RelB/p50 Complexes Regulate -Induced YKL-40 Expression Reetika Bhardwaj, Jessie W. Yester, Sandeep K. Singh, Debolina D. Biswas, Michael J. Surace, Michael R. Waters, This information is current as Kurt F. Hauser, Zhenqiang Yao, Brendan F. Boyce and of October 1, 2021. Tomasz Kordula J Immunol 2015; 194:2862-2870; Prepublished online 13 February 2015; doi: 10.4049/jimmunol.1400874 http://www.jimmunol.org/content/194/6/2862 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2015/02/13/jimmunol.140087

Material 4.DCSupplemental http://www.jimmunol.org/ References This article cites 64 articles, 17 of which you can access for free at: http://www.jimmunol.org/content/194/6/2862.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 © 2015 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

RelB/p50 Complexes Regulate Cytokine-Induced YKL-40 Expression

Reetika Bhardwaj,* Jessie W. Yester,* Sandeep K. Singh,* Debolina D. Biswas,* Michael J. Surace,* Michael R. Waters,* Kurt F. Hauser,† Zhenqiang Yao,‡ Brendan F. Boyce,‡ and Tomasz Kordula*,x

The secreted protein, YKL-40, has been proposed as a biomarker of a variety of human diseases characterized by ongoing inflam- mation, including chronic neurologic pathologies such as multiple sclerosis and Alzheimer’s disease. However, inflammatory mediators and the molecular mechanism responsible for enhanced expression of YKL-40 remained elusive. Using several mouse models of inflammation, we now show that YKL-40 expression correlated with increased expression of both IL-1 and IL-6. Furthermore, IL-1 together with IL-6 or the IL-6 family cytokine, oncostatin M, synergistically upregulated YKL-40 expression

in both primary human and mouse in vitro. The robust cytokine-driven expression of YKL-40 in astrocytes required Downloaded from both STAT3 and NF-kB binding elements of the YKL-40 promoter. In addition, YKL-40 expression was enhanced by constitu- tively active STAT3 and inhibited by dominant-negative IkBa. Surprisingly, cytokine-driven expression of YKL-40 in astrocytes was independent of the p65 subunit of NF-kB and instead required subunits RelB and p50. Mechanistically, we show that IL-1– induced RelB/p50 complex formation was further promoted by oncostatin M and that these complexes directly bound to the YKL- 40 promoter. Moreover, we found that expression of RelB was strongly upregulated during inflammation in vivo and by IL-1 in

astrocytes in vitro. We propose that IL-1 and the IL-6 family of regulate YKL-40 expression during sterile inflammation http://www.jimmunol.org/ via both STAT3 and RelB/p50 complexes. These results suggest that IL-1 may regulate the expression of specific anti-inflammatory in nonlymphoid tissues via the canonical activation of the RelB/p50 complexes. The Journal of Immunology, 2015, 194: 2862–2870.

n the past several years, YKL-40 (also known as 3– viocytes and chondrocytes (4), promotes adhesion and migration like protein 1, human cartilage gp39, and breast regression of vascular smooth muscle and endothelial cells (5–7), and is I protein 39) has attracted growing attention as a marker of a migration factor for astrocytes (8). Surprisingly, unchallenged ongoing inflammation and oncogenic transformation (1). This YKL-40 knockout mice appear normal; however, when challenged secreted glycoprotein, expressed in both invertebrates and verte- with Ag, they display impaired Th2-dependent immune responses, by guest on October 1, 2021 brates, belongs to the 18-glycosyl-hydrolase family of proteins but show diminished fibrosis and tissue inflammation, decreased ac- lacks glycolytic activity (2, 3). Although its biological functions cumulation of M2 , and increased of CD4+ are not fully understood, it is expressed by many cell types, in- T cells and macrophages (9). In contrast, these mice display an cluding macrophages, , synoviocytes, chondrocytes, exacerbated response to experimental autoimmune encephalo- smooth muscle cells, endothelial cells, , and astrocytes, myelitis (EAE) (10) and enhanced inflammatory responses to suggesting that its biological effects are not restricted to a partic- hyperoxia (11), suggesting that YKL-40 differentially affects in- ular cell type. Indeed, YKL-40 stimulates proliferation of syno- flammatory responses depending on the type of immune activation and tissue involved. In agreement with the role of YKL-40 during inflammatory responses in mice, YKL-40 levels are elevated in *Department of Biochemistry and Molecular Biology, Virginia Commonwealth Uni- patients with a wide array of human diseases characterized by † versity School of Medicine, Richmond, VA 23298; Department of Pharmacology ongoing inflammation, including rheumatoid arthritis, athero- and Toxicology, Virginia Commonwealth University School of Medicine, Richmond, VA 23298; ‡Department of Pathology and Laboratory Medicine, University of sclerosis, type 2 diabetes, pelvic inflammatory disease, chronic Rochester Medical Center, Rochester, NY 14642; and xMassey Cancer Center, Vir- pancreatitis and secondary diabetes, severe pediatric , cir- ginia Commonwealth University, Richmond, VA 23298. rhosis, Crohn’s disease, and others (12–19). Elevated levels of Received for publication April 4, 2014. Accepted for publication January 20, 2015. YKL-40 are also present in the cerebrospinal fluids of patients This work was supported by National Institutes of Health/National Institute of Al- with a variety of acute and chronic neurologic pathologies, such as lergy and Infectious Diseases Grant 1R01AI093718 (to T.K.) and the National Insti- tute of Arthritis and Musculoskeletal and Skin Diseases/National Institutes of Health multiple sclerosis (MS), Alzheimer’s disease, viral encephalitis, Grant AR43510 (to B.F.B.). HIV-associated , brain infarction, and traumatic brain Address correspondence and reprint requests to Dr. Tomasz Kordula, Department of injury (20–23), with activated microglia and reactive astrocytes Biochemistry and Molecular Biology, Virginia Commonwealth University, Rich- both producing YKL-40 in the CNS. Thus, during the past decade, mond, VA 23298. E-mail address: [email protected] it has become evident that elevated YKL-40 expression correlates The online version of this article contains supplemental material. with both infection-induced inflammation and sterile inflamma- Abbreviations used in this article: ChIP, chromatin immunoprecipitation; DOX, doxycycline; EAE, experimental autoimmune encephalomyelitis; GBM, glioblas- tion, a paradigm triggered by physical, chemical, or metabolic toma multiforme; IKK, IkB kinase; MS, multiple sclerosis; NIK, NF-kB–inducing noxia. Accordingly, YKL-40 is also expressed by several solid kinase, OSM, oncostatin M; qPCR, quantitative PCR; sIL-6R, soluble IL-6R; TAT, tumors, such as osteosarcoma, ovarian carcinoma, and glioblas- transactivator of transcription. toma multiforme (GBM) (24, 25), and promotes and Copyright Ó 2015 by The American Association of Immunologists, Inc. 0022-1767/15/$25.00 radioresistance of GBM tumors and angiogenesis of breast and www.jimmunol.org/cgi/doi/10.4049/jimmunol.1400874 The Journal of Immunology 2863 colon cancer cells (26, 27). Because elevated expression of YKL- Irradiation 40 correlates with ongoing inflammation that is a component of Irradiations were carried out using an MDS Nordion gamma-cell 40 re- many diseases, YKL-40 has been proposed as a biomarker for search irradiator with a 137Cs source delivering a dose of 1.05 Gy/min to many pathologies, including cardiovascular disease, asthma, the head only. Mice received a dose of 20 Gy. arthritis, MS, Alzheimer’s disease, and many cancers, including Clinical samples GBM, melanoma, hepatocellular carcinoma, and breast and pan- creatic cancers (20, 24). Biopsy samples of oligodendroglioma tumors were obtained from Virginia Commonwealth University’s brain tumor bank (Department of Neurosur- Despite the numerous reports documenting elevated expression gery, Richmond, VA). Normal cortical tissue samples were obtained from of YKL-40, relatively little is known about the inflammatory healthy brain regions of patients with oligodendroglioma or epileptic mediators and specific molecular mechanisms that control its patients. The biopsy samples represent primary grade II and III and re- expression. Proinflammatory cytokines, including TNF and IL-1, current tumors. induce expression of YKL-40 in chondrocytes (28) and astro- Cell culture cytes (8), however, to a much lower extent than the conditioned media of macrophages (29). Both IL-1 and TNF are known to Human U373-MG cells were obtained from American Type Culture Collection (Manassas, VA). Mouse astrocytes were prepared as trigger a classical IkB kinase (IKK)g-dependent activation of NF- described previously (36). Human cortical astrocytes were prepared from kB, which involves IKKb-mediated phosphorylation and subse- fetal tissue provided by Advanced Bioscience Resources and were cul- quent degradation of IkBs, followed by the release and subsequent tured as described previously (37, 38). Primary human chondrocytes nuclear translocation of the p65/p50 NF-kB complexes (30). In were purchased from PromoCell (Heidelberg, Germany). Cells were stimulated with 25 ng/ml IL-1, 25 ng/ml OSM, 25 ng/ml IL-6, 10 ng/ml contrast, these cytokines do not efficiently activate a noncanonical TNF, or 25 ng/ml soluble IL-6R (sIL-6R) (all from R&D Systems, Minne- Downloaded from IKKg-independent NF-kB pathway, which involves NF-kB–in- apolis, MN). ducing kinase (NIK), IKKa-dependent processing of NF-kB2 p100, and generation of RelB/p52 complexes (31–33). Concor- Knockdown of target genes dantly, the p65/p50 complex has been proposed to mediate TNF- Expression of RelB, cRel, p50, STAT3, p65, and p52 mRNAs was down- and IL-1–induced YKL-40 expression in chondrocytes (28). In regulated using SmartPool siRNAs transfected into astrocytes using contrast however, p65/p50 also has been shown to recruit histone Dharmafect 1 (all from Dharmacon, Lafayette, CO). http://www.jimmunol.org/ deacetylases 1 and 2 to the YKL-40 promoter and to repress its Quantitative PCR expression in response to TNF in GBM cells (34). In addition, Total RNA was prepared utilizing TRIzol (Life Technologies, Carlsbad, IL-6 and oncostatin M (OSM) moderately upregulate YKL-40 CA). One microgram of RNA was reverse transcribed using the high- expression in human astrocytes, which requires STAT3 and for- capacity cDNA Archive kit (Life Technologies). Premixed primer- mation of a complex between STAT3 and NFI-X3 on the YKL-40 probe sets and TaqMan Universal PCR Master Mix (Life Technolo- promoter (8). However, profound activation of YKL-40 expression gies) were employed to examine mRNA levels. cDNAs were diluted 10-fold (for the target genes) or 100-fold (for GAPDH) and amplified observed during ongoing inflammatory processes in vivo has not using the ABI7900HT cycler. expression levels were normalized to been adequately recapitulated in the in vitro experiments. There- GAPDH and presented as a fold induction with the mean values 6 SEM.

fore, in this study, we set up experiments to identify molecular A Power SYBRGreen PCR kit (Life Technologies) was used for chro- by guest on October 1, 2021 mechanisms that govern YKL-40 expression during inflammation. matin immunoprecipitation quantitative PCR (qPCR) assays as described previously (8). Materials and Methods Western blotting Mice The cells were lysed in 10 mM Tris (pH 7.4), 150 mM sodium chloride, 1 mM EDTA, 0.5% Nonidet P-40, 1% Triton X-100, 1 mM sodium The doxycycline (DOX)-inducible, brain-specific HIV-Tat transgenic 1–86 orthovanadate, 0.2 mM PMSF, and protease inhibitor mixture (Roche mice have been described previously (35). Tat expression was induced Applied Science, Indianapolis, IN). Samples were separated using SDS- with chow containing 6 mg/g DOX (Harlan, Indianapolis, IN) for 1 wk. PAGE and transferred onto nitrocellulose membranes (Schleicher and C57BL/6 mice were obtained from The Jackson Laboratory. Mice were Schuell, Dassel, Germany). The anti–YKL-40, anti-RelB, anti-p65, anti- housed at Virginia Commonwealth University according to guidelines of p50, anti–b-tubulin, and anti-GAPDH Abs were obtained from Santa the Institutional Animal Care and Use Committee of Virginia Common- Cruz Biotechnology (Santa Cruz, CA), and anti-Stat3 Ab was obtained wealth University. The mouse protocols were approved by the Institutional from Cell Signaling Technology (Danvers, MA). Ag–Ab complexes were Animal Care and Use Committee. Male and female mice 6–16 wk of age visualized by ECL using Immobilon Western blotting kit (Millipore, were used. Temecula, CA). Turpentine- and LPS-induced inflammation Coimmunoprecipitation Turpentine abscesses were initiated by s.c. injection of pure gum spirits of Protein lysates (200–300 mg), prepared as described above, were pre- turpentine (50 ml) into anesthetized age-matched male and female mice (6– cleared with 10 ml protein G–Sepharose beads (GE Healthcare, Pittsburgh, 8 wk of age). For LPS-induced inflammation, mice were injected s.c. with PA) for 1 h. The lysates were then incubated with 2 mg anti-RelB or anti- 50 mgLPS(in50ml PBS). Mice were killed after 8 h, and the skin and p50 Abs overnight at 4˚C, and 25 ml protein G–Sepharose beads were underlying muscles at the site of injection were collected for mRNA analysis. added and incubated for 1 h at 4˚C. The beads were washed extensively with the lysis buffer, and immunoprecipitated proteins were eluted in Experimental autoimmune encephalomyelitis sample buffer at 95˚C for 5 min.

Each mouse received intradermally 300 mgMOG35–55 peptide (AnaSpec, Synthetic oligonucleotides Fremont, CA) emulsified in CFA containing 500 mg Mycobacterium tu- berculosis H37Ra (Difco, Detroit, MI) and i.p. 200 ng pertussis toxin (Enzo The following oligonucleotides were synthesized to introduce mutations in Life Sciences, Farmingdale, NY). An additional dose of 200 ng pertussis the YKL-40 promoter: 2669 NF-KB, 59-CTGAATTCGATAGCTGTCTT- toxin was administered 2 d postimmunization. A booster dose MOG35–55/ TCCCTCTAA-39 and 59-ACAGCTATCGAA TTCAGAATGCTTTAAGC-39; CFA/M. tuberculosis H37Ra was given 7 d postimmunization. Mice were and 2717 NF-KB, 59-ATCTCGAGAATAAAACAGAAGCAAAAT AG-39 examined daily (the first day following booster injection was assigned as day and 59-TTATTCTCGAGATAAAGAGAGAGGATCTT-39. The following 1) and weighed, and the severity of the disease was quantitated using a five- oligonucleotides were used in EMSA: 2669 NF-KB probe, 59-GATCT- point scale: 0, no symptoms; 1, limp tail; 2, limp tail with loss of righting; 3, TTCTTGGGAATTTCCC TGTCA-39 and 59-GATCTGACAGGGAAATT- paralysis of single hind limp; 4, paralysis of both hind limps; and 5, death. CCCAAGAAA-39;and2717 NF-KB probe, 59-GATCTCTTTATGG GAAT- Tissues were collected at day 28 for RNA analysis. TTCAAAACAA-39 and 59-GATCTTGTTTTGAAATTCCCATAAAGA-39. 2864 RelB REGULATES YKL-40

Nuclear extracts and EMSA domly chosen mice were used per experimental group. SPSS Statistics 21 software was used for statistical analyses. A Bonferroni posthoc test was Nuclear extracts were prepared as described previously (39). Briefly, used for one-way ANOVA comparisons, with p , 0.05 being considered dsDNA fragments were labeled by filling in the 59-protruding ends with 32 statistically significant. Independent sample Student t test was used for Klenow enzyme using [a-[ P]]dCTP (3000 Ci/mmol). Five micrograms unpaired observations. of nuclear extracts and ∼10 fmol (10,000 cpm) probe were used. Anti- RelB, anti-p50, and anti-p65 (Santa Cruz Biotechnology, Santa Cruz, CA) Abs were used for supershift studies. Results Inflammation induces YKL-40, IL-1, and IL-6 expression Chromatin immunoprecipitation assay To model the induction of YKL-40 expression during inflamma- Chromatin immunoprecipitation (ChIP) assays were performed as previously tion, we first used a turpentine mouse model of irritant-induced described (8) with the following exceptions. The cells were cross-linked with 1% formaldehyde for 10 min at 37˚C and then washed with ice-cold PBS acute inflammation, which is IL-1R–dependent (41). Turpentine containing 125 mM glycine. RelB and p50 Abs or nonimmune rabbit serum causes tissue destruction at the site of injection, local cytokine were used to precipitate cross-linked proteins. The following primers to the production, and infiltration of proinflammatory cells (38). The YKL-40 promoter were used in the qPCR: 59-GTGCAGCCGCCCCGTAG- local reaction is subsequently followed by a systemic response, 39 and 59-GCCTGAAACTGAGCGCTCC-39. including the production of acute-phase proteins in the liver (42). Plasmids and transfections Indeed, YKL-40 mRNA expression was robustly activated in tis- The pYKL(21300)Luc reporter was provided by Dr. M. Rehli (University sues surrounding the site of turpentine injection, and its induction of Regensberg, Regensberg, Germany) (40). The site-directed mutagenesis was accompanied by a very strong induction of both IL-1 and IL-6 was performed using QuikChange II site-directed mutagenesis kit (Stra- mRNAs (Fig. 1A). To corroborate these findings in a CNS model tagene, La Jolla, CA) and the oligonucleotides described above and pub- of sterile inflammation, expression of YKL-40, IL-1, and IL-6 was Downloaded from lished previously (8). Plasmids expressing dominant-negative IkBa and subsequently analyzed using a mouse EAE model of MS. Im- constitutively active STAT3 were described previously (8). Human astro- cytes were transiently transfected in 12-well clusters using the FuGENE6 munization of mice with MOG35–50 peptide resulted in the in- transfection reagent (Roche Applied Science). Twenty-four hours post- duction of EAE and modest upregulation of YKL-40 mRNA transfection, the cells were stimulated with IL-1 or OSM for 20 h, and cell expression in the spinal cords of animals (Fig. 1B). The magnitude extracts were prepared. Luciferase assays were performed using a dual of YKL-40 induction was comparable to the results previously luciferase reporter assay kit (Promega Corporation, Madison, WI). Lu- ciferase activities were normalized to Renilla luciferase activities. reported by others (10). In addition, expression of both IL-1 and http://www.jimmunol.org/ IL-6 mRNAs also was enhanced. We also found an increase in the Statistical analysis expression of glial fibrillary acidic protein mRNA, which indi- All experiments were repeated at least three times with consistent results. cated activation of astrocytes during EAE induction as reported Data are presented as mean 6 SEM. For mouse studies, four to six ran- previously (43). Subsequently, we analyzed expression of YKL- by guest on October 1, 2021

FIGURE 1. Sterile inflammation induces YKL-40, IL-1, and IL-6 expression. (A) Mice were injected s.c. with 50 ml turpentine or PBS (n = 6). Tissue at the site of injection was collected after 8 h, and the expression of YKL-40, IL-1, and IL-6 mRNA was ana- lyzed by qPCR. (B) EAE was induced in mice (n = 6) and scored as described in Materials and Methods (left panel). Ex- pression of YKL-40, IL-1, IL-6, and GFAP mRNA was analyzed in spinal cords at day 28 by qPCR (right panels). (C) Expression of YKL-40, IL-1, and IL-6 was analyzed in the hippocampus of control (CON) or irra- diated (IR) mice after 24 h. (D) Tat expres- sion was induced by DOX in Tat transgenic mice, and the expression of YKL-40 mRNA was analyzed by qPCR. (E) Expression of YKL-40, IL-1, and IL-6 mRNA was ana- lyzed by qPCR in biopsy samples of oligo- dendroglioma tumors (OD) and healthy brain tissue. Lines indicate median values. *p , 0.05, ***p , 0.001 Student t test. The Journal of Immunology 2865

40, IL-1, and IL-6 in the brains of mice exposed to gamma irra- sion in human astrocytes. Downregulation of STAT3 (Fig. 3C) diation, which is known to induce inflammation (44). In agree- dramatically diminished IL-1/OSM–induced YKL-40 mRNA ex- ment with the turpentine and EAE data, levels of YKL-40, IL-1, pression (Fig. 3A). Surprisingly, knockdown of p65 had no effect and IL-6 mRNAs were significantly higher in the irradiated brains on YKL-40 mRNA expression (Fig. 3A) but did drastically di- (Fig. 1C). We also used the HIV transactivator of transcription minish the expression of IL-8 mRNA, which is p65-dependent (TAT) transgenic mouse model in which induction of TAT protein (Fig. 3B). To identify the mechanism of cytokine-induced YKL- expression in the brain induces local sterile inflammation and 40 expression, putative NF-kB and STAT3 binding sites of the enhances expression of IL-1 and IL-6 (45). Using the TAT model, YKL-40 promoter were mutated, and the generated reporters were we found that YKL-40 mRNA expression also was enhanced upon analyzed in astrocytes (Fig. 3D). Although cytokine-driven ex- activation of TAT expression (Fig. 1D). Last, we analyzed samples pression of YKL-40 is p65 independent (Fig. 3A), mutation of the of human oligodendroglioma tumors and found that the expression proximal NF-kB site (2669 to 2660 bp) alone substantially di- of YKL-40, IL-1, and IL-6 mRNAs was strongly induced in minished activation of the reporter, whereas mutation of the distal comparison with normal brain (Fig. 1E). Similarly to sterile in- NF-kB site (2717 to 2708 bp) alone had no effect (Fig. 3D). In flammation, YKL-40 mRNA expression also coincides with IL-1 addition, mutation of the STAT3 site also drastically diminished and IL-6 mRNA expression during inflammation induced by reporter activity. In another approach, we overexpressed dominant- bacterial LPS (Supplemental Fig. 1). We conclude that YKL-40 negative IkBa and constitutively active STAT3 and analyzed ex- expression is enhanced during inflammation, which coincides with pression of the YKL-40 reporter (Fig. 3E). Constitutively active the increased expression of IL-1 and IL-6. STAT3 enhanced, whereas dominant-negative IkBa diminished cytokine responsiveness. Collectively, these results suggest that the IL-1 and OSM synergistically induce YKL-40 expression in Downloaded from proximal NF-kB and STAT3 elements of the YKL-40 promoter astrocytes are required for full responsiveness to IL-1 and OSM; however, In the nervous system, both astrocytes and microglia express and in contrast to previous reports (34), p65 is dispensable for this YKL-40 (8, 21), and these cell types also respond to a broad response. range of inflammatory stimuli (46). We used astrocytes as model RelB and p50 regulate cytokine-induced YKL-40 expression cells to study activation of YKL-40 expression during sterile inflammation. Primary human astrocytes were stimulated with To determine which component of the NF-kB complex regulates http://www.jimmunol.org/ IL-1 and IL-6 together with sIL-6R (because human fetal cytokine-driven YKL-40 expression, expression of p65, cRel, astrocytes express limited amounts of IL-6R (37)). We also in- RelB, p50, and p52 was effectively knocked down in astrocytes cluded OSM, a member of the IL-6 cytokine family that signals (Fig. 4B). Knockdown of either RelB or p50 significantly dimin- via OSM receptors abundantly expressed by astrocytes. Al- ished cytokine-induced YKL-40 mRNA expression, whereas though IL-1, IL-6/sIL-6R, or OSM alone only moderately acti- knockdown of p65, cRel, and p52 had no effect (Fig. 4A). This vated YKL-40 mRNA expression, stimulation of astrocytes with finding implicates both RelB and p50 in YKL-40 regulation. In- IL-1 together with OSM resulted in synergistic activation terestingly, although human astrocytes constitutively express low (Fig. 2A). Because IL-1–treated astrocytes express IL-6, their levels of RelB, IL-1 induced dramatic RelB protein accumulation by guest on October 1, 2021 costimulation with IL-1 and sIL-6R also caused synergistic in- in these cells (Fig. 4C). Similarly, RelB mRNA expression also duction of YKL-40 mRNA expression, however, to a lesser ex- was upregulated by IL-1 in mouse astrocytes (Fig. 4D). RelB tent than OSM alone. Similar effects of IL-1 and OSM on YKL- mRNA expression also was strongly activated in vivo by turpen- 40 mRNA expression also were observed at the protein level tine in an IL-1–dependent model of irritant-induced sterile in- (Fig. 2A, inset) and in mouse primary astrocytes (Fig. 2B). These flammation (Fig. 4E). Because TNF also efficiently activates NF- data together with our in vivo studies suggest that YKL-40 ex- kB, we tested whether, similarly to IL-1, TNF regulates YKL-40 pression is coordinately regulated by IL-1 and cytokines of the expression by a RelB-dependent mechanism. Although TNF and IL-6 cytokine family during sterile inflammation. OSM could not induce YKL-40 mRNA expression in astrocytes, YKL40 mRNA expression was RelB dependent in response to k STAT3 but not p65 (NF- B) regulate cytokine-induced YKL-40 these cytokines in U373 glioma cells (Supplemental Fig. 2). These expression data suggest that RelB regulates expression of YKL-40 in response NF-kB and STAT3 are the major transcription factors activated by to proinflammatory cytokines, such as IL-1 and TNF, which induce IL-1 and OSM, respectively (30, 47). Because p65 subunit of NF- canonical activation of RelB/p50 complexes. In contrast to astro- kB and STAT3 were previously implicated in cytokine-induced cytes and U373 cells, we found that basal expression of YKL-40 expression of YKL-40 (8, 34), we knocked down their expres- and RelB mRNA was very high in primary human chondrocytes

FIGURE 2. Proinflammatory cytokines induce YKL-40 ex- pression in astrocytes. Primary human (A) and mouse (B) astrocytes were stimulated with IL-1, IL-6, OSM, and sIL-6R for 18 h, and expression of YKL-40 mRNA was analyzed by qPCR. Secreted YKL-40 was analyzed by Western blotting in the culture medium (inset). ***p , 0.001 one-way ANOVA. 2866 RelB REGULATES YKL-40 Downloaded from http://www.jimmunol.org/

FIGURE 3. Induction of YKL-40 expression by IL-1 and OSM is STAT3 dependent but p65 independent. Human astrocytes were transfected with the indicated small interfering (si)-RNAs. Forty hours posttransfection, cells were stimulated with IL-1 and OSM for 18 h, and expression of YKL-40 (A), IL-8 (B), and STAT3 (C) was analyzed by qPCR. (D) Astrocytes were transfected with the indicated reporters and stimulated with IL-1 and OSM for 18 h, and luciferase and renilla activities were determined. Data are presented in comparison with the induced wild-type reporter (set as 100%). (E) Astrocytes were transfected with pYKL(21300)Luc and plasmids expressing either vector, dominant-negative IkB(IkBSR), or constitutively active (CA-STAT3). Cells were stimulated with IL-1 and OSM for 18 h and processed as described in (D). **p , 0.01, ***p , 0.001 one-way ANOVA. and not stimulated by IL-1 or OSM (Supplemental Fig 3). In these RelB/p50 complexes can directly bind to the proximal NF-kB by guest on October 1, 2021 cells, basal expression of YKL-40 is RelB-, p65-, and STAT3- site of the YKL-40 promoter and are essential for the cytokine- independent, implicating other unknown factors. Cumulatively, induced YKL-40 expression. these data suggest that in cells expressing relatively low levels of OSM enhances IL-1–induced RelB/p50 heterodimer formation YKL-40, such as astrocytes, cytokine-driven RelB promotes YKL- 40 induction in vitro and in vivo. OSM and IL-1 efficiently regulated YKL-40 expression via STAT3 (Fig. 3A) and OSM promoted the recruitment of IL-1– RelB/p50 complexes bind to the YKL-40 promoter induced p50 to the YKL-40 promoter (Fig. 5B). Because YKL- To determine whether RelB regulates YKL-40 expression directly 40 expression depends on RelB/p50 (Fig. 4A), we subsequently or indirectly, binding of RelB and p50 to the YKL-40 promoter asked whether OSM promotes formation of IL-1–induced RelB/ was analyzed by ChIP (Fig. 5A). These experiments were per- p50 heterodimers. RelB and p50 were immunoprecipitated from formed in U373 glioma cells, which similarly to human and cytokine-treated cells and analyzed for the presence of coim- mouse astrocytes upregulate expression of both YKL-40 and munoprecipitated RelB and p50 (Fig. 6). Indeed, p50/RelB RelB in response to IL-1 and OSM, and this cytokine-induced complexes were formed in response to IL-1. More importantly, expression is diminished by the knockdown of p50 and RelB OSM significantly enhanced IL-1–induced formation of the p50/ (Supplemental Fig. 4). Although RelB and p50 did not bind to RelB complexes. These data suggest that in addition to STAT3 the YKL-40 promoter in unstimulated U373 cells, binding of activation, OSM enhances YKL-40 expression by promoting RelB and p50 was apparent in cytokine-treated cells (Fig. 5B), formation of the RelB/p50 complexes, which bind to the YKL- suggesting that RelB/p50 complexes directly regulate YKL-40 40 promoter. expression. To determine whether IL-1/OSM treatment induces RelB/p50 binding to the previously identified NF-kBsites, Discussion EMSAs were performed using oligonucleotides containing distal Although YKL-40 is upregulated in a broad range of human and proximal NF-kB sites. In agreement with the mutational diseases associated with ongoing sterile inflammation, its bio- analysis that showed the importance of the proximal NF-kBsite logical functions still remain elusive (48–50). Nevertheless, ex- (Fig. 3D), strong protein binding to the proximal, but not distal, acerbated EAE (10) and enhanced responses to hyperoxia (11) in NF-kB site of the YKL-40 promoter was induced by IL-1 alone YKL-40 knockout mice suggest that cytokine-induced YKL-40 or together with OSM (Fig. 5C). Cytokine stimulation facilitated is needed for the proper resolution of inflammation. Once in- the binding of protein complexes to the proximal NF-kBelement flammation is resolved, YKL-40 expression ceases. In contrast, in vitro, which supershifted with anti-p65, anti-RelB, and anti- chronic inflammation leads to continuous aberrant upregulation p50 Abs (Fig. 5D), indicating that the proximal NF-kBsitecan of YKL-40 expression, which is continuously driven by proin- bind both p65/p50 and RelB/p50 complexes. We conclude that flammatory cytokines. Our data from the turpentine, TAT The Journal of Immunology 2867 Downloaded from

FIGURE 4. Cytokine-induced RelB/p50 complexes regulate YKL-40 expression. Human astrocytes were transfected with the indicated small interfering (si)-RNAs. Forty hours posttransfection, cells were stimulated with IL-1 and OSM for 18 h. Expression of YKL-40 (A) and p100 and cRel (B, right panel) B C

was analyzed by qPCR. Expression of p65, RelB, p50, and b-tubulin was analyzed by Western blotting ( , left panel). ( ) Human astrocytes were http://www.jimmunol.org/ stimulated with IL-1 and OSM for the indicated times, and expression of RelB and GAPDH was analyzed by Western blotting. (D) Mouse astrocytes were stimulated with IL-1 and OSM for 18 h, and expression of RelB was analyzed by qPCR. (E) Mice were injected s.c. with 50 ml turpentine or PBS (n = 6). Tissue at the site of injection was collected after 24 h, and the expression of RelB mRNA was analyzed by qPCR. Lines indicate median values. **p , 0.01, ***p , 0.001 one-way ANOVA. overexpression, and EAE mouse models together with human oli- that YKL-40 also inhibits NF-kB activation and expression of godendroglioma data demonstrate that YKL-40 expression IL-6, IL-8, and MCP-1 by skeletal muscle cells by a protease correlates with the expression of both IL-1b and IL-6. Impor- activated receptor 2–dependent mechanism (57). by guest on October 1, 2021 tantly, IL-1a and IL-1b are both critical initiators of sterile It is generally accepted that IL-1 triggers a classical NF-kB inflammation released in response to broad range of danger- pathway in many cell types, which leads to the induction of p65/ associated molecular patterns. IL-1b is processed from its in- p50 target genes, including those encoding proinflammatory active precursor in response to inflammasome activation and cytokines such as IL-8 and IL-6 (58). It is also accepted that RelB subsequently secreted from activated cells, whereas active in- is activated in lymphoid cells, such as dendritic cells, by a non- tracellular IL-1a, which does not require processing, is liberated canonical NF-kB pathway and generation of RelB/p52 complexes from damaged cells (51). Because IL-1 efficiently activates ex- that are important for proper dendritic cell functions (59). This pression of IL-6 in many cell types (52, 53), it is not surprising noncanonical NF-kB activation pathway is activated by ligands that these two cytokines are found at the sites of local inflam- such as CD40 and lymphotoxin-b (60) but cannot be activated by mation. The synergistic upregulation of YKL-40 expression by IL-1. However, it recently has been recognized that in dendritic IL-1 and IL-6 (or OSM) in in vitro experiments therefore likely cells, RelB also can be activated by a canonical pathway in re- recapitulates the regulation of YKL-40 in vivo. Similarly to the sponse to TNF, LPS, and CpG, and this activation results in RelB/ regulation of YKL-40, IL-1 and IL-6 (or OSM) are known to p50 complex formation (60). This canonical RelB activation control expression of acute-phase proteins in the liver, which are depends on TRAF6 (61), which is further supported by findings released to the blood to limit inflammation-associated damage that the RelB/p50 complexes are bound to IkBa and IkBε in the and allow for the return to homeostasis (54). Thus, acute cyto- cytoplasm (60). It also has been suggested that canonical activa- kine-driven expression of YKL-40 is likely highly beneficial and tion of RelB in dendritic cells is possible because of a higher level allows for proper resolution of inflammation. Nevertheless, the of RelB expression in these cells and the constitutive activation of mechanism by which YKL-40 limits inflammation remains noncanonical pathway (60). The physiological importance of the elusive because of limited understanding of YKL-40 receptors RelB/p50 complexes is strongly supported by a more deleterious and their signaling. In fact, YKL-40 induces the interaction of phenotype of RelB-deficient than the NIK-deficient mice (62), avb3 integrins with syndecan-1 in endothelial cells (27), and it suggesting that RelB has NIK-independent functions that likely activatesERK,AKT,andWnt/b-catenin signaling in macro- can be attributed to RelB/p50 complexes. Nevertheless, the phages via IL-13Ra2–dependent mechanism (55) and activates mechanism of specific RelB/p50-dependent gene regulation is not ERK, AKT, and p38 via protease activated receptor 2 in bron- clear because RelB/p50 and p65/p50 complexes bind the same chial smooth muscle cells (56). It remains to be established regulatory sequences. whether these proposed receptors/mechanisms are important for Although existence of RelB/p50 complexes in nonlymphoid the biological functions of YKL-40, which are associated with tissues has been shown more than a decade ago (33), the function of inflammation. However, in agreement with the proposed role of these complexes remained elusive. Our data clearly demonstrate YKL-40 in limiting inflammation, it has recently been shown that RelB expression is strongly upregulated during sterile and 2868 RelB REGULATES YKL-40 Downloaded from

FIGURE 5. RelB/p50 complexes bind at the distal NF-kB site of YKL-40 promoter. (A) YKL-40 promoter. NF-kB sites are indicated by gray boxes. http://www.jimmunol.org/ Positions of ChIP primers and probes used in EMSA (2717 and 2669 probe) are indicated. (B) ChIP was performed using chromatin prepared from U373 cells treated with IL-1 and OSM for 2h. Binding of p50 and RelB to the YKL-40 promoter was analyzed using the Abs described in Materials and Methods. NRS indicates normal rabbit serum used for immunoprecipitation. Results are shown as normalized binding (binding of NRS-immunoprecipitated untreated samples were set as 1). Experiments were performed three times. (C and D) Nuclear extracts were prepared from U373, and glioma cells were stimulated with IL-1 and OSM for 8 h. The binding was then analyzed by EMSA using the 32P-labeled oligonucleotide probes derived from the 59-flanking region of the YKL-40 (2717 NF-kB and 2669 NF-kB, as indicated). (C) Binding to the 2717 NF-kB and 2669 NF-kB elements in U373 glioma cells. (D) Binding to the 2669 NF-kB probe in U373 cells. Specific Abs or NRS were added to the binding reaction. Asterisk indicates supershifted complexes. *p , 0.05, ***p , 0.001 two-way ANOVA. by guest on October 1, 2021

LPS-induced inflammation in vivo, which is likely mediated by p50 complexes by a mechanism, which currently remains un- IL-1. Furthermore, IL-1 strongly activates expression of RelB in clear, but these complexes do not contain STAT3 (data not astrocytes in vitro. RelB and YKL-40 expression also can be shown). upregulated strongly in cells responding to TNF. RelB subse- In summary, we propose that inflammation leads to the ex- quently forms complexes with p50 and activation of these com- pression of YKL-40, which is induced by IL-1 and cytokines of plexes is controlled by IkBa. Once the RelB/p50 complexes are the IL-6 family. Mechanistically, both STAT3 and the RelB/p50 activated, they regulate expression of YKL-40 and likely other complexes drive the expression of YKL-40 via the elements RelB/p50-responsive genes. Thus, it appears that RelB/p50 com- within the YKL-40 promoter. plexes may play critical role(s) during acute inflammation that normally leads to a return to homeostasis. However, chronic in- flammation also may lead to pathological RelB/p50-dependent . Indeed, YKL-40 recently has been proposed as a marker of a particularly aggressive mesenchymal subtype of GBM (63). In agreement with our current findings indicating that RelB regulates YKL-40 expression, RelB is strongly expressed in mesenchymal subtype of GBM and promotes expression of mes- enchymal genes, including YKL-40 (63). Furthermore, loss of RelB expression attenuates GBM cell survival, motility, and in- vasion (64). Our previous (8) and current data suggest that activation of YKL-40 expression requires both STAT3 and RelB. Although mechanistic details of YKL-40 promoter activation are not clear, robust cytokine-induced YKL-40 expression involves induction of RelB expression, binding of both STAT3 and RelB/p50 complexes to the YKL-40 promoter, and acetylation of histo- FIGURE 6. OSM enhances IL-1–induced RelB/p50 heterodimer for- nes. It remains to be established whether increased STAT3- mation. Human astrocytes were stimulated with IL-1 or OSM for 8 h, p50 dependent acetylation of histones (and opening of chromatin) or RelB was immunoprecipitated, and coimmunoprecipitated p50 and allows for efficient binding of newly formed RelB/p50 com- RelB were detected by Western blotting, as indicated. Expression of RelB plexes. Surprisingly, OSM also promotes formation of the RelB/ and p50 in the lysates is shown the lower panels (lysate). The Journal of Immunology 2869

Acknowledgments modulates the biological activity of basic fibroblast growth factor. Am. J. Pathol. 173: 130–143. We thank Dr. Kristoffer Valerie (Virginia Commonwealth University) for 22. Colton, C. A., R. T. Mott, H. Sharpe, Q. Xu, W. E. Van Nostrand, and irradiating mice. M. P. Vitek. 2006. Expression profiles for alternative activation genes in AD and in mouse models of AD. J. Neuroinflammation 3: 27. 23. Comabella, M., M. Ferna´ndez, R. Martin, S. Rivera-Vallve´, E. Borra´s, C. Chiva, Disclosures E. Julia`, A. Rovira, E. Canto´, J. C. Alvarez-Cermen˜o, et al. 2010. Cerebrospinal The authors have no financial conflicts of interest. fluid chitinase 3-like 1 levels are associated with conversion to multiple sclerosis. Brain 133: 1082–1093. 24. Kzhyshkowska, J., A. Gratchev, and S. Goerdt. 2007. Human and chitinase-like proteins as indicators for inflammation and cancer. Biomark. References Insights 2: 128–146. 1. Roslind, A., and J. S. Johansen. 2009. YKL-40: a novel marker shared by 25. Ling, H., and A. D. Recklies. 2004. The chitinase 3-like protein human cartilage chronic inflammation and oncogenic transformation. Methods Mol. Biol. 511: glycoprotein 39 inhibits cellular responses to the inflammatory cytokines 159–184. interleukin-1 and tumour necrosis factor-a. Biochem. J. 380: 651–659. 2. Kawamura, K., T. Shibata, O. Saget, D. Peel, and P. J. Bryant. 1999. A new 26. Shao, R., K. Hamel, L. Petersen, Q. J. Cao, R. B. Arenas, C. Bigelow, B. Bentley, family of growth factors produced by the fat body and active on Drosophila and W. Yan. 2009. YKL-40, a secreted glycoprotein, promotes tumor angio- imaginal disc cells. Development 126: 211–219. genesis. Oncogene 28: 4456–4468. 3. Hakala, B. E., C. White, and A. D. Recklies. 1993. Human cartilage gp-39, 27. Francescone, R. A., S. Scully, M. Faibish, S. L. Taylor, D. Oh, L. Moral, W. Yan, a major secretory product of articular chondrocytes and synovial cells, is B. Bentley, and R. Shao. 2011. Role of YKL-40 in the angiogenesis, radio- a mammalian member of a chitinase protein family. J. Biol. Chem. 268: 25803– resistance, and progression of glioblastoma. J. Biol. Chem. 286: 15332–15343. 25810. 28. Recklies, A. D., H. Ling, C. White, and S. M. Bernier. 2005. Inflammatory 4. De Ceuninck, F., S. Gaufillier, A. Bonnaud, M. Sabatini, C. Lesur, and cytokines induce production of CHI3L1 by articular chondrocytes. J. Biol. P. Pastoureau. 2001. YKL-40 (cartilage gp-39) induces proliferative events in Chem. 280: 41213–41221. cultured chondrocytes and synoviocytes and increases glycosaminoglycan syn- 29. Bonneh-Barkay, D., S. J. Bissel, J. Kofler, A. Starkey, G. Wang, and C. A. Wiley. thesis in chondrocytes. Biochem. Biophys. Res. Commun. 285: 926–931. 2012. and macrophage regulation of YKL-40 expression and cellular 5. Nishikawa, K. C., and A. J. Millis. 2003. gp38k (CHI3L1) is a novel adhesion response in neuroinflammation. Brain Pathol. 22: 530–546. Downloaded from and migration factor for vascular cells. Exp. Cell Res. 287: 79–87. 30. Hayden, M. S., and S. Ghosh. 2008. Shared principles in NF-kB signaling. Cell 6. Shackelton, L. M., D. M. Mann, and A. J. Millis. 1995. Identification of a 38- 132: 344–362. kDa heparin-binding glycoprotein (gp38k) in differentiating vascular smooth 31. Pomerantz, J. L., and D. Baltimore. 2002. Two pathways to NF-kB. Mol. Cell muscle cells as a member of a group of proteins associated with tissue remod- 10: 693–695. eling. J. Biol. Chem. 270: 13076–13083. 32. Oeckinghaus, A., M. S. Hayden, and S. Ghosh. 2011. Crosstalk in NF-kB sig- 7. Malinda, K. M., L. Ponce, H. K. Kleinman, L. M. Shackelton, and A. J. Millis. naling pathways. Nat. Immunol. 12: 695–708. 1999. Gp38k, a protein synthesized by vascular smooth muscle cells, stimulates 33. Derudder, E., E. Dejardin, L. L. Pritchard, D. R. Green, M. Korner, and V. Baud.

directional migration of human umbilical vein endothelial cells. Exp. Cell Res. 2003. RelB/p50 dimers are differentially regulated by tumor necrosis factor-a http://www.jimmunol.org/ 250: 168–173. and lymphotoxin-b receptor activation: critical roles for p100. J. Biol. Chem. 8. Singh, S. K., R. Bhardwaj, K. M. Wilczynska, C. I. Dumur, and T. Kordula. 278: 23278–23284. 2011. A complex of nuclear factor I-X3 and STAT3 regulates astrocyte and 34. Bhat, K. P., C. E. Pelloski, Y. Zhang, S. H. Kim, C. deLaCruz, M. Rehli, and glioma migration through the secreted glycoprotein YKL-40. J. Biol. Chem. 286: K. D. Aldape. 2008. Selective repression of YKL-40 by NF-kB in glioma cell 39893–39903. lines involves recruitment of histone deacetylase-1 and -2. FEBS Lett. 582: 9. Lee, C. G., D. Hartl, G. R. Lee, B. Koller, H. Matsuura, C. A. Da Silva, 3193–3200. M. H. Sohn, L. Cohn, R. J. Homer, A. A. Kozhich, et al. 2009. Role of breast 35. Bruce-Keller, A. J., J. Turchan-Cholewo, E. J. Smart, T. Geurin, A. Chauhan, regression protein 39 (BRP-39)/chitinase 3-like-1 in Th2 and IL-13-induced R. Reid, R. Xu, A. Nath, P. E. Knapp, and K. F. Hauser. 2008. Morphine causes tissue responses and apoptosis. J. Exp. Med. 206: 1149–1166. rapid increases in glial activation and neuronal injury in the striatum of inducible 10. Bonneh-Barkay, D., G. Wang, W. A. Laframboise, C. A. Wiley, and S. J. Bissel. HIV-1 Tat transgenic mice. Glia 56: 1414–1427. 2012. Exacerbation of experimental autoimmune encephalomyelitis in the ab- 36. Stiene-Martin, A., R. Zhou, and K. F. Hauser. 1998. Regional, developmental,

sence of breast regression protein 39/chitinase 3-like 1. J. Neuropathol. Exp. and cell cycle-dependent differences in m, d,andk-opioid receptor expression by guest on October 1, 2021 Neurol. 71: 948–958. among cultured mouse astrocytes. Glia 22: 249–259. 11. Sohn, M. H., M. J. Kang, H. Matsuura, V. Bhandari, N. Y. Chen, C. G. Lee, and 37. Kordula, T., R. E. Rydel, E. F. Brigham, F. Horn, P. C. Heinrich, and J. Travis. J. A. Elias. 2010. The chitinase-like proteins breast regression protein-39 and 1998. Oncostatin M and the interleukin-6 and soluble interleukin-6 receptor YKL-40 regulate hyperoxia-induced acute lung injury. Am. J. Respir. Crit. Care complex regulate a1-antichymotrypsin expression in human cortical astrocytes. Med. 182: 918–928. J. Biol. Chem. 273: 4112–4118. 12. Johansen, J. S., M. Stoltenberg, M. Hansen, A. Florescu, K. Hørslev-Petersen, 38. Harikumar, K. B., J. W. Yester, M. J. Surace, C. Oyeniran, M. M. Price, I. Lorenzen, and P. A. Price. 1999. Serum YKL-40 concentrations in patients W. C. Huang, N. C. Hait, J. C. Allegood, A. Yamada, X. Kong, et al. 2014. K63- with rheumatoid arthritis: relation to disease activity. Rheumatology 38: 618– linked polyubiquitination of transcription factor IRF1 is essential for IL-1‑in- 626. duced production of chemokines CXCL10 and CCL5. Nat. Immunol. 15: 231– 13. Rathcke, C. N., J. S. Johansen, and H. Vestergaard. 2006. YKL-40, a biomarker 238. of inflammation, is elevated in patients with type 2 diabetes and is related to 39. Paugh, B. S., L. Bryan, S. W. Paugh, K. M. Wilczynska, S. M. Alvarez, insulin resistance. Inflamm. Res. 55: 53–59. S. K. Singh, D. Kapitonov, H. Rokita, S. Wright, I. Griswold-Prenner, et al. 14. Rathcke, C. N., and H. Vestergaard. 2006. YKL-40, a new inflammatory marker 2009. Interleukin-1 regulates the expression of sphingosine kinase 1 in glio- with relation to insulin resistance and with a role in endothelial dysfunction and blastoma cells. J. Biol. Chem. 284: 3408–3417. atherosclerosis. Inflamm. Res. 55: 221–227. 40. Rehli, M., H. H. Niller, C. Ammon, S. Langmann, L. Schwarzfischer, 15. Hansen, M., A. R. Nielsen, T. Vilsbøll, A. Lund, T. Krarup, F. K. Knop, and R. Andreesen, and S. W. Krause. 2003. Transcriptional regulation of CHI3L1, H. Vestergaard. 2012. Increased levels of YKL-40 and interleukin 6 in patients a marker gene for late stages of macrophage differentiation. J. Biol. Chem. 278: with chronic pancreatitis and secondary diabetes. Pancreas 41: 1316–1318. 44058–44067. 16. Konradsen, J. R., A. James, B. Nordlund, L. E. Reinius, C. So¨derha¨ll, E. Mele´n, 41. Leon, L. R., C. A. Conn, M. Glaccum, and M. J. Kluger. 1996. IL-1 type I re- A. M. Wheelock, K. C. Lo¨drup Carlsen, M. Lidegran, M. Verhoek, et al. 2013. ceptor mediates acute phase response to turpentine, but not lipopolysaccharide, The chitinase-like protein YKL-40: a possible biomarker of inflammation and in mice. Am. J. Physiol. 271: R1668–R1675. airway remodeling in severe pediatric asthma. J. Allergy Clin. Immunol. 132: 42. Glibetic, M. D., and H. Baumann. 1986. Influence of chronic inflammation on 328–335, e5. the level of mRNA for acute-phase reactants in the mouse liver. J. Immunol. 137: 17. Erzin, Y., H. Uzun, A. Karatas, and A. F. Celik. 2008. Serum YKL-40 as 1616–1622. a marker of disease activity and stricture formation in patients with Crohn’s 43. Kothavale, A., D. Di Gregorio, F. P. Somera, and M. E. Smith. 1995. GFAP disease. J. Gastroenterol. Hepatol. 23: e357–e362. mRNA fluctuates in synchrony with chronic relapsing EAE symptoms in SJL/J 18. Lee, S. H., C. Y. Lin, P. H. Wang, C. P. Han, S. F. Yang, J. T. Chang, M. C. Lee, mice. Glia 14: 216–224. L. Y. Lin, and M. S. Lee. 2012. Significant association of elevated concentration 44. Lee, W. H., W. E. Sonntag, M. Mitschelen, H. Yan, and Y. W. Lee. 2010. Irra- of plasma YKL-40 with disease severity in patients with pelvic inflammatory diation induces regionally specific alterations in pro-inflammatory environments disease. J. Clin. Lab. Anal. 26: 136–142. in rat brain. Int. J. Radiat. Biol. 86: 132–144. 19. Johansen, J. S., S. Møller, P. A. Price, F. Bendtsen, J. Junge, C. Garbarsch, and 45. Fitting, S., S. Zou, W. Chen, P. Vo, K. F. Hauser, and P. E. Knapp. 2010. Regional J. H. Henriksen. 1997. Plasma YKL-40: a new potential marker of fibrosis in heterogeneity and diversity in cytokine and chemokine production by astroglia: patients with alcoholic cirrhosis? Scand. J. Gastroenterol. 32: 582–590. differential responses to HIV-1 Tat, gp120, and morphine revealed by multiplex 20. Craig-Schapiro, R., R. J. Perrin, C. M. Roe, C. Xiong, D. Carter, N. J. Cairns, analysis. J. Proteome Res. 9: 1795–1804. M. A. Mintun, E. R. Peskind, G. Li, D. R. Galasko, et al. 2010. YKL-40: a novel 46. Liberto, C. M., P. J. Albrecht, L. M. Herx, V. W. Yong, and S. W. Levison. 2004. prognostic fluid biomarker for preclinical Alzheimer’s disease. Biol. Psychiatry Pro-regenerative properties of cytokine-activated astrocytes. J. Neurochem. 89: 68: 903–912. 1092–1100. 21. Bonneh-Barkay, D., S. J. Bissel, G. Wang, K. N. Fish, G. C. Nicholl, 47. Heinrich, P. C., I. Behrmann, G. Muller-Newen,€ F. Schaper, and L. Graeve. 1998. S. W. Darko, R. Medina-Flores, M. Murphey-Corb, P. A. Rajakumar, J. Nyaundi, Interleukin-6-type cytokine signalling through the gp130/Jak/STAT pathway. et al. 2008. YKL-40, a marker of simian immunodeficiency virus encephalitis, Biochem. J. 334: 297–314. 2870 RelB REGULATES YKL-40

48. Le´tuve´, S., A. Kozhich, N. Arouche, M. Grandsaigne, J. Reed, M. C. Dombret, 57. Rallabhandi, P., Q. M. Nhu, V. Y. Toshchakov, W. Piao, A. E. Medvedev, P. A. Kiener, M. Aubier, A. J. Coyle, and M. Pretolani. 2008. YKL-40 is elevated M. D. Hollenberg, A. Fasano, and S. N. Vogel. 2008. Analysis of proteinase- in patients with chronic obstructive pulmonary disease and activates alveolar activated receptor 2 and TLR4 signal transduction: a novel paradigm for receptor macrophages. J. Immunol. 181: 5167–5173. cooperativity. J. Biol. Chem. 283: 24314–24325. 49. Kuepper, M., K. Bratke, and J. C. Virchow. 2008. Chitinase-like protein and 58. Matsusaka, T., K. Fujikawa, Y. Nishio, N. Mukaida, K. Matsushima, asthma. N. Engl. J. Med. 358: 1073–1075, author reply 1075. T. Kishimoto, and S. Akira. 1993. Transcription factors NF-IL6 and NF-kB 50. Johansen, J. S. 2006. Studies on serum YKL-40 as a biomarker in diseases with synergistically activate transcription of the inflammatory cytokines, interleukin 6 inflammation, tissue remodelling, fibroses and cancer. Dan. Med. Bull. 53: 172–209. and interleukin 8. Proc. Natl. Acad. Sci. USA 90: 10193–10197. 51. Gross, O., C. J. Thomas, G. Guarda, and J. Tschopp. 2011. The inflammasome: 59. Bonizzi, G., and M. Karin. 2004. The two NF-kB activation pathways and their an integrated view. Immunol. Rev. 243: 136–151. role in innate and adaptive immunity. Trends Immunol. 25: 280–288. 52. Norris, J. G., L. P. Tang, S. M. Sparacio, and E. N. Benveniste. 1994. Signal 60. Shih, V. F., J. Davis-Turak, M. Macal, J. Q. Huang, J. Ponomarenko, transduction pathways mediating astrocyte IL-6 induction by IL-1b and tumor J. D. Kearns, T. Yu, R. Fagerlund, M. Asagiri, E. I. Zuniga, and A. Hoffmann. necrosis factor-a. J. Immunol. 152: 841–850. 2012. Control of RelB during dendritic cell activation integrates canonical and 53. Sparacio, S. M., Y. Zhang, J. Vilcek, and E. N. Benveniste. 1992. Cytokine noncanonical NF-kB pathways. Nat. Immunol. 13: 1162–1170. regulation of interleukin-6 gene expression in astrocytes involves activation of an 61. Kobayashi, T., P. T. Walsh, M. C. Walsh, K. M. Speirs, E. Chiffoleau, C. G. King, NF-kB‑like nuclear protein. J. Neuroimmunol. 39: 231–242. W. W. Hancock, J. H. Caamano, C. A. Hunter, P. Scott, et al. 2003. TRAF6 is 54. Bode, J. G., U. Albrecht, D. Ha¨ussinger, P. C. Heinrich, and F. Schaper. 2012. a critical factor for dendritic cell maturation and development. Immunity 19: Hepatic acute phase proteins—regulation by IL-6- and IL-1-type cytokines in- 353–363. volving STAT3 and its crosstalk with NF-kB‑dependent signaling. Eur.J.Cell 62. Sun, S. C. 2012. The noncanonical NF-kB pathway. Immunol. Rev. 246: 125– Biol. 91: 496–505. 140. 55. He, C. H., C. G. Lee, C. S. Dela Cruz, C. M. Lee, Y. Zhou, F. Ahangari, B. Ma, 63. Bhat, K. P., V. Balasubramaniyan, B. Vaillant, R. Ezhilarasan, K. Hummelink, E. L. Herzog, S. A. Rosenberg, Y. Li, et al. 2013. Chitinase 3-like 1 regulates F. Hollingsworth, K. Wani, L. Heathcock, J. D. James, L. D. Goodman, et al. cellular and tissue responses via IL-13 receptor a2. Cell Reports 4: 830–841. 2013. Mesenchymal differentiation mediated by NF-kB promotes radiation re- 56. Bara, I., A. Ozier, P. O. Girodet, G. Carvalho, J. Cattiaux, H. Begueret, sistance in glioblastoma. Cancer Cell 24: 331–346. M. Thumerel, O. Ousova, R. Kolbeck, A. J. Coyle, et al. 2012. Role of YKL-40 64. Lee, D. W., D. Ramakrishnan, J. Valenta, I. F. Parney, K. J. Bayless, and in bronchial smooth muscle remodeling in asthma. Am. J. Respir. Crit. Care R. Sitcheran. 2013. The NF-kB RelB protein is an oncogenic driver of mesen- Med. 185: 715–722. chymal glioma. PLoS One 8: e57489. Downloaded from http://www.jimmunol.org/ by guest on October 1, 2021