African-Derived Genetic Polymorphisms in TNFAIP3 Mediate Risk for Autoimmunity James P. Lodolce, Lauren E. Kolodziej, Lesley Rhee, Silvia N. Kariuki, Beverly S. Franek, Nancy M. McGreal, Mark F. This information is current as Logsdon, Sarah J. Bartulis, Minoli A. Perera, Nathan A. of September 29, 2021. Ellis, Erin J. Adams, Stephen B. Hanauer, Meenakshi Jolly, Timothy B. Niewold and David L. Boone J Immunol 2010; 184:7001-7009; Prepublished online 7 May

2010; Downloaded from doi: 10.4049/jimmunol.1000324 http://www.jimmunol.org/content/184/12/7001

Supplementary http://www.jimmunol.org/content/suppl/2010/05/07/jimmunol.100032 http://www.jimmunol.org/ Material 4.DC1 References This article cites 45 articles, 10 of which you can access for free at: http://www.jimmunol.org/content/184/12/7001.full#ref-list-1

Why The JI? Submit online. by guest on September 29, 2021 • Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*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 © 2010 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

African-Derived Genetic Polymorphisms in TNFAIP3 Mediate Risk for Autoimmunity

James P. Lodolce,*,1 Lauren E. Kolodziej,*,1 Lesley Rhee,* Silvia N. Kariuki,* Beverly S. Franek,* Nancy M. McGreal,*,2 Mark F. Logsdon,* Sarah J. Bartulis,* Minoli A. Perera,* Nathan A. Ellis,* Erin J. Adams,† Stephen B. Hanauer,* Meenakshi Jolly,‡ Timothy B. Niewold,* and David L. Boone*

The TNF a-induced 3 (TNFAIP3) is an ubiquitin-modifying enzyme and an essential negative regulator of inflammation. Genome-wide association studies have implicated the TNFAIP3 in susceptibility to autoimmune disorders in European cohorts, including , coronary artery disease, psoriasis, celiac disease, type 1 diabetes, inflammatory bowel

disease, and systemic lupus erythematosus (SLE). There are two nonsynonymous coding polymorphisms in the deubiquitinating Downloaded from (DUB) domain of TNFAIP3: F127C, which is in high-linkage disequilibrium with reported SLE-risk variants, and A125V, which has not been previously studied. We conducted a case–control study in African-American SLE patients using these coding variants, along with tagging polymorphisms in TNFAIP3, and identified a novel African-derived risk haplotype that is distinct from previously reported risk variants (odds ratio = 1.6, p = 0.006). In addition, a rare protective haplotype was defined by A125V (odds ratio = 0.31, p = 0.027). Although A125V was associated with protection from SLE, surprisingly the same allele was

associated with increased risk of inflammatory bowel disease. We tested the functional activity of nonsynonymous coding poly- http://www.jimmunol.org/ morphisms within TNFAIP3, and found that the A125V coding-change variant alters the DUB activity of the protein. Finally, we used computer modeling to depict how the A125V amino acid change in TNFAIP3 may affect the three-dimensional structure of the DUB domain to a greater extent than F127C. This is the first report of an association between TNFAIP3 polymorphisms and autoimmunity in African-Americans. The Journal of Immunology, 2010, 184: 7001–7009.

he TNF a-induced protein 3 (TNFAIP3, also known as (RIP1) and TNF receptor-associated factor (TRAF) 6, respectively A20) is required for the negative regulation of inflammatory (2, 3). TNFAIP3 controls inflammation by removing the signaling responses (1). It is a dual function ubiquitin-modifying en- form of K63-linked polyubiquitin (deubiquitinating [DUB] activity)

T by guest on September 29, 2021 zyme capable of deubiquitination, as well as ubiquitin ligation, of and attaching degradative K48-linked polyubiquitin (ubiquitin ligase target (2). TNFAIP3 terminates TNF and TLR responses by activity) to cell signaling molecules like RIP1 and TRAF6. The modifying the signaling proteins receptor-interacting protein 1 combined actions of TNFAIP3 result in the elimination of key acti- vated signaling proteins in inflammatory pathways illustrating the importance of TNFAIP3 in the termination of inflammation down- *Department of Medicine and †Department of Biochemistry and Molecular Biophys- stream of multiple stimuli. Mice deficient in TNFAIP3 develop se- ‡ ics, University of Chicago; and Rush University Medical Center, Chicago, IL 60612 vere multiorgan inflammation that leads to premature death (4). 1J.P.L. and L.E.K. contributed equally to this work. Systemic inflammation occurs spontaneously and resembles many 2Current address: Duke University, Durham, NC. features of human autoimmune disease. Received for publication February 2, 2010. Accepted for publication April 8, 2010. The multiple potential roles of TNFAIP3 in regulating in- This work was supported by National Institutes of Health Grants DK42086 (to D.L.B.), flammation are not fully understood, and are likely to be cell and 1F32AI077235-01A1 (to J.P.L.), T32AI-07090 (to J.P.L.), DK07074 (to L.E.K.), context dependent. For example, TNFAIP3 mutations/deletions are K08AI083790 (to T.B.N.), and AI071651 (to T.B.N.), Crohn’s and Colitis Foundation of America (to D.L.B.), International Organization for the Study of Inflammatory Bowel frequently associated with the development of B cell lymphomas, Disease (to D.L.B. and S.B.H.), Arthritis National Research Foundation Eng Tan Scholar suggesting that loss of TNFAIP3 in this cell type is oncogenic, Award (to T.B.N.), Lupus Research Institute Novel Research Grant (to T.B.N.), University possibly due to increased NF-kB signals that are antiapoptotic/ of Chicago Clinical and Translational Science Award (to T.B.N.), and Collaborative Uni- versity of Chicago/Northshore University Health System Translational Research Pilot proproliferative in B lymphocytes (5–7). Conversely, TNFAIP3 is Grant UL1 RR024999 (to T.B.N.). required to protect thymocytes or fibroblasts from TNF-induced Address correspondence and reprint requests to Dr. Timothy B. Niewold, 924 East , despite prolonged NF-kB activation (4). TNFAIP3 57th Street, JFK R420, Chicago, IL 60637, or Dr. David L. Boone, 900 East 57th appears to be a central regulator of inflammatory responses, and Street, Mailbox 9, Chicago, IL 60637. E-mail addresses: [email protected]. uchicago.edu or [email protected] it is likely that multiple distinct variants of TNFAIP3 could differ- The online version of this article contains supplemental material. entially modulate risk of autoimmunity in different diseases and different ancestral backgrounds. Abbreviations used in this paper: CAD, coronary artery disease; CD, Crohn’s disease; CeD, celiac disease; CI, confidence interval; dbSNP, Single Nucleotide Polymorphism Systemic lupus erythematosus (SLE) is a complex autoimmune Database; DUB, deubiquitinating; IB, immunoblotted; IBD, inflammatory bowel disease; disease that is frequently characterized by inappropriate type I IFN IP, immunoprecipitation; K48R-Ub, K48R-ubiquitin; LD, linkage disequilibrium; OR, odds ratio; PVDF, polyvinylidene fluoride; RA, rheumatoid arthritis; RIP1, receptor- responses driven by small nuclear self-Ags (8). Humoral auto- interacting protein 1; SLE, systemic lupus erythematosus; SNP, single nucleotide poly- immunity is a hallmark of disease, and the formation of immune morphism; T1D, type 1 diabetes; TNFAIP3, TNF a-induced protein 3; TRAF, TNF re- complexes containing self-Ags leads to local and systemic inflam- ceptor-associated factor; YRI, Yoruba population in Ibadan, Nigeria; WT, wild-type. mation and subsequent clinical symptoms. The pathogenesis of Copyright Ó 2010 by The American Association of Immunologists, Inc. 0022-1767/10/$16.00 SLE likely involves both environmental and genetic triggers, the www.jimmunol.org/cgi/doi/10.4049/jimmunol.1000324 7002 TNFAIP3 AND AUTOIMMUNITY latter highlighted by the observation that siblings of affected indi- functional effect, greatly decreasing TNFAIP3-mediated deubiquiti- viduals have a 20- to 30-fold increased risk of developing SLE (9). nation. Finally, we used computer modeling to examine how these In addition, compared with European-Americans the incidence of amino acid changes may be affecting the predicted three-dimensional SLE is three to four times higher in African-Americans (10, 11). structure of the TNFAIP3 DUB domain. Genome-wide association studies have begun to identify that contribute to SLE susceptibility (12–17). Most of these stud- Materials and Methods ies have been conducted exclusively in European cohorts; thus, Patient samples little is known about whether variants in these genes are risk Patient genomic DNA samples were obtained from the Translational Re- factors in other ancestral backgrounds. Differences in autoimmune search Initiative in the Department of Medicine registry at the University of disease risk variants exist across different continental populations. Chicago (n = 1222) and from the Rush University Medical Center (n = 33). For example, the autoimmune disease risk variant of the protein A total of 217 self-identified African-American SLE patients, 24 African- tyrosine phosphatase type 22 (PTPN22) is almost absent in American IBD patients, 272 African-American nonautoimmune controls, and 615 self-identified white or European-American samples of mixed African-derived populations (18). disease status were studied. All SLE patients met four or more of the The TNFAIP3 gene region has been implicated in susceptibility American College of Rheumatology criteria for diagnosis (29). Diagnosis to multiple autoimmune diseases in populations of European an- of IBD was made histologically using intestinal biopsies. The control cestry. Single nucleotide polymorphisms (SNPs) ∼185 kb up- population consisted of adults recruited from outpatient clinics at the Uni- stream of the TNFAIP3 gene are associated with risk for versity of Chicago who were either normal healthy or had an unrelated condition diagnosed, such as cancer, type 2 diabetes, or hypertension. The rheumatoid arthritis (RA) (19, 20), type 1 diabetes (T1D) (21), and medical records of the control population were examined to exclude those celiac disease (CeD) (22). The Wellcome Trust Case Control with autoimmune or inflammatory disease diagnoses, including SLE, CD, Downloaded from Consortium identified a SNP upstream of TNFAIP3 in 6q23 that is sarcoidosis, autoimmune thyroid disease, psoriasis, CeD, RA, and T1D. associated with risk for Crohn’s disease (CD) (23). In addition, Subjects with documented CAD were also excluded given the results of a previous study suggesting an association between TNFAIP3 variants and this same chromosomal region was found as a candidate for sus- CAD (26). The institutional review boards at both institutions approved ceptibility in a study of 260 inflammatory bowel disease (IBD)- this study, and informed consent was obtained from all participants. affected relative pairs (24). Intronic TNFAIP3 SNPs are associated Genotyping with psoriasis (25) and coronary artery disease (CAD) in type 2 http://www.jimmunol.org/ diabetes patients (26). Individual patients were genotyped at the coding SNPs A125V (rs5029941) Two independent studies have found an association between and F127C (rs2230926) using direct sequencing. The close proximity of these TNFAIP3 and SLE in European ancestry. The same SNP identified two SNPs made probe-based genotyping difficult, and so direct sequencing ∼ was used instead. Briefly, a 303 bp fragment was amplified from genomic in RA and CeD that falls 185 kb upstream of TNFAIP3 was also DNA using PCR (forward primer 59-GAAAACCTTTGCTGGGTCTT- associated with SLE risk, and a second SLE risk haplotype was ACAT-39, reverse primer 59-CCATGGAGCTCTGTTAGTAGATAA-39). identified within the TNFAIP3 gene (15). Another study suggested The PCR product was treated with Exonuclease I and Shrimp Alkaline a complex contribution of three individual variants, including Phosphatase (USB, Cleveland, OH) prior to submission for ultra high- throughput sequencing at the University of Chicago Cancer Research Center a nonsynonymous F127C coding SNP (rs2230926) (17). Meta- DNA Sequencing Facility. Sequencing was performed on an Applied Bio- analysis and imputation has narrowed down the SLE-associated systems 3730XL capillary sequencer (Applied Biosystems, Foster City, CA). by guest on September 29, 2021 region of TNFAIP3 in patients of European ancestry to a single Genotype calls were determined by visual inspection of individual 109 kb haplotype centered on the TNFAIP3 gene, which includes peaks in each sequence trace file. The haplotype-tagging SNPs rs5029939, the F127C SNP (27). In functional studies, Musone et al. demon- rs5029953, rs2230926, rs610604, rs3757173, rs5029938, and rs582757 were chosen by focusing on the TNFAIP3 gene and the regions 10 kb upstream and strated decreased activity of the TNFAIP3 protein with the SLE- downstream of the coding sequence. A two and three marker-tagging algo- associated F127C amino acid change (rs2230926) (17). The iden- rithm was used with a correlation coefficient (r2) threshold of 0.8 to choose tification of a functionally defective risk variant may explain how SNPs with an allele frequency .10% in the International HapMap Project TNFAIP3 contributes to SLE pathogenesis. In people of European YRI population. Haplotype-tagging SNPs were genotyped using ABI Taq- Man primers and probes on the ABI 7900HT Fast Real-Time PCR System ancestry, the minor allele frequency of rs2230926 is relatively low (Applied Biosystems) per manufacturer’s protocol. All scatter plots were (0.05 in SLE cases and 0.02 in controls) compared with the examined visually to verify the accuracy of the genotype calls. None of the reported frequency of 0.447 in the Yoruba population in Ibadan, SNPs significantly deviated from predicted Hardy-Weinberg equilibrium in Nigeria (YRI) (28). In addition, the association between SLE and African-American control subjects (lowest p value = 0.0680) (Table I). Two the TNFAIP3 gene region appears to be complex, and not com- tagging SNPs (rs5029938 and rs582757) were genotyped in only 25% of cases and 48% of controls due to either low allele frequency and/or low pletely accounted for by the F127C polymorphism. Overall, these information content, whereas the remaining six SNPs were successfully studies demonstrate that SNPs within or near TNFAIP3 are asso- genotyped in .93% of controls and .99% of cases (Supplemental Table I). ciated with autoimmunity, and significant differences in TNFAIP3 Statistical analysis SNP frequencies between European and African populations sug- gest that disease-associated haplotypes in these two ancestral Odds ratios with 95% confidence intervals and p values using two-tailed x2 backgrounds may be dissimilar. tests were calculated with GraphPad Prism software (GraphPad, San Diego, CA). None of the reported p values were corrected for multiple We therefore assessed the contribution of TNFAIP3 variants to testing. The odds ratios (ORs) were adjusted to account for the “winner’s autoimmune risk in African-Americans. To better understand how curse” using WINNER version 1.1 (http://csg.sph.umich.edu/boehnke/win- alterations in TNFAIP3 affect autoimmune susceptibility, we con- ner/) written by Rui Xiao and Michael Boehnke (30). Ascertainment- ducted a case–control study to look at the association between hap- corrected ORs are listed in Supplemental Table I. In addition, the mini- lotype-tagging and coding-change SNPs within the TNFAIP3 gene mum genetic effect that could be significantly (p = 0.05) detected in this study with 80% power was calculated using an additive model in the CaTS and SLE in African-Americans. We identified a novel African- Power Calculator written by Andrew Skol (www.sph.umich.edu/csg/abe- derived SLE-risk haplotype and a protective haplotype for SLE de- casis/CaTS/) and Power for Association with Error, version 1.2, written by fined by the A125V polymorphism. We genotyped the A125V and Derek Gordon was used for the smaller cohorts that cannot be calculated F127C coding variants in African-American subjects with IBD and using the CaTS program. Power for Association with Error can be found at http://linkage.rockefeller.edu/joanne/pawe/pawe.cgi (31–33). In the Afri- found that the A125V variant was associated with risk for IBD. Next, can-American SLE cohort, this study had 80% power to detect an OR of we tested the functional effect of four nonsynonymous coding poly- $1.53 with a p value of 0.05 using an additive genetic model over the morphisms in TNFAIP3. The A125V SNP demonstrated a dramatic allele frequency range represented by the SNPs studied, except the rare The Journal of Immunology 7003 variant A125V (Supplemental Table I). In the case of A125V, we could were incubated overnight at 4˚C with anti-HA Ab and for 2 h at 4˚C with detect an OR of 0.33 or lower for A125V using the same parameters as a 50% slurry of Protein G (Pierce, Rockford, IL). Beads were recovered, above. In the African-American IBD cohort, we had 75% power to detect washed 3 times with lysis buffer, and mixed with equal volume 23 an OR of $3.7 for association of A125V using the same parameters as Laemmli sample buffer (with 50 mM DTT). For SDS-PAGE, whole cell above. The degree of pairwise linkage disequilibrium (LD) between SNPs lysates and immunoprecipitated protein were resolved on NuPAGE Novex (r2 values) and haplotype frequencies were determined with Haploview 4.1 4–12% Bis-Tris gels (Invitrogen) and transferred to polyvinylidene fluoride software. Linkage blocks were defined using the solid spine of LD method (PVDF) membranes (Millipore, Bedford, MA). Membranes were blocked implemented by the Haploview software. Multivariate and conditional for 45 min at room temperature with blocking buffer for near infra-red logistic regression models were performed using an online tool written fluorescent Western blotting (Rockland) diluted in PBS (1:1) before over- by John C. Pezzullo (http://statpages.org/logistic.html). night incubation at 4˚C with primary Ab. The following day membranes were incubated for 45 min with infrared secondary Abs and imaged using Expression plasmids IR fluorescence detection on the LI-COR Odyssey Infrared Imaging Sys- The wild-type (WT) human TNFAIP3 cDNA was PCR cloned from tem (Li-Cor Biosciences, Lincoln, NE). The intensity of the bands was HCT116 epithelial cell cDNA into a pENTR entry vector (Invitrogen, quantified by using the area under the curve method in ImageJ (http:// San Diego, CA). The C103A, A125V, F127C, R706Q, and A766P point rsbweb.nih.gov/ij/). mutants, as well as the A125V/F127C double mutant and the E3 ligase- deficient mutant (C607A, C612A, C624A, and C627A) were generated Results from WT TNFAIP3 according to the QuikChange Site-Directed Mutagen- esis Kit instructions (Stratagene, La Jolla, CA). The V5-tagged TNFAIP3 Polymorphisms in the TNFAIP3 locus are associated with SLE constructs were each generated by shuttling the gene into pcDNA3.1/nV5 risk in African-Americans using the Gateway Clonase LR II enzyme kit (Invitrogen). Cytokine re- A case–control study was conducted using a cohort of African- sponse modifier A and pCMV-Myc-ubiquitin expression plasmids were gifts from M. Peter and C. Maki, respectively (University of Chicago, American subjects to examine the association of TNFAIP3 var- Downloaded from Chicago, IL). WT Myc-ubiquitin was generated from Myc-K48R- iants with altered SLE risk. SLE cases (n = 217) and nonautoim- ubiquitin (K48R-Ub) using site-directed mutagenesis (Stratagene). All mune controls (n = 272) were genotyped with real-time PCR using plasmids were confirmed by sequencing through the University of Chicago TaqMan probe chemistry at six tagging SNPs within the TNFAIP3 Cancer Research Center DNA Sequencing Facility. gene region. In addition, the Single Nucleotide Polymorphism Cell culture and Abs Database (dbSNP) lists two SNPs within the DUB domain of

HEK293 cells (American Type Culture Collection, Manassas, VA) were TNFAIP3 (aa 92–263) that result in nonsynonymous coding http://www.jimmunol.org/ maintained in DMEM supplemented with 10% FBS, penicillin 100 mg/ml, changes in the protein sequence. These include rs5029941 and streptomycin 100 U/ml (Life Technologies, Rockville, MD) and in- (A125V) and rs2230926 (F127C), the latter associated with SLE cubated at 37˚C in a humidified, 5% CO2 atmosphere. Polyclonal rabbit risk in European populations (15, 17). Thus, in addition to the six anti-HA Ab (Y-11) and goat anti-actin were purchased from Santa Cruz tagging SNPs, we also genotyped these two coding polymor- Biotechnology (Santa Cruz, CA); monoclonal mouse anti-Myc and anti-V5 Abs were purchased from Invitrogen. Species-specific secondary Abs con- phisms in our patient samples using sequencing. jugated to IRDye or Alexa Fluor infrared dyes (680 nm or 800 nm) were We tested each of the eight TNFAIP3 SNPs for association with used at a 1:20,000 dilution (Rockland, Gilbertsville, PA, and Invitrogen). SLE and found that two tagging SNPs (rs5029939 and Transfections, immunoprecipitation, and immunoblot analysis rs5029953), along with the F127C coding SNP, were associated with SLE susceptibility (Table I). In Europeans, the tagging SNP by guest on September 29, 2021 6 For transient transfection assays, HEK293 cells were seeded at 2 3 10 cells rs5029939 (p = 0.0453, OR = 1.3) and the coding variant F127C per 10 cm plate (BD Falcon, Franklin Lakes, NJ) 18 h prior to transfection. Immediately before transfection, cells were rinsed in 13 PBS (Life (p = 0.0279, OR = 1.3) were identified as SLE risk variants, and Technologies) and supplemented with serum-free media. Plasmid DNA similar to the European studies, we find these two variants in very (7.5 mg total) and a 2 mg/ml polyethylenimine solution (Polysciences, high LD (r2 = 0.99) (15, 17). The most significant SLE risk- Warrington, PA) were each diluted in 150 ml final volume of 150 mM associated SNP we observed was not either of these two previ- NaCl and combined by adding the polymer solution dropwise to the plas- ously associated SNPs, but was instead the African-derived mid solution while vortexing. All transfections included 0.5 mg cytokine response modifier A expression vector to protect against cell death. After rs5029953 SNP (p = 0.008, OR = 1.6). Conditioning on 30 min the transfection mixture was added to the cells. The following day, rs5029953 abolished the association signals from the other SNPs cells were supplemented with complete media. Forty hours after transfec- (rs5029939 and F127C), indicating that they are not independently tion, samples were washed in PBS and lysed in 1% Triton lysis buffer contributing to risk (Table I). No SNP-SNP interactions were (50 mM HEPES, 150 mM NaCl, 1.5 mM MgCl2, 1 mM EGTA, 1% Triton X-100, 10% glycerol with 20 mM N-ethylmaleimide, 10 mM observed, and thus the simple model of a single risk variant 1,10-phenanthroline monohydrate, and 13 protease inhibitor mixture) marked by rs5029953 was the best fit (data not shown). In the (Roche, Basel, Switzerland). For immunoprecipitation (IP) studies, lysates dbSNP database the minor allele of rs5029953 has a reported

Table I. Single marker and conditional association analysis of TNFAIP3 SNPs in African-American SLE cases

Allele Ratio Counts Minor Allele Freq p Value (Conditioned SNP Minor Allele (Cases, Controls) (Cases, Controls) p Valuea OR (95% CI) on rs5029953) HWE p Valueb rs3757173 G 210:220, 226:302 0.488, 0.428 0.0621 1.276 (0.9875–1.648) 0.5069 0.0680 rs5029938 T 15:195, 16:124 0.071, 0.114 0.1668 0.5962 (0.2845–1.249) — 0.7609 rs5029939 G 166:268, 172:364 0.382, 0.321 0.0453 1.311 (1.005–1.709) 0.5667 0.2606 A125V T 5:429, 15:491 0.0115, 0.0296 0.0549 0.3815 (0.1375–1.059) — 0.3848 F127C G 167:267, 172:370 0.385, 0.317 0.0279 1.345 (1.032–1.753) 0.4489 0.2263 rs582757 C 66:144, 51:87 0.314, 0.370 0.2856 0.7819 (0.4975–1.229) — 0.3489 rs610604 T 156:278, 197:325 0.359, 0.377 0.5670 0.9258 (0.7108–1.206) — 0.7017 rs5029953 A 80:354, 66:470 0.184, 0.123 0.0080 1.609 (1.130–2.293) — 0.1297 The eight TNFAIP3 SNPs genotyped in this study are listed, along with the identity of the minor allele. Minor allele frequency in cases and controls is shown. The p value (from x2 analysis) and OR with 95% confidence interval (CI) was calculated with GraphPad Prism software. SNPs with a p value ,0.10 were conditioned on rs5029953 using logistic regression. The two predictor variables were the rs5029953 genotype and the genotype of the SNP being tested with the outcome variable being SLE. ap Values are uncorrected for multiple testing. bCalculated using the control population. 7004 TNFAIP3 AND AUTOIMMUNITY frequency ranging from 0.077 to 0.161 in African populations, and nonautoimmune controls), and the A125V and F127C genotypes this SNP has not been identified in Asian or European populations. were analyzed together as predictor variables in a logistic regression We genotyped a total of 324 white or European-American samples, model. Strikingly, the minor allele of A125V is significantly as- and all carried the major allele of rs5029953, supporting the idea sociated with susceptibility to IBD (Table II) (OR = 3.73, p = 0.027). that this SNP is indeed of African origin. As noted previously, this same allele is associated with protection Surprisingly, we found that the rare minor allele of A125V was from SLE (OR = 0.33, p = 0.032). Thus, the rare variant A125V has associated with protection from SLE in African-American subjects opposite effects on risk of IBD and SLE in African-Americans. (p = 0.05, OR = 0.38) (Table I). We did not detect this variant in Despite the low number of African-American IBD case samples, self-reported white or European-ancestry subjects (n = 291) (Sup- this study had 75% power to detect the large genetic effect identi- plemental Table II), and this variant was present in our nonau- fied (OR = 3.7) with a p value = 0.05 (Supplemental Table I). toimmune African-American controls at a frequency of 0.03 Furthermore, correcting for the “winner’s curse effect” does not (Table II), which is comparable to the frequency reported in Af- significantly decrease the estimated OR for this association (Sup- rican populations in public databases. The minor allele of A125V plemental Table I). Although the low number of patients limits our is in high LD with the minor allele of F127C, as the two patients confidence in the IBD data to some degree, these preliminary data identified as homozygous for the minor allele of A125V were also suggest a strong effect, which is almost directly opposite of that homozygous for the minor allele of F127C, and every A125V observed in SLE. heterozygote was either heterozygous or homozygous for the mi- Diminished functional activity of A125V TNFAIP3 nor allele of F127C. The complete distribution of genotypes by ethnicity and autoimmune disease for both A125V and F127C can The observation that coding SNPs in TNFAIP3 are associated with Downloaded from be found in Supplemental Table II. autoimmunity in African-Americans suggests that these amino Haplotype analysis in African-American subjects detected eight acid substitutions may be altering the enzymatic function of the haplotypes with a frequency of $1% in cases and controls (Fig. 1A). protein. Because the A125V and the F127C polymorphisms occur The rs5029953 SNP associated with SLE risk is unique to one together (A125V is never found without F127C; whereas, F127C haplotype, and thus marks the SLE risk-associated haplotype (hap- is often found without A125V), we tested the functional activity of

lotype 2, OR = 1.6, p = 0.0061). A linkage disequilibrium plot of the A125V/F127C TNFAIP3. In addition to the A125V/F127C double http://www.jimmunol.org/ SLE cases or controls shows that among the SNPs analyzed, mutant, four single nonsynonymous coding mutations (A125V, rs5029953 is most strongly associated with rs5029939 and F127C F127C, R706Q, and A766P) were made in the TNFAIP3 cDNA (Fig. 1B, Supplemental Fig. 1). In addition, the A125V rare variant and cloned into mammalian expression vectors. R706Q and is only found on one haplotype, identifying the protective haplotype A766P were chosen because they are listed in the dbSNP database (haplotype 8, OR = 0.3, p = 0.027), which does not include the SLE- as validated, nonsynonymous coding SNPs located near the ubiq- risk-associated allele. These findings taken together suggest that the uitin ligase domain of TNFAIP3. These variants were not studied F127C polymorphism is not a major determinant of SLE risk in in the genetic association study, as their probable low frequency in African-Americans, and that instead a model of risk, neutral, and human populations based on HapMap data would likely prevent protective haplotypes seems to apply. Interestingly, the protective significant conclusions given our sample size. A previous study by guest on September 29, 2021 haplotype contains a rare coding polymorphism (A125V) that could demonstrated that mutating the cysteine at amino acid position alter TNFAIP3 activity; whereas, the functional implications of the 103 to alanine inhibits the DUB activity of TNFAIP3 (3); thus, risk haplotype are not currently clear. C103A TNFAIP3 was used as a negative control. TRAF2 is a key A key clinical feature of SLE is the production of autoantibodies signaling molecule in several inflammatory pathways, is known to against nucleic acids (i.e., anti-dsDNA) and small nuclear proteins be ubiquitinated with K63-linked chains, and has been shown to that stably interact with RNA (RNA binding proteins or RBPs). be associated with TNFAIP3 (35–37). TNFAIP3 is a dual func- Other genes that confer risk to SLE, such as IRF5, are associated tioning enzyme that both deubiquitinates and induces the degra- with higher IFN responses in patients with either anti-dsDNA dation of its targets. The two activities of this enzyme are linked, or anti-RBP autoantibodies (34). We tested whether any of the in that TNFAIP3 must deubiquitinate targets before it can induce TNFAIP3 SNPs studied could predict the production of any SLE- their degradation. The ability of these TNFAIP3 mutants to induce related autoantibodies, and found no significant associations (data the degradation of TRAF2 was tested to evaluate the functional not shown). consequence of nonsynonymous coding variants in TNFAIP3.WT TNFAIP3 induces the degradation of TRAF2 in a dose-dependent The F127C and A125V TNFAIP3 coding SNPs are associated manner; whereas, the catalytic mutant C103A fails to do so (Fig. with autoimmunity in African-Americans 2). We found that the A125V/F127C double mutant TNFAIP3 is We next assessed the contribution of the A125V and F127C poly- less efficient at inducing the degradation of TRAF2 when com- morphisms to inflammatory disease susceptibility. The patient pared with WT TNFAIP3 (Fig. 2A). Of the four single polymor- samples were grouped according to diagnosis (IBD, SLE, or phisms, the A125V SNP appears to have a diminished ability to

Table II. Frequency of TNFAIP3 SNPs A125V and F127C in African-American autoimmune cases and association with disease

IBD SLE Controls Minor SNP Allele MAF p Value OR (95% CI) MAF p Value OR (95% CI) MAF A125V T 0.104 0.027 3.73 (1.16–11.97) 0.012 0.032 0.33 (0.12–0.91) 0.030 F127C G 0.354 0.932 0.97 (0.49–1.92) 0.385 0.013 1.42 (1.08–1.87) 0.317 African-American DNA samples were genotyped by sequencing the region centered on the A125V and F127C nonsynon- ymous SNPs in TNFAIP3. Allele frequencies are compared between patient groups and nonautoimmune controls using x2 analysis, and ORs with 95% CI are shown for each comparison. The Journal of Immunology 7005

ligase activity of TNFAIP3. The WT TNFAIP3/E3 ligase mutant had reduced TRAF2 degradation in the whole cell lysates prior to immunoprecipitation (pre-IP). After cotransfection of V5-tagged TNFAIP3 (WT or mutant), HA-tagged TRAF2, and Myc-tagged K48R-Ub, the lysates were immunoprecipitated with Ab against HA-TRAF2. By using the K48R ubiquitin construct, polyubiqui- tin chains formed with this protein cannot be linked via the lysine at position 48. Thus, the Myc tag marks monoubiquitinated and polyubiquitinated proteins linked at a position other than K48 (e.g., the K63 signaling form of polyubiquitin). The amount of non–K48-linked polyubiquitinated TRAF2 (IP) or total non–K48- linked polyubiquitinated proteins (pre-IP) present with each TNFAIP3 construct was determined by immunoblotting for Myc-K48R-Ub. Consistent with the reduced degradation of TRAF2 by the A125V variant (Fig. 2B), the A125V mutant had decreased DUB activity against immunoprecipitated TRAF2, as well as total non–K48-linked polyubiquitinated proteins in the pre-IP compared with WT TNFAIP3. Thus, the A125V mutation

decreases the ability of TNFAIP3 to cause the deubiquitination Downloaded from and subsequent degradation of the target signaling protein TRAF2. Computer modeling of TNFAIP3 structure Of the four naturally occurring TNFAIP3 polymorphisms tested, we

were surprised to find such a strong functional defect in the A125V http://www.jimmunol.org/ mutant. Although both the A125Vand F127C are located within the N-terminal DUB domain of TNFAIP3, it was not initially obvious why an alanine to valine amino acid change would have a greater effect on protein function compared with a phenylalanine to cyste- FIGURE 1. Haplotype and LD analysis of TNFAIP3 SNPs in African- ine change. The DUB domain of TNFAIP3 has been crystallized, so American SLE cases. A, Using Haploview 4.1 software, the eight haplo- we used computer modeling to determine why these changes may types with frequencies $1% in cases and controls were determined. Six of be affecting enzymatic activity (38). We found that after making the . the eight SNPs were genotyped in 90% of our samples and therefore alanine to valine change, the R-group of the valine was pointing into included in the haplotype analysis. Haplotype 2 is the only one that is the center of protein and potentially causing steric interference with by guest on September 29, 2021 significantly associated with SLE risk, and only this haplotype contains the isoleucines 29 and 202 (Fig. 4A). This could result in a conforma- minor allele of rs5029953. Haplotype 8 is the protective haplotype, and is the only one carrying the minor allele of rs5029941 (A125V). B,AnLD tional change that affects the nearby catalytic core of the DUB plot generated from African American SLE cases shows the relative link- domain. The F127C substitution pointed away from the protein age between the TNFAIP3 SNPs genotyped in this study. The numbers in (Fig. 4B). Such a change may not directly interfere with the enzy- each box represent the pairwise r2 value 3100 between SNPs, calculated matic activity of the protein, but could influence the ability of target by the Haploview 4.1 software. Darker shaded boxes represent higher r2 proteins to bind with TNFAIP3. This could also explain why we values between SNPs. The location of each SNP within the general intron/ failed to see a significant functional consequence of F127C in our exon structure of the TNFAIP3 locus is indicated above the plot. The 59 and assay; whereas, Musone et al. observed decreased regulation of 39 untranslated regions are colored dark gray; whereas, the coding regions NF-kB activation by this variant (17). are colored light gray. The corresponding exon number is indicated above each exonic block. Discussion Our results demonstrate the first association of TNFAIP3 with au- induce degradation of TRAF2 (Fig. 2B). This defect is not a com- toimmunity in African-Americans. We have shown that an African- plete loss of function, because high amounts of A125V TNFAIP3 derived rare variant in the TNFAIP3 gene (A125V) is associated lead to loss of TRAF2, but not to the same extent as WT TNFAIP3. with risk for IBD, yet is protective for SLE. However, the role of In addition, the A766P mutation may have a subtle effect on the TNFAIP3 appears complex in SLE, as we also describe a common functional activity of TNFAIP3 and future genetic and biochemical African-derived haplotype associated with risk for SLE, marked by studies will be needed to determine the significance of this obser- the tagging SNP rs5029953. Thus, we can identify risk, neutral, and vation. Thus, the A125V polymorphism is likely the main cause of protective haplotypes in TNFAIP3 in African-Americans, and these the functional deficiency in the A125V/F127C double mutant. haplotypes are defined by African-derived genetic elements. We However, the A125V/F127C mutant may have a more profound tested the effects of nonsynonymous coding SNPs on the enzymatic functional defect (compare levels of TRAF2 degradation of activity of TNFAIP3 and found that the A125V mutation causes A125V/F127C TNFAIP3 in Fig. 2A with A125V TNFAIP3 in Fig. a decrease in the ability of TNFAIP3 to deubiquitinate and degrade 2B). Nevertheless, the F127C mutation does not appear to correct or the target protein TRAF2. Although our functional data suggest that significantly enhance the A125V TNFAIP3 enzymatic defect. this mutation causes a reduction in the activity of TNFAIP3, it is To further explore the effect of the A125V mutation, we tested unclear whether this results in increased or decreased risk for de- the DUB activity of the A125V TNFAIP3 single mutant on veloping inflammation. In fact, our genetic data argue that the effect ubiquitinated TRAF2 (Fig. 3). To reduce the amount of TRAF2 of this polymorphism in inflammatory disease will be complex, as substrate degraded by TNFAIP3, this experiment was performed the same polymorphism is associated with risk of one condition and with ubiquitin ligase domain mutant constructs lacking the E3 susceptibility to another. 7006 TNFAIP3 AND AUTOIMMUNITY

We found that the minor allele of the African-American–specific tagging SNP rs5029953 marks the SLE-risk haplotype in 139 African-American samples (n = 485), but in none of the European ancestry samples (n = 324). Our study covered a range of ∼10 kb centered on the TNFAIP3 gene. A closer examination of the link- age of rs5029953 with nearby SNPs in the International HapMap Project database’s YRI population suggests that this polymor- phism lies within a linkage block of ∼7 kb (Supplemental Fig. 2). There are nine other SNPs in LD with rs5029953 (r2 . 0.50), including SNPs that fall within two LD blocks upstream and one LD block downstream of rs5029953 covering ∼68 kb of the ge- netic region in and around the TNFAIP3 gene. All nine SNPs in high linkage with rs5029953 are of African origin (frequency range in YRI between 9–19%), with none of the minor alleles found in Europeans (Supplemental Table III). Thus, the SLE- risk haplotype we identify is of African origin. The SLE-risk signal in African-Americans is centered on the TNFAIP3 gene in our study, similar to the recent SLE meta-analysis in Europeans

(27). The functional element on this African-derived SLE-risk Downloaded from haplotype causing SLE susceptibility remains unclear, and the most strongly associated SNP (rs5029953) has no predicted func- tion. However, it is possible that this intronic SNP affects mRNA splicing, stability, or the binding of small regulatory RNAs. Other candidate functional elements include TNFAIP3 coding variants

located within the same LD block as rs5029953 (F127C, R706Q, http://www.jimmunol.org/ P765, and A766P). Although we find that F127C is associated with SLE risk, it is not independently contributing to the risk associated with rs5029953. We also found no additive effect be- tween rs2230926 and rs5029953. Because we find limited func- tional or genetic effects with this SNP, it is possible that the result of this coding change in vivo is negligible or is compensated for in another way. Based on our analysis of the structure of the TNFAIP3 DUB domain, further functional studies that focus on the interaction of TNFAIP3 with various target proteins are by guest on September 29, 2021 needed to elucidate the extent of the F127C amino acid substitu- tion on the enzymatic function of the protein. Another coding region SNP (rs5029956) on the risk haplotype is a synonymous base pair substitution in the proline residue 765 (CCC-.CCT) in the ninth exon of TNFAIP3. This is a potential functional element, as protein levels can be modulated by synonymous base pair changes in codons. A variation in protein expression of ∼250- fold was found by engineering synonymous mutations in GFP expressed by Escherichia coli which resulted in differential mRNA stability and folding (39). The effect of the synonymous

of each protein band was determined, and the densitometry of the TRAF2 and TNFAIP3 bands are plotted using arbitrary units. One of three experi- ments is shown. B, The ability of TNFAIP3 single mutants to degrade the protein target TRAF2 was tested. Increasing amounts of either V5-tagged WT TNFAIP3 (1, 2.5, 3.5, or 5 mg), V5-tagged C103A TNFAIP3 (1, 2.5, 3.5, or 5 mg), V5-tagged A125V TNFAIP3 (1 or 2.5 mg), V5-tagged F127C TNFAIP3 (1, 2.5, or 3.5 mg), V5-tagged R706Q TNFAIP3 (1, 2.5, 3.5, or 5 mg), or V5-tagged A766P TNFAIP3 (1, 2.5, 3.5, or 5 mg) were cotransfected with HA-tagged TRAF2 (0.5 mg) into HEK 293 cells. FIGURE 2. A125V TNFAIP3 mediates diminished degradation of Whole cell lysates were transferred to PVDF membrane and IB for HA, TRAF2. A, The ability of the A125V/F127C double mutant to degrade the V5, or actin. WT TNFAIP3 caused degradation of TRAF2; whereas, the protein target TRAF2 was tested. Increasing amounts of either V5-tagged known catalytic mutant C103A TNFAIP3 did not. The A125V, F127C, WT or A125V/F127C TNFAIP3 (1, 2.5, or 3.5 mg), or V5-tagged C103A R706Q, and A766P TNFAIP3 mutants were tested and compared with TNFAIP3 (3.5 mg) were cotransfected with HA-tagged TRAF2 (0.5 mg) WT TNFAIP3. Equivalent protein loading was confirmed by immunoblot- into HEK293 cells. Whole cell lysates were transferred to PVDF mem- ting for actin. The relative intensity of the protein bands was determined, brane and immunoblotted (IB) for HA, V5, or actin. The A125V/F127C and the densitometry of the TRAF2 and TNFAIP3 bands are plotted in the mutant fails to degrade TRAF2 to the same extent as WT TNFAIP3, and is lower half of the panel using arbitrary units. On the bar graph for F127C, comparable to the known catalytic mutant C103A. Equivalent protein the highest dose of F127C TNFAIP3 (5 mg) was not determined (pND). loading was confirmed by immunoblotting for actin. The relative intensity One of three experiments is shown. The Journal of Immunology 7007 Downloaded from http://www.jimmunol.org/

FIGURE 4. The effect of nonsynonymous coding SNPs on the predicted crystal structure in the DUB domain of TNFAIP3. Using MacPyMOL, the reported crystal structure of the DUB domain was used to make the A125V and F127C amino acid changes. PDB accession number 2VFJ was used for this analysis. A, The R group of the valine group at position 125 (pink) has the potential of a steric clash with the R group of the isoleucine at position 29 and/or the isoleucine at position 202 (green). The relative position of by guest on September 29, 2021 the catalytic cysteine at position 103 is indicated in the background in orange. B, The relative positions of the A125V change (purple) pointing into the protein’s core compared with the F127C change (red) pointing away from the protein. The nearby catalytic DUB domain is marked by the cysteine at position 103 (orange).

P765 SNP on the expression of TNFAIP3 in vivo remains to be studied. Finally, the nonsynonymous coding SNPs rs3734553 (R706Q) and rs5029957 (A766P) are found within the SLE- FIGURE 3. A125V TNFAIP3 has reduced DUB activity on non–K48- associated LD block; however, the existence of these SNPs has linked polyubiquitinated target proteins. The ability of WT TNFAIP3 and not been fully verified, and these may be extremely rare variants. the TNFAIP3 mutants C103A and A125V to deubiquitinate the protein In addition, we found limited functional effects of these mutations. target TRAF2 was tested. Increasing amounts (1, 2.5, 5, or 7.5 mg) of V5- Further fine-mapping and sequencing along this ancestral risk tagged WT TNFAIP3, C103A TNFAIP3, or A125V TNFAIP3 were haplotype, along with in vitro functional assays, will help to iden- cotransfected with Myc-tagged K48R-ubiquitin (1 mg) and HA-tagged tify the precise functional element or elements that determine the TRAF2 (3 mg). All three TNFAIP3 constructs have point mutations in conserved cysteines within zinc-fingers 3 and 4 in the C-terminal ubiquitin association between the TNFAIP3 gene and increased SLE risk in ligase domain (C607A, C612A, C624A, and C627A), rendering them African-Americans. ubiquitin ligase deficient (E3 ligase deficient). This helped to reduce sub- We demonstrated that the A125V human polymorphism in sequent degradation of TRAF2 by TNFAIP3. Whole cell lysates were TNFAIP3 has a strong in vitro effect on protein function and also either transferred to PVDF membranes and IB for Myc, HA, V5, or actin shows significant and opposing genetic effects in two distinct (pre-IP), or IP with Ab against HA. These immunoprecipitated lysates autoimmune diseases (OR . 3 in IBD versus OR , 0.33 in SLE). were then transferred onto PVDF membranes and immunoblotted for There are precedents for genetic variants that are associated with Myc, HA, and V5. The A125V/E3 Ligase mutant demonstrates a reduced risk in one disease, but with protection in another. The non- ability to DUB TRAF2 in the IP samples, as well as total non–K48-linked synonymous variant R620W in the PTPN22 gene is associated polyubiquitinated protein in the pre-IP samples. Equivalent protein loading with risk for RA, T1D, SLE, and autoimmune thyroiditis (40). in the pre-IP samples was confirmed by immunoblotting for actin. The relative intensity of each protein band was determined, and the densitom- etry of the K48R-Ub and TRAF2 bands in the IP samples are presented on the left side; whereas, the K48R-Ub, TNFAIP3, and TRAF2 bands in the the samples immunoprecipitated for TRAF2 that has been normalized for pre-IP samples are plotted on the right side in the lower half of the figure the amount of TRAF2 present in the IP is located in the lower left bar using arbitrary units. Furthermore, a plot showing levels of K48R-Ub in graph. One of three experiments is shown. 7008 TNFAIP3 AND AUTOIMMUNITY

However, this same polymorphism is protective in CD and Beh- ubiquitin ligase domains of A20 downregulate NF-kappaB signalling. Nature 430: 694–699. cet’s disease (41, 42). The H131R variant in the Fc g receptor 2A 3. Boone, D. L., E. E. Turer, E. G. Lee, R. C. Ahmad, M. T. Wheeler, C. Tsui, (FCGR2A) gene is associated with risk for SLE, T1D, Guillain- P. Hurley, M. Chien, S. Chai, O. Hitotsumatsu, et al. 2004. The ubiquitin- Barre syndrome, and multiple sclerosis; however, it is protective modifying enzyme A20 is required for termination of Toll-like receptor responses. Nat. Immunol. 5: 1052–1060. in ulcerative colitis in the Japanese population (12, 43–45). Thus, 4. Lee, E. G., D. L. Boone, S. Chai, S. L. Libby, M. Chien, J. P. Lodolce, and there are several examples of coding variants that have opposing A. Ma. 2000. Failure to regulate TNF-induced NF-kappaB and cell death disease associations in autoimmunity. responses in A20-deficient mice. Science 289: 2350–2354. 5. Schmitz, R., M. L. Hansmann, V. Bohle, J. I. Martin-Subero, S. Hartmann, A central question that arises from this study is how a variant that G. Mechtersheimer, W. Klapper, I. Vater, M. Giefing, S. Gesk, et al. 2009. decreases the function of TNFAIP3 is associated with risk for IBD TNFAIP3 (A20) is a tumor suppressor gene in Hodgkin lymphoma and yet protection from SLE. Much of the previous work on TNFAIP3 primary mediastinal B cell lymphoma. J. Exp. Med. 206: 981–989. 6. Kato, M., M. Sanada, I. Kato, Y. Sato, J. Takita, K. Takeuchi, A. Niwa, Y. Chen, suggests that impairing its enzymatic activity results in increased K. Nakazaki, J. Nomoto, et al. 2009. Frequent inactivation of A20 in B-cell inflammation (1). Although this is in agreement with increased lymphomas. Nature 459: 712–716. 7. Compagno, M., W. K. Lim, A. Grunn, S. V. Nandula, M. Brahmachary, Q. Shen, susceptibility to a proinflammatory disease like IBD, it is not F. Bertoni, M. Ponzoni, M. Scandurra, A. Califano, et al. 2009. Mutations of consistent with protection from SLE. One hypothetical model that multiple genes cause deregulation of NF-kappaB in diffuse large B-cell lym- could explain this paradox involves differences in the types of phoma. Nature 459: 717–721. 8. Pascual, V., L. Farkas, and J. Banchereau. 2006. Systemic lupus erythematosus: cytokine milieus that are produced during the course of each all roads lead to type I interferons. Curr. Opin. Immunol. 18: 676–682. disease, thus influencing the Th1/2 skewing of T lymphocytes. 9. Tsao, B. P., J. M. Grossman, G. Riemekasten, N. Strong, J. Kalsi, D. J. Wallace, SLE is generally thought of as a Th2-type disease; whereas, IBD C. J. Chen, C. S. Lau, E. M. Ginzler, R. Goldstein, et al. 2002. Familiality and co-occurrence of clinical features of systemic lupus erythematosus. Arthritis is more Th1-mediated. Loss of TNFAIP3 function results in more Rheum. 46: 2678–2685. Downloaded from inflammation, which might alter the Th skewing of immune 10. Rahman, A., and D. A. Isenberg. 2008. Systemic lupus erythematosus. N. Engl. responses. Thus, inflammatory stimuli insufficient to initiate J. Med. 358: 929–939. 11. Lau, C. S., G. Yin, and M. Y. Mok. 2006. Ethnic and geographical differences in IBD (an intermediate Th1/2 response) are made sufficient (more systemic lupus erythematosus: an overview. Lupus 15: 715–719. toward a Th1 response) by reduced anti-inflammatory activity of 12. Harley, J. B., M. E. Alarco´n-Riquelme, L. A. Criswell, C. O. Jacob, R. P. Kimberly, K. L. Moser, B. P. Tsao, T. J. Vyse, C. D. Langefeld, TNFAIP3. Conversely, weaker inflammatory stimuli that might S. K. Nath, et al; International Consortium for Systemic Lupus Erythematosus lead to SLE are increased in individuals with less functional Genetics (SLEGEN). 2008. Genome-wide association scan in women with sys- http://www.jimmunol.org/ TNFAIP3, leading to more inflammation and less susceptibility temic lupus erythematosus identifies susceptibility variants in ITGAM, PXK, KIAA1542 and other loci. Nat. Genet. 40: 204–210. to SLE. This may involve Th1/2 skewing of T cells, such that 13. Hom, G., R. R. Graham, B. Modrek, K. E. Taylor, W. Ortmann, S. Garnier, a Th2 response that might lead to SLE is skewed toward a Th1 A. T. Lee, S. A. Chung, R. C. Ferreira, P. V. Pant, et al. 2008. Association of profile, reducing the risk for SLE. Our genetic findings are con- systemic lupus erythematosus with C8orf13-BLK and ITGAM-ITGAX. N. Engl. J. Med. 358: 900–909. sistent with the observation that very few people are diagnosed 14. Kozyrev, S. V., A. K. Abelson, J. Wojcik, A. Zaghlool, M. V. Linga Reddy, with both SLE and IBD (46). Studies correlating TNFAIP3 SNPs E. Sanchez, I. Gunnarsson, E. Svenungsson, G. Sturfelt, A. Jo¨nsen, et al. 2008. with cytokine profiles in African-Americans with SLE versus IBD Functional variants in the B-cell gene BANK1 are associated with systemic lupus erythematosus. Nat. Genet. 40: 211–216. are needed to provide a basis for this hypothetical model. 15. Graham, R. R., C. Cotsapas, L. Davies, R. Hackett, C. J. Lessard, J. M. Leon, The pathogenesis of inflammatory autoimmune diseases is com- N. P. Burtt, C. Guiducci, M. Parkin, C. Gates, et al. 2008. Genetic variants near by guest on September 29, 2021 plex, and likely involves both genetic and environmental factors. The TNFAIP3 on 6q23 are associated with systemic lupus erythematosus. Nat. Genet. 40: 1059–1061. identification of potential genetic associations has helped define 16. Nath, S. K., S. Han, X. Kim-Howard, J. A. Kelly, P. Viswanathan, possible mechanisms of disease initiation and progression. Genetic G. S. Gilkeson, W. Chen, C. Zhu, R. P. McEver, R. P. Kimberly, et al. 2008. A nonsynonymous functional variant in integrin-alpha(M) (encoded by ITGAM) variation in the TNFAIP3 gene has been associated with several is associated with systemic lupus erythematosus. Nat. Genet. 40: 152–154. types of autoimmune disease in patients of European ancestry, and 17. Musone, S. L., K. E. Taylor, T. T. Lu, J. Nititham, R. C. Ferreira, W. Ortmann, in this study we show novel genetic elements on African-derived N. Shifrin, M. A. Petri, M. I. Kamboh, S. Manzi, et al. 2008. Multiple poly- morphisms in the TNFAIP3 region are independently associated with systemic also modulate autoimmunity. Reconciliation of the lupus erythematosus. Nat. Genet. 40: 1062–1064. differences between European- and African-American genetic stud- 18. Chapman, S. J., C. C. Khor, F. O. Vannberg, N. A. Maskell, C. W. Davies, ies of TNFAIP3 around functional elements will be a critical step in E. L. Hedley, S. Segal, C. E. Moore, K. Knox, N. P. Day, et al. 2006. PTPN22 and invasive bacterial disease. Nat. Genet. 38: 499–500. understanding disease biology. Our finding that the A125V poly- 19. Plenge, R. M., C. Cotsapas, L. Davies, A. L. Price, P.I. de Bakker, J. Maller, I. Pe’er, morphism confers functional differences is an important advance in N. P. Burtt, B. Blumenstiel, M. DeFelice, et al. 2007. Two independent alleles at this effort. A better understanding of how polymorphisms in 6q23 associated with risk of rheumatoid arthritis. Nat. Genet. 39: 1477–1482. 20. Thomson, W., A. Barton, X. Ke, S. Eyre, A. Hinks, J. Bowes, R. Donn, TNFAIP3 affect risk and protection in autoimmunity will provide D. Symmons, S. Hider, I. N. Bruce, et al; Wellcome Trust Case Control Con- valuable mechanistic insights into the pathogenesis and treatment of sortium; YEAR Consortium. 2007. Rheumatoid arthritis association at 6q23. Nat. Genet. 39: 1431–1433. autoimmune disease. 21. Fung, E. Y., D. J. Smyth, J. M. Howson, J. D. Cooper, N. M. Walker, H. Stevens, L. S. Wicker, and J. A. Todd. 2009. Analysis of 17 autoimmune disease- associated variants in type 1 diabetes identifies 6q23/TNFAIP3 as a susceptibility Acknowledgments locus. Genes Immun. 10: 188–191. We thank Ken Onel and Eugene Chang for helpful comments and sugges- 22. Trynka, G., A. Zhernakova, J. Romanos, L. Franke, K. A. Hunt, G. Turner, tions; the Digestive Disease Research Core Center of the University of Chi- M. Bruinenberg, G. A. Heap, M. Platteel, A. W. Ryan, et al. 2009. Coeliac disease-associated risk variants in TNFAIP3 and REL implicate altered NF- cago, the Gastro-Intestinal Research Foundation, and the Gastro-Intestinal kappaB signalling. Gut 58: 1078–1083. Research Foundation Associates Board for support. 23. Wellcome Trust Case Control Consortium. 2007. Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature 447: 661–678. Disclosures 24. Barmada, M. M., S. R. Brant, D. L. Nicolae, J. P. Achkar, C. I. Panhuysen, The authors have no financial conflicts of interest. T. M. Bayless, J. H. Cho, and R. H. Duerr. 2004. A genome scan in 260 inflammatory bowel disease-affected relative pairs. Inflamm. Bowel Dis. 10: 513–520. 25. Nair, R. P., K. C. Duffin, C. Helms, J. Ding, P. E. Stuart, D. Goldgar, J. E. Gudjonsson, Y. Li, T. Tejasvi, B. J. Feng, et al; Collaborative Association References Study of Psoriasis. 2009. Genome-wide scan reveals association of psoriasis with 1. Coornaert, B., I. Carpentier, and R. Beyaert. 2009. A20: central gatekeeper in IL-23 and NF-kappaB pathways. Nat. Genet. 41: 199–204. inflammation and immunity. J. Biol. Chem. 284: 8217–8221. 26. Boonyasrisawat, W., D. Eberle, S. Bacci, Y. Y. Zhang, D. Nolan, E. V. Gervino, 2. Wertz, I. E., K. M. O’Rourke, H. Zhou, M. Eby, L. Aravind, S. Seshagiri, P. Wu, M. T. Johnstone, V. Trischitta, S. E. Shoelson, and A. Doria. 2007. Tag poly- C. Wiesmann, R. Baker, D. L. Boone, et al. 2004. De-ubiquitination and morphisms at the A20 (TNFAIP3) locus are associated with lower gene The Journal of Immunology 7009

expression and increased risk of coronary artery disease in type 2 diabetes. activation depends upon the E2/E3 complex Ubc13-Uev1A/TNF receptor- Diabetes 56: 499–505. associated factor 2 (TRAF2). J. Biol. Chem. 278: 15429–15434. 27. Bates, J. S., C. J. Lessard, J. M. Leon, T. Nguyen, L. J. Battiest, J. Rodgers, 37. Song, H. Y., M. Rothe, and D. V. Goeddel. 1996. The tumor necrosis factor- K. M. Kaufman, J. A. James, G. S. Gilkeson, J. A. Kelly, et al. 2009. Meta-analysis inducible zinc finger protein A20 interacts with TRAF1/TRAF2 and inhibits NF- and imputation identifies a 109 kb risk haplotype spanning TNFAIP3 associated kappaB activation. Proc. Natl. Acad. Sci. USA 93: 6721–6725. with lupus nephritis and hematologic manifestations. Genes Immun. 10: 470–477. 38. Komander, D., and D. Barford. 2008. Structure of the A20 OTU domain and 28. Frazer, K. A., D. G. Ballinger, D. R. Cox, D. A. Hinds, L. L. Stuve, R. A. Gibbs, mechanistic insights into deubiquitination. Biochem. J. 409: 77–85. J. W. Belmont, A. Boudreau, P. Hardenbol, S. M. Leal, et al; International 39. Kudla, G., A. W. Murray, D. Tollervey, and J. B. Plotkin. 2009. Coding-sequence HapMap Consortium. 2007. A second generation human haplotype map of over determinants of in Escherichia coli. Science 324: 255–258. 3.1 million SNPs. Nature 449: 851–861. 40. Chung, S. A., and L. A. Criswell. 2007. PTPN22: its role in SLE and auto- 29. Tan, E. M., A. S. Cohen, J. F. Fries, A. T. Masi, D. J. McShane, N. F. Rothfield, immunity. Autoimmunity 40: 582–590. J. G. Schaller, N. Talal, and R. J. Winchester. 1982. The 1982 revised criteria for 41. Barrett, J. C., S. Hansoul, D. L. Nicolae, J. H. Cho, R. H. Duerr, J. D. Rioux, the classification of systemic lupus erythematosus. Arthritis Rheum. 25: 1271– S. R. Brant, M. S. Silverberg, K. D. Taylor, M. M. Barmada, et al; NIDDK IBD 1277. Genetics Consortium; Belgian-French IBD Consortium; Wellcome Trust Case 30. Xiao, R., and M. Boehnke. 2009. Quantifying and correcting for the winner’s Control Consortium. 2008. Genome-wide association defines more than 30 dis- curse in genetic association studies. Genet. Epidemiol. 33: 453–462. tinct susceptibility loci for Crohn’s disease. Nat. Genet. 40: 955–962. 31. Gordon, D., M. A. Levenstien, S. J. Finch, and J. Ott. 2003. Errors and linkage 42. Baranathan, V., M. R. Stanford, R. W. Vaughan, E. Kondeatis, E. Graham, disequilibrium interact multiplicatively when computing sample sizes for genetic F. Fortune, W. Madanat, C. Kanawati, M. Ghabra, P. I. Murray, and case-control association studies. Pac. Symp. Biocomput. 490–501. G. R. Wallace. 2007. The association of the PTPN22 620W polymorphism with 32. Gordon, D., S. J. Finch, M. Nothnagel, and J. Ott. 2002. Power and sample size calculations for case-control genetic association tests when errors are present: Behcet’s disease. Ann. Rheum. Dis. 66: 1531–1533. application to single nucleotide polymorphisms. Hum. Hered. 54: 22–33. 43. Alizadeh, B. Z., G. Valdigem, M. J. Coenen, A. Zhernakova, B. Franke, 33. Gordon, D., S. C. Heath, X. Liu, and J. Ott. 2001. A transmission/disequilibrium test A. Monsuur, P. L. van Riel, P. Barrera, T. R. Radstake, B. O. Roep, et al. 2007. that allows for genotyping errors in the analysis of single-nucleotide polymorphism Association analysis of functional variants of the FcgRIIa and FcgRIIIa genes data. Am. J. Hum. Genet. 69: 371–380. with type 1 diabetes, celiac disease and rheumatoid arthritis. Hum. Mol. Genet. 34. Niewold, T. B., J. A. Kelly, M. H. Flesch, L. R. Espinoza, J. B. Harley, and 16: 2552–2559. Downloaded from M. K. Crow. 2008. Association of the IRF5 risk haplotype with high serum 44. Takai, T. 2002. Roles of Fc receptors in autoimmunity. Nat. Rev. Immunol. 2: interferon-alpha activity in systemic lupus erythematosus patients. Arthritis 580–592. Rheum. 58: 2481–2487. 45. Asano, K., T. Matsushita, J. Umeno, N. Hosono, A. Takahashi, T. Kawaguchi, 35. Wajant, H., and P. Scheurich. 2001. Tumor necrosis factor receptor-associated T. Matsumoto, T. Matsui, Y. Kakuta, Y. Kinouchi, et al. 2009. A genome-wide factor (TRAF) 2 and its role in TNF signaling. Int. J. Biochem. Cell Biol. 33: 19– association study identifies three new susceptibility loci for ulcerative colitis in 32. the Japanese population. Nat. Genet. 41: 1325–1329. 36. Shi, C. S., and J. H. Kehrl. 2003. Tumor necrosis factor (TNF)-induced germinal 46. Nitzan, O., M. Elias, and W. R. Saliba. 2006. Systemic lupus erythematosus and

center kinase-related (GCKR) and stress-activated protein kinase (SAPK) inflammatory bowel disease. Eur. J. Intern. Med. 17: 313–318. http://www.jimmunol.org/ by guest on September 29, 2021 African-derived genetic polymorphisms in TNFAIP3 mediate risk for autoimmunity

Lodolce, et al.

Supplemental Data

Supplemental Tables

Supplemental Table I. Power to detect and corrected OR for genotyped TNFAIP3 SNPsa

SNP disease % genotypedb 80% power to “winner’s curse” correctiond controls cases detect ORc naïve OR ascertainment- corrected OR rs3757173 SLE 97.1 99.1 1.47 1.2755 1.2757 rs5029939 SLE 98.5 100 1.46 1.3108 1.3090 A125V SLE 93.0 100 0.33 0.3800 0.7938 A125V IBD 93.0 100 ND 3.7907 3.7703 F127C SLE 99.6 100 1.46 1.3448 1.0884 F127C IBD 99.6 100 ND 1.1790 1.1821 rs610604 SLE 96.0 100 1.46 0.9258 0.9250 rs5029953 SLE 98.5 100 1.53 1.6093 1.3431 aThe six TNFAIP3 SNPs genotyped in >90% of total samples in this study are listed, along with the disease population (SLE vs. IBD). bTotal number of controls=272, SLE cases=217, and IBD cases=24. cThe minimum odds ratio (OR) that could be detected with 80% power with a significance of p=0.05 was calculated for each SNP genotyped based on the size of the SLE population. Values were determined using an additive model with the CaTS Power Calculator written by Andrew Skol (http://www.sph.umich.edu/csg/abecasis/CaTS/). Calculations were not determined (ND) for the IBD cases due to the small sample size. dThe adjusted odds ratios (ORs) were calculated with WINNER version 1.1 written by Rui Xiao and Michael Boehnke (http://csg.sph.umich.edu/boehnke/winner/) using a significance level of 0.05. Supplemental Table II. Genotypes of A125V and F127C in African-American and White/European-American Populationsa

White/ African-American European- IBD SLE controls SNP genotype American # freq # freq # freq # freq A125V TT 0 0.00 0 0.00 0 0.00 1 0.0040 TC 0 0.00 5 0.21 5 0.023 13 0.052 CC 291 1.00 19 0.79 212 0.98 238 0.94 F127C GG 1 0.0034 3 0.13 29 0.13 30 0.12 GT 21 0.072 11 0.46 109 0.50 100 0.40 TT 269 0.92 10 0.42 79 0.36 122 0.48 aBoth African-American and White/European-American samples were sequenced in an area centered on the A125V and F127C SNPs in TNFAIP3. The total number of patients with each genotype and the minor allele frequencies for each of these SNPs in both ancestral populations is shown.

Supplemental Table III. Population frequencies of SNPs in linkage (r2>0.50) with rs5029953a

SNP Position r2 w/ YRI YRI CEU CEU (Mb) rs5029953 freq. allele freq. allele counts counts rs7740811 138.194843 0.55 0.117 14/120 0 0/118 rs9494876 138.199541 0.60 0.125 15/120 0 0/120 rs6923085 138.208718 0.83 0.146 33/226 0 0/118 rs10223631 138.209746 0.55 0.112 25/224 0 0/118 rs10223636 138.209907 0.60 0.107 24/224 0 0/118 rs9484096 138.215016 0.55 0.117 14/120 0 0/118 rs5029953 138.242453 - 0.155 35/226 0 0/118 rs5029956 138.244071 0.66 0.088 20/226 0 0/114 rs9484098 138.266866 0.84 0.186 22/118 0 0/120 rs9484099 138.274376 0.74 0.177 40/226 0 0/120

aData from the YRI population was analyzed with Haploview 4.1 software to identify SNPs within 100kb upstream and downstream of rs5029953 that had r2 values >0.50. The frequency of the minor allele of the nine highly linked SNPs identified along with rs5029953 is shown for both African ancestry (YRI) and Utah residents with ancestry from northern and western Europe (CEU).

Supplemental Figures

Supplemental Figure 1. LD plot of TNFAIP3 SNPs in African-American non- autoimmune controls.

An LD plot generated from African-American non-autoimmune controls shows the relative linkage between the TNFAIP3 SNPs genotyped in this study. The numbers in each box represent the pairwise r2 value x100 between SNPs, calculated by the Haploview 4.1 software. Darker shaded boxes represent higher r2 values between SNPs. The location of each SNP within the general structure of the TNFAIP3 locus is indicated above the plot. The 5’ and 3’ untranslated regions are colored dark grey, while the coding regions are colored light grey. The corresponding exon number is indicated above each exonic block.

Supplemental Figure 1 Supplemental Figure 2. LD plot of a ~68 kb region spanning the TNFAIP3 gene from the International HapMap Project YRI population.

Using Haploview 4.1 software an LD plot was generated that includes the 7 kb LD block containing rs5029953, two LD blocks upstream (20 kb and 19 kb), and one LD block downstream (22 kb). The software defined the LD blocks using a solid spine of LD calculation. The numbers in each box represent the pairwise r2 value x100 between SNPs, calculated by the Haploview 4.1 software. Darker shaded boxes represent higher linkage between SNPs. The nine SNPs with an r2 value >0.50 are marked with an arrow, and the rs5029953 SNP is denoted with an arrow and an asterisk (*). 

Supplemental Figure 2