IL-21 Is a Double-Edged Sword in the Systemic Lupus ... · The Journal of Immunology IL-21 Is a Double-Edged Sword in the Systemic Lupus Erythematosus–like Disease of BXSB.Yaa Mice

of February 3, 2019.This information is current as

MiceYaaof BXSB.like Disease−Systemic Lupus Erythematosus

IL-21 Is a Double-Edged Sword in the

RoopenianGregory J. Christianson, Herbert C. Morse III and Derry C.Giljun Park, Martin P. Steinbuck, William H. Schott, Caroline G. McPhee, Jason A. Bubier, Thomas J. Sproule,

http://www.jimmunol.org/content/191/9/4581doi: 10.4049/jimmunol.1300439September 2013;

2013; 191:4581-4588; Prepublished online 27J Immunol 

MaterialSupplementary

9.DC1http://www.jimmunol.org/content/suppl/2013/09/27/jimmunol.130043

Referenceshttp://www.jimmunol.org/content/191/9/4581.full#ref-list-1

, 24 of which you can access for free at: cites 47 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. All rights reserved.1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on February 3, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on February 3, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

IL-21 Is a Double-Edged Sword in the Systemic LupusErythematosus–like Disease of BXSB.Yaa Mice

Caroline G. McPhee,* Jason A. Bubier,* Thomas J. Sproule,* Giljun Park,*

Martin P. Steinbuck,*,1 William H. Schott,* Gregory J. Christianson,*

Herbert C. Morse, III,† and Derry C. Roopenian*

The pleiotropic cytokine IL-21 is implicated in the pathogenesis of human systemic lupus erythematosus by polymorphisms in the

molecule and its receptor (IL-21R). The systemic lupus erythematosus-like autoimmune disease of BXSB.Yaa mice is critically

dependent on IL-21 signaling, providing a model for understanding IL-21/IL-21R signaling in lupus pathogenesis. In this study, we

generated BXSB.Yaa mice selectively deficient in IL-21R on B cells, on all T cells, or on CD8+ T cells alone and examined the

effects on disease. We found that IL-21 signaling to B cells is essential for the development of all classical disease manifestations,

but that IL-21 signaling also supports the expansion of central memory, CD8+ suppressor cells and broadly represses the cytokine

activity of CD4+ T cells. These results indicate that IL-21 has both disease-promoting and disease-suppressive effects in the

autoimmune disease of BXSB.Yaa mice. The Journal of Immunology, 2013, 191: 4581–4588.

Systemic lupus erythematosus (SLE) is a complex auto-immune disease in which a loss of tolerance to nucleicacids results in an abundance of autoantibodies that form

pathogenic immune complexes and cause multisystem clinicaldisease. CD4+ T cells and B cells acting in a cooperative mannerare critical drivers of all forms of SLE, but debate continues overtheir mechanisms of action. Further complicating the issue is therecognition that regulatory cells, including CD4+ regulatory Tcells (Tregs) and CD8+ Tregs/suppressor T cells (Ts), are able toslow the development of disease. Thus, the balance between dis-ease drivers and controllers determines the course and outcomesof SLE.IL-21, a member of the common g-chain cytokine family, has

gained increasing interest because of its potent effects on normalimmunological homeostasis and disease (reviewed in Refs. 1, 2).This cytokine is a major product of CXCR5hiICOShiCD4+ fol-licular Th cells (TFH) as well as a CD4+CXCR5lo/2ICOShi Thsubset, termed extrafollicular Th cells (TEFH), that is involved in

extrafollicular differentiation of plasmablasts and plasma cells (3–5). IL-21 is emerging as a critical driver for the development ofmany forms of autoimmune disease, including SLE (1, 2, 6–10).IL-21 signals through its receptor, IL-21R, which is expressed bymany types of lymphoid and myeloid cells (1, 2, 11). In keepingwith this expression pattern, there is evidence for the need for bothCD4+ T cell–intrinsic and B cell–intrinsic IL-21 signaling in thepromotion of SLE-like autoimmune diseases (12). IL-21 can act inan autocrine manner to promote the differentiation of naive CD4+

T cells to TFH (13–15) and suppress the functions of CD4+ Tregs(16). IL-21 can also potently promote Ag-experienced B cellsthrough follicular and extrafollicular development, class-switchrecombination, and plasma cell differentiation (17–22). IL-21 isalso a powerful inducer of the cytotoxic arm of immunity, stim-ulating the maturation and enhancing the activity of NK and CD8+

T cells, as well as acting with IL-15 to support the maintenance ofmemory CD8+ T cells (23–26). The downstream consequences ofIL-21 signaling can therefore be complex.BXSB male mice carrying the Y-linked autoimmune accelerator

mutation Yaa recapitulate key facets of severe human SLE (4, 27–30). Yaa is a duplication of genes, most importantly TLR7 (Tlr7),that normally reside on the X chromosome but were translocatedto the Y chromosome (31, 32). Therefore, male mice bearing Yaaexpress two copies of Tlr7, whereas females express only onecopy due to X inactivation. Heightened TLR7 signaling in re-sponse to ligation of ssRNA is a primary driver of SLE in BXSB.Yaa mice (31, 32). In cooperation with autosomal SLE predispo-sition alleles, TLR7 signaling promotes interactions among den-dritic cells, activated B cells, and CD4+ T cells, ultimately leadingto the production of copious amounts of anti-nuclear Abs (ANA),resulting in severe clinical disease (31, 33). Recent studies havealso identified an expansion of CD8+ T cells with a central memory(CM) phenotype that act to substantially retard the onset and se-verity of disease in BXSB.Yaa and B6.Yaa mice (27, 34).The pathogenesis of disease in BXSB.Yaa mice is characterized

by an expansion of TFH as well as TEFH, both of which expresshigh levels of IL-21 (4). In contrast, Th17 cells, which have beenidentified as another source of IL-21 (35, 36), are not expanded inthis strain (4, 27). BXSB.Yaa mice lacking the IL-21R are highly

*The Jackson Laboratory, Bar Harbor, ME 04609; and †Laboratory of Immunoge-netics, National Institute of Allergy and Infectious Diseases, National Institutes ofHealth, Rockville, MD 20852

1Current address: Department of Pathology and Laboratory Medicine, Boston Uni-versity School of Medicine, Boston, MA.

Received for publication February 14, 2013. Accepted for publication August 27,2013.

This work was supported in part by the Intramural Research Program of the NationalInstitutes of Health, National Institute of Allergy and Infectious Diseases (to H.C.M.)and by the Alliance for Lupus Research (to D.C.R.). C.G.M. was supported in part bya National Institutes of Health Training Grant. J.A.B. was supported by the ArthritisFoundation.

Address correspondence and reprint requests to Dr. Derry C. Roopenian or Dr.Herbert C. Morse, III, The Jackson Laboratory, 600 Main Street, Bar Harbor, ME04609 (D.C.R.) or National Institute of Allergy and Infectious Diseases, NationalInstitutes of Health, Room 1421, 5640 Fishers Lane, Rockville, MD 20852 (H.C.M.).E-mail addresses: [email protected] (D.C.R.) or [email protected] (H.C.M.)

The online version of this article contains supplemental material.

Abbreviations used in this article: ANA, anti-nuclear Ab; BM, bone marrow; CM,central memory; EM, effector memory; IL-21r Ø, IL-21R–deficient; MZ, marginalzone; PD-1, programmed cell death 1; qPCR, quantitative PCR; SLE, systemic lupuserythematosus; TEFH, extrafollicular Th cell; TFH, follicular Th cell; Treg, regulatoryT cell; Ts, suppressor T cell; Wt, wild-type.

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1300439

by guest on February 3, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

resistant to lupus (4, 27). Despite this, the administration of an IL-21R–Fc fusion protein known to bind and remove IL-21 was notan effective treatment (37). Early in the course of treatment, somemice developed more severe disease but, as treatment continued,the disease became less severe and a protective effect was ob-served. It was suggested that the biphasic response to this inter-vention could be ascribed to an early negative impact of reducedIL-21 signaling on CD8+ Ts, whereas the later protective effectcould be attributed to limiting IL-21–driven B cell proliferationand differentiation. In this study, we directly address the lymphocytepopulations targeted by IL-21 that drive or restrict the lupus-likedisease of these mice.

Materials and MethodsMice

Mice were bred and maintained in a specific pathogen-free mouse colonyat The Jackson Laboratory, including BXSB/MpJ (designated BXSB.Yaa)and BXSB.B6-Yaa+/J (designated BXSB.B6Y). All experiments wereperformed under protocol 01022 approved by the Institutional AnimalCare and Use Committee. BXSB.Yaa mice carrying the Il21r null allelewere bred to N11 (4). BXSB-Cd8a2/2 mice were as described (27). Bcell–deficient BXSB-Ighmtm1Cgn (38) and lymphocyte-deficient BXSB.Rag1tm1Mom mice (both N11) were produced similarly. T cell–deficientBXSB-Tcratm1Mjo/TheoJ mice (N8) were generously provided by DwightKono (Scripps Institute) (39).

Mixed bone marrow chimeras

Mixed bone marrow (BM) chimeras were generated and validated as de-scribed in Supplemental Table I and Supplemental Fig. 1. Recipient micewere irradiated twice with 550–600 cGY 3 to 4 h apart and were theninjected i.v. with a 1:1 mixture of 2.5 3 106 BM cells from each donor.

Flow cytometry

Flow cytometry was performed using established procedures and Absagainst B220, CD8, ICOS, CD11b, IL-21R, programmed cell death 1 (PD-1) (103227, 100730, 313516, 101216, 131906, and 109109; BioLegend),CD4 (MCD0430; Invitrogen), CD44, CD62L (27-0441, 25-0621; eBio-science), and CD122 and FAS (552818, 557653; BD Pharmingen). Studieswere performed on a two-laser/four-color FACSCalibur or a four-laser/13-color BD LSRII analytical cytometer (both from BD Biosciences) andanalyzed with FlowJo software (Tree Star).

Serum Abs

IgG1, IgG2b, IgG2c, and IgM serum levels were quantified by ELISAtechniques with data expressed as concentration of Ig per milliliter byinterpolation using a standard curve based on the titration of purified mouseIgs as described (40). Anti-DNA Abs were quantified by ELISA as de-scribed (41). ANA abundance was determined by their binding to Hep-2cells (Antibodies, Inc.) using 15 ml 1:40 dilution of serum following thesupplier’s protocol. Intensity was quantified based on a scale of 0–4 asdescribed (4, 27).

Renal pathology

Formalin-fixed, paraffin-embedded kidneys were stained with H&E andperiodic acid-Schiff. The sections were scored in a blinded manner usingcriteria described in Weening et al. (42). For immunohistochemistry, kidneysections were incubated with anti-IgG (1030-01; Southern BiotechnologyAssociates) as previously described (27). Slides were scanned on a Nano-zoomer slide scanner at 340, and images were obtained using NDP.view2software (Hamamatsu).

Gene expression analysis

Spleen cells were isolated and stained with anti-CD4, IL-21R, and B220(clones denoted above), and propidium iodine was used to assess viability.Using an FACSAria (BD Biosciences), cells were gated and sorted forviable B2202CD4+IL-21R+ and B2202CD4+IL-21R2. Total RNA wasextracted with the RNeasy Micro kit (Qiagen) followed by cDNA synthesisusing the QuantiTect Reverse Transcription kit (Qiagen) according to themanufacturer’s protocol. The cDNAwas combined with SYBR Green RT-PCR master mix (Applied Biosystems) and oligonucleotide primers. Oli-gonucleotide primers for RT-quantitative PCR (qPCR): Il2, forward 59-

AGCAGGCCACAGAATTGAAAG-39 and reverse 59-CTCATCATCG-AATTGGCACTCA-39; Il6, forward 59-TAGTCCTTCCTACCCCAATTTCC-39 and reverse 59-TTGGTCCTTAGCCACTCCTTC-39; IL10, for-ward 59-CCAAGCCTTATCGGAAATGAT C-39 and reverse 59-TTTC-TGGGCCATGCTTCTCT-39; IL21, forward 59-GAAGATGGCAATGAA-AGCCTG T-39 and reverse 59-AGGATGTGGGAG AGGAGACTGA-39;IL21R, forward 59-CCTTCTCAGGACGCTATGATATCTC-39 and reverse59-CTTGCCCCTCAGCACGTAGT-39; Ifng, forward 59-CAAGCGGCT-GACTGAACTCA-39 and reverse 59-CACTGCAGCTCTGAATGTTTC-TTATT-399; Tnfa, forward 59-TATGGCTCAGGGTCCAACTC-39 andreverse 59-CTCCCTTTGCAGAACTCAGG-39; IL17a, forward 59-GAA-GAT GCTGGTGGGTGTGG-39 and reverse 59-AGCCGCGGGTCTC-TGTTTAG-39; Stat3, forward 59-GAC CCGCCAACAAATTAAGA-39and reverse 59-TCGTGGTAAACTGGACACCA-39; Stat5a, forward 59-GTGAAGCGCTCAACATGAAA-39 and reverse 59-ACTGGGACCAG-GACACAGAC-39; Stat5b, forward 59-CCAGATGCAGGCCTTGTACG-39and reverse 59-AGGCTTGGCTTTCGATCCAC-39; 18sRNA, forward 59-CCGCAGCTAGGAATAATGGAAT-39 and reverse 59-CGAACCTCCG-ACTTT CGTTCT-39; and Gapdh, forward 59-TGTGTCCGTCGTGGAT-CTGA-39 and reverse 59-TTGCTGTTG AAGTCGCAGGAG-39. RT-qPCRgene expression quantification (43) was performed in 384-microwell plates(Applied Biosystems) using ViiA7 Real-Time PCR system (Applied Bio-systems).

Statistical analyses

The two-tailed heteroscedastic Student t test was used for two samplecomparisons. The log-rank test was used for survival analysis.

ResultsExpression of the IL-21R is increased on B cells andCD8+ T cells but not CD4+ T cells early in the course ofthe BXSB.Yaa disease

BXSB.Yaa mice represent a unique model of SLE in that theirCD4+ T cells express high levels of IL-21, a cytokine absolutelyrequired for the lupus-like disease that characterizes this strain(4, 17, 27). Importantly, however, the cell types responsive to IL-21in this disease have not been defined. As IL-21 can induce targetcells to upregulate their IL-21Rs (44), we quantified levels of IL-21R expression on subpopulations of spleen cells from 6- and 14-wk-old BXSB.Yaa and control BXSB mice carrying a C57BL/6-derived Y chromosome (BXSB.B6Y) (Fig. 1A, 1B). The elevatedexpression of the IL-21R was seen on all subsets of B220+ B cellsexamined, including the dominant follicular B cell subset (Fig.1A, 1B) as well as T1 B cells, marginal zone (MZ) B cells, andgerminal center B cells (not shown). In contrast, Yaa was notassociated with heightened expression of IL-21R on total CD4+

T cells or the ICOShi subsets that are the primary sources of IL-21.However, Yaa consistently caused the upregulation of IL-21R onCD8+ T cells from mice of both ages. These data suggest thatelevated expression of IL-21R on Yaa B cells and CD8+ T cellsmay position them for heightened responses to IL-21.

IL-21R–positive B cells are required for the development of thelupus-like disease of BXSB.Yaa mice

The disease of BXSB.Yaa mice, like other systemic humoralautoimmune disorders, is characterized by expanded numbers ofactivated B cells, hypergammaglobulinemia, and high levels ofcirculating autoantibodies clearly defining a pathogenic role forB cells. These mice also develop a characteristic monocytosis andsubstantially increased numbers of ICOShi TFH/TEFH that are con-sidered to be critical cellular drivers of this disease (4, 27). Wefirst asked if B cells are necessary for development of disease. Todo this, we introduced a null allele (Ø) allele of Ighm onto theBXSB genetic background to generate B cell–deficient Yaa mice.Survival studies demonstrated that BXSB.Yaamice lacking B cellswere highly resistant to the lethal effects of BXSB.Yaa disease(Fig. 2A). Resistance to disease in the absence of B cells wasassociated with remarkably low numbers of CD4+ T cells and

4582 IL-21, A DOUBLE-EDGED SWORD IN LUPUS

by guest on February 3, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

CD11b+ monocytes (Fig. 2B). Importantly, there was no increasein the numbers of ICOShi CD4+ T cells that are elevated in B cell–competent BXSB.Yaa mice, suggesting that B cells are requiredfor the development of TFH/EFH cells (Fig. 2B). These resultsdemonstrated that B cells are central to several cellular manifes-tations as well as the lethal aspects of the autoimmune diseasecharacteristic of BXSB.Yaa mice.In view of the critical requirement for B cells in disease de-

velopment, their heightened expression of the IL-21R, and therequirement of IL-21 signaling for all aspects of disease, we askedif expression of the IL-21R on B cells contributed to the devel-opment of disease. Mixed BM chimeras were produced in whichlethally irradiated female BXSB mice were reconstituted with a 1:1mixture of B cell–deficient (Ighm2/2) and IL-21R–deficient (Il21rØ)BXSB.Yaa BM cells or with a mixture of Il21r Ø Ighm wild-type(Wt) and Il21r Wt BXSB.Yaa BM cells. The former pairingresulted in donor-derived splenic populations in which B cellslacked the IL-21R, with all other cell types being a mixture ofIl21rWt and Il21r Ø genotypes, whereas the latter pairing resultedin chimeras with all donor BM-derived compartments being com-prised of both the Il21r Wt and Ø genotypes (Supplemental Fig.1A, Supplemental Table I). Survival studies showed that by 32 wkafter reconstitution, 79% of mice with IL-21R–competent B cells

had died. In contrast, only 9% of mice reconstituted with donorIL-21R–deficient B cells succumbed at late time points (Fig. 3A).The late deaths in this group were likely due to the fact that asimilar percentage of this cohort exhibited partial reconstitutionby radioresistant host IL-21R+ B cells (up to 5%; data not shown).Even considering partial host cell reconstitution in some mice, thiscohort had significantly reduced serum levels of IgM, IgG1,IgG2b, IgG2c, anti-DNA Abs, and ANA as compared with chi-meras that received IL-21R B cell–competent BM (Fig. 3B).Histological analysis of kidneys from surviving 32-wk chimericmice were consistent with an IL-21 dependence for glomerularpathology in that mice lacking IL-21R+ B cells had minimalglomerular deposition of IgG and glomerulonephritis (Supple-mental Fig. 2).Further comparisons showed that a deficit in IL-21R–expressing

B cells was associated with substantially reduced numbers of totalspleen cells and CD19+ B cells (Fig. 3C). Previous studies showedthat the low frequencies of splenic MZ B cells characteristic ofBXSB.Yaa mice were normalized in BXSB.Yaa mice globallylacking the IL-21R (23). The frequencies of splenic MZ B cellswere significantly increased in mice with a deficit restricted to IL-21R–expressing B cells, indicating that direct signaling by IL-21to B cells is a requirement for the reduced numbers of MZ B cells

FIGURE 1. Yaa acts early to cause the upregulation of IL-21R on B cells and CD8+ T cells but not on CD4+ T cells. (A) Results compare IL-21R

expression of spleen cells from BXSB.Yaa and BXSB.B6Y control mice (n = 5) at 6 and 14 wk of age. The mean fluorescence intensity (MFI) + SEM of

spleen cells from IL-21r2/2 BXSB controls (,100 MFI in this experiment) is shown. (B) Representative histograms of 14-wk-old BXSB.Yaa and BXSB.

B6Y splenic B and T cells. *p , 0.05, **p , 0.01.

FIGURE 2. BXSB.Yaa mice lacking B cells are resistant to autoimmune disease. (A) Survival of Ighm2/2 and Wt BXSB.Yaa mice (n. 42). (B) The lack

of B cells was associated with substantial reductions in the numbers of splenic CD4+ T cells and CD11b+ cells and the expression of ICOS on CD4+ T cells.

n = 5 at 17 wk of age. *p , 0.05, **p , 0.01.

The Journal of Immunology 4583

by guest on February 3, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

in BXSB.Yaa mice (Fig. 3C). We also found that the frequenciesof splenic monocytes, which progressively increase in BXSB.Yaamice, were significantly lower in mice with a deficit in IL-21R–expressing B cells, indicating that monocytosis is also dependenton B cells responding to IL-21 (Fig. 3C). In contrast, this deficithad no effect on the total numbers of splenic CD4+ T cells (Fig.3C), the levels of ICOS expressed by these cells were only min-imally reduced, and the expression of PD-1, another marker forTFH, was unchanged (Fig. 3D). These patterns contrast with theconsistently reduced expression of ICOS and PD-1 typically ob-served for CD4+ T cells from BXSB.Yaa fully lacking the IL-21Rcompared with Il21R Wt BXSB.Yaa mice (Fig. 3D). Taken to-gether, these results support a critical role for IL-21 signaling toB cells in the development of multiple pathological manifestationsof BXSB.Yaa disease, including their greatly shortened survival,while still enabling a high state of CD4+ T cell activation.

Development of BXSB.Yaa SLE-like disease does not requireIL-21 signaling to T cells

Observations indicating that IL-21 promotes the differentiation ofTFH suggest that IL-21 may act in an autocrine manner to supportthe expansion and function of these cells and thereby contributeto disease (13–15, 45). To examine this possibility, we generatedmixed BM chimeras distinguished only by the presence or ab-sence of IL-21R–competent T cells. This was accomplished by thetransfer of mixtures of Tcra Ø and Il21r2 Ø BXSB.Yaa BM cellsinto lethally irradiated BXSB female recipients that were alsoRag1 Ø to negate the possibility of host lymphocyte repopulation(Supplemental Fig. 1B, Supplemental Table I). Survival studiesrevealed no significant differences between these cohorts through

34 wk of observation (Fig. 4A). Despite the fact that the survivalof these cohorts was essentially identical, we unexpectedly foundthat serum IgG1, IgG2b, and ANA levels were significantly in-creased in mice lacking IL-21R+ T cells (Fig. 4B). However, thesereadouts of heightened B cell activation in mice with IL-21R–deficient T cells were not accompanied by manifestations of ac-celerated systemic disease as the cohorts did not differ signifi-cantly for renal scores or the extent of glomerular Ig deposition.(Supplemental Fig. 2). In addition, they did not differ significantlyin studies of PBL for expression of FAS on B cells (anothermarker of activation), levels of ICOS on CD4+ T cells (Fig. 4C), orthe frequencies of monocytes (Fig. 4C), B cells, or CD4+ T cells(Fig. 4D). The only significant difference in T cells between thecohorts was a reduction in the frequencies of CD8+ T cells in micelacking expression of IL-21R on all T cells (Fig. 4D). As thesedata provided evidence to suggest that a failure of CD4+ T cells tosignal through the IL-21R promoted certain disease biomarkers,we sought to determine the consequences of expression of the IL-21R by CD4+ T cells more directly. To do so, we set up a BMreconstitution experiment in which lethally irradiated BXSB.YaaRag1 Ø females were reconstituted with 1:1 mixture of Il21r Øand Il21r Wt BXSB.Yaa BM. The CD4+ T cells that develop inthese mixed BM chimeras will have comparable exposure to au-toimmune processes with the only differential being their abilityto receive signals through the IL-21R. Fourteen weeks after re-constitution, IL-21R–positive and –negative CD4+ T cells werepurified by FACS (Fig 5A). RNA was isolated and analyzed byRT-qPCR to determine the effects of IL-21 signaling on expres-sion of a selected set of genes (Fig. 5B). Surprisingly, transcriptsfor Il2, Il10, Il21, Ifng, Tnfa, Stat3, and Stat5a were significantly

FIGURE 3. IL-21R+ B cells are required for development of BXSB.Yaa disease. (A) Survival of mixed BM chimeras with IL-21R expression varied only

in B cells (n = 19–23). (B) Expression of IL-21R on B cells is required for the elevation of serum Igs, anti-ssDNA and ANA. n = 5–8, 32 wk after BM

transplantation (BMT). (C) Total numbers of spleen cells, CD19+ B cells, MZ B cells (B220+, CD21hi, CD23lo), and percentages of splenic monocytes

(CD11b+F4/802) are affected by the expression of IL-21R on B cells. n = 5, 23 wk after BMT. (D) Expression of IL-21R on B cells has a small effect on the

expression of TFH markers. n = 7–10, 32 wk after BMT. Results from spleen cells from nonreconstituted Il21r Wt (n = 7) and Il21r2/2 (n = 9) BXSB.Yaa

mice at 30 wk of age and evaluated in parallel with the BMT samples are shown for comparison. *p , 0.05, **p , 0.01.

4584 IL-21, A DOUBLE-EDGED SWORD IN LUPUS

by guest on February 3, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

increased in IL-21R–deficient CD4+ T cells. These results indi-cated that IL-21 signaling to CD4+ T cells broadly represses theexpression of cytokines, including those considered to promote(IL-21, IFN-g, and TNF-a) and limit (IL-2/IL-10) autoimmunedisease (46, 47).

IL-21 promotes the disease suppressive activity of CD8+ T cellsand the development of CM CD8+ T cells

We recently identified a potent regulatory axis comprised of CD8+

T cells and NK cells that actively suppresses disease progressionin BXSB.Yaa mice (27). A similar population of CD8+ Ts witha CM phenotype was shown to be active in B6.Yaa mice (34). Thefact that the expression of IL-21R is increased on CD8+ T cells inBXSB.Yaa mice as young as 6 wk (Fig. 1) and that the frequenciesof CD8+ T cells were selectively reduced in PBL of BM chimeraslacking IL-21R on T cells (Fig. 4C) prompted us to directly testwhether IL-21R signaling of CD8+ T cells can influence BXSB.Yaa disease. To do this, we generated BM chimeras distinguishedonly by the presence or absence of the IL-21R on CD8+ T cells(Supplemental Fig. 1C, Supplemental Table I). In contrast to

chimeras in which all T cells lacked the IL-21R, the survival ofchimeric mice was significantly reduced when a deficiency in theIL-21R was restricted to CD8+ T cells (Fig. 6A). Consistent withobservations made for chimeras in which all T cells lacked the IL-21R, the lack of expression of the IL-21R on CD8+ T cells wasassociated with increased levels of serum IgGs and IgM, whereasANA levels were also elevated but to a lesser extent (Fig. 6B). Thetwo cohorts expressed similar levels of ICOS on CD4+ T cells andFAS on B cells in PBL (Fig. 6C). Similarly, the frequencies ofB cells and CD4+ T cells in PBL were not significantly different,but the frequency of CD8+ T cells was significantly lower in micehaving only IL-21R–deficient CD8+ T cells (Fig. 6D). An ex-panded population of CD122hiCD44hiCD62Lhi CM CD8+ T cellsis correlated strongly with splenic CD8+ Ts in BXSB.Yaa as wellas B6.Yaa mice (27, 34). However, although some mice with IL-21R–competent CD8+ T cells showed a considerable elevationin the expression of CD122, the levels were not significantlydifferent from those of mice lacking IL-21R–expressing CD8+

T cells (Fig. 6E). The lack of IL-21R–expressing CD8+ T cellsdid, however, result in a significant shift in the relative frequencies

FIGURE 4. IL-21R+ T cells are not

required for disease progression in

BXSB.Yaamice. (A) Survival of mixed

BM chimeras with IL-21R expression

varied only on T cells (n = 13 to 14).

(B) Serum Ig and ANA levels 15 wk

after BM transplantation (BMT) (n =

18–20). (C) Percentages of monocytes

and activated CD4+ T and B cells in

peripheral blood at 16 wk following

BMT (n = 9–13). Data from PBL of

Yaa (n = 4), B6Y (n = 6), and Il21r2/2

Yaa BXSB mice (n = 2) 14 wk of age

evaluated in parallel with the BMT

samples are shown for comparison. (D)

Relative proportions of B220+ B cells,

CD4+ T cells, and CD8+ T cells among

PBL at 16 wk after BM transfer (n =

10–13). (A–D) Data are representative

of two experiments. *p, 0.05, **p,0.01.

FIGURE 5. IL-21 signaling prevents CD4+ T cells from acquiring a more pathogenic phenotype. Lethally irradiated Rag2/2 BXSB female mice were

reconstituted with a 1:1 mixture of BM from BXSB.Yaa Il21r Ø and BXSB.Yaa Il21rWt mice. (A) Fourteen weeks postreconstitution, IL-21R–positive and

–negative CD4+ T cells were sort-purified by FACS, and gene expression analysis of RNAwas performed by RT-qPCR procedures. (B) DD cycle threshold

(CT) values of samples run in technical triplicates of three to four independently analyzed recipient mice using 18S RNA as a normalizer. Similar results

were found when GAPDH was used as a normalizer (not shown). *p , 0.05, **p , 0.01, ***p , 0.001.

The Journal of Immunology 4585

by guest on February 3, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

of circulating CD8+ T cell subsets, evidenced by reductions in CMand naive CD8+ T cells and an increase in effector memory (EM)cells (Fig. 6F). Taken together, these results are consistent withthe proposition that a population of CM CD8+ Ts that retard thedevelopment of lethal autoimmune disease in BXSB.Yaa mice isIL-21R dependent.

DiscussionRecent studies defined three cardinal features of the severe SLE-like disease of BXSB.Yaa mice: the mice express high levels ofIL-21 (17); IL-21 signaling is critically required for the develop-ment of all measured parameters of disease (4, 27); and a regula-tory axis comprised of CM CD8+ Ts and NK cells retards earlydisease development and progression (27, 34). The data presentedin this study unify these observations by showing that IL-21functions as a double-edged sword in this disease, driving theexpansion of autoreactive B cells while also retarding the diseaseby promoting the development of CD8+ Ts with a CM phenotype.First, we found that the Yaa mutation results in enhanced ex-

pression of the IL-21R on splenic B cells and CD8+ T cells but noton CD4+ T cells in mice as young as 6 wk, well before the de-velopment of overt signs of autoimmunity. This suggests thatheightened expression of the IL-21R may serve as a novel earlybiomarker of impending disease. The early increase in IL-21R ex-pression on CD8+ T cells is unlikely to be a direct consequence ofcell-intrinsic effects of the Yaa mutation because among lympho-cytes, TLR7 expression is limited to B cells. Enhanced expressionof IL-21R both on B cells and CD8+ T cells is more likely to beinduced by cell-extrinsic stimuli, potentially including a paracrineresponse to IL-21 (44).

Second, our results show that direct signaling of IL-21 to B cellsis the primary mechanism by which IL-21 drives the autoimmunedisease of BXSB.Yaa mice. BXSB.Yaa mice that lack B cells werehighly resistant to disease and failed to exhibit hematopoieticmarkers of disease including expansions of ICOShi TFH/TEFH cellsand monocytes. Thus, B cells contribute globally to the patho-genesis of this disease. To a surprising degree, the selective absenceof the IL-21R on B cells resulted in the same disease-protectiveeffects. BXSB.Yaa mice lacking IL-21R–positive B cells experi-enced greatly prolonged survival, even including the small numberof mice in which low levels of host cell chimerism with IL-21R–positive B cells was evident. In keeping with disease resistance,the lack of IL-21R–expressing B cells broadly impacted andnormalized the cellular and serological features that characterizethe SLE-like disease of BXSB.Yaa mice. The effects of a selec-tive lack of the IL-21R on B cells parallels findings made inBXSB.Yaa mice with a global deficit in expression of the IL-21R(4, 27), with the exception that it did not limit the development ofICOShi TFH/TEFH cells, which are the primary sources of IL-21.These results serve to highlight the centrality of B cell–intrinsicIL-21 signaling in the development of this lethal SLE-like auto-immune disease.Third, our studies, to our knowledge, provide the first evidence

to support an immunosuppressive role for IL-21 signaling of CD8+

T cells in autoimmune diseases. In contrast to the stringent re-quirement for IL-21 signaling to B cells, the selective ablation ofIL-21R expression on CD8+ T cells was disease protective. Micelacking IL-21R selectively on CD8+ T cells showed increases inserum Ig levels and ANA, and, most importantly, acceleratedmorbidity. Although this selective deficiency had little impact on

FIGURE 6. CD8+ T cells in BXSB.Yaa mice require IL-21R expression to suppress lethal disease and to expand CM CD8+ T cells. (A) Survival of mixed

BM chimeras with IL-21R varied only on CD8+ T cells (n = 18–20). (B) Serum IgM, IgG, and ANA levels 13 wk after BMT (n = 10–20). (C) Percentages of

B220+ B cells, CD4+, CD8+ T cells from PBL of chimeras with IL-21R varied only on CD8+ T cells. (D) Expression levels of FAS on B cells and ICOS on

CD4+ T cells in PBL. (E) MFI of CD122 on CD8+ T cells in PBL. (F) Relative frequencies of CM (CD44hiCD62Llo), EM (CD44hiCD62Llo) and naive

(CD44lo CD62Lhi) CD8+ T cells in PBL of chimeric mice. (C–F) Results from 15 wk after BMT (n = 18–20). (B–E) Data are representative of two

experiments. *p , 0.05, **p , 0.01.

4586 IL-21, A DOUBLE-EDGED SWORD IN LUPUS

by guest on February 3, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

the activation status of CD4+ T cells and B cells, it resulted ina reduction in the frequency of CD8+ T cells with a CM phenotypeand naive CD8+ cells, whereas the proportion of EM CD8+ T cellswas increased. Other studies have clearly demonstrated the im-portance of IL-21 in maintaining CM CD8+ T cell populations(23, 25, 26). Additional studies, including results from BXSB.Yaamice, strongly support a regulatory role for CM CD8+ cells thatretard the development of autoimmune disease by eliminating orotherwise controlling autoreactive B cells and/or TFH (27, 34).Although unlikely and not in keeping with the literature, the in-crease in naive CD8+ T cells seen in the presence of IL-21 sig-naling to CD8+ T cells could also be exerting a suppressive effect.Although the exact nature of these CD8+ suppressor T cells remainsto be determined, our results provide strong evidence that theyrequire IL-21 to mediate their immunosuppressive effect.Fourth, our findings strongly argue that IL-21 signaling to CD4+

T cells does not contribute significantly to the overall pathogenesisof autoimmune disease in BXSB.Yaa mice. These results obtainedusing the TLR7-dependent BXSB.Yaa model differ from recentstudies of an alloantigen-induced chronic graft-versus-host diseasemodel in which IL-21 signaling by CD4+ T cells contributedsignificantly to disease pathogenesis (12). Although the lack of IL-21R on all T cells in BXSB.Yaa mice did not result in acceleratedmortality, it was associated with increased serum levels of IgGsand ANA, tendencies toward monocytosis, and increased expres-sion of activation markers on B and T cells in peripheral blood.Moreover, gene expression analyses of IL-21R–positive and IL-21R–negative CD4+ T cells isolated from mixed BM chimericBXSB.Yaa hosts clearly indicated that IL-21 signaling to CD4+

T cells broadly reduced the levels of transcripts of both immune-regulatory (Il2 and Il10) and -promoting (Il21, Ifng, and Tnfa)cytokines. Autocrine IL-21 signaling to CD4+ T cells may thusbroadly restrain the amount and types of cytokines they produceand thereby alter the balance of downstream immunologicalprocesses both positively and negatively. Further studies are re-quired to unravel the mechanisms by which IL-21 signaling toCD4+ T cells impacts the overall pathogenesis of this disease.Although genetic and functional studies support the involvement

of IL-21 signaling in the pathogenesis of multiple autoimmunediseases of humans and mice, our findings interject a note of cau-tion for designing therapeutic interruptions of this signaling path-way. Blockade may lead to unpredicted effects, inhibiting thedevelopment of autoreactive B cells and autoantibody-secretingplasma cells, while at the same time impairing the development ofnaturally occurring CD8+ Ts that limit disease progression andbroadly restraining the spectrum of cytokines produced by CD4+

T cells.

AcknowledgmentsWe thank Rosalinda Doty for histological interpretations.

DisclosuresThe authors have no financial conflicts of interest.

References1. Spolski, R., and W. J. Leonard. 2008. Interleukin-21: basic biology and impli-

cations for cancer and autoimmunity. Annu. Rev. Immunol. 26: 57–79.2. Ettinger, R., S. Kuchen, and P. E. Lipsky. 2008. The role of IL-21 in regulating

B-cell function in health and disease. Immunol. Rev. 223: 60–86.3. Odegard, J. M., B. R. Marks, L. D. DiPlacido, A. C. Poholek, D. H. Kono, C. Dong,

R. A. Flavell, and J. Craft. 2008. ICOS-dependent extrafollicular helper T cells elicitIgG production via IL-21 in systemic autoimmunity. J. Exp. Med. 205: 2873–2886.

4. Bubier, J. A., T. J. Sproule, O. Foreman, R. Spolski, D. J. Shaffer, H. C. Morse,III, W. J. Leonard, and D. C. Roopenian. 2009. A critical role for IL-21 receptorsignaling in the pathogenesis of systemic lupus erythematosus in BXSB-Yaamice. Proc. Natl. Acad. Sci. USA 106: 1518–1523.

5. Craft, J. E. 2012. Follicular helper T cells in immunity and systemic autoim-munity. Nat Rev Rheumatol 8: 337–347.

6. Li, J., H. F. Pan, H. Cen, J. Tian, Y. Ma, J. H. Tao, and D. Q. Ye. 2011.Interleukin-21 as a potential therapeutic target for systemic lupus erythematosus.Mol. Biol. Rep. 38: 4077–4081.

7. Nakou, M., E. D. Papadimitraki, A. Fanouriakis, G. K. Bertsias, C. Choulaki,N. Goulidaki, P. Sidiropoulos, and D. T. Boumpas. 2013. Interleukin-21 is in-creased in active systemic lupus erythematosus patients and contributes to thegeneration of plasma B cells. Clin. Exp. Rheumatol. 31: 172–179.

8. Yu, D., and C. G. Vinuesa. 2010. Multiple checkpoints keep follicular helperT cells under control to prevent autoimmunity. Cell. Mol. Immunol. 7: 198–203.

9. Sarra, M., and G. Monteleone. 2010. Interleukin-21: a new mediator of inflam-mation in systemic lupus erythematosus. J. Biomed. Biotechnol. 2010: 294582.

10. Terrier, B., N. Costedoat-Chalumeau, M. Garrido, G. Geri, M. Rosenzwajg,L. Musset, D. Klatzmann, D. Saadoun, and P. Cacoub. 2012. Interleukin 21correlates with T cell and B cell subset alterations in systemic lupus eryth-ematosus. J. Rheumatol. 39: 1819–1828.

11. Monteleone, G., F. Pallone, and T. T. Macdonald. 2009. Interleukin-21 as a newtherapeutic target for immune-mediated diseases. Trends Pharmacol. Sci. 30:441–447.

12. Nguyen, V., I. Luzina, H. Rus, C. Tegla, C. Chen, and V. Rus. 2012. IL-21promotes lupus-like disease in chronic graft-versus-host disease through bothCD4 T cell- and B cell-intrinsic mechanisms. J. Immunol. 189: 1081–1093.

13. Vogelzang, A., H. M. McGuire, D. Yu, J. Sprent, C. R. Mackay, and C. King.2008. A fundamental role for interleukin-21 in the generation of T follicularhelper cells. Immunity 29: 127–137.

14. Nurieva, R. I., Y. Chung, D. Hwang, X. O. Yang, H. S. Kang, L. Ma, Y. H. Wang,S. S. Watowich, A. M. Jetten, Q. Tian, and C. Dong. 2008. Generation of Tfollicular helper cells is mediated by interleukin-21 but independent of T helper1, 2, or 17 cell lineages. Immunity 29: 138–149.

15. Vinuesa, C. G., and J. G. Cyster. 2011. How T cells earn the follicular rite ofpassage. Immunity 35: 671–680.

16. Attridge, K., C. J. Wang, L. Wardzinski, R. Kenefeck, J. L. Chamberlain,C. Manzotti, M. Kopf, and L. S. Walker. 2012. IL-21 inhibits T cell IL-2 pro-duction and impairs Treg homeostasis. Blood 119: 4656–4664.

17. Ozaki, K., R. Spolski, R. Ettinger, H. P. Kim, G. Wang, C. F. Qi, P. Hwu,D. J. Shaffer, S. Akilesh, D. C. Roopenian, et al. 2004. Regulation of B celldifferentiation and plasma cell generation by IL-21, a novel inducer of Blimp-1and Bcl-6. J. Immunol. 173: 5361–5371.

18. Ettinger, R., G. P. Sims, A. M. Fairhurst, R. Robbins, Y. S. da Silva, R. Spolski,W. J. Leonard, and P. E. Lipsky. 2005. IL-21 induces differentiation of humannaive and memory B cells into antibody-secreting plasma cells. J. Immunol. 175:7867–7879.

19. Linterman, M. A., L. Beaton, D. Yu, R. R. Ramiscal, M. Srivastava, J. J. Hogan,N. K. Verma, M. J. Smyth, R. J. Rigby, and C. G. Vinuesa. 2010. IL-21 actsdirectly on B cells to regulate Bcl-6 expression and germinal center responses. J.Exp. Med. 207: 353–363.

20. Zotos, D., J. M. Coquet, Y. Zhang, A. Light, K. D’Costa, A. Kallies,L. M. Corcoran, D. I. Godfrey, K. M. Toellner, M. J. Smyth, et al. 2010. IL-21regulates germinal center B cell differentiation and proliferation through a Bcell-intrinsic mechanism. J. Exp. Med. 207: 365–378.

21. Lee, S. K., R. J. Rigby, D. Zotos, L. M. Tsai, S. Kawamoto, J. L. Marshall,R. R. Ramiscal, T. D. Chan, D. Gatto, R. Brink, et al. 2011. B cell priming forextrafollicular antibody responses requires Bcl-6 expression by T cells. J. Exp.Med. 208: 1377–1388.

22. Bessa, J., M. Kopf, and M. F. Bachmann. 2010. Cutting edge: IL-21 and TLRsignaling regulate germinal center responses in a B cell-intrinsic manner. J.Immunol. 184: 4615–4619.

23. Allard, E. L., M. P. Hardy, J. Leignadier, M. Marquis, J. Rooney, D. Lehoux, andN. Labrecque. 2007. Overexpression of IL-21 promotes massive CD8+ memoryT cell accumulation. Eur. J. Immunol. 37: 3069–3077.

24. Nguyen, H., and N. P. Weng. 2010. IL-21 preferentially enhances IL-15-mediated homeostatic proliferation of human CD28+ CD8 memory T cellsthroughout the adult age span. J. Leukoc. Biol. 87: 43–49.

25. Johnson, L. D., and S. C. Jameson. 2009. Immunology. A chronic need for IL-21.Science 324: 1525–1526.

26. Tagaya, Y. 2010. Editorial: memory CD8 T cells now join “Club 21”. J. Leukoc.Biol. 87: 13–15.

27. McPhee, C. G., T. J. Sproule, D. M. Shin, J. A. Bubier, W. H. Schott,M. P. Steinbuck, L. Avenesyan, H. C. Morse, III, and D. C. Roopenian. 2011.MHC class I family proteins retard systemic lupus erythematosus autoimmunityand B cell lymphomagenesis. J. Immunol. 187: 4695–4704.

28. Murphy, E. D., and J. B. Roths. 1979. A Y chromosome associated factor instrain BXSB producing accelerated autoimmunity and lymphoproliferation.Arthritis Rheum. 22: 1188–1194.

29. Izui, S., N. Ibnou-Zekri, L. Fossati-Jimack, and M. Iwamoto. 2000. Lessons fromBXSB and related mouse models. Int. Rev. Immunol. 19: 447–472.

30. Baccala, R., R. Gonzalez-Quintial, R. D. Schreiber, B. R. Lawson, D. H. Kono,and A. N. Theofilopoulos. 2012. Anti-IFN-a/b receptor antibody treatment ame-liorates disease in lupus-predisposed mice. J. Immunol. 189: 5976–5984.

31. Pisitkun, P., J. A. Deane, M. J. Difilippantonio, T. Tarasenko, A. B. Satterthwaite,and S. Bolland. 2006. Autoreactive B cell responses to RNA-related antigens dueto TLR7 gene duplication. Science 312: 1669–1672.

32. Subramanian, S., K. Tus, Q. Z. Li, A. Wang, X. H. Tian, J. Zhou, C. Liang,G. Bartov, L. D. McDaniel, X. J. Zhou, et al. 2006. A Tlr7 translocationaccelerates systemic autoimmunity in murine lupus. Proc. Natl. Acad. Sci. USA103: 9970–9975.

The Journal of Immunology 4587

by guest on February 3, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

33. Deane, J. A., P. Pisitkun, R. S. Barrett, L. Feigenbaum, T. Town, J. M. Ward,R. A. Flavell, and S. Bolland. 2007. Control of toll-like receptor 7 expression isessential to restrict autoimmunity and dendritic cell proliferation. Immunity 27:801–810.

34. Kim, H. J., X. Wang, S. Radfar, T. J. Sproule, D. C. Roopenian, and H. Cantor.2011. CD8+ T regulatory cells express the Ly49 Class I MHC receptor and aredefective in autoimmune prone B6-Yaa mice. Proc. Natl. Acad. Sci. USA 108:2010–2015.

35. Wei, L., A. Laurence, K. M. Elias, and J. J. O’Shea. 2007. IL-21 is produced byTh17 cells and drives IL-17 production in a STAT3-dependent manner. J. Biol.Chem. 282: 34605–34610.

36. Nurieva, R., X. O. Yang, G. Martinez, Y. Zhang, A. D. Panopoulos, L. Ma,K. Schluns, Q. Tian, S. S. Watowich, A. M. Jetten, and C. Dong. 2007. Essentialautocrine regulation by IL-21 in the generation of inflammatory T cells. Nature448: 480–483.

37. Bubier, J. A., S. M. Bennett, T. J. Sproule, B. L. Lyons, S. Olland, D. A. Young,and D. C. Roopenian. 2007. Treatment of BXSB-Yaa mice with IL-21R-Fcfusion protein minimally attenuates systemic lupus erythematosus. Ann. N. Y.Acad. Sci. 1110: 590–601.

38. Kitamura, D., J. Roes, R. Kuhn, and K. Rajewsky. 1991. A B cell-deficientmouse by targeted disruption of the membrane exon of the immunoglobulinmu chain gene. Nature 350: 423–426.

39. Lawson, B. R., S. I. Koundouris, M. Barnhouse, W. Dummer, R. Baccala,D. H. Kono, and A. N. Theofilopoulos. 2001. The role of alpha beta+ T cells andhomeostatic T cell proliferation in Y-chromosome-associated murine lupus. J.Immunol. 167: 2354–2360.

40. Roopenian, D. C., G. J. Christianson, T. J. Sproule, A. C. Brown, S. Akilesh,N. Jung, S. Petkova, L. Avanessian, E. Y. Choi, D. J. Shaffer, et al. 2003. TheMHC class I-like IgG receptor controls perinatal IgG transport, IgG homeostasis,and fate of IgG-Fc-coupled drugs. J. Immunol. 170: 3528–3533.

41. Christianson, G. J., R. L. Blankenburg, T. M. Duffy, D. Panka, J. B. Roths,A. Marshak-Rothstein, and D. C. Roopenian. 1996. beta2-microglobulin de-pendence of the lupus-like autoimmune syndrome of MRL-lpr mice. J. Immunol.156: 4932–4939.

42. Weening, J. J., V. D. D’Agati, M. M. Schwartz, S. V. Seshan, C. E. Alpers,G. B. Appel, J. E. Balow, J. A. Bruijn, T. Cook, F. Ferrario, et al. 2004. Theclassification of glomerulonephritis in systemic lupus erythematosus revisited.J. Am. Soc. Nephrol. 15: 241–250.

43. Akilesh, S., D. J. Shaffer, and D. Roopenian. 2003. Customized molecularphenotyping by quantitative gene expression and pattern recognition analysis.Genome Res. 13: 1719–1727.

44. Leonard, W. J., R. Zeng, and R. Spolski. 2008. Interleukin 21: a cytokine/cytokine receptor system that has come of age. J. Leukoc. Biol. 84: 348–356.

45. Lu, K. T., Y. Kanno, J. L. Cannons, R. Handon, P. Bible, A. G. Elkahloun,S. M. Anderson, L. Wei, H. Sun, J. J. O’Shea, and P. L. Schwartzberg. 2011.Functional and epigenetic studies reveal multistep differentiation and plasticityof in vitro-generated and in vivo-derived follicular T helper cells. Immunity 35:622–632.

46. Spolski, R., H. P. Kim,W. Zhu, D. E. Levy, andW. J. Leonard. 2009. IL-21mediatessuppressive effects via its induction of IL-10. J. Immunol. 182: 2859–2867.

47. Spolski, R., and W. J. Leonard. 2010. IL-21 is an immune activator that alsomediates suppression via IL-10. Crit. Rev. Immunol. 30: 559–570.

4588 IL-21, A DOUBLE-EDGED SWORD IN LUPUS

by guest on February 3, 2019http://w

ww

.jimm

unol.org/D

ownloaded from