Reducing CD73 Expression by IL1b-Programmed …...by Th17TGFb1 cells subset with CD39 þCD73...

Microenvironment and Immunology

Reducing CD73 Expression by IL1b-Programmed Th17 CellsImproves Immunotherapeutic Control of Tumors

Shilpak Chatterjee1, Krishnamurthy Thyagarajan1, Pravin Kesarwani1, Jin H. Song2,Myroslawa Soloshchenko1, Jianing Fu3, Stefanie R. Bailey3, Chenthamarkshan Vasu3,Andrew S. Kraft2, Chrystal M. Paulos3, Xue-Zhong Yu3, and Shikhar Mehrotra1

AbstractT cells of the T helper (Th)17 subset offer promise in adoptive T-cell therapy for cancer. However, current

protocols for ex vivo programming of Th17 cells, which include TGFb exposure, increase the expression ofCD39 and CD73, two cell surface ATP ectonucleotidases that reduce T-cell effector functions and promoteimmunosuppression. Here, we report that ATP-mediated suppression of IFNg production by Th17 cells canbe overcome by genetic ablation of CD73 or by using IL1b instead of TGFb to program Th17 cells ex vivo. Th17cells cultured in IL1b were also highly polyfunctional, expressing high levels of effector molecules andexhibiting superior short-term control of melanoma in mice, despite reduced stem cell-like properties. TGFbaddition at low doses that did not upregulate CD73 expression but induced stemness properties drasticallyimproved the antitumor effects of IL1b-cultured Th17 cells. Effector properties of IL1b-dependent Th17 cellswere likely related to their high glycolytic capacity, since ex vivo programming in pyruvate impaired glycolysisand antitumor effects. Overall, we show that including TGFb in ex vivo cultures used to program Th17 cellsblunts their immunotherapeutic potential and demonstrate how this potential can be more fully realized foradoptive T-cell therapy. Cancer Res; 74(21); 1–12. �2014 AACR.

IntroductionAdoptive T-cell therapy (ACT), which involves the isolation

of antigen-specific T cells, followed by their ex vivo expansionand then infusion into autologous tumor-bearing host, is apromising approach for treating patients with advancedmalig-nancies (1). New strategies to improve adoptive immunother-apy are now emerging; including blocking inhibitorymolecules(CD28, 4-1BB, OX-40, ICOS, and VISTA), engaging costimula-tory molecules (2, 3), expanding T cells in different cytokines(IL2, IL15, IL12, IL21, and IL27; ref. 4), and generating distinct Thelper (Th) cell subsets (Th9 and Th17) with enhanced in vivopersistence (5, 6). However, recent studies show that immu-nosuppressive mechanisms induced by the tumor, such asindoleamine-2, 3-dioxygenase, PD-L1/B7-H, and FoxP3þ reg-ulatory T cells (Treg), might serve as negative feedbackmechanisms that follows rather than precedes the infiltrationof T cells into the tumor (7). These results underscore the need

to understand the T-cell–derived factors that aid in promotingan immunosuppressive tumor microenvironment, and to usethis knowledge in designing cellular therapies that effectivelytreat patients with advanced malignancies.

There has been a recent resurgence of the CD4þ T-cellsubsets (Th1, Th9, and Th17) in tumor immunotherapy (5, 6).Although studies have shown that Th17 cells do promotetumor growth (8, 9), an effective antitumor property of Th17cells can be observed when they coexpress key Th1 cytokineIFNg (5). These hybrid Th17þTh1 phenotype–bearing T cellsdisplay enhanced persistence and robust memory responseto tumors compared with Th1 cells when infused into micebearing melanoma (5). This implies that although antitumoreffector function of hybrid Th17þTh1 cell depends on Th1cytokine IFNg , the other Th17 properties of "stemness,"which may contribute to persistence (10, 11), or reducedsusceptibility to activation induced cell death (AICD) may bedependent specifically on Th17-programming conditions(12). Given that Th17 cells can also convert into a regulatoryTh17þFoxP3þ phenotype under inflammatory conditions inthe tumor microenvironment (13), it is crucial to understandwhich cytokines are responsible for regulating the pro-versus antitumorogenic properties of Th17 cells. Our datacharacterizing the hybrid Th17þTh1 cells generated usingTGFb-dependent (i.e., Th17TGFb1 with TGFb1/IL6) or TGFb-independent (i.e., Th17IL1b with IL1b/IL6) culture conditionsshow that Th17IL1b cells exhibit a dominant Th1 phenotypewith nominal expression of ectonucleotidase CD73, high effec-tor (i.e., T-bethi, Granzyme Bhi, and IL10lo) and glycolyticeffector phenotype (i.e., IFNghi, CD107ahi, and HIF1ahi)

1Department of Surgery, Hollings Cancer Center, Medical University ofSouth Carolina, Charleston, South Carolina. 2Department of Biochemistryand Molecular Biology, Hollings Cancer Center, Medical University ofSouth Carolina, Charleston, South Carolina. 3Department of Microbiologyand Immunology, Hollings Cancer Center, Medical University of SouthCarolina, Charleston, South Carolina.

Corresponding Author: Shikhar Mehrotra, Department of Surgery, Hol-lings Cancer Center (HO 512H), Medical University of South Carolina, 86Jonathan Lucas Street, Charleston, SC 29425. Phone: 843-792-9195; Fax:843-792-2556; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-14-1450

�2014 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org OF1

Research. on February 17, 2020. © 2014 American Association for Cancercancerres.aacrjournals.org Downloaded from

Published OnlineFirst September 9, 2014; DOI: 10.1158/0008-5472.CAN-14-1450

http://cancerres.aacrjournals.org/

as compared with Th17TGFb1 cells, which translates intoimproved in vivo tumor control. We believe this strategy willhelp us to design conditions for ex vivo expansion that willminimize Tregs property, maximize Th1 features while main-taining Th17 phenotype, and potentiate the long-term antitu-mor response after ACT.

Materials and MethodsMice

C57BL/6, CD73�/� (B6.129S1-Nt5etm1Lft/J), B6-Thy1.1 (B6.PL-Thy1a/CyJ), andOT-II (B6.Cg-Tg (Tcra Tcrb 425Cbn/J)micewere obtained from The Jackson Laboratory. Development ofh3T transgenic mouse bearing T-cell receptors (TCR) reactiveto the human tyrosinase 368-376 (YMDTMSQV) epitope hasbeen described recently (14). OT-II-GFP-FoxP3 mice coexpres-sing EGFP and FoxP3 were kind gift from Dr. C. Vasu, MedicalUniversity of South Carolina (MUSC, Charleston, SC). Animalswere maintained in pathogen-free facilities and proceduresapproved by the Institutional Animal Care andUseCommittee.

Reagents and cell linesOva (ova323-339; ISQAVHAAHAEINEAGR) andMART-1 (ELA-

GIGILTV) peptides were purchased fromGenScript. Penicillin,streptomycin, glucose-free RPMI-1640, and Iscove's modifiedDulbecco's medium (IMDM) were purchased from Life Tech-nologies. FBS was procured from BioAbChem Inc. All therecombinant cytokines except IL2 (Shenandoah Biotechnolo-gy) and fluorochrome-conjugated anti-mouse CD4 (GK1.5),CD73 (TY/11.8), CD26 (H194-112), CD44 (IM7), CD62L(MEL-14), IFNg (XMG1.2), IL17a (TC11-18H10.1), IL22(Poly5164), IL2 (JES6-5H4), TNFa (MP6-XT22), CD25 (PC61),and T-bet (4B10) were purchased from BioLegend. Fluoro-chrome-conjugated anti-mouse Vb5.1,5.2 (MR9-4), IRF-4 (3E4),CD39 (24DMS1), and ROR-gt (AFKJS-9) were obtained fromeBiosciences. Anti-human Vb12 was from Thermo Scientific.Purified anti-CD3, anti-CD28, anti-IFNg , and anti-IL4 wereobtained from UCSF mAb core. Anti-mouse pS6 conjugatedwith Alexa647 was purchased from Cell Signaling Technology.Tumor cells tested for antibody production were obtainedfrom our collaborators as follows: 624-MEL (Dr. MichaelNishimura, Loyola University, Chicago, IL), EL4 (Dr. Zihai Li,MUSC), and B16-ova (Dr. Mark Rubinstein, MUSC).

Culture conditionMagnetic-bead sorted CD4þ T cells (>95%) from B6 sple-

nocytes or in some cases FACS (MoFlo Astrios; BeckmanCoulter) sorted CD4þ T cells from h3T splenocytes weredifferentiated into either Th17TGFb1 (3 ng/mL TGF-b1 and25 ng/mL IL6) or Th17IL1b (20 ng/mL IL1b and 25 ng/mL IL6)or Th17IL1bþTGFb (20 ng/mL IL1b, 25 ng/mL IL6, and 250 pg/mLTGF-b1) in presence of plate-bound anti-CD3 (5 mg/mL) andanti-CD28 (5 mg/mL). mAb to IFNg (XMG1.2; 10 mg/mL) andmAb to IL4 (11B11; 10 mg/mL) were added to the polarizingmedia and after 48 hours cells were fed with IL2 (50 IU/mL).Complete IMDM (cIMDM) media containing 10% FBS, penicil-lin, streptomycin were used for T cells differentiation. However,in some cases, glucose and pyruvate-free RPMI-1640 mediasupplemented either with 20 mmol/L glucose plus 1 mmol/L

pyruvate or 20 mmol/L pyruvate plus 3-Mercaptopicolinic acid(gluconeogenesis blocker; Sigma) were used for Th17IL1b polar-ization. Purified CD4þ T cells stimulatedwith plate-bound anti-CD3 (5 mg/mL) and anti-CD28 (5 mg/mL) were termed as Th0 inthis study. In some experiments, T cells polarization wasperformed either in presence or absence of 50 mmol/L ATP(Sigma). On day 3 of culture, T cells were harvested and eitherprocessed for intracellular cytokine analysis, RNA preparationusing TRIzol (Life Technologies) or used for adoptive celltherapy. For ova-specific generation of different Th17 cells,total splenocytes from OT-II TCR transgenic mice were stim-ulated with 1 mg/mL ova323-339 peptide in presence of above-mentioned polarizing conditions.

Adoptive T-cell protocolMouse melanoma tumor (B16-F10-ova), human melano-

ma (624-MEL), and T cells lymphoma cells (EL-4) weremaintained in vitro in IMDM. EL-4 cells (0.25 � 106) wereinjected i.p. into C57BL/6 mice, and on day 12 a total of 1 �106 Th17 cells (either Th17TGFb1 or Th17IL1b) were trans-ferred i.p. into the tumor site. Following 48 hours of T-celltransfer, peritoneal ascites fluid was drawn and donor cellswere tracked using congenic Thy1.1 marker. B16-F10-ova(0.25 � 106) and 624-MEL (2.5 � 106) were injected s.c. intoleft flank of C57BL/6 or Rag1�/� C57BL/6 mice or NSG-A2mice, respectively. Twenty-four hours before adoptive trans-fer of T cells (CD4þVb5þ ova–specific Th17TGFb1, Th17IL1b,or Th17IL1bþ TGFb) on day seventh, the recipient mice wereinjected with cyclophosphamide (4 mg/mice). Tumors bear-ing C57BL/6 or Rag1�/� C57BL/6 mice were either keptuntreated or adoptively transferring with either CD4þVb5þ

(1� 106) ova–specific Th17TGFb1, Th17IL1b, or Th17IL1bþ TGFb

cells (1 � 106 cells/mice) on day 7. For xenograft tumorexperiment, 15 days subcutaneously established 624-MEL inNSG-A2 mice were either kept untreated or treated witheither 0.2 � 106 CD4þVb12þ Th17TGFb1 or Th17IL1bþ TGFb

cells.

Activation induced T-cell deathDifferentiated ova-specific Th17 (Th17TGFb1, Th17IL1b, or

Th17IL1bþTGFb) restimulated for 4 hours with either cognateantigen (ova323-339) or nonspecific antigen (MART-1) loadedirradiated C57BL/6 splenocytes at the 5:1 (T cells:B6 spleno-cytes) ratio. Apoptosis was measured by Annexin V (BDBiosciences) versus 7AAD staining according to the manufac-turer's protocol, followed by flow cytometry. Data were ana-lyzed with FlowJo software (Tree Star).

Cytotoxicity assayB16-F10-ova (specific target) or EL-4 (nonspecific target)

cells labeled with CFSE (carboxyfluorescein diacetate succini-midyl ester; Life Technologies) and cocultured with differen-tiated Th17TGFb1 and Th17IL1b cells were used to determinecytotoxic potential, as detailed in Supplementary Methods.

Flow cytometry and qPCRDetailed protocols for staining the cells for surface markers

and intracellular cytokines have been described earlier (15),

Chatterjee et al.

Cancer Res; 74(21) November 1, 2014 Cancer ResearchOF2




and in Supplementary Methods. Detailed methodology forqPCR is provided in Supplementary Methods.

Glucose uptake, oxygen consumption, and glycolyticfluxOxygen consumption rate (OCR) and extracellular acidifi-

cation rate (ECAR) were evaluated as described earlier (16).Glucose uptake was determined by 2NBDG (Cayman Chem-ical) uptake assay according to themanufacture's protocol andas described in Supplementary Methods.

Statistical analysisAll data reported are the arithmetic mean from three or five

independent experiments performed in triplicate �SD unlessstated otherwise. The unpaired Student t test was used toevaluate the significance of differences observed betweengroups, accepting P < 0.05 as a threshold of significance. Dataanalyseswere performed using the Prism software (GraphPad).

ResultsTGFb-induced CD73 expression on Th17 cells increasessusceptibility for IFNg suppression

CD39 and CD73 are two ectonucleotidases that sequentiallycleave ATP to produce adenosine and are expressed at higherlevels in TGFb-polarized Th17 (Th17TGFb1) cells as comparedwith the unpolarizedTh0 cells (Fig. 1A). Because ATP is presentat high concentration in the tumormicroenvironment (17) andthe ATP-byproduct adenosine (a potent immune suppressive)subverts antitumor immunity (18), we therefore tested wheth-er ATP could affect the functionality of Th17TGFb1 cells. Flowanalysis revealed that ATP (50 mmol/L) significantly sup-pressed IFNg productionwithout affecting the IL17 productionby Th17TGFb1 cells subset with CD39þCD73þ phenotype, ascompared with Th0 cells that express minimal CD73 on theircell surface (Fig. 1B). These data imply that CD73 expressioncould lead to impairment of IFNg production by Th17TGFb1

Pre-transfer

Post-transfer

C

Th

17IL

1b

Pre-transfer

Post-transfer

G

Th

17IL

1b

–ATP (50 mmol/L)

(50

mmo

l/L)

(50 mmol/L)

F

ATP +−

ATh0 Th17TGFb1

TGFb1

E Th0 Th17TGFb1 Th17IL1b

Th17Th0

–AT

P

+AT

P

B

ATP +− +−

Th0Th17

ATP +− +−

WtCD73-/-

Th17

Wt

CD

73-/

-

–ATP

+ATPD

+ATP

TGFb1

TGFb1

CD73

CD73

IL17

IL17

IL17 IL17

IL17

CD

39C

D39

IFN

g

IFN

g

IFN

g

IFN

g

IFN

g

20

15

10

5

0

20

15

10

5

0

% o

f IF

Ng-

pro

du

cin

g c

ells

% o

f IF

Ng-

pro

du

cin

g c

ells

% o

f IF

Ng-

pro

du

cin

g c

ells

15

10

5

0

Figure 1. TGFb-inducedCD73 expression on Th17 cells increases susceptibility for IFNg suppression. A, flow-cytometric analysis of ectonucleotidases (CD39and CD73) expression by unpolarized (Th0) or TGFb1-mediated Th17 (Th17TGFb1) cells. Data are representative of five independent experiments. B–D,intracellular IFNg and IL17 production in presence or absence of ATP (50 mmol/L) by Th17TGFb1 or unpolarized (Th0) cells (B); Thy1.1þTh17TGFb1 cells retrievedfrom the tumor site of C57BL/6 (Thy1.2þ) mice (n¼ 4) bearing EL-4 ascetic tumor following 48 hours of T cells transfer (C); and Th17TGFb1-polarized cells fromeither wt or CD73�/� C57BL/6 mice (D). Cumulative data from three different experiments are represented in bar diagram alongside the dot-plot forthe percentage of cells producing IFNg in presence or absence of ATP (50 mmol/L). E, flow-cytometric analysis (right) of CD39 and CD73 expression by Th0,Th17TGFb1, and Th17IL1b cells. F and G, intracellular IFNg and IL17 secretion in presence or absence of ATP by Th17IL1b cells (F) and Thy1.1þ Th17IL1b

cells retrieved from the tumor site of C57BL/6 (Thy1.2þ) mice (n ¼ 4) bearing EL-4 ascetic tumor following 48 hours of T cells transfer (G). Results arerepresentative of three (E) and five (F and G) independent experiments; ��, P < 0.0001.

CD73 Expression Dampens Antitumor Th17 Response

www.aacrjournals.org Cancer Res; 74(21) November 1, 2014 OF3




cells inATP-rich tumormicroenvironment. To further evaluatethe functional fate of Th17TGFb1 cells coexpressing CD39 andCD73 in tumor microenvironment, purified CD4þ T cells fromcongenic Thy1.1 mice were differentiated toward Th17TGFb1

and injected into C57BL/6 (Thy1.2) mice bearing EL-4 ascitestumor. Donor cells retrieved from the tumor site after 48 hoursshowed decreased IFNg production following restimulation(Fig. 1C). Furthermore, to confirm the role of ectonucleotidasesCD73 in ATP-mediated suppression of IFNg production byTh17TGFb1 cells, CD4þT cells were purified fromwild-type (Wt)or CD73�/�mice (CD73 converts AMP to adenosine) and thendifferentiated to Th17 phenotype in presence or absence ofATP. In contrast with Wt Th17TGFb1 cells, CD73�/� Th17TGFb1

cells showed no decrease in IFNg secretion when cultured inATP (Fig. 1D). These data support a role of CD73 in suppressingIFNg production by Th17TGFb1 cells exposed to ATP.

Next, we tested whether the polarization of Th17 cells inabsence of TGFb1, but in presence of IL1b that is reported tohave lower cell surface expression of CD39 and CD73 (19, 20),would affect the persistence and function of antitumor Th17cells (Fig. 1E).We thus tested the susceptibility of Th17IL1b cellsto IFNg suppression in ATP-rich environment. Na€�ve CD4þ Tcells were polarized to Th17IL1b in presence or absence of 50mmol/L ATP, and cytokines production was analyzed. Wefound that Th17IL1b cells express less CD39 and CD73 on cellsurface, and were resistant to suppression of IFNg productionin presence of ATP (Fig. 1F). Furthermore, Th17IL1b cellssignificantly retain their IFNg production at the tumor sitewhen retrieved from the tumor site after 48 hours of celltransfer (Fig. 1G). These data establish that Th17IL1b cellscould be better than Th17TGFb1 cells in retaining their func-tionality in tumor microenvironment.

Distinct functionality of Th17TGFb1 and Th17IL1b cellsIFNg plays a pivotal role in Th17-mediated control of tumor

growth (5). Because both Th17IL1b versus Th17TGFb1 cells differin their susceptibility to suppression of IFNg production at thetumor site, we further characterized these cells. On comparingthe effector cytokines secretion ability between by Th17TGFb1

and Th17IL1b cells in vitro, we observed that about approxi-mately 1.5-fold higher proportion of Th17IL1b cells (12%)secreted IFNg than Th17TGFb1 cells (8%; Fig. 2A). Similarly,Th17IL1b cells also showed the higher percentage of TNFa (80%vs. 65%) and IL22 (6.5% vs. 0.5%) as compared with Th17TGFb1

cells. The increased secretion of Th1 effector cytokines (IL2,IFNg , and TNFa) and reduction in IL17 by Th17IL1b cellscorrelated with the higher expression level of transcriptionfactor T-bet, IRF-4, and reduced level of ROR-g , respectively(Fig. 2B). Interestingly, evaluation of the cell surface markersshowed that Th17IL1b cells express high CD62LþCD44þ centralmemory phenotype (88%) as compared with Th17TGFb1 cells(50%; Fig. 2C, i), and display a prominent active phenotype aselucidated by CD25 expression (Fig. 2C, ii). Th17IL1b cells alsoshowed high surface expression of CD26 (Fig. 2C, iii), whichhas been shown to be associated with high IL22, GM-CSF,and IL23R (21). Th17IL1b cells also exhibit higher IL23R (thatstabilizes Th17 phenotype; ref. 22),GzmB (encodes for cytolyticmolecule granzyme B), CSF2 (encodes for GM-CSF), cytokines

IL22, and IL3 (Fig. 2D, i). It has been recently shown that IL23enhances/stabilizes Th17 phenotype, and similar to GM-CSFthat also increases effector function, leading to pathogenicphenotype in the experimental autoimmune encephalomyeli-tis (EAE) model (23, 24). Importantly, the expression of cyto-kine IL10, that could suppress T-cell response, was decreasedin Th17IL1b cells as compared with Th17TGFb1 cells. Further-more, a qPCR array–based analysis also revealed that tran-scription factors regulating TCR signaling, Erk1/2, Notch, andEGF pathway were highly expressed in Th17IL1b cells as com-pared with Th17TGFb1 cells (Fig. 2D, ii). In addition, a signaltransduction array analysis also revealed that Th17IL1b cellsengage inmultiple signaling pathway than Th17TGFb1 cells (Fig.2D, iii), which may be responsible for their overall enhancedeffector capability (25).

Enhanced antitumor function of Th17IL1b versusTh17TGFb1 cells

Because Th17IL1b cells are polyfunctional and produce anarray of various cytokines/effector molecules, we next inves-tigated their potential to control tumor growth. Using CD4 Tcells fromOT-II mice that were programmed to Th17TGFb1 andTh17IL1b phenotype, first we tested their ability to lyse murinemelanoma cell, in vitro. Our data show that Th17IL1b cells canlyse the tumor cells directly and are more cytolytic thanTh17TGFb1 cells (Fig. 3A and Supplementary Fig. S1A). Next,adoptive transfer of 1 � 106 ova–specific Th17TGFb1 andTh17IL1b cells (i.v.) to immunocompetent C57BL/6 mice bear-ing established B16-F10-ova melanoma showed a markeddelay in tumor growth in the group receiving Th17IL1b versusTh17TGFb1 cells (Fig. 3B). At the experimental endpoint, acytokine analysis of adoptively transferred donor cellsretrieved from various sites (tumor and nontumor) showedthat the IFNg secretion was dramatically reduced in Th17TGFb1

cells retrieved from the tumor site (Fig. 3C, lower left), ascompared with Th17IL1b cells that maintained their IFNgsecretion at the tumor site (Fig. 3C, bottom right). Importantly,donor cells retrieved from the lymph node, spleen, and blooddid not show significant decrease in IFNg secretion uponrestimulation. These data establish that Th17TGFb1 cells withhigh CD73 expression are more prone to loosing effectorcytokine IFNg in tumor microenvironment as compared withTh17IL1b cells. In addition, it is possible that higher expressionof CD25 on Th17IL1b cells resulted in their increased homeo-static proliferation as compared with Th17TGFb1 cells (26), andthereby improved antitumor effect. Next, to determine wheth-er the antitumor potential of Th17IL1b cells are independent ofany endogenous T cells, we transferred 1 � 106 of either OT-IITh17TGFb1 or Th17IL1b cells to Rag�/� C57BL/6 mice bearingB16-F10-ovamelanoma.We obtained similar tumor regressionby Th17IL1b cells in tumor-bearing Rag�/� C57BL/6mice as weobserved for immunocompetent mice, suggesting thatTh17IL1b cells could control the tumor growth independentof CD8 T cells (data not shown).

To further address the issue of Th17 plasticity, that is, theability to convert to FoxP3þ cells, which is considered a keyreason for the failure of Th17 cells in tumor immunotherapy,we tested whether there is a differential susceptibility for

Chatterjee et al.





Th17TGFb1 Th17IL1bTh17TGFb1 Th17IL1b

Th17TGFb1

Th17IL1b

Th17TGFb1

Th17IL1bTh17TGFb1

Th17IL1b

CS

F2

IL3

IL22

IL23

r

IL10

Gzm

B

IFN

g

IFNg

–5 0 –30 –20 –10 0 10 205 10

IFNg+IL17TNFa+IL17IL2IL2+IL17

IL22+IL17IL22

IL2

TNFa

IL22

A

IL17

C (i)

T-bet RORγIRF-4

Tbx21

Fol

d ch

ange

Fol

d ch

ange

Fol

d ch

ange

10

8

6

4

2

0

Fo

ld c

han

ge

Fold change Fold change

5040302010

3

2

1

0

4

3

2

1

0

1.5

1.0

0.5

0.0

RORg

ns

IRF-4B

C (ii)CD62L

CD

44

CD25 D (i)

D (ii) D (iii)

170485

7,04222,906

CD26

ciii

689 1,078

1,059904

1,6353,341

TN

Fa

Figure 2. Distinct functionality of Th17TGFb1 andTh17IL1bcells. A, naïveCD4þTcells fromC57BL/6miceweredifferentiated towardeither Th17TGFb1 or Th17IL1b

and intracellular production of various cytokines was analyzed. The percentage of cells producing different cytokines is also represented in pi-diagram(50,000 cells/group were analyzed to draw pi-diagram). B, qPCR analysis (top) and flow-cytometric analysis (bottom) of various Th17 signature transcriptionfactors expression by Th17TGFb1 and Th17IL1b cells. C and D, flow-cytometric analysis of CD62L versus CD44 expression (C, i) and CD25 expressionTh17TGFb1 andTh17IL1b cells at day 3of polarization (C, ii). D, i, qPCRanalysis of expressionof key effector genes in Th17TGFb1 and Th17IL1b cells after 3 days ofpolarization. Data represent three independent experiments; ��,P < 0.005; ��,P < 0.0001. Transcription factors array (D,ii) and signal transduction array (D, iii)were performed using the 84-Gene qPCR-Based Array Kit (SABiosciences) as per the manufacturer's recommendation. Fold upregulation (blue) ordownregulation (red) of Th17IL1b over Th17TGFb1.






conversion to FoxP3þTh17 cells between the Th17TGFb1 andTh17IL1b cells. Using CD4þ T cells from GFP-FoxP3 developedon OT-II background for programming to Th17TGFb1 andTh17IL1b cells, we noted an increased percentage of FoxP3þ

GFPþ cells in conventional Th17TGFb1 cultures (Fig. 3D). Inaccordance with the recent study that established the inhib-itory role of transcription repressor growth factor independent1 (Gfi-1) in nTreg generation, our data show that conversion toiTreg phenotype is decreased in Th17IL1b cells as comparedwith Th17TGFb1 ones, potentially due to higher Gfi-1 expression(Fig. 3D, right). These sets of data establish that Th17IL1b cellspossess better antitumor properties and are less "plastic" thanconventional Th17TGFb1 cells.

Improved antitumor ability of Th17IL1b cells correlateswith increased glycolysis

Recent evidences suggest that T-cell energy metabolismis not merely a cellular phenomenon, rather it can deter-mine the cytokine productions and functional outcome ofT cells (27, 28). Because IFNg secretion has also beenshown to be dependent on glucose consumption (29, 30),we reasoned that Th17IL1b cells could be more glycolytic. Acomparison of glucose consumption reveals that Th17IL1b

cells were more glycolytic than Th17TGFb1 cells (Fig. 4A). Tofurther address this issue, ECAR, an indicator of glycolysis,and OCR, an indicator of oxidative phosphorylation(OXPHOS), were measured in Th17IL1b and Th17TGFb1 cellsusing seahorse bioanalyzer. Our data show that Th17IL1b

cells had higher ECAR value as compared with Th17TGFb1

cells, indicating their high glycolytic capacity (Fig. 4B, i).Furthermore, the basal OCR was less in Th17IL1b cells ascompared with Th17TGFb1 cells (Fig. 4B, ii), leading to ahigher OCR/ECAR ratio in Th17TGFb1 cells (Fig. 4B, iii).To further ascertain the metabolic difference betweenTh17IL1b and Th17TGFb1 cells, we analyzed mRNA expres-sion of various glycolysis-associated genes using qPCR. Ourdata show that the mRNA expression levels of all glycolyticgenes evaluated were several fold higher in Th17IL1b cells ascompared with Th17TGFb1 cells (Fig. 4C). The increasedexpression at mRNA level also translated to increasedexpression of glycolytic proteins, as represented by hexo-kinase II (HKII) expression (Fig. 4D). The increased glycol-ysis in Th17IL1b cells also correlated with increased acti-vation of the mTOR pathway, as determined by phosphor-ylation of S6 (Fig. 4E). These data support that highereffector function and antitumor control displayed by

D

A B

Sp

leen

Lym

ph

no

de

Pre

tran

sfer

Pos

ttra

nsfe

r

Per

iph

eral

blo

od

Tum

or

site

CTh17TGFb1

Th17IL1b

Th17TGFb1

1.83 7.74 13.9 4.89

43.5 46.9 61.3 19.45.09 0.36 9.61 0.288

88.1 6.42 87.1 3.04

25.5 1.28 32.5 0.428

66.6 6.6 61.6 5.32

5.81 1 6.5 1.23

91.7 1.5 90.8 1.47

0.958 0.43 17.2 1.14

89.2 9.39 77.3 4.4

Th17IL1b

Th17TGFb1 Th17IL1b

Th17TGFb1

Th17IL1b

Th17TGFb1

Control

Th17IL1b

0 5 10 15 20Days

Gfi

25

IFN

g

IL17

Fo

ld c

han

ge

% o

f B

16-O

va c

ell l

ysis

Tu

mo

r si

ze (

mm

2 )

Vb5

FoxP3-GFP

100

80

60

40

20

0

500

400

300

200

100

0

5

4

3

2

1

0

Figure 3. Enhanced antitumor function of Th17IL1b versus Th17TGFb1 cells. A, B16-F10-ova cells labeled with CFSE were cocultured at the 1:5 ratio with eitherTh17TGFb1 or Th17IL1b for 6 hours, and decrease in the number of cells expressing CFSE was analyzed by using flow cytometry. B, wt C57BL/6 mice (n ¼ 5mice/group) were inoculated (s.c.) with 0.25 � 106 B16-F10-ova murine melanoma cells and treated with cyclophosphamide (4 mg/mouse) after 7 days.Cyclophosphamide-treated mice were either kept untreated as control or adoptively transferred 1 day later with either 1 � 106 ova–specific Th17TGFb1 orTh17IL1b cells. Tumor growth wasmeasured using digital calipers every fourth day. Data, mean tumor size at each time point from one of the two experimentswith similar results. C, intracellular cytokine production of ova-specific donor Th17TGFb1 and Th17IL1b cells after retrieving from either lymph nodes, spleen,peripheral blood, or tumor site of 21 days tumor-bearing mice (n ¼ 4). Cytokines production of donor cells was compared with nontransferred cells. Datarepresent two independent experiments. D, naïve CD4þ T cells fromOT-II GFP-FoxP3mice were polarized to Th17TGFb1 and Th17IL1b type in presence of ovaand the percentage of cells expressing GFP (indicative of FoxP3-expressing cells) was analyzed after 3 days of polarization using flow cytometer (left). Right,qPCR analysis of Gfi-1 expression by Th17TGFb1 and Th17IL1b cells. Data represent three independent experiments; ��, P < 0.005 and ��, P < 0.0001.

Chatterjee et al.





Th17IL1b cells as compared with Th17TGFb1 cells could beattributed to higher glycolytic commitment.

Shift from glycolysis dampens Th17IL1b cells effectorfunctionsNext, we tested whether the increase in glycolytic commit-

ment observed in Th17IL1b cells is key to increased expressionof effector molecules that are responsible for better antitumoreffector response. For this purpose, the Th17IL1b cells werepolarized in complete media containing glucose that supportsglycolysis, or media with no glucose but pyruvate with gluco-neogenesis blocker (that supports oxidative phosphorylation).We observed that cells grew at same rate (Fig. 5A, i) and alsohad similar activation profile (Supplementary Fig. S1B) inpyruvate-containing (without glucose) media as comparedwith completemedia. On analyzing the Th17IL1b cells polarizedin different media for signature cytokines production afterstimulation with PMA/Ionomycin, we found that blocking theglycolytic pathway by culturing in pyruvate-containing mediamarkedly inhibited the IFNg and IL17 production implyingthat the these cytokines production is dependent on glycolysis.However, TNFa secretion remained unaffected by pyruvate(Fig. 5A, ii). It has been documented that IFNg andTNFahave acomplimentary relationship, and it is likely that similar to the

differences in secretion pattern between IFNg and TNFa ON/OFF cycle (31, 32), different metabolic requirements exist forthese cytokines to compensate for effector function underchanging microenvironment. To evaluate whether inhibitingthe glycolytic pathway could also affect the cytolytic capacity ofTh17IL1b cells, mRNA expression of GzmB was analyzed afterpolarization of cells in pyruvate-containing media. Expressionanalysis using qPCR revealed that key effector molecules(GzmB, GM-CSF, T-bet, IL22, and IL3) were significantly down-regulated in Th17IL1b cells cultured in presence of only pyru-vate-containing media as compared with complete media(Fig. 5B). Moreover, we also found that antigen-specific killingof tumor cells B16-F10-ova was greatly reduced when Th17IL1b

cells were polarized in pyruvate-containing media (Fig. 5C andSupplementary Fig. S1C). However, no correlation betweenglycolysis and CD39/CD73 expression by Th17IL1b cells wasestablished, because culturing Th17IL1b cells in pyruvate-con-taining media (glucose free) did not affect the CD39/CD73expression (Supplementary Fig. S2A). Next, to further confirmthat glycolysis is necessary to mount proper antitumorresponse by Th17 cells, Rag�/� C57BL/6 mice with 7 daysestablished B16-F10-ova tumor were treated by adoptivelytransferring 1� 106 ova–specific Th17IL1b cells polarized eitherin normal or only pyruvate-containing media. Antitumor

A B (ii)

E

B (iii)B (i)

C D

HKII

b-Actin

p102

p45

0 2 4 6

*

Fold change

Th17TGFb1

Th17IL1bTh17TGFb1

Th17IL1b

Th17TGFb1

Th17IL1b

Th17TGFb1

Th17IL1b

Th17TGFb1

Th17IL1b

pS6

Th17TGFb1

Th17IL1b

Glu

t1

Glucose

transport

Positive re

gulator

glycolysis

Negative re

gulator

oxphosGlycolysis

HIF

1a

HK

II

Pfk

p

Pg

am1

Ld

hA

Pd

k1

Th17TGFb1

Th17IL1b

Th17TGFb1 Th17IL1b

Fo

ld c

han

ge

EC

AR

(m

pH

/min

)

OC

R (

pm

ol O

2/m

in)

2NBDG

80

60

40

20

0

OC

R/E

CA

R

80

60

40

20

0

2,000

1,500

1,000

500

0

20

15

10

5

0

Figure 4. Th17TGFb1 and Th17IL1b cells aremetabolically different. Differences in glycolysis between Th17TGFb1 and Th17IL1b cells were observed using glucoseuptake using fluorescent glucose (2-NBDG; A); basal ECAR (B, i); basal OCR (B, ii); the basal OCR/ECAR ratio (B, iii); qPCR analysis of the expressionof key genes associated with glycolysis (C). D, top, Western blot analysis for HKII; bottom, blot quantification of HKII. E, flow-cytometric analysis ofphosphorylation ofS6 (pS6) ribosomalprotein. Results inAandBare representative of four and inCandDare representative of three independent experimentswith similar results. �, P < 0.05; ��, P < 0.005; ��, P < 0.0001.






potential of Th17IL1b cells was severely impaired when cellswere polarized in pyruvate-containing media as determinedby the rapid progression of tumor growth in treated mice(Fig. 5D). However, mice treated with Th17IL1b cells polarizedin normal media markedly delayed tumor growth. These datatogether strongly suggest that antitumor potential of Th17cells is highly dependent on their metabolic commitment, andhigher glycolysis may have contributed to improved antitumorresponse observed with Th17IL1b cells.

Lack of stemness in Th17IL1b cells could be restored byvery low dose of TGFb

Despite superior antitumor activity of adoptively transferredTh17IL1b cells in vivo, we noticed that the Th17IL1b-recipientmice could not control tumor growth after 30 to 35 days. Theinability to control tumor growth long-term was also corre-lated with the poor persistence of these Th17IL1b cells in vivo(Supplementary Fig. S2B). Because long-term persistence of

conventional Th17TGFb1 cells is attributed to "stemness" sig-nature (10, 11), we analyzed various stemness-associated genesin Th17IL1b versus Th17TGFb1 cells. Real-time PCR analysisrevealed that expression of various stem cell–associated genes(b-catenin, Bcl6, Tcf7, and Lef1) was significantly lower inTh17IL1b cells as compared with Th17TGFb1 cells (Fig. 6A). Wealso found greater induction of AICD in Th17IL1b cells ascompared with Th17TGFb1 cells following antigen restimula-tion (data not shown). These data together indicate thatTh17IL1b cells exert the characteristics of terminally differen-tiated populationwith profound antitumor potential, however,lack stemness features, and thus persist for a short period oftime in tumor-bearing host. Because TGFb has been shown toinduce expression of various stem cell–associated genes (33),we argued that it may be responsible for the stemness signa-ture reported in conventionally programmed Th17TGFb1 cells(10, 11). We thus titrated for a minimum dose of TGFb thatcould impart stemness without inducing the expression of

IL17

Completemedia

–Glu+Pyruvate

A (ii)

B

C D

0 5 10Days

15 20 25

**

A (i)

CFSE

(complete media)

(–Glu +Pyru)

*****

**

******

**

Th17IL1b

Th17IL1b

Th17IL1b

(–Glu +Pyru)Th17IL1b

(complete media)

Th17IL1b

(–Glu +Pyru)

Th17IL1b

(complete media)

Control

Th17 (IL1b mediate)

B6 SpleenCD4+ T cells

Complete media

–Glucose+Pyruvate (20 mmol/L)

CS

F2

IL3

IL22

IL10

Gzm

B

Tb

x21

IFN

gIL

2T

NF

aIL

22

Fo

ld c

han

ge

6

4

2

% o

f B

16-O

va c

ell l

ysis

Tu

mo

r si

ze (

mm

2 )

100

80

60

40

20

0

500

400

300

200

100

0

0.6

0.4

0.2

0.0

Figure 5. Shift from glycolysisdampens Th17IL1b cells effectorfunctions. A, i, schematic diagramof the culture conditions used togenerate the Th17IL1b cells. ii,intracellular staining of variouscytokines. B, qPCR analysis of theexpression of key effector genes.C, cytolysis of B16-F10-ova cellswas evaluated using Th17IL1b cellspolarized either in complete media(green bars) or in 20 mmol/Lpyruvate (no glucose)-containingmedia (brown bars). D, C57BL/6Rag1�/� mice (n ¼ 5 mice/group)were inoculated (s.c.) with0.25 � 106 B16-F10-ova, and after7 days, mice were either keptuntreated as control or adoptivelytransferred with either 1� 106 ova–specific Th17IL1b OT-II cellspolarized either in complete mediaor 20 mmol/L pyruvate (noglucose)-containing media. Tumorgrowth was measured using digitalcalipers every fourth day. Data,mean tumor size at each time pointin one of the two experiments withsimilar results; �, P < 0.05;��, P < 0.005; and ��, P < 0.0001.

Chatterjee et al.





ectonucleotidase CD73. Our dose titration data showed that250 pg/mL of TGFb minimally upregulates CD73 expression(Fig. 6B). Th17IL1b cells differentiated in presence of 250 pg/mL

of TGFb (referred to as Th17IL1bþTGFb) alsomarkedly increasedthe mRNA transcripts of various stemness genes as well as thegenes associated with T-cell memory (Fig. 6C). However,

B

Lef1 8

6

4

2

0

8

6

4

2

0

8

6

4

2

0

3

2

1

0

500

400

300

200

100

0

600

400

200

0

Tcf7 Bcl6 b-Catenin A

C Glycolysis gene Stemness gene

Annexin V

Ova

323-

339

Ova

323-

339

Mar

t-1

Mar

t-1

Mar

t-1

Ova

323-

339

Ova

323-

339

7AA

D

Mar

t-1

D

E

0 20 40Days

60

0 20 40Days

60

80

F

Fo

ld c

han

ge

Fo

ld c

han

ge

Fo

ld c

han

ge

Fo

ld c

han

ge

Th17 IL1bTh17 TGFb1

CD73

CD

39

Th17IL1b

Th17IL1b

Th17IL1b

ns

Th17IL1b+TGFb

TGFb

Control

Th17TGFb1

Th17TGFb1

Th17IL1b+TGFb

Th17IL1b+TGFb

Th17IL1bTh17TGFb1

0.13 5.85 0.21 7.61 0.14

77.9 16.1 74.5 17.7 75.9 17.3

0.737.090.11 31.5 0.61

6.68

21.5

42.531.561.3 25.2 56.9 21.0

Th17IL1b+TGFb

Th17TGFb1

Th17IL1b+TGFb

Th17IL1b

Th17TGFb1

Th17IL1b+TGFb

Lef

1

Tcf

7

Bcl

6%

of

An

nex

in V

+ +

An

nex

in+ 7

AA

D+

cells

b-C

aten

in

Control

3 ng/mL7.32 58.7 59.5 26.9 79.5 9.96 86.4 0.95 87.7 2.96

6.04 28.0 10.6 3.01 9.69 0.84 12.6 0.052 9.15 0.16

1 ng/mL 0.5 ng/mL 0.25 ng/mL 0 ng/mL

Glu

t1

HIF

1a

HK

II

Pfk

p

Pg

am1

Ld

hA

Fo

ld c

han

ge

Tu

mo

r si

ze (

mm

2 )T

um

or

size

(m

m2 )

15

10

5

0

80

60

40

20

0

Fo

ld c

han

ge

15

10

5

0

Figure 6. Low dose of TGFb induces stem cell–like phenotype in Th17IL1b cells. A, qPCR analysis for expression of key memory and stemness-associatedgenes in Th17TGFb1 and Th17IL1b cells. Cumulative data from three independent experiments are presented. B, flow-cytometric analysis for CD39 and CD73expression onCD4-gated Tcells after 3days of culture in presenceof various concentrationof TGFb. C, qPCRanalysis of keyglycolysis regulatinggenes (left),and memory/stemness-associated genes (right) in either Th17TGFb1, Th17IL1b cells, or Th17IL1b cells cultured in presence of 250 pg/mL TGFb (i.e.,Th17IL1bþTGFb cells). D, OT-II CD4þ T cells were polarized toward different Th17 types and restimulated with either cognate antigen (ova323-339) ornonspecific antigen (MART-1) for 4 hours. Cell death was determined by evaluating Annexin V versus 7AAD by flow cytometry (left) as detailed in theSupplementary Methods. Bar diagram (right), the percentage of Annexin V– and 7AAD-positive cells from three different experiments. E, C57BL/6 Rag1�/�

mice (n ¼ 4–5 mice/group) were inoculated (s.c.) with 0.25 � 106 B16-F10-ova murine melanoma cells and after 7 days, mice were either kept untreated ascontrol or adoptively transferred with either 1 � 106 ova–specific Th17IL1b or Th17IL1bþTGFb OT-II (Vb5þCD4þ) cells. Tumor growth was measuredusing digital calipers every fourth day. Data, mean tumor size at each time point in one of the three experiments with similar results. F, NSG-A2 mice(n¼5mice/group)were inoculatedwith 2.5�106HLA-A2þhumanmelanoma 624-MELcells and after 15 days,micewere either kept untreated or treatedwithh3Tmouse-derived 0.2� 106 human tyrosinase epitope–reactive Th17TGFb1 or Th17IL1bþTGFb cells. Tumor growthwasmeasured using digital calipers every 3day. Data, mean tumor size at each time point; �, P < 0.05; ��, P < 0.005; and ��, P < 0.0001.






comparative analysis of glycolytic pathway molecules andstemness gene signature between three different Th17 popula-tions revealed that Th17IL1bþTGFb cells exhibited intermediateglycolysis (Th17IL1b >Th17IL1bþTGFb >Th17TGFb1), and stem-ness (Th17TGFb1 >Th17IL1bþTGFb >Th17IL1b) gene signature(Fig. 6C, left and right). In addition, a TCR restimulation–induced AICD was decreased in Th17IL1bþTGFb cells as indi-cated by the lower percentage of cells with Annexin V and7AAD positivity (Fig. 6D). These data also confirm a recentobservation that T cells differentiated in presence of glycolysisinhibitor 2-deoxy glucose programs for better antitumor con-trol and persistence (34). Importantly, the Th17IL1bþTGFb cellsalso showed significant improvement in the ability to controlboth B16 murine melanoma (Fig. 6E), and 624-MEL humanmelanoma (Fig. 6F). The tumors in mice treated with theTh17IL1bþTGFb group did not reach half the tumor endpoint(<100mm2) until sacrificed on day 70, indicating the long-termpersistence of Th17IL1bþTGFb cells in vivo (Supplementary Fig.S2C). Thus, Th17 cells generated ex vivo with IL1b and lowconcentrations of TGFb program antitumor T cells optimallyfor metabolic commitment and persistence, which in turnaffect the ability to control tumor growth long-term.

Overall, our data suggest that modifying ex vivo cultureconditions to generate an effective hybrid Th17þTh1 cells foradoptive immunotherapy would benefit from strategies thattarget to increase effector signature, glycolytic potential, per-sistence, and concomitantly reducing exonucleotidase CD73expression–induced immunosuppression (Supplementary Fig.S3). We believe that this study will uncover important aspectsthat need to be considered when generating potent tumor-specific T cells, to improve future T-cell cancer immunother-apy trials.

DiscussionCellular cancer therapies based on stimulating the immune

systemof the patient represent an important treatmentmodal-ity, but much remains to be understood to optimize their use.Several biologic mechanisms may account for their failure toachieve efficient immune protection (35). One important con-founding factor is immunosuppression by the tumor micro-environment (35, 36). In addition, recent studies have shownthat after T cells infiltrate to the tumor, there is a heightenedimmunosuppression in the host due to effector T cells them-selves (7). As cancer immunotherapy develops, it is, thus,particularly important to understand the impact of ACTtreatments in tumor immunosuppression and immunity.

Recent development in T-cell–mediated therapeuticapproach to control tumor growth has led to the developmentof various protocols to ex vivo program not only Th1/Tc1 cells,but other subsets as Th9/Tc9 and Th17/Tc17 for tumor control(5, 6, 37). Although Th17 cells have shown promise in control-ling tumor growth (5), the stable phenotype of these cells hasbeen called in question due to their ability to convert to FoxP3-expressing regulatory Th17 cells (13). Studies have shown thatTGFb-cultured Th17 cells express ectonucleotidase CD73 andexhibit the ability to suppress the effector T cells (19, 38). Inaddition, the TGFb-mediated expression of ectonucleotidaseon Th17 cells is also found in commensal bacteria–rich intes-

tinal lamina propria with high concentration of luminal ATP,which may be responsible for their maintenance and patho-genicity (39). However, a recent study has addressed that thecytokine requirements for Th17 cell polarization in vivodepends on the site of priming, and revealed key differencesby which the systemic, mucosal, and cutaneous immunesystems guide Th17 cell lineage commitment with IL1b in anirreplaceable role (40). Given these issues of Th17 plasticity andconversion to regulatory phenotype, along with the variabilityof cytokine environment in regulating Th17 generation, weaddressed as to how a stable Th17 cell with ability to controlthe tumor growth long-term could be programmed ex vivo. Ourdata substantiate the potential contribution of ectonucleoti-dase CD73 expression in self-suppression of the effector Th17cells generated using the conventional methodwith TGFb, andpropose the strategies to program long-lived effector Th17 cellsby combining inflammatory cytokine IL1b along with a lowdose of TGFb (that does not induce CD73, but upregulatesstemness genes). This, we believe could be an important stepforward to generate robust hybrid Th17þTh1 effector cells thatcould be readily translated to clinics when treating patientswith melanoma or other cancers.

Expression of ectonucleotidases CD39 and CD73 on tumorcells has been shown to contribute the immunosuppression bytheir sequential action of converting ATP to adenosine (41).Although blocking CD73 expression on tumor cells has shownto improve tumor control (42, 43), the engraftment of tumorand its metastases was also reported to be lower in the CD73-KO mice (44). Similarly, the role of adenosine generationcatalyzed by CD39/CD73–expressing Tregs in immunosup-pression is also established (45). A recent study has also shownthat combining anti-CD73 treatment with anti–CTLA-4 andanti-PD1 antibody (negative regulators of T-cell activation)results in improved tumor control (42). These studies implythat expression of either host-derived CD73 on tumors or itsexpression on the adaptive T-cell subsets may be a key con-tributor in tumor progression. Our data now show that theexpression of CD73 on T cell itself could also lead to itsincreased susceptibility to suppression and loss of effectorfunction. Because TGFb is the key contributor of CD73 expres-sion, we evaluated herein the T-cell subsets that are pro-grammed in TGFb, that is, Th17 cells. Ex vivo-generated Th17cells that are programmed with TGFb have been shown to bebetter than Th1 cells at controlling tumors (5), paradoxicallythey also express CD73 (19). We thus compared the differencesin Th17 cells thatwere programmed either in presence of TGFbor in absence of TGFb (but with IL1b). The improved ability ofIL1b-programmed Th17 cells to secrete higher level of IFNg ,express enhanced level effector molecules, and control tumorcould be due to the direct effect of IL1b on both CD4 andCD8Tcells that leads to activation of multiple pathways, as reportedpreviously (46, 47). Studies using autoimmune EAE modelshave shown that IL1b treatment increases the expression ofpathogenic genes that results in increased incidence of disease(20, 48). The increased expression of IL23R and other effectormolecules that resulted in increased pathogenesis in theautoimmune model may have been responsible for renderingtumor epitope–specific Th17IL1b more efficacious than

Chatterjee et al.





Th17TGFb cells in targeting "self" but tumor-associated epitopeand controlling tumor growth. Our data also show that IL1b-cultured Th17 cells exhibit increased level of glucose trans-porter Glut-1 that correlates to increased glucose consump-tion, which fuels the metabolic need of the rapidly dividingglycolytically active effector Th17IL1b cells. Our data alsosuggest that the expression of CD39/CD73 on Th17IL1b cellsdoes not inversely correlates with the level of glycolysis in Tcells, as we did not notice any increase in expression of theseectonucleotidases with the decrease in glycolysis (Supplemen-tary Fig. S2A, left). Contrary to the earlier studies that showCD73 expression correlates closely with HIF1a expressioneither in intestinal epithelia or gastric carcinoma (49, 50); exvivo programming of T cells with IL1b (that promotes HIF1a)in absence of TGFb inhibited the expression of CD39/73(Supplementary Fig. S2A, right). Importantly, although theseglycolytically active Th17IL1b cells do exhibit increased abilityto control tumors in the short term, they do not persist well inthe host . The decreased persistence of Th17IL1b cells correlateswith decreased expression of "stemness" signature that hasbeen otherwise shown to be a key feature of Th17 cells (10, 11).Our data show that it may be the presence of TGFb in theculture conditions that renders the "stemness" to Th17TGFb1

cells, and thus Th17IL1b cells do not exhibit stemness. However,addition of low concentration of TGFb (250 pg/mL) as opposedto the normally used 3 to 5 ng/mL does not results in upregula-tion of CD73, but increases stemness in the Th17IL1b cells. Thedetailed analysis also shows that even at 250 pg/mL the level ofglycolysis also drops, and these cells with intermediate level ofglycolysis and stemness are better than the cells that are eitherhighly glycolytic with low stemness (Th17IL1b) or exhibit highstemness but lower glycolysis (Th17TGFb1).Finally, our approach to use the alternative strategy for

generating antitumor helper T cells by combining Th17IL1b

culture conditions with low-dose TGFb could have transla-tional potential owing to long-termpersistence and substantial

improvement in tumor control. The difference in antitumoreffector phenotype based on metabolic commitment alsoenforces a key role of cellular energy requirements in regulatingantitumor function. Overall, this study may significantly for-ward our understanding of the factors that control long-termand stable antitumor T-cell functions.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception anddesign: S. Chatterjee, K. Thyagarajan, C.M. Paulos, S.MehrotraDevelopment of methodology: S. Chatterjee, P. Kesarwani, J. Fu, S. MehrotraAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): K. Thyagarajan, P. Kesarwani, M. Soloshchenko,S.R. Bailey, A.S. Kraft, C.M. PaulosAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): S. Chatterjee, K. Thyagarajan, P. Kesarwani,J.H. Song, J. Fu, X.-Z. Yu, S. MehrotraWriting, review, and/or revision of the manuscript: S. Chatterjee,P. Kesarwani, J.H. Song, J. Fu, S.R. Bailey, C. Vasu, C.M. Paulos, X.-Z. Yu,S. MehrotraAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): P. Kesarwani, M. Soloshchenko,S. MehrotraStudy supervision: S. Mehrotra

AcknowledgmentsThe authors acknowledge help from Drs. Bei Liu and Radhika Gudi at MUSC

for reagents and their valuable suggestions while preparing this article.

Grant SupportThe work was supported in part by funds from Department of Surgery

(MUSC) and NIH R01CA138930, R01AR057643, and PO1 CA154778. Flow Cyto-metry and Cell Sorting unit is supported in part by NIH P30 CA138313 grant toHollings Cancer Center at MUSC.

The costs of publication of this article were defrayed in part by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received May 15, 2014; revised July 11, 2014; accepted July 29, 2014;published OnlineFirst September 9, 2014.

References1. Rosenberg SA. Raising the bar: the curative potential of human cancer

immunotherapy. Sci Transl Med 2012;4:127ps8.2. Ngiow SF, Teng MW, Smyth MJ. Prospects for TIM3-targeted antitu-

mor immunotherapy. Cancer Res 2011;71:6567–71.3. Peggs KS, Quezada SA, Allison JP. Cancer immunotherapy: co-

stimulatory agonists and co-inhibitory antagonists. Clin Exp Immunol2009;157:9–19.

4. Klebanoff CA, Finkelstein SE, Surman DR, Lichtman MK, Gattinoni L,Theoret MR, et al. IL-15 enhances the in vivo antitumor activity oftumor-reactive CD8þ T cells. Proc Natl Acad Sci U S A 2004;101:1969–74.

5. Muranski P, Boni A, Antony PA, Cassard L, Irvine KR, Kaiser A, et al.Tumor-specific Th17-polarized cells eradicate large established mel-anoma. Blood 2008;112:362–73.

6. Lu Y, Hong S, Li H, Park J, Hong B, Wang L, et al. Th9 cells promoteantitumor immune responses in vivo. J Clin Invest 2012;122:4160–71.

7. Spranger S, Spaapen RM, Zha Y, Williams J, Meng Y, Ha TT, et al. Up-regulation of PD-L1, IDO, and T(regs) in the melanoma tumor micro-environment is driven by CD8(þ) T cells. Sci Transl Med 2013;5:200ra116.

8. Kryczek I, Banerjee M, Cheng P, Vatan L, Szeliga W, Wei S, et al.Phenotype, distribution, generation, and functional and clinical rele-vance of Th17 cells in the human tumor environments. Blood2009;114:1141–9.

9. Wang L, Yi T, Kortylewski M, Pardoll DM, Zeng D, Yu H. IL-17 canpromote tumor growth through an IL-6-Stat3 signaling pathway. J ExpMed 2009;206:1457–64.

10. Muranski P, BormanZA, Kerkar SP, Klebanoff CA, Ji Y, Sanchez-PerezL, et al. Th17 cells are long lived and retain a stem cell–like molecularsignature. Immunity 2011;35:972–85.

11. Kryczek I, Zhao E, Liu Y, Wang Y, Vatan L, Szeliga W, et al. HumanTH17 cells are long-lived effector memory cells. Sci Transl Med2011;3:104ra00.

12. YuY, IclozanC,Yamazaki T,YangX,AnasettiC,DongC,et al. Abundantc-Fas-associated death domain–like interleukin-1-converting enzymeinhibitory protein expression determines resistance of T helper 17 cellsto activation-induced cell death. Blood 2009;114:1026–8.

13. Tartar DM, VanMorlan AM, Wan X, Guloglu FB, Jain R, Haymaker CL,et al. FoxP3þRORgammat þT helper intermediates display suppres-sive function against autoimmune diabetes. J Immunol 2010;184:3377–85.






14. Mehrotra S, Al-Khami AA, Klarquist J, Husain S, Naga O, Eby JM,et al. A coreceptor-independent transgenic human TCR mediatesanti-tumor and anti-self immunity in mice. J Immunol 2012;189:1627–38.

15. Chatterjee S, Eby JM, Al-Khami AA, Soloshchenko M, Kang HK, KaurN, et al. A quantitative increase in regulatory T cells controls devel-opment of vitiligo. J Invest Dermatol 2014;134:1285–94.

16. Ferrick DA, Neilson A, Beeson C. Advances in measuring cellularbioenergetics using extracellular flux. Drug Discov Today 2008;13:268–74.

17. Pellegatti P, Raffaghello L, Bianchi G, Piccardi F, Pistoia V, Di Virgilio F.Increased level of extracellular ATP at tumor sites: in vivo imaging withplasma membrane luciferase. PLoS ONE 2008;3:e2599.

18. Ghiringhelli F, Bruchard M, Chalmin F, Rebe C. Production of aden-osine by ectonucleotidases: a key factor in tumor immunoescape.J Biomed Biotechnol 2012;2012:473712.

19. Chalmin F, Mignot G, Bruchard M, Chevriaux A, Vegran F, Hichami A,et al. Stat3 and Gfi-1 transcription factors control Th17 cell immuno-suppressive activity via the regulation of ectonucleotidase expression.Immunity 2012;36:362–73.

20. Ghoreschi K, Laurence A, Yang XP, Tato CM, McGeachy MJ, KonkelJE, et al. Generation of pathogenic T(H)17 cells in the absence of TGF-beta signalling. Nature 2010;467:967–71.

21. Bengsch B, Seigel B, Flecken T, Wolanski J, Blum HE, Thimme R.Human Th17 cells express high levels of enzymatically active dipepti-dylpeptidase IV (CD26). J Immunol 2012;188:5438–47.

22. McGeachy MJ, Chen Y, Tato CM, Laurence A, Joyce-Shaikh B,Blumenschein WM, et al. The interleukin 23 receptor is essential forthe terminal differentiation of interleukin 17-producing effector T helpercells in vivo. Nat Immunol 2009;10:314–24.

23. Mus AM, Cornelissen F, Asmawidjaja PS, van Hamburg JP, Boon L,Hendriks RW, et al. Interleukin-23 promotes Th17 differentiation byinhibiting T-bet and FoxP3 and is required for elevation of interleukin-22, but not interleukin-21, in autoimmune experimental arthritis. Arthri-tis Rheum 2010;62:1043–50.

24. Duhen R, Glatigny S, Arbelaez CA, Blair TC, Oukka M, Bettelli E.Cutting edge: the pathogenicity of IFN-gamma-producing Th17 cellsis independent of T-bet. J Immunol 2013;190:4478–82.

25. Ghoreschi K, Laurence A, Yang XP, Hirahara K, O'Shea JJ. T helper 17cell heterogeneity and pathogenicity in autoimmune disease. TrendsImmunol 2011;32:395–401.

26. Starbeck-Miller GR, Xue HH, Harty JT. IL-12 and type I interferonprolong the division of activated CD8 T cells by maintaining high-affinity IL-2 signaling in vivo. J Exp Med 2014;211:105–20.

27. Pearce EL, Pearce EJ. Metabolic pathways in immune cell activationand quiescence. Immunity 2013;38:633–43.

28. Sena LA, Li S, Jairaman A, Prakriya M, Ezponda T, Hildeman DA, et al.Mitochondria are required for antigen-specific T-cell activationthrough reactive oxygen species signaling. Immunity 2013;38:225–36.

29. Chang CH, Curtis JD, Maggi LB Jr, Faubert B, Villarino AV, O'SullivanD, et al. Posttranscriptional control of T-cell effector function byaerobic glycolysis. Cell 2013;153:1239–51.

30. Cham CM, Gajewski TF. Glucose availability regulates IFN-gammaproduction and p70S6 kinase activation in CD8þ effector T cells.J Immunol 2005;174:4670–7.

31. Badovinac VP, Corbin GA, Harty JT. Cutting edge: OFF cycling of TNFproduction by antigen-specific CD8þ T cells is antigen independent.J Immunol 2000;165:5387–91.

32. Corbin GA, Harty JT. T cells undergo rapid ON/OFF but not ON/OFF/ON cycling of cytokine production in response to antigen. J Immunol2005;174:718–26.

33. Karlsson S. Is TGF-beta a stemness regulator? Blood 2009;113:1208.34. Sukumar M, Liu J, Ji Y, Subramanian M, Crompton JG, Yu Z, et al.

Inhibiting glycolytic metabolism enhances CD8þ T cell memory andantitumor function. J Clin Invest 2013;123:4479–88.

35. Rabinovich GA, Gabrilovich D, Sotomayor EM. Immunosuppressivestrategies that are mediated by tumor cells. Annu Rev Immunol2007;25:267–96.

36. DunnGP,Old LJ, Schreiber RD. The three Esof cancer immunoediting.Annu Rev Immunol 2004;22:329–60.

37. LuY,HongB, LiH, ZhengY, ZhangM,WangS, et al. Tumor-specific IL-9-producingCD8þ Tc9 cells are superior effector than type-I cytotoxicTc1 cells for adoptive immunotherapy of cancers. Proc Natl Acad SciU S A 2014;111:2265–70.

38. Regateiro FS, Howie D, Nolan KF, Agorogiannis EI, Greaves DR,Cobbold SP, et al. Generation of anti-inflammatory adenosine byleukocytes is regulated by TGF-beta. Eur J Immunol 2011;41:2955–65.

39. Atarashi K, Nishimura J, Shima T, Umesaki Y, YamamotoM, OnoueM,et al. ATP drives lamina propria T(H)17 cell differentiation. Nature2008;455:808–12.

40. HuW,Troutman TD, Edukulla R, PasareC. Primingmicroenvironmentsdictate cytokine requirements for T helper 17 cell lineage commitment.Immunity 2011;35:1010–22.

41. Zhang B. CD73: a novel target for cancer immunotherapy. Cancer Res2010;70:6407–11.

42. AllardB,PommeyS,SmythMJ,StaggJ. TargetingCD73enhances theantitumor activity of anti-PD-1 and anti-CTLA-4 mAbs. Clin CancerRes 2013;19:5626–35.

43. Zhang B. Opportunities and challenges for anti-CD73 cancer therapy.Immunotherapy 2012;4:861–5.

44. Stagg J, Beavis PA, Divisekera U, Liu MC, Moller A, Darcy PK, et al.CD73-deficient mice are resistant to carcinogenesis. Cancer Res2012;72:2190–6.

45. Deaglio S, Dwyer KM, Gao W, Friedman D, Usheva A, Erat A, et al.Adenosine generation catalyzed by CD39 and CD73 expressed onregulatory T cells mediates immune suppression. J Exp Med 2007;204:1257–65.

46. Ben-Sasson SZ, Hogg A, Hu-Li J, Wingfield P, Chen X, Crank M, et al.IL-1 enhances expansion, effector function, tissue localization, andmemory response of antigen-specific CD8 T cells. J Exp Med 2013;210:491–502.

47. Ben-Sasson SZ, Hu-Li J, Quiel J, Cauchetaux S, Ratner M, Shapira I,et al. IL-1 acts directly on CD4 T cells to enhance their antigen-drivenexpansion and differentiation. Proc Natl Acad Sci U S A 2009;106:7119–24.

48. Lovett-Racke AE, Yang Y, Racke MK. Th1 versus Th17: are T-cellcytokines relevant in multiple sclerosis? Biochim Biophys Acta2011;1812:246–51.

49. Synnestvedt K, Furuta GT, Comerford KM, Louis N, Karhausen J,EltzschigHK, et al. Ecto-50-nucleotidase (CD73) regulation byhypoxia-inducible factor-1 mediates permeability changes in intestinal epithe-lia. J Clin Invest 2002;110:993–1002.

50. LuXX,ChenYT, FengB,MaoXB,YuB,ChuXY.Expressionandclinicalsignificance of CD73 and hypoxia-inducible factor-1alpha in gastriccarcinoma. World J Gastroenterol 2013;19:1912–8.


Chatterjee et al.




Published OnlineFirst September 9, 2014.Cancer Res Shilpak Chatterjee, Krishnamurthy Thyagarajan, Pravin Kesarwani, et al. Improves Immunotherapeutic Control of Tumors

-Programmed Th17 CellsβReducing CD73 Expression by IL1

Updated version

10.1158/0008-5472.CAN-14-1450doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2014/09/06/0008-5472.CAN-14-1450.DC1

Access the most recent supplemental material at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. (CCC)Click on "Request Permissions" which will take you to the Copyright Clearance Center's

.http://cancerres.aacrjournals.org/content/early/2014/10/15/0008-5472.CAN-14-1450To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-14-1450

http://cancerres.aacrjournals.org/content/suppl/2014/09/06/0008-5472.CAN-14-1450.DC1

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/early/2014/10/15/0008-5472.CAN-14-1450


Reducing CD73 Expression by IL1b-Programmed …...by Th17TGFb1 cells subset with CD39 þCD73...

Documents

Transcript of Reducing CD73 Expression by IL1b-Programmed …...by Th17TGFb1 cells subset with CD39 þCD73...