inhibition of mRNA expression, of the IL-1 type II re- bacteria: similarities and differences in the TLR2-mediated response between ceptor, with concomitant up-regulation of the type I receptor and induction of Downloaded from mammalian vertebrates and teleost fish. J. Immunol. 184: 2355–2368. incompletely spliced transcripts. J. Immunol. 162: 2931–2938. 48. Rebl, A., H. Rebl, S. Liu, T. Goldammer, and H. M. Seyfert. 2011. Salmonid Tollip 71. Qin, J. Z., Y. C. Qian, J. H. Yao, C. Grace, and X. X. Li. 2005. SIGIRR inhibits and MyD88 factors can functionally replace their mammalian orthologues in TLR- interleukin-1 receptor- and toll-like receptor 4-mediated signaling through dif- mediated trout SAA promoter activation. Dev. Comp. Immunol. 35: 81–87. ferent mechanisms. J. Biol. Chem. 280: 25233–25241. 49. Burge, C., and S. Karlin. 1997. Prediction of complete gene structures in human 72. Schreuder, H., C. Tardif, S. Trump-Kallmeyer, A. Soffientini, E. Sarubbi, genomic DNA. J. Mol. Biol. 268: 78–94. A. Akeson, T. Bowlin, S. Yanofsky, and R. W. Barrett. 1997. A new cytokine- 50. Falquet, L., M. Pagni, P. Bucher, N. Hulo, C. J. Sigrist, K. Hofmann, and receptor binding mode revealed by the crystal structure of the IL-1 receptor with

A. Bairoch. 2002. The PROSITE database, its status in 2002. Nucleic Acids Res. an antagonist. Nature 386: 194–200. http://www.jimmunol.org/ 30: 235–238. 73. Lo´pez-Castejo´n, G., M. P. Sepulcre, F. J. Roca, B. Castellana, J. V. Planas, 51. Schultz, J., F. Milpetz, P. Bork, and C. P. Ponting. 1998. SMART, a simple J. Meseguer, and V. Mulero. 2007. The type II interleukin-1 receptor (IL-1RII) of modular architecture research tool: identification of signaling domains. Proc. the bony fish gilthead seabream Sparus aurata is strongly induced after infection Natl. Acad. Sci. USA 95: 5857–5864. and tightly regulated at transcriptional and post-transcriptional levels. Mol. 52. Krogh, A., B. Larsson, G. von Heijne, and E. L. Sonnhammer. 2001. Predicting Immunol. 44: 2772–2780. transmembrane protein topology with a hidden Markov model: application to 74. Lech, M., C. Garlanda, A. Mantovani, C. J. Kirschning, D. Schlo¨ndorff, and complete genomes. J. Mol. Biol. 305: 567–580. H. J. Anders. 2007. Different roles of TiR8/Sigirr on toll-like receptor signaling 53. Sonnhammer, E. L., G. von Heijne, and A. Krogh. 1998. A hidden Markov in intrarenal antigen-presenting cells and tubular epithelial cells. Kidney Int. 72: model for predicting transmembrane helices in protein sequences. Proc. Int. 182–192. Conf. Intell. Syst. Mol. Biol. 6: 175–182. 75. Nasevicius, A., and S. C. Ekker. 2000. Effective targeted gene ‘knockdown’ in 54. Bendtsen, J. D., H. Nielsen, G. von Heijne, and S. Brunak. 2004. Improved zebrafish. Nat. Genet. 26: 216–220.

prediction of signal peptides: SignalP 3.0. J. Mol. Biol. 340: 783–795. 76. Fire, A., S. Xu, M. K. Montgomery, S. A. Kostas, S. E. Driver, and C. C. Mello. by guest on October 1, 2021 55. de Castro, E., C. J. Sigrist, A. Gattiker, V. Bulliard, P. S. Langendijk-Genevaux, 1998. Potent and specific genetic interference by double-stranded RNA in E. Gasteiger, A. Bairoch, and N. Hulo. 2006. ScanProsite: detection of PROSITE Caenorhabditis elegans. Nature 391: 806–811. signature matches and ProRule-associated functional and structural residues in 77. Tuschl, T., P. D. Zamore, R. Lehmann, D. P. Bartel, and P. A. Sharp. 1999. proteins. Nucleic Acids Res. 34(Web Server issue): W362. Targeted mRNA degradation by double-stranded RNA in vitro. Genes Dev. 13: 56. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: im- 3191–3197. proving the sensitivity of progressive multiple sequence alignment through se- 78. Elbashir, S. M., J. Harborth, W. Lendeckel, A. Yalcin, K. Weber, and T. Tuschl. quence weighting, position-specific gap penalties and weight matrix choice. 2001. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured Nucleic Acids Res. 22: 4673–4680. mammalian cells. Nature 411: 494–498. 57. Fang, W., J. Z. Shao, and L. X. Xiang. 2007. Molecular cloning and charac- 79. Hannon, G. J. 2002. RNA interference. Nature 418: 244–251. terization of IL-15R alpha gene in rainbow trout (Oncorhynchus mykiss). Fish 80. Sioud, M. 2004. Therapeutic siRNAs. Trends Pharmacol. Sci. 25: 22–28. Shellfish Immunol. 23: 119–127. 81. Brummelkamp, T. R., R. Bernards, and R. Agami. 2002. A system for stable 58. Jin, H. J., L. X. Xiang, and J. Z. Shao. 2007. Molecular cloning and identification expression of short interfering RNAs in mammalian cells. Science 296: 550–553. of macrophage migration inhibitory factor (MIF) in teleost fish. Dev. Comp. 82. Abbas-Terki, T., W. Blanco-Bose, N. De´glon, W. Pralong, and P. Aebischer. Immunol. 31: 1131–1144. 2002. Lentiviral-mediated RNA interference. Hum. Gene Ther. 13: 2197–2201. 59. Saitou, N., and M. Nei. 1987. The neighbor-joining method: a new method for 83. Aouadi, M., G. J. Tesz, S. M. Nicoloro, M. X. Wang, M. Chouinard, E. Soto, reconstructing phylogenetic trees. Mol. Biol. Evol. 4: 406–425. G. R. Ostroff, and M. P. Czech. 2009. Orally delivered siRNA targeting mac- 60. Kumar, S., K. Tamura, and M. Nei. 2004. MEGA3: Integrated software for rophage Map4k4 suppresses systemic inflammation. Nature 458: 1180–1184. Molecular Evolutionary Genetics Analysis and sequence alignment. Brief. Bio- 84. de Fougerolles, A., H. P. Vornlocher, J. Maraganore, and J. Lieberman. 2007. inform. 5: 150–163. Interfering with disease: a progress report on siRNA-based therapeutics. Nat. 61. Polentarutti, N., G. P. Rol, M. Muzio, D. Bosisio, M. Camnasio, F. Riva, C. Zoja, Rev. Drug Discov. 6: 443–453. A. Benigni, S. Tomasoni, A. Vecchi, et al. 2003. Unique pattern of expression 85. Kim, D. H., and J. J. Rossi. 2007. Strategies for silencing human disease using and inhibition of IL-1 signaling by the IL-1 receptor family member TIR8/ RNA interference. Nat. Rev. Genet. 8: 173–184. SIGIRR. Eur. Cytokine Netw. 14: 211–218. 86. Bumcrot, D., M. Manoharan, V. Koteliansky, and D. W. Sah. 2006. RNAi 62. Liu, Q., U. Berchner-Pfannschmidt, U. Mo¨ller, M. Brecht, C. Wotzlaw, H. Acker, therapeutics: a potential new class of pharmaceutical drugs. Nat. Chem. Biol. 2: K. Jungermann, and T. Kietzmann. 2004. A Fenton reaction at the endoplasmic 711–719. reticulum is involved in the redox control of hypoxia-inducible gene expression. 87. Whitehead, K. A., R. Langer, and D. G. Anderson. 2009. Knocking down bar- Proc. Natl. Acad. Sci. USA 101: 4302–4307. riers: advances in siRNA delivery. Nat. Rev. Drug Discov. 8: 129–138.