ARTHRITIS & RHEUMATISMVol. 60, No. 8, August 2009, pp 2402–2407DOI 10.1002/art.24658© 2009, American College of Rheumatology

A Polymorphism Within IL21R Confers Risk forSystemic Lupus Erythematosus

Ryan Webb,1 Joan T. Merrill,1 Jennifer A. Kelly,2 Andrea Sestak,2 Kenneth M. Kaufman,3

Carl D. Langefeld,4 Julie Ziegler,4 Robert P. Kimberly,5 Jeffrey C. Edberg,5

Rosalind Ramsey-Goldman,6 Michelle Petri,7 John D. Reveille,8 Graciela S. Alarcon,5

Luis M. Vila,9 Marta E. Alarcon-Riquelme,10 Judith A. James,1 Gary S. Gilkeson,11

Chaim O. Jacob,12 Kathy L. Moser,2 Patrick M. Gaffney,2 Timothy J. Vyse,13 Swapan K. Nath,2

Peter Lipsky,14 John B. Harley,3 and Amr H. Sawalha3

Objective. Interleukin-21 (IL-21) is a member ofthe type I cytokine superfamily that has a variety of

effects on the immune system, including B cell activa-tion, plasma cell differentiation, and immunoglobulinproduction. The expression of IL-21 receptor (IL-21R)is reduced in the B cells of patients with systemic lupuserythematosus (SLE), while serum IL-21 levels areincreased both in lupus patients and in some murinelupus models. We recently reported that polymorphismswithin the IL21 gene are associated with increasedsusceptibility to SLE. The aim of this study was toexamine the genetic association between single-nucleo-tide polymorphisms (SNPs) within IL21R and SLE.

Methods. We genotyped 17 SNPs in the IL21Rgene in 2 large cohorts of lupus patients (a European-derived cohort and a Hispanic cohort) and in ethnicallymatched healthy controls.

Results. We identified and confirmed the associ-ation between rs3093301 within the IL21R gene and SLEin the 2 cohorts (meta-analysis odds ratio 1.16 [95%confidence interval 1.08–1.25], P � 1.0 � 10�4).

Conclusion. Our findings indicate that IL21R is anovel susceptibility gene for SLE.

Supported by the NIH (grants R03-AI-076729 and P20-RR-015577). Dr. Langefeld and Ms Ziegler’s work was supported by theWake Forest University Health Sciences Center for Public HealthGenomics and the Alliance for Lupus Research. Dr. Kimberly’s workwas supported by the NIH (grants P01-AR-49084 and P60-AR-48095).Dr. Ramsey-Goldman’s work was supported by the NIH (grantsP01-AR-49084, K24-AR-002138, and M01-RR-00048). Dr. Alarcon’swork was supported by the NIH (P01-AR-49084). Dr. James’ work wassupported by the NIH (grants RR-15577 and AR-053483). Dr Jacob’swork was supported by the Alliance for Lupus Research (grant 52104).Dr. Moser’s work was supported by the NIH (grant AR-043274-13)and the Lupus Foundation of Minnesota. Dr. Gaffney’s work wassupported by the NIH (grants AR-052125-04 and AI-063274-05) andthe Lupus Foundation of Minnesota. Dr. Nath’s work was supportedby the NIH (grants P30-AR-053483 and P20-RR-020143). Dr. Harleyis recipient of a Kirkland Scholar award, and his work was supportedby the NIH (R01-AR-42460, R01-AI-31584, N01-AR-12253, R01-DE-015223, P20-RR-020143, R37-AI-24717, N01-AR-62277, AR-048940,and AR-0490084) and the Alliance for Lupus Research. Dr. Sawalha’swork was supported by the NIH (grants R03-AI-076729 and P20-RR-015577), the Arthritis National Research Foundation, the LupusFoundation of America, the Department of Veterans Affairs, and theUniversity of Oklahoma College of Medicine.

1Ryan Webb, BS, Joan T. Merrill, MD, Judith A. James, MD,PhD: Oklahoma Medical Research Foundation, and University ofOklahoma Health Sciences Center, Oklahoma City; 2Jennifer A. Kelly,MPH, Andrea Sestak, MD, PhD, Kathy L. Moser, PhD, Patrick M.Gaffney, MD, Swapan K. Nath, PhD: Oklahoma Medical ResearchFoundation, Oklahoma City; 3Kenneth M. Kaufman, PhD, John B.Harley, MD, PhD, Amr H. Sawalha, MD: Oklahoma Medical Re-search Foundation, University of Oklahoma Health Sciences Center,and VAMC, Oklahoma City, Oklahoma; 4Carl D. Langefeld, PhD,Julie Ziegler, MA: Wake Forest University Health Sciences, Winston-Salem, North Carolina; 5Robert P. Kimberly, MD, Jeffrey C. Edberg,PhD, Graciela S. Alarcon, MD, MPH: University of Alabama atBirmingham; 6Rosalind Ramsey-Goldman, MD, DrPH: NorthwesternUniversity, Chicago, Illinois; 7Michelle Petri, MD, MPH: Johns Hop-kins University School of Medicine, Baltimore, Maryland; 8John D.Reveille, MD: University of Texas Health Science Center at Houston;9Luis M. Vila, MD: University of Puerto Rico Medical SciencesCampus, San Juan; 10Marta E. Alarcon-Riquelme, MD, PhD: Okla-homa Medical Research Foundation, Oklahoma City, and Uppsala

University, Uppsala, Sweden; 11Gary S. Gilkeson, MD: Medical Uni-versity of South Carolina, Charleston; 12Chaim O. Jacob, MD, PhD:University of Southern California School of Medicine, Los Angeles;13Timothy J. Vyse, MA, MBBS, MRCP, PhD: Imperial CollegeHealthcare National Health Service Trust, London, UK; 14PeterLipsky, MD: National Institute of Arthritis and Musculoskeletal andSkin Diseases, Bethesda, Maryland.

Dr. Harley has received consulting fees, speaking fees, and/orhonoraria from Bio-Rad, Immunovision, and IVAX Diagnostics (morethan $10,000 each) and owns stock or stock options in IVAX Diag-nostics.

Address correspondence and reprint requests to Amr H.Sawalha, MD, 825 NE 13th Street, MS #24, Oklahoma City, OK73104. E-mail: amr-sawalha@omrf.ouhsc.edu.

Submitted for publication January 15, 2009; accepted inrevised form April 7, 2009.

2402

Systemic lupus erythematosus (SLE) is a chronicsystemic autoimmune disease that is associated withinvolvement of multiple organs. The etiology and patho-genesis of SLE involve genetic, epigenetic, and environ-mental factors, and SLE is a heterogeneous disease withhighly variable presentation. The one feature commonto nearly all lupus patients is the presence of antinuclearautoantibodies, which is considered a hallmark of thedisease (1).

A large body of evidence supports the role ofgenetics in SLE. More than 100 possible genetic riskfactors for SLE have been identified through casestudies, linkage analyses of multiplex families, and can-didate gene case–control studies (2). Epidemiologicstudies have demonstrated that familial clustering doesoccur in SLE and that family members of individualswith SLE have an increased risk of developing thedisease (3,4). There is, for example, a �10-fold increasein the concordance rate of SLE in monozygotic twinscompared with dizygotic twins (5).

Interleukin-21 (IL-21) is a recently discoveredmember of the type I cytokine superfamily and isexpressed primarily by activated CD4� T cells andnatural killer (NK) cells (6). Its receptor, IL-21R, isexpressed on a variety of cells including B cells, T cells,NK cells, and monocyte-derived dendritic cells (7–9).IL-21 exerts a variety of effects on the immune systemand has an important role in B cell responsiveness,proliferation, plasma cell differentiation, and immuno-globulin production (10,11). Furthermore, IL-21 is in-volved in Th17 cell differentiation (12) and modulatesthe function of both dendritic cells and NK cells (13).

IL-21 overproduction has been reported in sev-eral murine lupus models, including the BXSB-Yaamouse (11). Excessive production of IL-21 and a severelupus-like autoimmune phenotype have been reportedin the sanroque mouse strain, which has a mutation in aRING–type ubiquitin ligase protein family member (ro-quin) (14). In addition, an IL-21R–Fc fusion proteinsignificantly improves disease in MRL/lpr lupus-pronemice, suggesting that blocking IL-21 might be a potentialtherapeutic approach in patients with SLE (15).

In lupus, IL-21R is underexpressed on total,naive, and memory B cells, and on plasmablasts, com-pared with controls, but there is no difference in theexpression levels of IL-21R in T cells (8). Decreasedexpression of IL-21R is also associated with nephritisand a high titer of anti–double-stranded DNA (anti-dsDNA) antibody in lupus (8). Patients with SLE havehigher plasma concentrations of IL-21, although plasma

IL-21 levels do not appear to correlate with lupusdisease activity (6).

We recently reported that polymorphisms withinIL21 are associated with SLE in 2 independent cohorts(16). In the present study, we evaluated genetic poly-morphisms in IL21R in a large group of SLE patientsand controls. We identified a genetic association be-tween a single-nucleotide polymorphism (SNP) withinthe IL21R gene and SLE in a European-derived cohortand confirm our findings in a second independent cohortof Hispanic SLE patients and controls.

PATIENTS AND METHODS

Patients and controls. SLE patients and healthy con-trols were recruited and enrolled in the lupus genetics studiesat the Oklahoma Medical Research Foundation and at collab-orating sites in the US, the UK, and Sweden. One cohortconsisted of 2,573 unrelated SLE patients and 3,075 healthy,unrelated, ethnically matched controls of European ancestry(referred to below as the European-derived cohort). A secondcohort consisted of 657 unrelated SLE patients and 265healthy, unrelated, ethnically matched controls of Hispanicancestry (referred to below as the Hispanic cohort). Allpatients met the 1997 ACR classification criteria for SLE(17,18). Clinical and serologic data were acquired as describedpreviously (19).

Genotyping. Genotyping was performed on DNA iso-lated from peripheral blood mononuclear cells, lymphoblas-toid cell lines, or buccal cell swabs as described previously (20).In the 2 cohorts, we genotyped 17 SNPs located within theIL21R gene. These SNPs were selected from the publishedSNP databases (http://www.ncbi.nlm.nih.gov/projects/SNP/) tocover the entire length of the IL21R gene. They were allvalidated SNPs and had been previously tested successfullyusing the same genotyping platform that we utilized in thisstudy: the BeadStation 500GX with Illumina Infinium IIgenotyping assays according to the recommendations of themanufacturer (Illumina, San Diego, CA).

Statistical analysis. Principal component analyseswere used to examine sample homogeneity in both cohorts. Tofurther exclude population substratification, we used thegenomic control (GC) method and calculated an inflationfactor (�) using 2,218 “null” SNPs in the European-derivedcohort and 2,196 “null” SNPs in the Hispanic cohort.

This study was conducted utilizing a population-basedcase–control study design. Pearson’s chi-square measures werecalculated to determine whether allele frequencies were dif-ferent between cases and controls. Uncorrected P values arereported with 95% confidence intervals (95% CIs). Permuta-tion testing was used to correct allele frequency differences formultiple testing, using Haploview 4.1 (21). Meta-analysis wasperformed using the Cochran-Mantel-Haenszel test includedin SAS (SAS Institute, Cary, NC). Logistic regression analysiswas used to test for gene–gene interaction between the lupus-associated IL21 and IL21R SNPs.

IL21R POLYMORPHISM AND SLE RISK 2403

RESULTS

We genotyped 17 SNPs located within the IL21Rgene in 2 independent, ethnically divergent populationsof SLE patients and controls. The European-derivedcohort included 2,573 unrelated SLE patients and 3,075healthy controls, and the Hispanic cohort included 657unrelated SLE patients and 265 healthy controls. Prin-cipal component analyses were used to examine samplehomogeneity in both cohorts. A total of 234 and 156samples were identified as outliers in the European-derived and Hispanic cohorts, respectively, and were

excluded from further analysis. Subsequently, GC ana-lysis was performed and revealed no significant popula-tion substratification (� � 1.13 in the European-derivedcohort, and � � 1.17 in the Hispanic cohort). Fifteen ofthe 17 SNPs genotyped had a minor allele frequency of�1%, which we required for analysis. The genotypingsuccess rates were �99.2% and �97.8% in theEuropean-derived and Hispanic cohorts, respectively.

We found a genetic association betweenrs3093301 (A/G) and SLE in both the European-derivedand Hispanic cohorts. The lupus-associated allele (A)

Figure 1. Linkage disequilibrium plots of the IL21R single-nucleotide polymorphisms analyzed in the systemic lupus erythematosus patients andcontrols in the European-derived cohort (A) and the Hispanic cohort (B). Values are pairwise correlation coefficients (r2).

Table 1. Genetic association between SNPs within IL21R and SLE in the cohort of European-derived SLE patients (2,245 females and 239 males)and controls (2,009 females and 921 males)*

SNPAssociated

allele

Associated allele frequency

�2 OR (95% CI) PP for

permutationP for

HWENo. (%) of

casesNo. (%) of

controls

rs12934152 G 1,254 (25.6) 1,481 (25.3) 0.09 1.01 (0.93–1.11) 0.76 1.00 0.98rs3093301 A 3,264 (66.6) 3,696 (63.7) 9.42 1.13 (1.05–1.23) 0.0022 0.015 0.06rs179760 A 1,168 (23.6) 1,333 (22.7) 0.99 1.05 (0.96–1.14) 0.32 0.93 0.79rs8057464 A 1,102 (22.2) 1,282 (21.9) 0.16 1.02 (0.93–1.12) 0.69 1.00 0.75rs3093310 G 2,845 (57.4) 3,329 (56.8) 0.37 1.02 (0.95–1.11) 0.55 1.00 0.65rs3093319 G 2,183 (44.2) 2,587 (44.2) 0.00 1.00 (0.93–1.08) 1.00 1.00 0.15rs3093341 G 479 (9.7) 554 (9.5) 0.20 1.03 (0.91–1.17) 0.65 1.00 0.90rs3093349 G 472 (9.5) 552 (9.4) 0.03 1.01 (0.89–1.15) 0.86 1.00 0.98rs3093359 A 428 (8.6) 504 (8.6) 0.00 1.00 (0.88–1.15) 0.98 1.00 0.55rs3093363 G 1,597 (32.5) 1,811 (30.9) 3.01 1.07 (0.99–1.17) 0.08 0.46 0.88rs3093364 G 425 (8.6) 501 (8.6) 0.00 1.00 (0.88–1.15) 0.97 1.00 0.59rs963154 C 1,570 (31.9) 1,781 (30.5) 2.26 1.06 (0.98–1.16) 0.13 0.63 0.82rs3093375 C 1,602 (32.6) 1,814 (31.0) 3.16 1.08 (0.99–1.17) 0.08 0.43 0.80rs3093379 A 1,592 (32.3) 1,807 (30.9) 2.36 1.07 (0.98–1.16) 0.12 0.61 0.94rs2285452 A 1,229 (24.8) 1,351 (23.1) 4.27 1.10 (1.00–1.20) 0.039 0.24 0.76

* Only single-nucleotide polymorphisms (SNPs) with minor allele frequencies of �1% were analyzed. SLE � systemic lupus erythematosus; OR �odds ratio; 95% CI � 95% confidence interval; HWE � Hardy-Weinberg equilibrium.

2404 WEBB ET AL

had a frequency of 66.6% among patients in theEuropean-derived cohort compared with 63.7% amongcontrols in this cohort (odds ratio [OR] 1.13, �2 � 9.42,P � 0.0022) (Table 1 and Figure 1A). This associationwas confirmed in the Hispanic cohort, which had anassociated allele frequency of 64.4% in patients com-pared with 55.9% in controls (OR 1.43, �2 � 9.16, P �0.0025) (Table 2 and Figure 1B). Using the Cochran-Mantel-Haenszel test, the meta-analysis OR for the

associated allele in rs3093301 in patients compared withcontrols was 1.16 (95% CI 1.08–1.25, P � 1.0 � 10�4) inthe combined European-derived and Hispanic samples.

IL21R is located in proximity to the IL4R gene,another important candidate in autoimmunity. How-ever, IL21R and IL4R are located in separate haplotypeblocks in the European-derived samples (Centred’Etude du Polymorphisme Humain from Utah sam-ples) genotyped in the International HapMap Project

Table 2. Genetic association between SNPs within IL21R and SLE in the cohort of Hispanic SLE patients (494 females and 63 males) and controls(167 females and 42 males)*

SNPAssociated

allele

Associated allele frequency

�2 OR (95% CI) PP for

permutationP for

HWENo. (%) of

casesNo. (%) of

controls

rs12934152 G 202 (18.2) 58 (14.0) 3.74 1.37 (0.99–1.70) 0.053 0.37 0.38rs3093301 A 702 (64.4) 228 (55.9) 9.16 1.43 (1.13–1.80) 0.0025 0.021 0.76rs179760 A 292 (26.3) 100 (23.9) 0.87 1.13 (0.87–1.47) 0.35 0.96 0.49rs8057464 A 268 (24.1) 87 (20.8) 1.80 1.21 (0.92–1.58) 0.18 0.78 0.76rs3093310 A 382 (34.4) 125 (29.9) 2.71 1.23 (0.96–1.56) 0.10 0.57 0.08rs3093319 G 376 (33.8) 129 (30.9) 1.20 1.14 (0.90–1.46) 0.27 0.91 0.13rs3093341 G 75 (6.7) 21 (5.0) 1.51 1.36 (0.83–2.24) 0.22 0.84 0.37rs3093349 G 68 (6.1) 19 (4.5) 1.41 1.37 (0.81–2.31) 0.23 0.87 0.56rs3093359 A 62 (5.6) 16 (3.8) 1.90 1.48 (0.84–2.60) 0.17 0.77 0.81rs3093363 G 307 (27.7) 105 (25.1) 0.99 1.14 (0.88–1.47) 0.32 0.94 0.57rs3093364 G 55 (5.0) 16 (3.8) 0.96 1.33 (0.75–2.34) 0.33 0.95 1.00rs963154 C 295 (26.7) 94 (22.7) 2.55 1.24 (0.95–1.62) 0.11 0.59 0.60rs3093375 C 308 (27.7) 105 (25.1) 1.03 1.14 (0.88–1.48) 0.31 0.94 0.60rs3093379 A 308 (27.6) 105 (25.1) 0.99 1.14 (0.88–1.47) 0.32 0.94 0.59rs2285452 A 266 (23.9) 92 (22.0) 0.59 1.11 (0.85–1.46) 0.44 0.98 0.45

* Only single-nucleotide polymorphisms (SNPs) with minor allele frequencies of �1% were analyzed. See Table 1 for other definitions.

Figure 2. Linkage disequilibrium plot of the IL21R–IL4R region, depicting the haplotype structure in the Centre d’Etude du PolymorphismeHumain from Utah samples genotyped in the International HapMap Project. The arrow points to the location of the IL21R single-nucleotidepolymorphism (SNP) rs3093301, which was associated with systemic lupus erythematosus in the 2 cohorts included in this study. Chr � chromosome.

IL21R POLYMORPHISM AND SLE RISK 2405

(Figure 2). To further confirm that the genetic effect inrs3093301 that we describe was not a surrogate for agenetic association in IL4R, we genotyped 5 SNPs withinthe IL4R gene in both the European-derived and His-panic cohorts. None of the genotyped SNPs in IL4R(rs2107356, rs2234895, rs1805011, rs1805013, rs2074570)showed an association with SLE in either cohort (P �0.05).

To determine whether the lupus-risk allele inrs3093301 is more frequently associated with any clinicalor serologic manifestation in SLE patients, we deter-mined the frequency of the various manifestations ofSLE in female patients who were homozygous for therisk allele (A/A) (n � 197) compared with the frequencyin female patients homozygous for the protective allele(G/G) (n � 62) in the European-derived cohort. Vari-ables tested included malar rash, discoid rash, photosen-sitivity, oral ulcers, arthritis, pericarditis, pleuritis, pro-teinuria, urine cellular casts, seizures, psychosis,hemolytic anemia, leukopenia, lymphopenia, thrombo-cytopenia, and the presence of anti-dsDNA, anti-Sm,anti–nuclear RNP, anti-Ro, anti-La, and anti–ribosomalP antibodies. We found that, in female patients in theEuropean-derived cohort, the A/A genotype was associ-ated with the presence of malar rash (60.9%; versus35.5% for the G/G genotype) (OR 2.83 [95% CI �1.56–5.13], �2 � 12.31, P � 0.00045). This associationremained significant after Bonferroni correction formultiple testing (corrected P � 0.0095). It was previouslyreported that reduced expression of IL-21R in peri-pheral blood B cells is associated with nephritis in lupuspatients (8). We found no difference in the frequency ofproteinuria (P � 0.51) or urine cellular casts (P � 0.14)between SLE patients with the A/A genotype comparedwith those patients with the G/G genotype in rs3093301.

Our group has previously reported a geneticassociation between IL21 and SLE (22). In that study,rs907715 within the IL21 gene was the most significantlyassociated SNP in the European-derived cohort (22). Totest if there is gene–gene interaction between IL21 andIL21R in lupus susceptibility, we performed logisticregression analysis to determine if the simultaneouspresence of the risk alleles in the IL21 lupus-associatedSNP (rs907715) and the IL21R lupus-associated SNP(rs3093301) could explain the increased susceptibility toSLE. We found no evidence of a synergistic effect of thepresence of the risk alleles in rs907715 and rs3093301(P � 0.58), which suggests that there is no geneticepistasis between the IL21 and IL21R loci in lupussusceptibility.

DISCUSSION

IL-21 is a pleiotropic cytokine that has multipleeffects on a number of immunocompetent cells. It iscritically involved in B cell activation, differentiation,and immunoglobulin production (6). Herein, we reporton the genetic association between a SNP in IL21R(rs3093301) and SLE in 2 ethnically divergent cohorts.This SNP is located on chromosome 16p11 in the firstintron of IL21R, just upstream of the translational startsite (Figure 2).

In our European-derived cohort, the risk of ma-lar rash was �3 times greater in female patients whowere homozygous for the rs3093301 risk allele (A/A)than in female patients who were homozygous for theprotective IL21R allele (G/G). However, this findingshould be considered preliminary, and it requires furtherreplication in other independent cohorts of SLE pa-tients. In a previously studied cohort of SLE patients,plasma IL-21 levels were elevated, but there was nocorrelation between IL-21 levels and malar rash or anyother symptom or sign of SLE (Lipsky PE: unpublishedobservations). IL21R is located on chromosome 16p11,which is close to 16p12, a region that has been previouslylinked to SLE susceptibility using linkage analysis stud-ies in familial SLE (22). The 16p11 region also harborsthe ITGAM gene, which has a confirmed associationwith SLE (23,24). The ITGAM and IL21R genes are notin linkage disequilibrium. IL21R is, therefore, a likelycandidate gene contributing to the genetic linkage effecton chromosome 16p12 along with the ITGAM gene.

In summary, we report and confirm for the firsttime the genetic association between SLE and IL21R.Further studies are needed to identify the causal SNPfor this association, and to determine the functionalconsequences of this polymorphism in SLE susceptibil-ity. These findings, along with the findings of ourprevious study on the association between IL21 and SLE(16), indicate that the IL-21/IL-21R pathway is impor-tant in understanding the pathogenesis of this disease.However, there is no evidence for gene–gene interaction(epistasis) between the lupus-associated SNPs in IL21and IL21R in the susceptibility to SLE.

ACKNOWLEDGMENT

We are thankful to Dr. Peter Gregersen for providingDNA control samples for our study.

AUTHOR CONTRIBUTIONSAll authors were involved in drafting the article or revising it

critically for important intellectual content, and all authors approved

2406 WEBB ET AL

the final version to be published. Dr. Sawalha had full access to all ofthe data in the study and takes responsibility for the integrity of thedata and the accuracy of the data analysis.Study conception and design. Kaufman, Langefeld, Reveille, Jacob,Lipsky, Harley, Sawalha.Acquisition of data. Webb, Merrill, Kelly, Sestak, Kaufman, Kimberly,Edberg, Ramsey-Goldman, Petri, Reveille, Alarcon, Vila, Alarcon-Riquelme, James, Gilkeson, Jacob, Moser, Gaffney, Vyse, Harley,Sawalha.Analysis and interpretation of data. Webb, Kaufman, Langefeld,Ziegler, Nath, Sawalha.

REFERENCES

1. Sawalha AH, Harley JB. Antinuclear autoantibodies in systemiclupus erythematosus. Curr Opin Rheumatol 2004;16:534–40.

2. Sestak A, Nath S, Sawalha A, Harley J. Current status of lupusgenetics [review]. Arthritis Res Ther 2007;9:210.

3. Vyse TJ, Todd JA. Genetic analysis of autoimmune disease. Cell1996;85:311–8.

4. Sestak AL, Shaver TS, Moser KL, Neas BR, Harley JB. Familialaggregation of lupus and autoimmunity in an unusual multiplexpedigree. J Rheumatol 1999;26:1495–9.

5. Reichlin M, Harley JB, Lockshin MD. Serologic studies ofmonozygotic twins with systemic lupus erythematosus. ArthritisRheum 1992;35:457–64.

6. Ettinger R, Kuchen S, Lipsky PE. The role of IL-21 in regulatingB-cell function in health and disease. Immunol Rev 2008;223:60–86.

7. Parrish-Novak J, Dillon SR, Nelson A, Hammond A, Sprecher C,Gross JA, et al. Interleukin 21 and its receptor are involved in NKcell expansion and regulation of lymphocyte function. Nature2000;408:57–63.

8. Mitoma H, Horiuchi T, Kimoto Y, Tsukamoto H, Uchino A,Tamimoto Y, et al. Decreased expression of interleukin-21 recep-tor on peripheral B lymphocytes in systemic lupus erythematosus.Int J Mol Med 2005;16:609–15.

9. Brandt K, Bulfone-Paus S, Foster DC, Ruckert R. Interleukin-21inhibits dendritic cell activation and maturation. Blood 2003;102:4090–8.

10. Kuchen S, Robbins R, Sims GP, Sheng C, Phillips TM, Lipsky PE,et al. Essential role of IL-21 in B cell activation, expansion, andplasma cell generation during CD4� T cell-B cell collaboration.J Immunol 2007;179:5886–96.

11. Ozaki K, Spolski R, Ettinger R, Kim HP, Wang G, Qi CF, et al.Regulation of B cell differentiation and plasma cell generation byIL-21, a novel inducer of Blimp-1 and Bcl-6. J Immunol 2004;173:5361–71.

12. Yang L, Anderson DE, Baecher-Allan C, Hastings WD, Bettelli E,

Oukka M, et al. IL-21 and TGF-� are required for differentiationof human T(H)17 cells. Nature 2008;454:350–2.

13. Maeda M, Yanagawa Y, Iwabuchi K, Minami K, Nakamaru Y,Takagi D, et al. IL-21 enhances dendritic cell ability to induceinterferon-� production by natural killer T cells. Immunobiology2007;212:537–47.

14. Vinuesa CG, Cook MC, Angelucci C, Athanasopoulos V, Rui L,Hill KM, et al. A RING-type ubiquitin ligase family memberrequired to repress follicular helper T cells and autoimmunity.Nature 2005;435:452–8.

15. Herber D, Brown TP, Liang S, Young DA, Collins M, Dunussi-Joannopoulos K. IL-21 has a pathogenic role in a lupus-pronemouse model and its blockade with IL-21R.Fc reduces diseaseprogression. J Immunol 2007;178:3822–30.

16. Sawalha AH, Kaufman KM, Kelly JA, Adler AJ, Aberle T,Kilpatrick J, et al. Genetic association of interleukin-21 polymor-phisms with systemic lupus erythematosus. Ann Rheum Dis 2008;67:458–61.

17. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, RothfieldNF, et al. The 1982 revised criteria for the classification of systemiclupus erythematosus. Arthritis Rheum 1982;25:1271–7.

18. Hochberg MC, for the Diagnostic and Therapeutic Criteria Com-mittee of the American College of Rheumatology. Updating theAmerican College of Rheumatology revised criteria for the clas-sification of systemic lupus erythematosus [letter]. ArthritisRheum 1997;40:1725.

19. Jeffries M, Hamadeh F, Aberle T, Glenn S, Kamen DL, Kelly JA,et al. Haemolytic anaemia in a multi-ethnic cohort of lupuspatients: a clinical and serological perspective. Lupus 2008;17:739–43.

20. Gray-McGuire C, Moser KL, Gaffney PM, Kelly J, Yu H, OlsonJM, et al. Genome scan of human systemic lupus erythematosus byregression modeling: evidence of linkage and epistasis at 4p16-15.2. Am J Hum Genet 2000;67:1460–9.

21. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis andvisualization of LD and haplotype maps. Bioinformatics 2005;21:263–5.

22. Lee YH, Nath SK. Systemic lupus erythematosus susceptibility locidefined by genome scan meta-analysis. Hum Genet 2005;118:434–43.

23. Nath SK, Han S, Kim-Howard X, Kelly JA, Viswanathan P,Gilkeson GS, et al. A nonsynonymous functional variant inintegrin-�(M) (encoded by ITGAM) is associated with systemiclupus erythematosus. Nat Genet 2008;40:152–4.

24. Han S, Kim-Howard X, Deshmukh H, Kamatani Y, ViswanathanP, Guthridge JM, et al. Evaluation of imputation-based associationin and around the integrin-�-M (ITGAM) gene and replication ofrobust association between a non-synonymous functional variantwithin ITGAM and systemic lupus erythematosus (SLE). HumMol Genet 2009;18:1171–80.

IL21R POLYMORPHISM AND SLE RISK 2407