of September 14, 2018.This information is current as

Transcription Factor T-bet CTL Activity via the+IL-21 Promotes CD8

Sue M. Liu, Vijay K. Kuchroo and Michael J. GrusbyAndrew P. R. Sutherland, Nicole Joller, Monia Michaud,

http://www.jimmunol.org/content/190/8/3977doi: 10.4049/jimmunol.1201730March 2013;

2013; 190:3977-3984; Prepublished online 11J Immunol 

MaterialSupplementary

0.DC1http://www.jimmunol.org/content/suppl/2013/03/11/jimmunol.120173

Referenceshttp://www.jimmunol.org/content/190/8/3977.full#ref-list-1

, 20 of which you can access for free at: cites 44 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. Immunologists, Inc. All rights reserved.Copyright © 2013 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on September 14, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on September 14, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

IL-21 Promotes CD8+ CTL Activity via the TranscriptionFactor T-bet

Andrew P. R. Sutherland,*,†,‡ Nicole Joller,*,† Monia Michaud,‡ Sue M. Liu,*,†,x

Vijay K. Kuchroo,*,† and Michael J. Grusby‡

CD8+ T cells are fundamental for immune-mediated clearance of viral infections and contribute to immune pathology in auto-

immune diseases such as type 1 diabetes. To execute these functions, CD8+ T cells must differentiate into CTLs, a process that

is precisely regulated by a variety of cytokines, costimulatory molecules, and transcription factors. IL-21 is an IL-2 family

cytokine and a growth factor for multiple lymphocyte effector lineages, including cytotoxic CD8+ T cells. Recent studies demon-

strate that loss of IL-21 signaling results in reduced viral clearance in models of lymphocytic choriomeningitis virus infection, and

also protection from type 1 diabetes in the NOD model. This is most likely the result of impaired CD8+ CTL function in the

absence of IL-21 signaling. Currently, the mechanisms by which IL-21 promotes CTL differentiation in CD8+ T cells remain

unclear, particularly the identity of the relevant transcription factor(s). We show that IL-21 promotes CTL function in vitro and

killing of pancreatic islets in vivo via the use of transgenic mice expressing IL-21 in pancreatic b cells. We demonstrate that IL-21

induces the expression of the transcription factor T-bet in CD8+ T cells, predominantly via STAT1, and that T-bet is required for

the induction of cytolytic molecules, including perforin and granzyme B in response to IL-21. Finally, we show that IL-21–induced

CTL function is T-bet dependent, as T-bet deficiency results in defective IL-21–dependent cytotoxicity in CD8+ T cells in vitro

and in vivo. Thus, IL-21 drives CD8+ CTL differentiation via the actions of the transcription factor T-bet. The Journal of

Immunology, 2013, 190: 3977–3984.

CD8+ T cells are a fundamental component of the cellularimmune system and promote the clearance of intracel-lular pathogens such as viruses and bacteria (1). The

absence of CD8+ T cells is associated with a variety of immunedeficiencies (2, 3), and, conversely, aberrant CD8+ T cell effectorfunction can cause autoimmune pathology, particularly in thecontext of diseases such as type 1 diabetes (T1D) (4).To elicit the killing of target cells, CD8+ T cells are required to

differentiate into CTLs, a process controlled in part by the actionsof costimulatory molecules and cytokines (5). Cytokines can beprovided by Th cells (e.g., IL-2) or activated APCs (e.g., IL-12,IL-27, type 1 IFN) and serve to promote CTL differentiation byinducing cytotoxic functions, effector cytokines, and survival mol-ecules (5, 6). These transcriptional programs are controlled bymultiple transcription factors, the best studied of which are T-bet,eomesodermin, and Blimp-1 (7-10). Both T-bet and eomeso-dermin are able to bind to the promoters of IFN-g, perforin, andgranzyme B and induce the expression of these genes, whereasloss-of-function experiments indicate that both are essential foroptimal CTL differentiation (11–14). T-bet expression levels are

positively correlated with the development of short-lived effectorcells during immune responses to lymphocytic choriomeningitis

virus (15), and T-bet–deficient mice display impaired CD8+ T cell

responses and reduced disease onset in a model of virally induced

T1D (16). Mice that are doubly deficient in T-bet and eomeso-

dermin have more severely impaired CTL function and viral

clearance postinfection compared with mice deficient in either

transcription factor alone (17), indicating that T-bet and eomeso-

dermin have partially redundant functions during the differentia-

tion of CTLs (18). The expression of T-bet and eomesodermin is

regulated by qualitatively distinct signals, with T-bet expression

being induced by inflammatory factors such as IL-12 and CpG,

whereas eomesodermin is induced by the Th cytokine IL-2 (19).IL-21 is an IL-2 family cytokine that is highly expressed by Th

cell lineages and signals via a heterodimeric receptor complex

comprised of the specific IL-21R subunit and the common

receptor g-chain (20). IL-21 stimulates the function of multiple

lymphocyte subsets, including Th17 cells, follicular Th cells,

B cells, NK cells, and CD8+ T cells (20). IL-21 promotes CD8+

T cell responses in the context of tumor immunity and is required

for the clearance of chronic viral infections in animal models,

suggesting that IL-21 plays a critical role in the regulation of

effector CTL responses (20–23). IL-21 is also required for de-

velopment of T1D, as IL-21R–deficient mice crossed onto the

NOD background are almost completely resistant to the devel-

opment of T1D; conversely, transgenic mice overexpressing IL-21

in pancreatic islets develop spontaneous diabetes on a resistant

genetic background (24–29). To date, the molecular mechanisms

by which IL-21 promotes the development of CD8+ CTL re-

sponses have remained unclear. Additionally, the identity of the

transcription factors that are required for IL-21–induced CTL

function has not been defined. Thus, the aim of our present study

was to define the mechanisms by which IL-21 drives effector

CD8+ T cell responses.

*Center for Neurologic Diseases, Brigham and Women’s Hospital and Harvard Med-ical School, Boston, MA 02115; †Harvard Medical School, Boston, MA 02115;‡Department of Immunology and Infectious Diseases, Harvard School of PublicHealth, Boston, MA 02115; and xDepartment of Immunology, Garvan Institute ofMedical Research, Sydney, New South Wales 2010, Australia

Received for publication June 21, 2012. Accepted for publication February 13, 2013.

A.P.R.S. is the recipient of a Juvenile Diabetes Research Foundation postdoctoralfellowship. N.J. is supported by the Swiss Multiple Sclerosis Society.

Address correspondence and reprint requests to Dr. Andrew P.R. Sutherland at thecurrent address: St. Vincent’s Institute of Medical Research, 9 Princes Street, Fitzroy,VIC 3065, Australia. E-mail address: asutherland@svi.edu.au

The online version of this article contains supplemental material.

Abbreviations used in this article: 7-AAD, 7-aminoactinomycin D; T1D, type 1 diabetes.

Copyright� 2013 by The American Association of Immunologists, Inc. 0022-1767/13/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1201730

by guest on September 14, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

Materials and MethodsMice

All mice were housed in microisolator cages under specific pathogen-freeconditions at the Harvard Institutes of Medicine, and all animal studies wereperformed according to institutional and National Institutes of Healthguidelines for animal use and care. In accordance with these guidelines,blood glucose levels were monitored weekly using a handheld AscensiaContour glucometer (Bayer). Diabetic incidence was calculated per groupas two consecutive readings .250 mg/dl, and displayed as percentageof diabetic mice per group. IL-21R–deficient mice were generated as de-scribed (30). IL-21 transgenic mice were generated as described (27).STAT4-deficient mice were generated as described (31). C57BL/6J, OT-I(C57BL/6-Tg(TcraTcrb)1100Mjb/J), RIP-OVAlow (C57BL/6-Tg(Ins2-OVA)307Wehi/WehiJ), RIP-mOVA (C57BL/6-Tg(Ins2-TFRC/OVA)296Wehi/WehiJ), and IFN-g–deficient (B6.129S7-Ifngtm1Ts/J) mice were purchasedfrom The Jackson Laboratory (Bar Harbor, MA). The 129S6/SvEvTac,STAT1 deficient (129S6/SvEv-Stat1tm1Rds), and RAG2 deficient (B6.129S6-Rag2tm1Fwa N12) were purchased from Taconic. T-bet–deficient mice(B6.129S6-Tbx21tm1Glm/J) were provided by A. Lichtmann (Brigham andWomen’s Hospital).

Lymphocyte preparation, isolation, and activation

Single-cell suspensions were prepared from spleen and peripheral lymphnodes by mechanical disruption. Cells were filtered through a 70-mm cellstrainer (BD Biosciences, San Jose, CA), and then subjected to erythrocytelysis using ACK buffer (0.15 M NH4Cl, 10 mM KHCO3, 0.1 mM Na2EDTA[pH 7.2–7.4]). CD8+ T cells were isolated using magnetic separation(Miltenyi Biotec) and cultured at 37˚C in 10% CO2 in DMEM supplementedwith 10% heat-inactivated FCS, sodium pyruvate, L-glutamine, penicillin/streptomycin, nonessential amino acids, arginine/asparagine, folic acid,vitamins, and 2-ME. CD8+ T cells were activated at 1 3 106 cells/well in24-well plates previously coated with anti-CD3 (clone 145-2C11) and anti-CD28 (clone PV-1) Abs at 2 mg/ml. IL-12, IL-21, and IL-27 were used at25 ng/ml and purchased from R&D Systems. Lymphocytes were isolatedfrom pancreatic tissue by fine dissection with razor blades, followed bycollagenase digestion in HBSS for 20 min at 37˚C.

RNA extraction, cDNA synthesis, and quantitative RT-PCR

Total RNA was isolated with Qiagen RNeasy Plus Minikit. A quantityamounting to 1 mg total RNAwas used to prepare cDNA using the iScriptcDNA synthesis kit (Bio-Rad). Quantitative RT-PCR was performed usingTaqMan probes and the 7500 Fast Real-Time PCR System (AppliedBiosystems). All samples were normalized to b-actin internal control.

CFSE labeling and cell transfers

CD8+ T cells were resuspended at 2 3 107/ml in 0.1% BSA in PBS. CFSE(Invitrogen) was added to a final concentration of 10 mM, and cells wereincubated at 37˚C for 15 min. Cells were washed once with 10% FCS inPBS, twice with 0.1% BSA in PBS, and then resuspended in PBS fortransfer to recipient animals. A total of 5 3 106 OT-1 cells was transferredto IL-21Tg 3 RIP-OVAlow double-transgenic or control mice for CFSEproliferation and diabetes incidence experiments. A total of 1 3 106 ac-tivated OT-1 cells was transferred to RIP-mOVA mice for diabetes inci-dence studies.

Flow cytometry

Cells were harvested and resuspended in PBS containing 0.5% BSA and0.01% sodium azide (FACS buffer). A total of 1 3 105 cells was used perstain, with 1 mg/ml test mAb, or control mAb. Primary mAb staining wasperformed for 25 min at 4˚C, after which cells were washed twice withFACS buffer. Secondary staining was performed after resuspension in100 ml containing appropriate secondary staining reagents. After incubatingfor 15 min at 4˚C, cells were washed twice in FACS buffer and analyzedusing a BD FACSCalibur flow cytometer (BD Biosciences). Intracellularstaining was performed using BD Cytofix/Cytoperm kit. FITC-, PE-, PerCP-,and allophycocyanin- conjugated mAbs to CD3, CD4, CD8, B220, NK1.1,and T-bet were obtained from BioLegend (San Diego, CA) and usedaccording to the manufacturer’s instructions. mAbs to pSTAT1, pSTAT3,pSTAT4, and pSTAT5 were obtained from BD Biosciences.

Antiphospho-STAT staining

Cells were harvested and fixed in 4% paraformaldehyde for 12 min at 37˚C,rinsed with PBS, and fixed in 90% methanol solution on ice for 30 min.Cells were washed twice with Perm buffer (eBiosciences) and then blocked

in Perm buffer containing 2% FBS and Fc block (1:100) for 10 min atroom temperature. Ab staining was performed in Perm buffer for 45 min atroom temperature, and cells were then washed three times with PermBuffer and resuspended in FACS buffer for flow cytometry.

In vitro CTL assays

OVA-expressing (EG7) and a nontransfected control cell line (EL4) wereseeded at 1 3 105 cells/well into 96-well plates. CTLs were added at arange of concentrations and incubated with target cells overnight, afterwhich cells were stained with PE-conjugated anti-CD8 mAb and 7-ami-noactinomycin D (7-AAD) in FACS buffer and dead cells identified by 7-AAD positivity.

Statistical analysis

Statistical significance was determined for in vivo studies using one-wayANOVA analysis, log rank test, and Bonferroni’s multiple comparisonposttest using t test. Statistical significance for other studies was deter-mined using Student t test with relation to comparison data. Statisticalsignificance values are indicated as follows: *p , 0.05, **p , 0.01, and***p , 0.005.

ResultsIL-21 has previously been shown to promote the differentiation ofCD8+ CTLs in vitro and in vivo (20). To confirm these data, weisolated CD8+ T cells from wild-type mice and performed in vitroactivations with anti-CD3/CD28 mAbs in the presence of IL-21 orcontrol cytokines IL-27 and IL-12 (both of which induce CTLdifferentiation) for 72 h. Cells were harvested and mRNA levelswere measured by quantitative RT-PCR. IL-21 induced the ex-pression of cytotoxic molecules, including perforin (Fig. 1A),granzyme B (Fig. 1B), and granzyme A (Fig. 1C). To determinewhether IL-21 stimulation also resulted in increased CTL func-tion, CD8+ T cells were isolated from OT-I mice and activatedin vitro with anti-CD3/CD28 mAbs for 5 d in the presence of IL-21 or control cytokines IL-27 and IL-12. Cells were harvested, andin vitro CTL activity was quantified against an OVA-expressingcell line, which demonstrated that IL-21 increased the cytolyticactivity of the cultured CD8+ T cells (Fig. 1D).We have previously reported that transgenic mice expressing IL-

21 in pancreatic b-cells develop spontaneous T1D (27); however,the cellular mechanisms responsible for this phenotype have notbeen defined. Given that IL-21 can promote the development ofCTL function, we hypothesized that CD8+ T cells could play animportant role in the T1D in this model. To address this issue, IL-21 transgenic mice were crossed to the RAG2-deficient back-ground to determine whether T or B cells were required to initiateautoimmunity in IL-21 transgenic mice. RAG2-deficient micewere completely protected from the development of T1D in IL-21transgenic mice, clearly demonstrating the dependence on adap-tive immunity (Fig. 1E). The pancreatic infiltrates from predia-betic IL-21 transgenic mice were analyzed to identify cells thatmay be involved in the induction of diabetes. Most notably, weobserved an increased abundance of CD8+ T cells and B cells inthe pancreas of transgenic mice compared with littermate controls(Fig. 1F), with smaller increases in CD4+ T cells and NK cells alsoobserved. CD8+ T cells were isolated from spleen and lymphnodes, and the expression of perforin, granzyme A, granzyme B,and IFN-g was quantified by RT-PCR analysis. This revealed in-creased expression of these effector molecules in IL-21 transgenicmice, suggesting increased CTL differentiation (Fig. 1G). To testwhether diabetogenic CTL activation and differentiation were en-hanced by overexpression of IL-21, IL-21 transgenic mice werecrossed to RIP-OVAlow mice to generate double-transgenic mice(IL-21Tg 3 RIP-OVAlow) and relevant controls. OVA-specificTCR transgenic OT-I T cells were purified and labeled withCFSE, and 5 3 106 unactivated cells were transferred i.p. intodouble-transgenic recipient mice and controls. Mice were sacri-

3978 IL-21–INDUCED CTL DIFFERENTIATION IS T-bet DEPENDENT

by guest on September 14, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

ficed 10 d later, lymphoid organs were isolated, and division oftransferred cells was quantified by CFSE dye dilution (Fig. 1H).This demonstrated activation and proliferation of OT-I cells indouble-transgenic mice, and, as expected, no proliferation wasobserved in single-transgenic controls. In addition, transfer ofOT-I cells into double- but not single-transgenic recipient miceresulted in the onset of T1D from 10 d posttransfer (Fig. 1I anddata not shown). In contrast, IL-21R–deficient OT-I cells were notable to induce T1D upon transfer to double-transgenic recipientmice (Fig. 1I), demonstrating that CD8+ T cells require directstimulation by IL-21 to induce T1D in this system. Together, thesedata confirm previous findings that IL-21 can promote the ex-pression of cytolytic molecules and CTL differentiation in CD8+

T cells and demonstrate the IL-21 transgenic mice have enhancedCTL responses, resulting in destruction of pancreatic islets andonset of T1D.IL-21 is therefore a potent stimulator of CTL differentiation;

however, the underlying transcriptional events mediating this

process remain unclear. The transcription factors T-bet andeomesodermin are central regulators of CD8+ CTL differentiation,which coordinate the expression of important effector molecules,including perforin, granzyme B, and IFN-g, and are required foroptimal CTL function (7, 12, 19). We therefore hypothesized thatIL-21 may induce CTL function in CD8+ T cells via induction ofT-bet and/or eomesodermin. To test this hypothesis, wild-typeCD8+ T cells were activated in vitro with anti-CD3/CD28 alone orin the presence of IL-21 (and IL-27 or IL-12 as positive controls),and the expression of T-bet and eomesodermin was analyzed overa 72-h time course. These experiments did not reveal any signif-icant regulation of T-bet by IL-21 during this time period (Fig.2A); in contrast, IL-21 significantly suppressed the expression ofeomesodermin (Fig. 2B). To determine whether IL-21 was exert-ing positive effects at an earlier time point, we repeated the pre-vious experiments over a 24-h time course. These experimentsrevealed that IL-21 induced a transient 70% increase in T-betmRNA compared with controls at 4 h poststimulation (Fig. 2C);

FIGURE 1. IL-21 promotes CD8+ CTL function

in vitro and in vivo. CD8+ T cells were isolated from

spleen and lymph nodes of C57BL/6 mice via mag-

netic separation, activated in the presence of IL-21,

control cytokines (IL-27 and IL-12), or no cytokine

(control), and quantitative RT-PCR was performed for

(A) perforin, (B) granzyme B, and (C) granzyme A

(n = 6 for all groups). (D) CD8+ T cells were isolated

from OT-I mice and activated for 5 d in the presence of

IL-21, IL-27, IL-12, or no cytokine (control). CTLs

were incubated overnight with OVA-expressing cell

line (EG7), and cell death was measured by 7-AAD

incorporation. A representative experiment is shown.

(E) IL-21 transgenic mice were crossed to the RAG2-

deficient background, and blood glucose levels were

monitored weekly. Diabetes incidence was scored as

survival curve data and is different by log rank test

(+/+, n = 16; +/2, n = 16; 2/2, n = 7). (F) Pancreatic

infiltrating lymphocytes were isolated by collagenase

digestion of pancreata and enumerated by flow cy-

tometry (n = 12 per group). (G) CD8+ T cells were

isolated from spleen and lymph nodes via magnetic

separation, and transcript levels for the indicated genes

were measured by quantitative RT-PCR (n = 3 per

group). (H) CD8+ T cells were isolated from spleen

and lymph nodes of OT-I mice via magnetic separation

and labeled with CFSE, and 5 3 106 cells were

transferred i.p. into IL-21Tg 3 RIP-OVAlow double-

transgenic recipients and relevant controls. Lymphoid

organs were isolated 10 d later and analyzed by flow

cytometry. Numerical value represents divided cells as

a percentage of total cell number (a representative plot

is shown, n . 5). (I) CD8+ T cells were isolated from

OT-I and IL-21R–deficient OT-I mice, and 5 3 106

cells were transferred i.p. into IL-21Tg3 RIP-OVAlow

double-transgenic recipients. Blood glucose levels

were measured daily. Diabetes incidence was scored

as survival curve data and is different by log rank test

(n = 10 for both groups). *p , 0.05, ***p , 0.005.

The Journal of Immunology 3979

by guest on September 14, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

suppression of eomesodermin by IL-21 was again observed duringthis time course (Fig. 2D). We then analyzed T-bet protein ex-pression by flow cytometry, which demonstrated that IL-21 in-duced a 30% increase in T-bet protein expression at 8 h (Fig. 2E).CD8+ T cells produce high levels of IFN-g in response to TCRactivation and costimulation and, because IFN-g is a potent driverof T-bet expression (32), we reasoned that the true capacity for IL-21–dependent induction of T-bet may be obscured in our in vitroculture system by excessive IFN-g production. To test this hy-pothesis, we measured T-bet expression in response to IL-21stimulation in IFN-g knockout mice. We observed a 6-fold in-duction of T-bet mRNA at 4 h (Fig. 2F), indicating that themagnitude of IL-21–dependent T-bet induction was drasticallyincreased in the context of IFN-g deficiency. Analysis of T-betprotein expression by flow cytometry demonstrated that the mag-nitude of T-bet protein induction was enhanced in the absence ofIFN-g, with a 2-fold increase in T-bet protein observed at 8 h (Fig.2G, 2H). Similar results were observed when wild-type CD8+

T cells were cultured in the presence of IL-21 and IFN-g blockingAbs (data not shown). Thus, IL-21 can rapidly induce T-bet ex-pression during the activation of CD8+ T cells, and T-bet inductionis even more pronounced in the absence of IFN-g.IL-21 mediates many of its effects via the function of STAT

family transcription factors (20); thus, we next determined whichSTAT molecules were required for induction of T-bet expressionby IL-21. CD8+ T cells were activated in vitro in the presence ofIL-21, IL-27 as a positive control, or no cytokine (control). Cellswere harvested at the indicated time points, and levels of STAT

phosphorylation were quantified by intracellular staining. Theseexperiments demonstrated that IL-21 induced the phosphorylationof STAT1, 3, and 5, as previously described (20), but unexpectedlyalso STAT4 (Fig. 3A). The specificity of pSTAT4 detection in thisassay system was confirmed by the complete lack of signal ob-served in STAT4 knockout mice (Supplemental Fig. 1A). Becauseboth STAT1 and STAT4 are able to induce the expression of T-bet(7, 33), we next determined whether STAT1 and/or STAT4 wererequired for IL-21–dependent induction of T-bet. CD8+ T cellsfrom STAT1-deficient mice and wild-type controls were acti-vated in vitro in the presence of IL-21 or no cytokine (control) andT-bet mRNA and protein levels quantified at the indicated timepoints. STAT1 knockout CD8+ T cells exhibited a 50% decrease inIL-21–induced T-bet mRNA at 4 h (Fig. 3B) and a corresponding35% reduction in T-bet protein level at 8 h poststimulation (Fig.3C). Similarly, CD8+ T cells from STAT4-deficient mice and wild-type controls were activated in vitro in the presence of IL-21 or nocytokine (control). These data showed a 20% decrease in IL-21–induced T-bet mRNA at 4 h (Fig. 3D) and a corresponding 20%reduction in T-bet protein level at 8 h poststimulation in STAT4knockout CD8+ T cells (Fig. 3E), neither of which reached sta-tistical significance. However, IL-21–induced IFN-g productionwas STAT4 dependent over the same time course, indicating thatIL-21 can induce functional effects via activation of STAT4 in thissystem (Supplemental Fig. 1B). Together, these data indicate thatIL-21 induction of T-bet is predominantly STAT1 dependent.IL-21 induces the expression of cytotoxic molecules such as

perforin and granyzme B (20), which are targets of T-bet and are

FIGURE 2. IL-21 induces the transcription factor T-

bet in CD8+ T cells. CD8+ T cells were isolated from

spleen and lymph nodes of C57BL/6 mice via magnetic

separation and activated with plate-bound anti-CD3/

CD28 in the presence of IL-21, IL-27, or no cytokine.

Cells were harvested over a 72-h time course, and

quantitative RT-PCR was performed for (A) T-bet and

(B) eomesodermin (n = 3 per group). The experiment

was then performed over a 24-h time course, and

quantitative RT-PCR was performed for (C) T-bet and

(D) eomesodermin (n = 4 per group). (E) T-bet protein

expression was measured by intracellular staining and

flow cytometry at 8 h (n = 4 for all groups). CD8+

T cells were isolated from C57BL/6 and IFN-g–defi-

cient mice and activated, and T-bet expression was

determined over a 24-h time course by (F) quantitative

RT-PCR or (G) intracellular staining and flow cytom-

etry. A representative histogram is shown at 8 h (black

line = control, gray line = IL-21 stimulation), and (H)

T-bet induction at 8 h was compared between IFN-g–

deficient mice and wild-type controls (n = 2 for all

groups). *p , 0.05, **p , 0.01.

3980 IL-21–INDUCED CTL DIFFERENTIATION IS T-bet DEPENDENT

by guest on September 14, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

upregulated during CTL differentiation. Thus, we tested whetherthe IL-21–induced expression of T-bet target genes, such as per-forin and granzyme B, was in fact dependent on T-bet. We stim-ulated CD8+ T cells isolated from T-bet–deficient mice and wild-type controls with IL-21 for 72 h and assayed gene induction byRT-PCR. This demonstrated that IL-21–induced expression ofHlx-1 (Fig. 4A), IL-12Rb2 (Fig. 4B), perforin (Fig. 4C), andgranzyme B (Fig. 4D) was dependent on T-bet (7, 32, 34). Inter-estingly, similar levels of granzyme A expression were observedin T-bet knockout mice and wild-type controls after IL-21 stim-ulation (Fig. 4E), indicating that IL-21 also regulates T-bet–in-dependent transcriptional circuits that are required for optimalcytolytic molecule expression.We then tested whether IL-21–induced CTL function required

T-bet. To this end, OT-I mice were crossed to the T-bet–deficientbackground, activated in vitro with anti-CD3/CD28 for 5 d in thepresence of IL-21 or control cytokines IL-27 and IL-12 (both ofwhich induce T-bet–dependent CTL differentiation). Cells wereharvested, and in vitro CTL activity was quantified against anOVA-expressing cell line. IL-21 enhanced in vitro CTL activity inwild-type OT-I, but this effect was completely ablated in the ab-sence of T-bet (Fig. 5A). In keeping with previous studies, CTLactivity induced by IL-27 and IL-12 was also dependent on T-bet(Supplemental Fig. 2A). Quantification of perforin, granzyme B,and granzyme A expression after 5 d of culture showed that up-regulation of perforin and granzyme B by IL-21 was T-bet de-pendent (Supplemental Fig. 2B–D), as observed at earlier time

points (Fig. 4). To test whether T-bet was required for IL-21–in-duced cytotoxicity in vivo, the cultured wild-type and T-bet–de-ficient OT-1 cells were transferred to RIP-mOVA recipient mice,and blood glucose levels were monitored to determine diabetesonset. IL-21–stimulated OT-I cells were able to induce diabetesafter transfer into R-IP-mOVA recipient mice (Fig. 5B); however,in contrast, IL-21–stimulated T-bet–deficient OT-I cells were un-able to induce diabetes (Fig. 5B). Wild-type and T-bet–deficientOT-I cells stimulated with IL-27 or IL-12 demonstrated a similarpattern of T-bet dependency. To corroborate this finding, we useda second model, in which unactivated T-bet–deficient OT-I cells orwild-type OT-I controls were transferred to IL-21Tg 3 RIP-OVAlow double-transgenic recipient mice. In contrast to OT-Icontrols, recipients of T-bet–deficient OT-I cells were almostcompletely protected from the onset of T1D (Fig. 5C). As thismodel is completely dependent on IL-21R signaling in transferredCD8+ T cells (as shown in Fig. 1I), these data indicate that T-bet isrequired for IL-21–induced CTL activity in vivo. Thus, our datademonstrate that IL-21–induced expression of cytotoxic mole-cules and in vitro and in vivo CTL function is dependent on thetranscription factor T-bet.

DiscussionCommon g-chain cytokines can regulate both effector and mem-ory differentiation in CD8+ T cells. Previous studies demonstratethat IL-2 is a differentiation factor for effector CD8+ T cells (19),whereas IL-7 and IL-15 play critical roles in the generation and

FIGURE 3. IL-21 induces T-bet expression pre-

dominantly via STAT1. (A) CD8+ T cells were iso-

lated from spleen and lymph nodes of 129S6/SvEvTac

mice via magnetic separation and activated with plate-

bound anti-CD3/CD28 in the presence of IL-21, IL-

27, or no cytokine (control). Cells were harvested at

the indicated time points, and intracellular staining

and flow cytometry were performed with a panel of

phospho-STAT Abs (n = 3 for all groups). CD8+

T cells were isolated from 129S6/SvEvTac and

STAT1-deficient mice and activated, and T-bet ex-

pression was quantified by (B) quantitative RT-PCR

and (C) intracellular protein staining and flow cytom-

etry at 8 h (n = 2 for all groups). Similarly, CD8+

T cells were isolated from C57BL/6 and STAT4-de-

ficient mice and activated, and T-bet expression was

quantified by (D) quantitative RT-PCR and (E) intra-

cellular protein staining and flow cytometry at 8 h (n =

2 for all groups). *p , 0.05.

The Journal of Immunology 3981

by guest on September 14, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

maintenance of memory CD8+ T cells (35). IL-21 stimulationpromotes the development of both effector and memory charac-teristics in CD8+ T cells, inducing CTL differentiation in someexperimental systems (20, 36, 37), while promoting the acquisi-tion of a memory phenotype in others (38–40). Our present studiesfurther support the notion that IL-21 is a differentiation factor foreffector CD8+ T cells. We demonstrate that IL-21 promotes thedevelopment of CD8+ CTLs both in vitro and in vivo by inducingthe transcription factor T-bet, which is critical for the developmentof effector CD8+ T cells (7). Thus, IL-21 can be grouped witha panel of other cytokines, such as IL-27 and IL-12, which induceeffector function in CD8+ T cells via T-bet. Whereas these cyto-kines all induce T-bet expression, there are qualitative differencesbetween their functions. We suggest that these differences couldbe ascribed in part to their differing kinetics of T-bet induction.IL-21 induces a more rapid and transient induction of T-betcompared with that observed with IL-27, and particularly IL-12,

which does not show significant induction until at least 48 h afteractivation. This rapid and transient induction of T-bet by IL-21results in the upregulation of T-bet–dependent genes required forCTL differentiation, such as perforin and granzyme B; however,the lack of sustained T-bet expression may be an important pre-requisite for the subsequent development of a memory phenotypein response to IL-21 in some experimental systems (38–40). Pre-vious studies demonstrate that CD8+ T cells lacking eomesoderminhave impaired memory CD8+ T cell development, indicating thateomesodermin is important for the differentiation of memoryCD8+ T cells (18). However, our experiments, and those of othergroups, indicate that IL-21 is a potent inhibitor of eomesodermin(39, 41), thus suggesting that IL-21 would prevent memory CD8+

T cell development. It should be noted that these studies were per-formed at relatively early time points (3–5 d postactivation), and it ispossible that IL-21 suppression of eomesodermin is subsequentlyrelieved, which could allow memory CD8+ T cell development toproceed in response to IL-21 stimulation.One mode of signaling resulting from the engagement of the IL-

21R is the activation of the JAK/STAT pathway, particularly JAK1and JAK3 and STAT1, STAT3, and STAT5 (20, 42, 43). Our studieshave also identified a previously unrecognized potential for IL-21to induce STAT4 phosphorylation and STAT4-dependent gene

FIGURE 4. IL-21 induces T-bet–dependent gene expression. CD8+

T cells were isolated from spleen and lymph nodes of C57BL/6 and T-bet–

deficient mice via magnetic separation and activated in the presence of IL-

21 or no cytokine (control), and quantitative RT-PCR was performed for

(A) Hlx-1, (B) IL-12Rb2, (C) perforin, (D) granzyme B, and (E) granzyme

A (n = 2 for all groups). *p , 0.05.

FIGURE 5. IL-21–induced CD8+ CTL activity is T-bet dependent.

CD8+ T cells were isolated from OT-I and T-bet–deficient OT-I mice and

activated for 5 d in the presence of IL-21, IL-27, IL-12, or no cytokine

(control). (A) CTLs were incubated overnight with OVA-expressing cell

line (EG7), and cell death was measured by 7-AAD incorporation. A

representative experiment is shown. (B) A total of 1 3 106 cells was

transferred i.p. into RIP-mOVA recipients, and blood glucose levels were

monitored daily (n = 5 for IL-21 from two independent experiments, n = 2

for all other groups). (C) Unactivated CD8+ T cells were isolated from OT-I

and T-bet–deficient OT-I mice via magnetic separation; 5 3 106 cells were

transferred i.p. into IL-21Tg 3 RIP-OVAlow double-transgenic recipients;

and blood glucose levels were monitored daily (OT-I, n = 5; T-bet–defi-

cient OT-I, n = 9 from two independent experiments). *p , 0.05, **p ,0.01.

3982 IL-21–INDUCED CTL DIFFERENTIATION IS T-bet DEPENDENT

by guest on September 14, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

expression in CD8+ T cells. Our experiments demonstrate that IL-21 can induce STAT4-dependent IFN-g expression, indicating thatthe activation of STAT4 by IL-21 has functional effects. In con-trast, STAT4 was not required for the optimal induction of T-betexpression in response to IL-21. This is in keeping with previousstudies demonstrating that IL-21–induced CTL function in CD8+

T cells is STAT4 independent (37), and, along with experimentsperformed in STAT1-deficient mice, indicates that STAT1 is theprimary inducer of T-bet in response to IL-21. The exact mech-anisms of STAT4 activation and the relative importance of STAT4-mediated gene induction in response to IL-21 remain to be de-termined.Recent studies indicate that IL-21R–deficient mice fail to clear

chronic viral infections (21–23) and that IL-21R NOD mice areprotected from T1D (24–28), most likely via impaired CD8+ CTLfunction (26, 29, 44). Our data suggest that this generalized im-pairment of CTL function in the absence of IL-21 signaling mayresult from impaired induction of T-bet and reduced expression ofT-bet–dependent transcriptional programs that promote CTLfunction. In keeping with this hypothesis, T-bet–deficient micehave attenuated CD8+ T cell responses and are protected from thedevelopment of T1D in a model of virally induced T1D and on theNOD background (16, 45). IL-21 is located in the Idd3 disease-susceptibility locus, and protection conferred by this congenicinterval is correlated with reduced expression of IL-21 (24). Wewould hypothesize that this may cause NOD.Idd3 mice to haveimpairments in CTL responses due to reduced IL-21–dependentexpression of T-bet, and that this may account in part for theprotection from diabetes that is observed in NOD.Idd3 mice.In conclusion, we demonstrate that IL-21 promotes CTL func-

tion in vitro and in vivo, as overexpression of IL-21 in pancreaticb cells leads to CD8+ CTL differentiation and T1D. We showthat IL-21 induces the expression of the transcription factor T-betin CD8+ T cells and the expression of cytolytic molecules, and thefunctional differentiation of CTLs in response to IL-21 is T-betdependent. These data illuminate a mechanistic basis for thehelper effects of this cytokine during the development of effectorCD8+ T cell responses, coupling IL-21 production by CD4+ Thcells to the induction of the gene program associated with CD8+

CTL differentiation via the action of T-bet.

AcknowledgmentsWe thank Maggie Tarrio and Andrew Lichtman for provision of T-bet–

deficient mice and Jenna Sullivan for technical support.

DisclosuresThe authors have no financial conflicts of interest.

References1. Wong, P., and E. G. Pamer. 2003. CD8 T cell responses to infectious pathogens.

Annu. Rev. Immunol. 21: 29–70.2. Buckley, R. H. 2004. Molecular defects in human severe combined immuno-

deficiency and approaches to immune reconstitution. Annu. Rev. Immunol. 22:625–655.

3. Sugamura, K., H. Asao, M. Kondo, N. Tanaka, N. Ishii, K. Ohbo, M. Nakamura,and T. Takeshita. 1996. The interleukin-2 receptor gamma chain: its role in themultiple cytokine receptor complexes and T cell development in XSCID. Annu.Rev. Immunol. 14: 179–205.

4. Liblau, R. S., F. S. Wong, L. T. Mars, and P. Santamaria. 2002. Autoreactive CD8T cells in organ-specific autoimmunity: emerging targets for therapeutic inter-vention. Immunity 17: 1–6.

5. Williams, M. A., and M. J. Bevan. 2007. Effector and memory CTL differen-tiation. Annu. Rev. Immunol. 25: 171–192.

6. Zhang, N., and M. J. Bevan. 2011. CD8(+) T cells: foot soldiers of the immunesystem. Immunity 35: 161–168.

7. Glimcher, L. H., M. J. Townsend, B. M. Sullivan, and G. M. Lord. 2004. Recentdevelopments in the transcriptional regulation of cytolytic effector cells. Nat.Rev. Immunol. 4: 900–911.

8. Kallies, A., A. Xin, G. T. Belz, and S. L. Nutt. 2009. Blimp-1 transcription factoris required for the differentiation of effector CD8(+) T cells and memoryresponses. Immunity 31: 283–295.

9. Rutishauser, R. L., G. A. Martins, S. Kalachikov, A. Chandele, I. A. Parish,E. Meffre, J. Jacob, K. Calame, and S. M. Kaech. 2009. Transcriptional repressorBlimp-1 promotes CD8(+) T cell terminal differentiation and represses the ac-quisition of central memory T cell properties. Immunity 31: 296–308.

10. Cox, M. A., and A. J. Zajac. 2010. Shaping successful and unsuccessful CD8T cell responses following infection. J. Biomed. Biotechnol. 2010. DOI: 10.1155/2010/159152.

11. Szabo, S. J., B. M. Sullivan, C. Stemmann, A. R. Satoskar, B. P. Sleckman, andL. H. Glimcher. 2002. Distinct effects of T-bet in TH1 lineage commitment andIFN-gamma production in CD4 and CD8 T cells. Science 295: 338–342.

12. Pearce, E. L., A. C. Mullen, G. A. Martins, C. M. Krawczyk, A. S. Hutchins,V. P. Zediak, M. Banica, C. B. DiCioccio, D. A. Gross, C. A. Mao, et al. 2003.Control of effector CD8+ T cell function by the transcription factor eomeso-dermin. Science 302: 1041–1043.

13. Sullivan, B. M., A. Juedes, S. J. Szabo, M. von Herrath, and L. H. Glimcher.2003. Antigen-driven effector CD8 T cell function regulated by T-bet. Proc.Natl. Acad. Sci. USA 100: 15818–15823.

14. Townsend, M. J., A. S. Weinmann, J. L. Matsuda, R. Salomon, P. J. Farnham,C. A. Biron, L. Gapin, and L. H. Glimcher. 2004. T-bet regulates the terminalmaturation and homeostasis of NK and Valpha14i NKT cells. Immunity 20: 477–494.

15. Joshi, N. S., W. Cui, A. Chandele, H. K. Lee, D. R. Urso, J. Hagman, L. Gapin,and S. M. Kaech. 2007. Inflammation directs memory precursor and short-livedeffector CD8(+) T cell fates via the graded expression of T-bet transcriptionfactor. Immunity 27: 281–295.

16. Juedes, A. E., E. Rodrigo, L. Togher, L. H. Glimcher, and M. G. von Herrath.2004. T-bet controls autoaggressive CD8 lymphocyte responses in type 1 dia-betes. J. Exp. Med. 199: 1153–1162.

17. Intlekofer, A. M., A. Banerjee, N. Takemoto, S. M. Gordon, C. S. Dejong,H. Shin, C. A. Hunter, E. J. Wherry, T. Lindsten, and S. L. Reiner. 2008.Anomalous type 17 response to viral infection by CD8+ T cells lacking T-betand eomesodermin. Science 321: 408–411.

18. Banerjee, A., S. M. Gordon, A. M. Intlekofer, M. A. Paley, E. C. Mooney,T. Lindsten, E. J. Wherry, and S. L. Reiner. 2010. Cutting edge: the transcriptionfactor eomesodermin enables CD8+ T cells to compete for the memory cellniche. J. Immunol. 185: 4988–4992.

19. Pipkin, M. E., J. A. Sacks, F. Cruz-Guilloty, M. G. Lichtenheld, M. J. Bevan, andA. Rao. 2010. Interleukin-2 and inflammation induce distinct transcriptionalprograms that promote the differentiation of effector cytolytic T cells. Immunity32: 79–90.

20. Spolski, R., and W. J. Leonard. 2008. Interleukin-21: basic biology and impli-cations for cancer and autoimmunity. Annu. Rev. Immunol. 26: 57–79.

21. Frohlich, A., J. Kisielow, I. Schmitz, S. Freigang, A. T. Shamshiev, J. Weber,B. J. Marsland, A. Oxenius, and M. Kopf. 2009. IL-21R on T cells is critical forsustained functionality and control of chronic viral infection. Science 324: 1576–1580.

22. Yi, J. S., M. Du, and A. J. Zajac. 2009. A vital role for interleukin-21 in thecontrol of a chronic viral infection. Science 324: 1572–1576.

23. Elsaesser, H., K. Sauer, and D. G. Brooks. 2009. IL-21 is required to controlchronic viral infection. Science 324: 1569–1572.

24. King, C., A. Ilic, K. Koelsch, and N. Sarvetnick. 2004. Homeostatic expansion ofT cells during immune insufficiency generates autoimmunity. Cell 117: 265–277.

25. Spolski, R., M. Kashyap, C. Robinson, Z. Yu, and W. J. Leonard. 2008. IL-21signaling is critical for the development of type I diabetes in the NOD mouse.Proc. Natl. Acad. Sci. USA 105: 14028–14033.

26. McGuire, H. M., S. Walters, A. Vogelzang, C. M. Lee, K. E. Webster, J. Sprent,D. Christ, S. Grey, and C. King. 2011. Interleukin-21 is critically required inautoimmune and allogeneic responses to islet tissue in murine models. Diabetes60: 867–875.

27. Sutherland, A. P., T. Van Belle, A. L. Wurster, A. Suto, M. Michaud, D. Zhang,M. J. Grusby, and M. von Herrath. 2009. Interleukin-21 is required for the de-velopment of type 1 diabetes in NOD mice. Diabetes 58: 1144–1155.

28. Datta, S., and N. E. Sarvetnick. 2008. IL-21 limits peripheral lymphocytenumbers through T cell homeostatic mechanisms. PLoS One 3: e3118.

29. McGuire, H. M., A. Vogelzang, C. S. Ma, W. E. Hughes, P. A. Silveira,S. G. Tangye, D. Christ, D. Fulcher, M. Falcone, and C. King. 2011. A subset ofinterleukin-21+ chemokine receptor CCR9+ T helper cells target accessoryorgans of the digestive system in autoimmunity. Immunity 34: 602–615.

30. Kasaian, M. T., M. J. Whitters, L. L. Carter, L. D. Lowe, J. M. Jussif, B. Deng,K. A. Johnson, J. S. Witek, M. Senices, R. F. Konz, et al. 2002. IL-21 limits NKcell responses and promotes antigen-specific T cell activation: a mediator of thetransition from innate to adaptive immunity. Immunity 16: 559–569.

31. Kaplan, M. H., Y. L. Sun, T. Hoey, and M. J. Grusby. 1996. Impaired IL-12responses and enhanced development of Th2 cells in Stat4-deficient mice. Na-ture 382: 174–177.

32. Afkarian, M., J. R. Sedy, J. Yang, N. G. Jacobson, N. Cereb, S. Y. Yang,T. L. Murphy, and K. M. Murphy. 2002. T-bet is a STAT1-induced regulator ofIL-12R expression in naive CD4+ T cells. Nat. Immunol. 3: 549–557.

33. Yang, Y., J. C. Ochando, J. S. Bromberg, and Y. Ding. 2007. Identification ofa distant T-bet enhancer responsive to IL-12/Stat4 and IFNgamma/Stat1 signals.Blood 110: 2494–2500.

34. Mullen, A. C., A. S. Hutchins, F. A. High, H. W. Lee, K. J. Sykes, L. A. Chodosh,and S. L. Reiner. 2002. Hlx is induced by and genetically interacts with T-bet topromote heritable T(H)1 gene induction. Nat. Immunol. 3: 652–658.

The Journal of Immunology 3983

by guest on September 14, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

35. Schluns, K. S., and L. Lefrancois. 2003. Cytokine control of memory T-celldevelopment and survival. Nat. Rev. Immunol. 3: 269–279.

36. Zeng, R., R. Spolski, S. E. Finkelstein, S. Oh, P. E. Kovanen, C. S. Hinrichs,C. A. Pise-Masison, M. F. Radonovich, J. N. Brady, N. P. Restifo, et al. 2005.Synergy of IL-21 and IL-15 in regulating CD8+ T cell expansion and function. J.Exp. Med. 201: 139–148.

37. Casey, K. A., and M. F. Mescher. 2007. IL-21 promotes differentiation ofnaive CD8 T cells to a unique effector phenotype. J. Immunol. 178: 7640–7648.

38. 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.

39. Hinrichs, C. S., R. Spolski, C. M. Paulos, L. Gattinoni, K. W. Kerstann,D. C. Palmer, C. A. Klebanoff, S. A. Rosenberg, W. J. Leonard, andN. P. Restifo. 2008. IL-2 and IL-21 confer opposing differentiation programs toCD8+ T cells for adoptive immunotherapy. Blood 111: 5326–5333.

40. Cox, M. A., L. E. Harrington, and A. J. Zajac. 2011. Cytokines and the inceptionof CD8 T cell responses. Trends Immunol. 32: 180–186.

41. Suto, A., A. L. Wurster, S. L. Reiner, and M. J. Grusby. 2006. IL-21 inhibits IFN-gamma production in developing Th1 cells through the repression of eomeso-dermin expression. J. Immunol. 177: 3721–3727.

42. Asao, H., C. Okuyama, S. Kumaki, N. Ishii, S. Tsuchiya, D. Foster, andK. Sugamura. 2001. Cutting edge: the common gamma-chain is an indispensablesubunit of the IL-21 receptor complex. J. Immunol. 167: 1–5.

43. Murray, P. J. 2007. The JAK-STAT signaling pathway: input and output inte-gration. J. Immunol. 178: 2623–2629.

44. Van Belle, T. L., S. Nierkens, R. Arens, and M. G. von Herrath. 2012.Interleukin-21 receptor-mediated signals control autoreactive T cell infiltrationin pancreatic islets. Immunity 36: 1060–1072.

45. Esensten, J. H., M. R. Lee, L. H. Glimcher, and J. A. Bluestone. 2009. T-bet-deficient NOD mice are protected from diabetes due to defects in both T cell andinnate immune system function. J. Immunol. 183: 75–82.

3984 IL-21–INDUCED CTL DIFFERENTIATION IS T-bet DEPENDENT

by guest on September 14, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from