Function CBLB Links Genetic Risk with Type I IFN Associated Variant Of

A Multiple Sclerosis−Associated Variant of CBLB Links Genetic Risk with Type I IFN Function

This information is current as Klarissa Hanja Stürner, Uwe Borgmeyer, Christian Schulze, of September 27, 2021. Ole Pless and Roland Martin J Immunol 2014; 193:4439-4447; Prepublished online 26 September 2014; doi: 10.4049/jimmunol.1303077

http://www.jimmunol.org/content/193/9/4439 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2014/09/26/jimmunol.130307 Material 7.DCSupplemental http://www.jimmunol.org/ References This article cites 51 articles, 12 of which you can access for free at: http://www.jimmunol.org/content/193/9/4439.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on September 27, 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 © 2014 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

A Multiple Sclerosis–Associated Variant of CBLB Links Genetic Risk with Type I IFN Function

Klarissa Hanja Sturner,*€ Uwe Borgmeyer,† Christian Schulze,‡ Ole Pless,x,1 and Roland Martin*,{,1

Multiple sclerosis (MS) is an autoimmune disease of the CNS, and autoreactive CD4+ T cells are considered important for its pathogenesis. The etiology of MS involves a complex genetic trait and environmental triggers that include viral infections, particularly the EBV. Among the risk alleles that have repeatedly been identiﬁed by genome-wide association studies, three are located near the Casitas B-lineage lymphoma proto-oncogene b gene (CBLB). The CBLB protein (CBL-B) is a key regulator of peripheral immune tolerance by limiting T cell activation and expansion and hence T cell–mediated autoimmunity through its ubiquitin E3-ligase activity. In this study, we show that CBL-B expression is reduced in CD4+ T cells from relapsing-remitting MS

(RR-MS) patients during relapse. The MS risk-related single nucleotide polymorphism of CBLB rs12487066 is associated with Downloaded from diminished CBL-B expression levels and alters the effects of type I IFNs on human CD4+ T cell proliferation. Mechanistically, the CBLB rs12487066 risk allele mediates increased binding of the transcription factor C/EBPb and reduced CBL-B expression in human CD4+ T cells. Our data suggest a role of the CBLB rs12487066 variant in the interactions of a genetic risk factor and IFN function during viral infections in MS. The Journal of Immunology, 2014, 193: 4439–4447.

ultiple sclerosis (MS) is a complex genetic disorder. and hence indicate perturbations of T cell activation and function http://www.jimmunol.org/ Multiple genes and environmental factors, such as EBV as an important aspect of MS pathogenesis. M infection, reduced vitamin D3 levels, and smoking, can Among the known MS risk alleles, three are located near the contribute to the disease (1, 2). Genome-wide association studies CBLB gene (3, 8, 9). The CBL-B protein limits T cell responses (GWAS) have been highly successful in identifying an increasing through ubiquitinylation and subsequent proteasomal degradation number of common genetic variants and single nucleotide poly- of the TCR, p85 signaling proteins, the regulatory subunit of morphisms (SNPs) (3, 4), but knowledge regarding their func- PI3K, and phospholipase C-g 1 (PLC-g1); the protein is consid- tional contribution to MS pathogenesis remains limited, with the ered a key negative regulator of peripheral T cell tolerance (10, exception of MS risk-associated SNPs that lead to differential 11). CBL-B also controls the costimulatory signal by CD28 during splicing of the IL-7 receptor and of TNFR1 (5-7). Furthermore, T cell activation (11), which is required particularly for the acti- by guest on September 27, 2021 even less is known about mechanisms explaining the interaction vation of naive T cells but is less important for memory cells (12). and interplay of genetic and environmental factors on disease. A In experimental models, absence of CBL-B or functional defi- considerable fraction of the now more than 100 MS risk variants ciency of its E3-ligase activity triggers diabetes in rats (13) or that have been identified by large GWAS studies (3, 4) are in- mice carrying a TCR transgene specific for a foreign Ag expressed volved in T cell proliferation or setting their activation threshold, in the pancreas (14). Furthermore, CBL-B–deficient animals show increased susceptibility to peptide-induced autoimmunity (15) that is caused by releasing the requirement for CD28-mediated co- *Institute of Neuroimmunology and Multiple Sclerosis, Center for Molecular Neurobi- ology Hamburg, University Medical Center Hamburg-Eppendorf, 20251 Hamburg, stimulation and results in disproportionate T cell activation even Germany; †Institute for Molecular and Cellular Cognition, Center for Molecular Neu- to weak stimuli. In MS, diminished CBLB mRNA and protein robiology Hamburg, University Medical Center Hamburg-Eppendorf, 20251 Hamburg, ‡ expression in whole PBMCs of patients with relapsing-remitting Germany; Systems Biology, Center for Molecular Neurobiology Hamburg, University Medical Center Hamburg-Eppendorf, 20251 Hamburg, Germany; xFraunhofer Institute MS (RR-MS) have been described previously (16). Based on the for Molecular Biology and Applied Ecology, ScreeningPort, 22525 Hamburg, Germany; above data and our observation of increased autologous prolifer- and {Neuroimmunology and MS Research Section, Department of Neurology, Univer- sity Hospital Zurich,€ CH-8091 Zurich,€ Switzerland ation in MS patients carrying the most important MS risk gene, the 1 HLA-DR15 haplotype (17), we considered CBL-B a particularly O.P. and R.M. contributed equally to this work. interesting candidate for further study. Received for publication December 2, 2013. Accepted for publication August 29, 2014. The Institute of Neuroimmunology and Multiple Sclerosis was supported by the Materials and Methods € Gemeinnutzige Hertie Stiftung. Study population Address correspondence and reprint requests to Prof. Roland Martin, Neuroimmu- nology and MS Research Section, Department of Neurology, University Hospital All patients had MS as defined by the 2010 McDonald criteria and were Zurich,€ Frauenklinikstrasse 26, CH-8091 Zurich,€ Switzerland. E-mail address: recruited from the MS outpatient clinic of the Institute of Neuro- [email protected] immunology and Multiple Sclerosis at the University Medical Center The online version of this article contains supplemental material. Hamburg-Eppendorf. All patients were without any immunomodulatory or immunosuppressive treatment for at least 3 mo before participating in our Abbreviations used in this article: ChIP, chromatin immunoprecipitation; EDSS, study. Blood samples during relapse were collected before the start of Expanded Disability Status Scale; GWAS, genome-wide association study; HD, healthy control donor; LD, linkage disequilibrium; MedFI, median fluorescence in- steroid treatment. Treatment response to IFN-b was defined according to tensity; MS, multiple sclerosis; PLC-g1, phospholipase C-g 1; RR-MS, relapsing- published criteria (18); one relapse or at least one point of progression in remitting MS; siRNA, small interfering RNA; SNP, single nucleotide polymorphism. the Expanded Disability Status Scale (EDSS) score both in combination with magnetic resonance imaging activity during 2 y of IFN-b treatment Copyright Ó 2014 by The American Association of Immunologists, Inc. 0022-1767/14/$16.00 was used to define nonresponder status. “Intermediate responders” were www.jimmunol.org/cgi/doi/10.4049/jimmunol.1303077 4440 VARIANT OF CBLB LEADS TO IMPAIRED CONTROL OF CD4+ T CELLS defined as patients suffering from a relapse within the first 3 mo of vation if their purity was .96%. For in vitro stimulation CD4+ T cells treatment or solely inflammatory magnetic resonance imaging activity were seeded in 96-well microplates at 150,000 cells/well in X-VIVO, and without any relapse or EDSS progression. Healthy control donors (HDs) these CD4+ T cells were then stimulated with human anti-CD3 (clone were recruited from coworkers at the University Medical Center Hamburg- OKT3; BioXCell) 5 mg/ml and plate-bound human anti-CD28 (R&D Eppendorf. This study was approved by the local ethics committee (Ethik- Systems) 2.5 mg/ml 6 IFN-b-1a (100 IU/ml Rebif [Merck Serono] or 100 Kommission der A¨ rztekammer Hamburg, No. 2758), and written informed IU/ml Avonex [Biogen Idec]) for a maximal 72 h. Media without stimuli consent was obtained from all study subjects. was included as a negative control for each donor. CC and TT allele carriers were matched for every experiment—that is, for every stimulation Genotyping condition, CD4+ T cells were collected and purified, and experiments were performed in parallel from a CC and its matching TT allele carrier partner. We performed two-color TaqMan assays (Applied Biosystems) for the identification of homozygote allele carriers using a FAM-labeled probe for Chromatin immunoprecipitation the T allele (risk allele) and a VIC-labeled probe for the C allele for the rs12487066 and the rs9657904 SNP. Chromatin immunoprecipitation was performed as described previously (21) with minor modifications. PCR primers (BioMers) were chosen Linkage disequilibrium 100 bp upstream and downstream of the SNP. Sequences were as Genotype data from the 1000 genomes project (19), European super follows: for the rs12487066-F 59-CTCCGCAAGTACAGGGATATT-39, population, for a genomic region comprising rs12487066, rs9657904, and rs12487066-R 59-CCTCTTATTTGGACTAGTTG-39; and for the control rs2028597 plus 1 kb upstream and downstream were used to calculate rs9657904-F 59-CTCTCAATCTTTACACACAG-39 and rs9657904-R 59- linkage disequilibrium (HaploView 4.2). ACCTTCAGTTGTTCTGGTAT-39. Prediction of transcription factor binding sites Flow cytometric analysis of CBL-B expression Downloaded from Match is a weight matrix–based tool for searching putative transcription Measurements from whole blood samples of patients with RR-MS were factor binding sites (20). We used it for analyzing the DNA sequence performed using 250 ml EDTA blood. For surface marker analysis, cells starting 50 bp upstream and downstream of the CBLB SNP-containing were stained with mAbs for 30 min at room temperature. Erythrocytes region, focusing on the transcription factor binding prediction for se- were lysed using lysis solution (BD Biosciences) for 10 min at room quence differences at the CBLB SNP base (C versus T) in the first instance. temperature. The following Abs were used: anti-CD4 (APC, clone RPA- Only candidates binding to a sequence that included the CBLB SNP base T4; eBioscience), anti-CD8 (PacBlue, clone RPA-T8; BD Biosciences), were considered for further analysis. Identification of putatively binding anti-CD14 (PE, clone M5E2; BD Biosciences), anti-CD19 (PE Cy 5.5, + transcription factors was based on their expression in CD4 T cells, on clone SJ25C1; BD Biosciences), anti-CD25 (APC, clone BC96; eBio- http://www.jimmunol.org/ high core and matrix match values provided by Match, and on differences science), anti-CD127 (PE Cy 5.5, clone R34.34; Beckmann Coulter), anti- in the match values for the different CBLB alleles. CD45RA (APC, clone HI100; BD Biosciences), anti-CCR7 (PE-Cy7, clone 3D12; BD Biosciences). For in vitro experiments, CD4+ T cells EMSA were washed in PBS and Fc-fragments blocked with human IgG (Jackson ImmunoResearch) after 72 h of in vitro incubation. All additional steps Single-stranded oligonucleotides (Metabion, Martinsried, Germany) were were performed in the dark. For CD4+, surface staining anti-CD4 annealed in 10 mM Tris-HCl (pH 7.5) and 60 mM NaCl and were stored (PacBlue, clone MT310; Dako/BIOZOL) was used for 30 min at room 2 at 20˚C. Double-stranded oligonucleotides had 59 overhangs of five temperature. PBMCs and CD4+ T cells (from blood or in vitro activation, nucleotides on both strands. For EMSAs, double-stranded oligonucleotides respectively) were fixed in 100 ml fixation buffer (eBioscience) for 20 min were labeled using Klenow polymerase (Roche Applied Science, Mann- 32 and then permeabilized in permeabilization buffer (PBS, 5% FCS, 2% heim, Germany) with [a-[ P]]dATP (Hartmann Analytics, Braunschweig, BSA, 0.1% Triton), containing a second Fc-block with human IgG by guest on September 27, 2021 Germany). Unincorporated nucleotides were removed with gel filtration on (Jackson ImmunoResearch). For intracellular staining, either anti-human Sephadex G25 spin columns (Roche Applied Science). In vitro translation CBL-B (polyclonal rabbit IgG; Proteintech Group) at 0.005 mg/ml (1:50 was performed using the TNT Coupled Reticulocyte Lysate System diluted) or rabbit IgG as isotype control (Jackson ImmunoResearch) at (Promega) according to the manufacturer’s instructions and stored at –70˚C. 0.005 mg/ml was added for 35 min at room temperature. After washing, Labeled oligonucleotides were stored at 4˚C in 10 mM Tris-HCl, pH a FITC-labeled anti-rabbit IgG secondary Ab (Dianova, Hamburg, Ger- 7.5, 1 mM EDTA, and 60 mM NaCl. Binding reactions were performed in many) was applied for 35 min at room temperature. All analyses were a total volume of 10 ml consisting of 20 mM HEPES, pH 7.4, 80 mM performed with an LSRII (BD Biosciences) flow cytometer. Data analysis NaCl, 20 mM KCl, 3 mM DTT, 0.1 mg Cot-1 DNA (Roche Applied was performed with BD FACS Diva for analyzing median fluorescence Science), and 2 ml reticulocyte lysate. A 30-min preincubation was fol- intensities (MedFI) for all data presented. lowed by the addition and binding of 1 ml of the labeled oligonucleotides for 30 min at room temperature. Complexes were resolved by non- Knockdown of C/EBPb in primary human CD4+ T cells denaturing PAGE in 0.53 TBE (45 mM Tris base, 45 mM boric acid, 1 mM EDTA) at 4˚C at 20 V/cm for 4 h. The gels were dried and analyzed Small interfering RNA (siRNA) knockdown was performed according to with the Fujix BAS 2000 bioimaging system using the Tina software the manufacturer’s instructions using Accell SMARTpool siRNA duplexes (Raytest). The sequences used for EMSA were: CBLB-WT-F 59-AACT- (Dharmacon, Lafayette, CO) targeting the 39UTR and open reading frame TAACGACTAAAAGTGATTTCACTGACAAGGTGCTGT-39; CBLB-WT-rev of C/EBPb and compared with a mixture of nontargeting control siRNAs. 59-TGCTACAGCACCTTGT CAGTGAAATCACTTTTAGTCGTT-39; CBLB- Briefly, 125,000 isolated human CD4+ T cells per well were resuspended mut-F 59-AACTTAACGACTAAAAGTGATCTCACTGACAAGGTGC- in X-VIVO, seeded in 96-well plates, and incubated with 1 mM of the TGT-39;andCBLB-mut-rev 59-TGCTACAGCACCTTGTCAGTGAGA- siRNA pool for C/EBPb or the nontargeting pool. After administering the TCACTTTTAGTCGTT-39. siRNA, CD4+ T cells were rested for 30 min at 37˚C and were then A luciferase template (Promega) served as a negative control. C/EBPb, stimulated with 75,000 anti-CD3, anti-CD2, anti-CD28–coated beads SRF, and GFI1 were generated from pCDNA3.1 mC/EBPb (Addgene (Miltenyi Biotec) and IFN-b for 72 h. As a control, siRNA with a non- #12557), pCGN-SRF (Addgene # 11977), and pCMV-SPORT6-hGFI1 sense/scrambled sequence from the Accell siRNA Control Kit was used. (Source BioScience, IMAGE ID 5585794), respectively. Knockdown efficiency was evaluated with real-time quantitative PCR (Supplemental Fig. 2A, 2B). Isolation of CD4+ T cells and in vitro CD4+ T cell activation Real-time quantitative PCR PBMCs were prepared from EDTA blood by density gradient with His- topaque (PAA, Pasching, Austria). Immediately after separation, CD4+ After 72 h of in vitro stimulation of CD4+ T cells, cells were washed once T cells were isolated from PBMCs by negative selection using the BD with PBS and resuspended in TRIZOL before snap-freezing in liquid ni- IMag Kit (BD Biosciences) according to the manufacturer’s instructions. trogen and storage at 280˚C. RNA was extracted according to published CD4+ T cell purity was confirmed with flow cytometry. Cells were washed protocols (22). RNA was purified from PBMCs of RR-MS patients and in PBS, and Fc-receptors were blocked with human IgG (Jackson Immuno- HDs. Real-time quantitative PCR was performed using an ABI real-time Research). All subsequent steps were then performed in the dark. Cells PCR 7900HF Fast System (Applied Biosystems). The TaqMan primer and were stained with mAbs for 30 min at room temperature using anti-CD4 probe set for C/EBPb (Hs00270923_s1), CBLB (Hs00180288_m1), and (PacBlue, clone MT310; Dako/BIOZOL) and anti-CD3 (FITC, clone 18s (Hs03003631_g1) rRNA were purchased from Applied Biosystems UCHT1; BD Biosciences). CD4+ cells were used for in vitro T cell acti- and used according to the manufacturer’s instructions. 18S rRNA was used The Journal of Immunology 4441 as endogenous control, and the relative gene expression was calculated Next, we examined whether the MS risk-associated CBLB SNPs ΔΔ + with the CT method using untreated CD4 T cells as a calibrator. are involved in the immune alterations in MS. To date, three SNPs Cytokine determination have been associated with MS risk: rs12487066 was identified in the first GWAS study for MS (8) and was reproduced in a family IFN-b was measured in serum samples of HDs and patients with untreated case-control study (23), rs9657904 was identified in the Sardinian RR-MS during relapse or remission using an IFN-b ELISA Kit (Cata- logue No. KAC1201; Life Technologies) according to the manufac- population (9), and rs2028597 was identified in the second major turer’s instructions. IL-2 was detected in culture supernatants of CD4+ GWAS study for MS (3). Using data from the 1000 Genomes T cells after 72 h of in vitro culture usinganIL-2ELISAset(Human Project (19), we examined the linkage disequilibrium (LD) of ELISA MAX Deluxe; BioLegend) according to the manufacturer’s these three reported SNPs. There is no indication for linkage instructions. disequilibrium between rs12487066 and rs9657904/rs2028597, Phosphorylation of PLC-g1 by flow cytometry supporting the idea that rs12487066 is independent of these two PLC-g1 expression and phosphorylation was measured using the Phosflow alleles (Fig. 2A). Because at the time of our study the rs12487066 protocol (BD Biosciences) and the following Abs: anti-CD4 (PacBlue, was the only reproduced MS-associated SNP and because the clone MT310; Dako/BIOZOL), anti-PLC-g1 (PE, clone 10/PLC-g1; BD populations in which the rs12487066 was identified were more Biosciences) and anti-PLC-g1 pY783 (Alexa Fluor 647, clone 27/PLC; BD similar to our Northern German cohort, we focused on the Biosciences). All analyses were performed on an LSRII (BD Biosciences) rs12487066 that seems to confer MS risk independently of the flow cytometer. Data analysis was performed with BD FACS Diva for analyzing MedFI, and PLC-g1 pY783 expression was normalized for PLC- other two SNPs. g1 expression [MedFI PLC-g1 pY783/(MedFI PLC-g1/MedFI PLC-g1 Because we observed a decrease in CBL-B expression in CD4+ medium control)]. T cells during relapse, probably caused by the active phase of the Downloaded from Proliferation disease, we only considered patients during remission for the analysis of CBL-B expression concerning the corresponding ge- + One hundred twenty-five thousand isolated human CD4 T cells per well notype. MS patients carrying the risk allele (T) of the first de- were resuspended in X-VIVO, seeded in 96-well plates, and stimulated with 75,000 anti-CD3, anti-CD2, anti-CD28-coated beads (Miltenyi Bio- scribed CBLB-attributed SNP rs12487066 showed significantly + tec) and IFN-b for 72 h. After 72 h of in vitro stimulation of CD4+ T cells, lower CBL-B expression in CD4 T cells than patients lacking this 3 cells were pulsed overnight with 1 mCi [methyl- H]thymidine (Hartmann allele (Fig. 1B). There was no difference in CBL-B expression http://www.jimmunol.org/ Analytics, Braunschweig, Germany) per well. Cells were harvested after between heterozygous and homozygous allelic state in MS 15 h of incubation, and incorporation of radioactivity was measured in patients, indicating a dominant effect of the T allele and a lack of a scintillation counter (Wallac 1450; PerkinElmer, Rodgau-Jurgesheim,€ Germany). a gene dosing effect. Surprisingly, when we applied the genotyping to healthy rs12487066 allele carriers, we were not able to Statistical analysis reproduce the difference in CBL-B expression between T and non- Statistical analyses were performed using Graph Pad Prism 5.0 Software T allele carriers (Fig. 1C). Consequently, we hypothesized that the (GraphPad Software, La Jolla, CA), conducting one-way ANOVA followed rs12487066-associated differences in CBL-B expression are likely by Bonferroni multiple comparison test. Allele frequency was calculated to be related to the disease state—that is, MS, and that some as yet 2 using the Hardy-Weinberg equation and tested with the Pearson x test. unknown factors are responsible. To identify potential mediators, by guest on September 27, 2021 we examined the genomic location of the CBLB risk SNP Results rs12487066 in more detail. As shown in Fig. 2A, rs12487066 is When investigating its role in MS we observed lower CBL-B located ∼320 kb upstream of the CBLB coding region and sur- expression ex vivo in CD4+ T cells during relapses of RR-MS rounded by regulatory and epigenetic modification sites (methyl- patients (Fig. 1A) but not in other lymphocyte subsets (Sup- ation and acetylation) specifically reported for CD4+ T cells plemental Fig. 1A–D). Further analysis of the CD4+ T cell sub- according to the Ensembl database. sets revealed that the decrease in CBL-B expression was primarily Using an unbiased bioinformatics analysis that applied the seen in naive CD4+ T cells (Supplemental Fig. 1E) and in CD4+ weight matrix–based tool Match to predict transcription factor CD25+ T regulatory cells (Supplemental Fig. 1F). binding to the risk versus non-risk allele SNP sequence (20) gave

FIGURE 1. CBL-B expression is diminished in CD4+ T cells of RR-MS patients during relapse and in CD4+ cells of RR-MS patients carrying the rs12487066 risk allele (T). (A) CBL-B expression is significantly diminished in MS patients during relapse (RR-MS/R*) as measured by flow cytometry in comparison with HDs and MS patients during remission (RR-MS, n = 15). HD n = 15; RR-MS n = 15; RR-MS/R* n = 10. One-way ANOVA, p = 0.0122. (B) CBLB rs12487066 risk allele carriers (T) show a reduced CBL-B expression in CD4+ T cells in comparison with homozygote non-risk allele carriers (CC) in RR-MS patients. CC n =8;CTn = 14; TT n = 15. One-way ANOVA, p = 0.012. (C) CBL-B expression is not altered in CD4+ T cells from HDs of CBLB rs12487066 risk allele carriers. CC n =7;CTn = 14; TT n = 13. One-way ANOVA, p = 0.9721. All values are presented as MedFI normalized to the isotype control (MedFI/Isotype control 3 100). n.s., not significant. 4442 VARIANT OF CBLB LEADS TO IMPAIRED CONTROL OF CD4+ T CELLS Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 2. C/EBPb binds exclusively to the risk-associated but not to the non–risk-associated configuration of the CBLB MS-associated allele rs12487066. (A) A schematic overview of the chromosomal region near the MS-associated CBLB-related SNPs rs12487066 (*), rs2028597, and rs9657904 relative to the protein coding region of CBL-B (light gray), and reported regulatory regions in CD4+ T cells (dark gray). Data are retrieved from the Ensembl database. Below, the linkage disequilibrium between the rs12487066, rs2028597, and rs9657904 is shown (Haploview, LD plot, r2 values). (B) MatchTM value (core matrix score) predicting the binding of C/EBP to the T allele configuration of CBLB SNPs rs12487066, rs9657904, and rs2028597. (C) EMSA for C/EBPb and Gfi1 together with the CBLB rs12487066 risk allele sequence (T) and the non-risk allele sequence. C/EBPb shows better binding to the T versus the C oligonucleotide sequence. Luciferase (Luc) and SRF serve as negative controls. (D) PCR of the rs12487066 SNP containing region using ChIP DNA isolated from CD4+ T cells of one representative donor pair (homozygote non-risk: CC versus risk allele carriers: TT) as a template. Binding of C/EBPb to the risk allele sequence of CD4+ T cells could be detected only after IFN-b stimulation. Donor data are presented in Supplemental Table I. (E) Quantification of ChIP assays by quantitative real-time PCR of CBLB genomic regions carrying MS-associated risk and non-risk alleles rs12487066 is depicted as percent of input. C/EBPb-associated genomic DNA was immunoprecipitated from primary CD4+ T cells after stimulation with aCD3, aCD28, and IFN-b for 72 h for the homozygote allele carriers (CC non-risk, TT risk allele). Stronger binding to the homozygote CBLB MS risk allele (TT) could be detected in a C/EBPb Ab-dependent manner. Results are pooled from three independent experiments of three donor pairs. (F) C/EBPb isoforms are induced after 24 h of stimulation with IFN-b as shown by Western blot from CD4+ T cells. IFN-b treatment further increases the basal C/EBPb induction by anti-CD3/CD28. (G) C/EBPb is upregulated in MS patients during remission and in particular during relapse. Real-time quantitative PCR from PBMC-derived RNA. HD n = 10; RR-MS n = 10; RR-MS/R* n = 10. One-way ANOVA, *** p , 0.001. (H) IFN-b is increased in ELISA of serum samples of MS patients during relapse. HD n = 20; RR-MS n = 20; RR-MS/R* n = 20. One-way ANOVA, ** p , 0.01. as the top-scoring candidate transcription factors C/EBP b binding site if it contained the risk (T) allele of rs12487066 (C/EBPb) and Growth factor independent protein-1 (Gfi-1; data not (Fig. 2B), whereas the prediction for rs9657904 and rs2028597 shown). Interestingly, binding was predicted only for the C/EBPb showed no differential binding for C/EBPb. C/EBPb and Gfi-1 The Journal of Immunology 4443 were subsequently tested for differential binding to risk and non- fail to upregulate CBL-B upon exposure to IFN-b,andsince risk allele sequences of the rs12487066 in EMSAs and, consistent CBL-B deficiency causes increased phosphorylation of PLC-g1, with the aforementioned prediction, C/EBPb shifts oligonucleo- elevated production of IL-2 as well as hyperproliferation (11), we tides containing the risk allele (T) significantly better than the studied the influence of IFN-b on these functional aspects in CD4+ non-risk oligonucleotide (Fig. 2C). Gfi-1 showed no binding to T cells in risk versus non-risk allele carriers. In the presence of either oligonucleotide. IFN-b, CD4+ T cells of non-risk allele carriers showed less C/EBPb is involved in inflammation, immune regulation, and phosphorylation of PLC-g1 and diminished secretion of IL-2 cy- T cell biology (24). In particular, C/EBPb has been shown to be tokine (Fig. 3B, 3C). The differences between the carriers of the upregulated significantly in MS tissue samples in a study exam- non-risk versus risk allele in all parameters could be abrogated by ining gene expression in MS brain tissue (25). Furthermore, it has the use of siRNA against C/EBPb, whereas the nontargeting been associated with viral infections, among which is EBV (26), siRNA had no effect. Consistent with this result, proliferation was the strongest environmental risk factor for MS (1). IFN-b is inhibited by IFN-b only in CBLB non-risk allele carriers (Fig. produced by a broad range of cells in response to viral infection; it 3D), indicating that inhibition of T cell proliferation by IFN-b is limits T cell proliferation and cytokine production, is used as compromised in CBLB risk allele carriers. a first-line drug to treat MS, and is known to induce C/EBPb (27). Consequently, we tested whether the rs12487066 is associated We therefore considered IFN-b and treatment with this cytokine with treatment responsiveness to IFN-b in RR-MS patients and as a meaningful model for the inflammatory host response causing therefore might be clinically relevant. We genotyped 125 patients C/EBPb induction in vitro. We were able to confirm and expand with RR-MS who received IFN-b for at least 2 y with regular previous observations by showing increased expression of the full- follow-up at our outpatient unit. We found that the rs12487066 Downloaded from length C/EBPb isoform LAP (Fig. 2F) upon stimulation of naive non-risk allele (C) was significantly more prominent in responders + and activated CD4 cells with IFN-b. We then examined binding to IFN-b treatment (Supplemental Fig. 3A) and that homozygote of endogenously expressed C/EBPb to the MS risk-associated non-risk allele carriers (CC) showed a lower degree of disability + CBLB allele rs12487066 in CD4 T cells of homozygote risk as measured by the EDSS (Supplemental Fig. 3B). versus non-risk allele carriers with and without exposure to IFN-b

by chromatin immunoprecipitation (ChIP). C/EBPb binds to the http://www.jimmunol.org/ rs12487066 risk allele sequence in human CD4+ T cells only upon Discussion Our study indicates an important role of the CBLB SNP exposure to IFN-b, whereas binding could not be detected in + homozygous non-risk allele carriers under this condition and rs12487066 risk allele in lowering the activation threshold of CD4 could not be quantified accordingly (Fig. 2D, 2E). When analyz- T cells and provides an interesting link to a genetic variant of ing RNA of PBMCs from patients with untreated RR-MS, we a gene that is involved in immune regulation, which might be found that C/EBPb RNA expression was significantly upregulated modulated by environmental factors such as viral infections. During the course of RR-MS, CBL-B expression is diminished already during remission and clearly elevated during relapse + (Fig. 2G). Consistent with this observation and previously pub- only during relapses in CD4 T cells and, while more pronounced + + lished experiments, serum levels of IFN-b are significantly ele- in naive than in effector CD4 T cells, it is also observed in CD4 by guest on September 27, 2021 + vated in patients with untreated relapsing RR-MS (Fig. 2H) (28, CD25 regulatory T cells. Previously, diminished CBL-B mRNA 29). We then hypothesized that C/EBPb decreases CBL-B ex- and protein expression in whole PBMCs of patients with RR-MS pression in risk allele carriers and thus lowers the activation have both been described (16). In experimental models, the ab- threshold or activation requirements, or both, of CD4+ T cells in sence of CBL-B or a functional deficiency of its E3-ligase activity the presence of IFN-b. Consequently, CBLB risk allele carriers triggers diabetes in rats (13, 14) or mice carrying a TCR transgene might less stringently control T cell activation, cytokine produc- specific for a foreign Ag expressed in the pancreas (14). Because + tion, and proliferation in the presence of IFN-b. autoreactive CD4 T cells are critical for the development of MS, Genotyping of 60 healthy donors for the MS risk alleles it is notable that we found a reduction of CBL-B during MS + rs12487066 (T, non-risk allele C) and rs9657904 was performed to relapses in CD4 T cells. A comparably reduced CBL-B expres- + identify donors for functional CD4+ T cell studies. We identified sion in CD3 T cells has been observed in patients with systemic four individuals per group for the rs12487066 SNP (Supplemental lupus erythematosus (31). CBL-B–deficient animals show in- Table I) who were matched by gender and age. CBL-B expression creased susceptibility to peptide-induced autoimmunity (15) that in CD3/CD28-stimulated purified CD4+ T cells did not differ is caused by removing the requirement for CD28-mediated significantly in healthy SNP rs12487066 risk (TT) versus non-risk costimulation and results in disproportionate T cell activation (CC) allele carriers (Supplemental Fig. 2C). Fig. 3 shows the IFN- even to weak stimuli. Our data indicate reduced dependence of + b–mediated effects on purified CD4+ T cells of healthy SNP CD4 T cells on CD28-mediated costimulation with differen- + rs12487066 risk (TT) versus non-risk (CC) allele carriers stimu- tiation from naive to effector CD4 T cells (12). As a conse- lated with anti-CD2/anti-CD3/anti-CD28 in the absence versus quence, naive CD4+ T cells of patients with RR-MS could be presence of IFN-b. No upregulation of CBL-B is seen in risk al- more prone to T cell activation during relapse (i.e., the active lele carriers after exposure of activated CD4+ T cells to IFN-b phase of MS). CBL-B is involved in regulating Foxp3 expres- (Fig. 3A), whereas non-risk allele carriers express significantly sion (32, 33), the key transcription factor for CD4+CD25+ T higher levels of CBL-B in the presence of IFN-b. To ensure that regulatory cell function. In line with our above observation of this effect is dependent on C/EBPb, we used siRNA to diminish reduced CBL-B expression in CD4+CD25+ T regulatory cells, it the expression level of C/EBPb in primary human CD4+ cells has been reported that during an MS relapse, expression of (Supplemental Fig. 2A, 2B). siRNA for C/EBPb abrogated the Foxp3 and the suppressive function of CD4+CD25+ T regula- IFN-b–mediated upregulation of CBLB in non-risk allele carriers tory cells are impaired in patients with MS (34). Based on these (Fig. 3A, right panel). Therefore, CBL-B expression appears to findings and our data we assume that, although the activation depend on C/EBPb in non-risk allele carriers. IFN-b inhibits the threshold of CD4+CD25+ T regulatory cells is probably de- proliferation of T cells in vitro through multiple pathways and creased in RRMS patients in relapse as indicated by the lower mechanisms (30). After observing that CBLB risk allele carriers CBL-B expression, their functional compromise does not 4444 VARIANT OF CBLB LEADS TO IMPAIRED CONTROL OF CD4+ T CELLS

compensate for the easier activation of the entire population of naive CD4+ T cells. To date, three CBLB MS risk SNPs have been described: rs12487066 (8, 23), rs9657904 from the promoter region of CBLB (9) and in LD, and the more recent rs2028597 from a large GWAS study (3). None of the mentioned SNPs could be replicated in the recent ImmunoChip study (4). This ﬁnding might be explained by the wide ethnic origin of the populations contributing to this study, because it appears that the ethnic origin of the population inﬂu- ences the results of GWAS studies and hence causes different outcomes depending on the analyzed population (35). This result can lead to false-positive and false-negative results for a single variant because of the complex statistical methods used in GWAS studies. A complex genetic disease like MS is the common result of a process that is initiated, shaped, and maintained by several different pathways and therefore network-based gene-gene interactions (36). Although the single variant contributes only to a minor extent to the complex phenotype of MS, its importance

and function for disease manifestation increases in certain Downloaded from genetic and environmental surroundings. Our SNP of interest, rs12487066, was found in a GWAS of individuals recruited from the U.K. and United States and was the only one of the three CBLB attributed SNPs to be reproduced in a northern Caucasian cohort (23), whereas the rs9657904 SNP was described initially in

Southern European individuals and in particular in the genetically http://www.jimmunol.org/ unique Sardinian population. Based on the assumption that the MS and control population of the former study is likely to be more similar to our Northern German study cohort, we decided to pursue the effect of rs12487066. The weak LD between the rs12487066 and the two other MS-associated CBLB SNPs (4, 8, 30) indicates that the rs12487066 will contribute independently to MS risk. In this study, we identify a functional role for the CBLB SNP rs12487066. Identifying the causal variant of a SNP risk allele remains a challenge (37, 38), and the potential importance by guest on September 27, 2021 of noncoding variants that represent ∼88% of trait-associated variants is often underestimated (39). MS is an example of a complex genetic disease showing familial aggregation because of shared genetic risk factors that derive from a cumulative contribution of different variants with variable expressivity and in- complete penetrance. We cannot rule out a functional interplay among the reported independent CBLB SNPs associated with MS risk; however, our data indicate that the non-coding rs12487066 is FIGURE 3. CBLB rs12487066 risk allele carriers differ in their phe- functionally relevant in the context of MS. A more detailed picture notype of CBL-B and CBL-B–dependent T cell functions in the presence on eventually tagged additional sequence variations could evolve of IFN-b.(A) CBL-B protein expression in FACS analysis is significantly from high resolution sequencing data. enhanced in the presence of IFN-b in CD4+ T cells of CBLB rs12487066 Our data demonstrate that environmental factors such as virus- non-risk allele carriers (CC), whereas risk allele carriers (TT) fail to up- induced type I IFN, which is used as a first-line treatment in RR- regulate CBL-B. The upregulation of CBL-B can be abrogated by the use MS, at least in part modulates T cell activation via CBL-B. The B of siRNA for C/EBPb (right). One-way ANOVA, p = 0.0004. ( ) Phos- induction of CBL-B expression appears to be dependent on C/EBPb phorylation of PLC-g1 in flow cytometric analysis is significantly de- as CBL-B is upregulated in MS patients carrying the homozygote creased in the presence of IFN-b in CD4+ T cells of CBLB rs12487066 + non-risk allele carriers (CC) in contrast to the risk allele carriers (TT). The non-risk allele (CC) and in CD4 T cells of healthy, non-risk allele effect of IFN-b in can be abrogated by the use of siRNA for C/EBPb carriers exposed to IFN-b in vitro. In CBLB rs12487066 risk + (right). One-way ANOVA, p = 0.0029. (C) Production of IL-2 is signifi- allele carriers (T), CD4 T cells fail to upregulate CBL-B upon cantly decreased in the presence of IFN-b in CD4+ T cells of CBLB rs12487066 non-risk allele carriers (CC), whereas IFN-b fails to decrease IL-2 production in risk allele carriers (TT). The effect of IFN-b can be beads without IFN-b versus CD4+ T cells stimulated with anti-CD3, anti- abrogated by the use of siRNA for C/EBPb (right). One-way ANOVA, p = CD2, and anti-CD28–coated beads with IFN-b from four independent 0.0078. (D) Proliferation of CD4+ T cells (3H-thymidine incorporation) of experiments for the untreated and for three independent experiments for CBL-B rs12487066 non-risk allele carriers (CC) is significantly dimin- the siRNA with a nonsense/scrambled sequence (ssRNAi) and siRNA ished in the presence of IFN-b compared with risk allele carriers (TT). C/EBPb conditions. Response to stimulation with anti-CD3, anti-CD2, and One-way ANOVA, p , 0.0001. Risk allele carriers (TT, n = 4) for the anti-CD28–coated beads between the different allele carriers showed equal rs12487066 are indicated in black. Non-risk allele carriers (CC, n = 4) for results in mean, SEM, and variances (Supplemental Fig. 2C–F). Statistical the rs12487066 are indicated in white. Data are presented as d between methods used one-way ANOVA followed by Bonferroni multiple com- CD4+ T cells stimulated with anti-CD3, anti-CD2, and anti-CD28–coated parison test. *p , 0.05, **p , 0.01. n.s., not significant. The Journal of Immunology 4445

TCR-mediated activation in the presence of IFN-b. These obser- such as N-CoR (45) or SMRT (46), of histone-modifying enzymes vations indicate that CBLB risk allele carriers might express lower like lysine or arginine methyltransferases (21, 47) or by alteration of levels of CBL-B during viral infections because of a type I IFN- homodimerization or heterodimerization states of C/EBP com- induced repression by C/EBPb. As a result, T cells of CBLB risk plexes (48). It remains to be determined whether one of the above or allele carriers probably have a lower T cell activation threshold and yet another mechanism is involved in the altered transcription of can be activated more easily, even by low affinity interactions with CBLB in the context of the rs12487066 MS risk allele. In the self-antigens (e.g., myelin peptides in MS) (40). Interestingly, it is context of the treatment of RR-MS with IFN-b, it is interesting that well known from epidemiologic studies that viral infections fre- only approximately half respond to IFN-b,while∼50% are in- quently precede MS relapses (41) and that at least half of the complete responders or nonresponders (18). A study that investi- patients with RR-MS show an increase in the expression of type I gated SNPs that might contribute to IFN-b treatment insufficiency IFN-related genes (42), corresponding to our data showing elevated or failure did not include the CBLB SNP rs12487066 (49), for IFN-b serum levels during relapse and our observed association which we have seen differences in the response to IFN-b.This with a more frequent allele presence of the risk allele in non- study revealed SNPs in genes of the IL-6 signaling pathway, which responders to IFN-b therapy. Risk allele carriers might be more is consistent with its deregulation in the monocytes of RR-MS prone to T cell activation by low-affinity interactions because of patients who respond weakly to IFN-b (28). A major transcrip- their lower CBL-B expression levels in CD4+ T cells during viral tion factor in the IL-6 signaling pathway is C/EBPb (50), for infections that induce a type I IFN host response and consequently which we found differential binding to the CBLB MS-associated RR-MS patients carrying the risk allele seem to respond less to IFN-b rs12487066 risk allele. Supporting our hypothesis, a recent report therapy, which should be validated in a larger cohort of patients shows that CBLB RNA expression is significantly lowered in Downloaded from with IFN-b–treated RR-MS. Accordingly, we propose a model in regulatory T cells of RR-MS patients and that IFN-b treatment which C/EBPb binds selectively to the rs12487066 MS risk allele was not able to reconstitute CBLB RNA expression in this patient upstream of CBLB and thereby alters the transcription and induc- cohort, for whom no genotyping information is available (51). tion of CBL-B by C/EBPb in the promoter region (Fig. 4). We were Therefore, our experiments indicate that genetic variants of CBLB not able to show a preferential induction of the repressive LIP are relevant in controlling T cell–mediated immune mechanisms.

isoform of C/EBPb that has been discussed previously as a possible Our findings also indicate that genetic risk traits can only unmask http://www.jimmunol.org/ mechanism for C/EBPb induced repression (43). The C/EBPb long their functional role in a specific context, such as an infection and isoforms LAP* and LAP, which contain transactivation domains, subsequent IFN production, and then maintain their effects during can also act as context-dependent repressors (e.g., by recruitment of the disease. This finding could explain why we were not able to chromatin remodeling complexes) (44) of repressor complexes, detect the phenotype of the rs12487066 in CD4+ T cells of healthy by guest on September 27, 2021

FIGURE 4. Proposed model for the influence of the CBLB variant rs12487066 on CBLB gene expression and T cell function. Upper and lower left panel, During T cell stimulation, C/EBPb induces CBL-B independent of the CBLB allele variant for the rs12487066. T cells show enhanced PLC-g1 phosphorylation, IL-2 production, and enhanced proliferation. Upper right panel, During T cell stimulation in the presence of a type I IFN (i.e., IFN-b), C/EBPb is increasingly induced in the non-risk allele (C) carrier, CBL-B expression is enhanced, and consecutively T cell activation (as determined by PLC-g1 phosphorylation, IL-2 production, and proliferation) is inhibited. Lower right panel, During T cell stimulation in the presence of a type I IFN (i.e., IFN-b), C/EBPb is still increasingly induced in the risk allele (T) carrier, but the binding to the allelic variant inhibits CBL-B expression by a still unknown mechanism. As a result, T cell activation (PLC-g1 phosphorylation, IL-2 production, and CD4+ T cell proliferation) is not inhibited. 4446 VARIANT OF CBLB LEADS TO IMPAIRED CONTROL OF CD4+ T CELLS individuals, because they lack the proinflammatory mediators that 16. Zhou, W. B., R. Wang, Y. N. Deng, X. B. Ji, G. X. Huang, and Y. Z. Xu. 2008. Study of Cbl-b dynamics in peripheral blood lymphocytes isolated from patients induce C/EBPb and thereby mediate the upregulation of CBL-B in with multiple sclerosis. Neurosci. Lett. 440: 336–339. + CD4 T cells of CC allele carriers, but only in an inflammatory 17. Mohme, M., C. Hotz, S. Stevanovic, T. Binder, J. H. Lee, M. Okoniewski, setting like in CD4+ T cells of patients with RR-MS. To our T. Eiermann, M. Sospedra, H. G. Rammensee, and R. Martin. 2013. HLA-DR15- derived self-peptides are involved in increased autologous T cell proliferation in knowledge, this is the first report of a differential transcriptional multiple sclerosis. Brain 136: 1783–1798. regulation by type I IFNs in association with an MS risk-related 18. Rıó, J., C. Nos, M. Tintore´,N.Te´llez, I. Galań, R. Pelayo, M. Comabella, and genetic variant. Future studies should examine whether others of X. Montalban. 2006. Defining the response to interferon-beta in relapsing- remitting multiple sclerosis patients. Ann. Neurol. 59: 344–352. the recently reported MS risk alleles act in concert with the CBLB 19. 1000 Genomes Project Consortium, G. R. Abecasis, A. Auton, L. D. Brooks, risk SNPs and whether such gene-environment interactions also M. A. DePristo, R. M. Durbin, R. E. Handsaker, H. M. Kang, G. T. Marth, and G. A. McVean. 2012. An integrated map of genetic variation from 1,092 human exist for other MS risk genes. genomes. Nature 491: 56–65. 20. Kel, A. E., E. Go¨ssling, I. Reuter, E. Cheremushkin, O. V. Kel-Margoulis, and E. Wingender. 2003. MATCH: A tool for searching transcription factor binding Acknowledgments sites in DNA sequences. Nucleic Acids Res. 31: 3576–3579. We thank M. Kolster and N. Verse for technical assistance, Christoph 21. Pless, O., E. Kowenz-Leutz, M. Knoblich, J. Lausen, M. Beyermann, M. J. Walsh, and A. Leutz. 2008. G9a-mediated lysine methylation alters the Heesen for help in patient recruitment, Mireia Sospedra for discussions function of CCAAT/enhancer-binding protein-beta. J. Biol. Chem. 283: 26357– of methodologic aspects, Dr. Stephen Smith for critical reading of the 26363. manuscript, Peter Johnson for providing the C/EBP expression plasmids 22. Chomczynski, P., and N. Sacchi. 2006. The single-step method of RNA iso- through Addgene (Cambridge, MA), and Merck Serono and Biogen Idec lation by acid guanidinium thiocyanate-phenol-chloroform extraction: twenty- something years on. Nat. Protoc. 1: 581–585. b for providing the IFN- -1a. 23. D’Netto, M. J., H. Ward, K. M. Morrison, S. V. Ramagopalan, D. A. Dyment, G. C. DeLuca, L. Handunnetthi, A. D. Sadovnick, and G. C. Ebers. 2009. Downloaded from Risk alleles for multiple sclerosis in multiplex families. Neurology 72: 1984– Disclosures 1988. The authors have no financial conflicts of interest. 24. Ramji, D. P., and P. Foka. 2002. CCAAT/enhancer-binding proteins: structure, function and regulation. Biochem. J. 365: 561–575. 25. Lock, C., G. Hermans, R. Pedotti, A. Brendolan, E. Schadt, H. Garren, A. Langer-Gould, S. Strober, B. Cannella, J. Allard, et al. 2002. Gene-microarray References analysis of multiple sclerosis lesions yields new targets validated in autoimmune

1. Ramagopalan, S. V., R. Dobson, U. C. Meier, and G. Giovannoni. 2010. Multiple encephalomyelitis. Nat. Med. 8: 500–508. http://www.jimmunol.org/ sclerosis: risk factors, prodromes, and potential causal pathways. Lancet Neurol. 26. Bristol, J. A., A. R. Robinson, E. A. Barlow, and S. C. Kenney. 2010. The 9: 727–739. Epstein-Barr virus BZLF1 protein inhibits tumor necrosis factor receptor 1 ex- 2. Sospedra, M., and R. Martin. 2005. Immunology of multiple sclerosis. Annu. pression through effects on cellular C/EBP proteins. J. Virol. 84: 12362–12374. Rev. Immunol. 23: 683–747. 27. Ravimohan, S., L. Gama, S. A. Barber, and J. E. Clements. 2010. Regulation of 3. International Multiple Sclerosis Genetics Consortium, Wellcome Trust Case SIV mac 239 basal long terminal repeat activity and viral replication in mac- Control Consortium 2, S. Sawcer, G. Hellenthal, M. Pirinen, C. C. Spencer, rophages: functional roles of two CCAAT/enhancer-binding protein beta sites in N. A. Patsopoulos, L. Moutsianas, A. Dilthey, Z. Su, C. Freeman, S. E. Hunt, activation and interferon beta-mediated suppression. J. Biol. Chem. 285: 2258– et al. 2011. Genetic risk and a primary role for cell-mediated immune mecha- 2273. nisms in multiple sclerosis. Nature 476: 214–219. 28. Comabella, M., J. D. Lunemann, J. Rio, A. Sanchez, C. Lopez, E. Julia, 4. International Multiple Sclerosis Genetics Consortium, A. H. Beecham, M. Fernandez, L. Nonell, M. Camina-Tato, F. Deisenhammer, E. Caballero, N. A. Patsopoulos, D. K. Xifara, M. F. Davis, A. Kemppinen, C. Cotsapas, M. T. Tortola, M. Prinz, X. Montalban, and R. Martin. 2009. A type I interferon

T. S. Shah, C. Spencer, D. Booth, A. Goris, et al. 2013. Analysis of immune- signature in monocytes is associated with poor response to interferon-beta in by guest on September 27, 2021 related loci identifies 48 new susceptibility variants for multiple sclerosis. Nat. multiple sclerosis. Brain 132: 3353–3365. Genet. 45: 1353–1360. 29. Rudick, R. A., M. R. Rani, Y. Xu, J. C. Lee, J. Na, J. Shrock, A. Josyula, 5. Gregory, A. P., C. A. Dendrou, K. E. Attfield, A. Haghikia, D. K. Xifara, E. Fisher, and R. M. Ransohoff. 2011. Excessive biologic response to IFNb is F. Butter, G. Poschmann, G. Kaur, L. Lambert, O. A. Leach, et al. 2012. TNF associated with poor treatment response in patients with multiple sclerosis. PLoS receptor 1 genetic risk mirrors outcome of anti-TNF therapy in multiple scle- ONE 6: e19262. rosis. Nature 488: 508–511. 30. Huber, J. P., and J. D. Farrar. 2011. Regulation of effector and memory T-cell 6. Gregory, S. G., S. Schmidt, P. Seth, J. R. Oksenberg, J. Hart, A. Prokop, functions by type I interferon. Immunology 132: 466–474. S. J. Caillier, M. Ban, A. Goris, L. F. Barcellos, et al; Multiple Sclerosis 31. Go´mez-Martıń, D., M. Ibarra-Sańchez, J. Romo-Tena, J. Cruz-Ruı´z, J. Esparza- Genetics Group. 2007. Interleukin 7 receptor alpha chain (IL7R) shows Lo´pez, M. Galindo-Campos, M. Dıáz-Zamudio, and J. Alcocer-Varela. 2013. allelic and functional association with multiple sclerosis. Nat. Genet. 39: Casitas B lineage lymphoma b is a key regulator of peripheral tolerance in 1083–1091. systemic lupus erythematosus. Arthritis Rheum. 65: 1032–1042. 7. Kreft, K. L., E. Verbraak, A. F. Wierenga-Wolf, M. van Meurs, B. A. Oostra, 32. Harada, Y., Y. Harada, C. Elly, G. Ying, J. H. Paik, R. A. DePinho, and Y. C. Liu. J. D. Laman, and R. Q. Hintzen. 2012. Decreased systemic IL-7 and soluble IL- 2010. Transcription factors Foxo3a and Foxo1 couple the E3 ligase Cbl-b to the 7Ra in multiple sclerosis patients. Genes Immun. 13: 587–592. induction of Foxp3 expression in induced regulatory T cells. J. Exp. Med. 207: 8. International Multiple Sclerosis Genetics Consortium, D. A. Hafler, A. Compston, 1381–1391. S. Sawcer, E. S. Lander, M. J. Daly, P. L. De Jager, P. I. de Bakker, S. B. Gabriel, 33. Qiao, G., Y. Zhao, Z. Li, P. Q. Tang, W. Y. Langdon, T. Yang, and J. Zhang. D. B. Mirel, A. J. Ivinson, et al. 2007. Risk alleles for multiple sclerosis identified 2013. T cell activation threshold regulated by E3 ubiquitin ligase Cbl-b determines by a genomewide study. N. Engl. J. Med. 357: 851–862. fate of inducible regulatory T cells. J. Immunol. 191: 632–639. 9. Sanna, S., M. Pitzalis, M. Zoledziewska, I. Zara, C. Sidore, R. Murru, 34. Viglietta, V., C. Baecher-Allan, H. L. Weiner, and D. A. Hafler. 2004. Loss of M. B. Whalen, F. Busonero, A. Maschio, G. Costa, et al. 2010. Variants within functional suppression by CD4+CD25+ regulatory T cells in patients with the immunoregulatory CBLB gene are associated with multiple sclerosis. Nat. multiple sclerosis. J. Exp. Med. 199: 971–979. Genet. 42: 495–497. 35. Simoń-Sańchez, J., and A. Singleton. 2008. Genome-wide association studies in 10. Bachmaier, K., C. Krawczyk, I. Kozieradzki, Y. Y. Kong, T. Sasaki, A. Oliveira- neurological disorders. Lancet Neurol. 7: 1067–1072. dos-Santos, S. Mariathasan, D. Bouchard, A. Wakeham, A. Itie, et al. 2000. 36. Baranzini, S. E., N. W. Galwey, J. Wang, P. Khankhanian, R. Lindberg, Negative regulation of lymphocyte activation and autoimmunity by the molec- D. Pelletier, W. Wu, B. M. Uitdehaag, L. Kappos, C. H. Polman, et al; GeneMSA ular adaptor Cbl-b. Nature 403: 211–216. Consortium. 2009. Pathway and network-based analysis of genome-wide asso- 11. Chiang, Y. J., H. K. Kole, K. Brown, M. Naramura, S. Fukuhara, R. J. Hu, ciation studies in multiple sclerosis. Hum. Mol. Genet. 18: 2078–2090. I. K. Jang, J. S. Gutkind, E. Shevach, and H. Gu. 2000. Cbl-b regulates the CD28 37. McCarthy, M. I., G. R. Abecasis, L. R. Cardon, D. B. Goldstein, J. Little, dependence of T-cell activation. Nature 403: 216–220. J. P. Ioannidis, and J. N. Hirschhorn. 2008. Genome-wide association studies for 12. Dubey, C., M. Croft, and S. L. Swain. 1996. Naive and effector CD4 T cells complex traits: consensus, uncertainty and challenges. Nat. Rev. Genet. 9: 356– differ in their requirements for T cell receptor versus costimulatory signals. 369. J. Immunol. 157: 3280–3289. 38. Wang, K., S. P. Dickson, C. A. Stolle, I. D. Krantz, D. B. Goldstein, and 13. Yokoi, N., K. Komeda, H. Y. Wang, H. Yano, K. Kitada, Y. Saitoh, Y. Seino, H. Hakonarson. 2010. Interpretation of association signals and identification of K. Yasuda, T. Serikawa, and S. Seino. 2002. Cblb is a major susceptibility gene causal variants from genome-wide association studies. Am. J. Hum. Genet. 86: for rat type 1 diabetes mellitus. Nat. Genet. 31: 391–394. 730–742. 14. Hoyne, G. F., E. Flening, M. Yabas, C. Teh, J. A. Altin, K. Randall, C. B. Thien, 39. Edwards, S. L., J. Beesley, J. D. French, and A. M. Dunning. 2013. Beyond W. Y. Langdon, and C. C. Goodnow. 2011. Visualizing the role of Cbl-b in GWASs: illuminating the dark road from association to function. Am. J. Hum. control of islet-reactive CD4 T cells and susceptibility to type 1 diabetes. Genet. 93: 779–797. J. Immunol. 186: 2024–2032. 40. Kondo, T., I. Cortese, S. Markovic-Plese, K. P. Wandinger, C. Carter, M. Brown, 15. Paolino, M., and J. M. Penninger. 2010. Cbl-b in T-cell activation. Semin. S. Leitman, and R. Martin. 2001. Dendritic cells signal T cells in the absence of Immunopathol. 32: 137–148. exogenous antigen. Nat. Immunol. 2: 932–938. The Journal of Immunology 4447

41. Sibley, W. A., C. R. Bamford, and K. Clark. 1985. Clinical viral infections and 47. Kowenz-Leutz, E., O. Pless, G. Dittmar, M. Knoblich, and A. Leutz. 2010. multiple sclerosis. Lancet 1: 1313–1315. Crosstalk between C/EBPbeta phosphorylation, arginine methylation, and SWI/ 42. van Baarsen, L. G., T. C. van der Pouw Kraan, J. J. Kragt, J. M. Baggen, SNF/Mediator implies an indexing transcription factor code. EMBO J. 29: 1105– F. Rustenburg, T. Hooper, J. F. Meilof, M. J. Fero, C. D. Dijkstra, C. H. Polman, 1115. and C. L. Verweij. 2006. A subtype of multiple sclerosis defined by an activated 48. Parkin, S. E., M. Baer, T. D. Copeland, R. C. Schwartz, and P. F. Johnson. 2002. immune defense program. Genes Immun. 7: 522–531. Regulation of CCAAT/enhancer-binding protein (C/EBP) activator proteins by 43. Welm, A. L., S. L. Mackey, L. T. Timchenko, G. J. Darlington, and heterodimerization with C/EBPgamma (Ig/EBP). J. Biol. Chem. 277: 23563– N. A. Timchenko. 2000. Translational induction of liver-enriched transcriptional 23572. inhibitory protein during acute phase response leads to repression of CCAAT/ 49. Comabella, M., D. W. Craig, C. Morcillo-Sua´rez, J. Rıó, A. Navarro, enhancer binding protein alpha mRNA. J. Biol. Chem. 275: 27406–27413. M. Fernańdez, R. Martin, and X. Montalban. 2009. Genome-wide scan of 44. Kowenz-Leutz, E., and A. Leutz. 1999. A C/EBP beta isoform recruits the SWI/ 500,000 single-nucleotide polymorphisms among responders and nonresponders SNF complex to activate myeloid genes. Mol. Cell 4: 735–743. to interferon beta therapy in multiple sclerosis. Arch. Neurol. 66: 972–978. 45. Ki, S. H., I. J. Cho, D. W. Choi, and S. G. Kim. 2005. Glucocorticoid receptor 50. Akira, S., H. Isshiki, T. Sugita, O. Tanabe, S. Kinoshita, Y. Nishio, T. Nakajima, (GR)-associated SMRT binding to C/EBPbeta TAD and Nrf2 Neh4/5: role of T. Hirano, and T. Kishimoto. 1990. A nuclear factor for IL-6 expression (NF- SMRT recruited to GR in GSTA2 gene repression. Mol. Cell. Biol. 25: 4150–4165. IL6) is a member of a C/EBP family. EMBO J. 9: 1897–1906. 46. Raghav, S. K., S. M. Waszak, I. Krier, C. Gubelmann, A. Isakova, 51. Sellebjerg, F., M. Krakauer, M. Khademi, T. Olsson, and P. S. Sørensen. 2012. T. S. Mikkelsen, and B. Deplancke. 2012. Integrative genomics identifies the FOXP3, CBLB and ITCH gene expression and cytotoxic T lymphocyte antigen corepressor SMRT as a gatekeeper of adipogenesis through the transcription 4 expression on CD4(+) CD25(high) T cells in multiple sclerosis. Clin. Exp. factors C/EBPb and KAISO. Mol. Cell 46: 335–350. Immunol. 170: 149–155. Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021