Regeneration of CD8ab T Cells from Derived iPSC Imparts ...Cancer Res; 76(23); 6839–50. 2016 AACR. Introduction During the last several years, cancer immunotherapy using a variety

Microenvironment and Immunology

Regeneration of CD8ab T Cells fromT-cell–Derived iPSC Imparts Potent TumorAntigen-Specific CytotoxicityTakuya Maeda1,2, Seiji Nagano1,2, Hiroshi Ichise1, Keisuke Kataoka3, Daisuke Yamada4,Seishi Ogawa3, Haruhiko Koseki4, Toshio Kitawaki2, Norimitsu Kadowaki5,Akifumi Takaori-Kondo2, Kyoko Masuda1, and Hiroshi Kawamoto1

Abstract

Although adoptive transfer of cytotoxic T lymphocytes (CTL)offer a promising cancer therapeutic direction, the generation ofantigen-specific CTL from patients has faced difficulty in efficientexpansion in ex vivo culture. To resolve this issue, several groupshave proposed that induced pluripotent stem cell technology beapplied for the expansion of antigen-specific CTL, which retainexpression of the same T-cell receptor as original CTL.However, inthese previous studies, the regenerated CTL are mostly of theCD8aaþ innate type and have less antigen-specific cytotoxicactivity than primary CTL. Here we report that, by stimulating

purified iPSC-derived CD4/CD8 double-positive cells with anti-CD3 antibody, T cells expressing CD8ab were generated andexhibited improved antigen-specific cytotoxicity compared withCD8aaþ CTL. Failure of CD8ab T-cell production using theprevious method was found to be due to killing of double-positive cells by the double-negative cells in the mixed cultures.We found that WT1 tumor antigen-specific CTL regenerated bythis method prolonged the survival of mice bearing WT1-expres-sing leukemic cells. Implementation of our methods may offer auseful clinical tool. Cancer Res; 76(23); 6839–50. �2016 AACR.

IntroductionDuring the last several years, cancer immunotherapy using a

variety of strategies has progressed remarkably, for example,blockade of inhibitory signals in the immune system (1–3),adoptive transfer of cytotoxic T lymphocytes (CTL; refs. 4–6),transfer of T-cell receptor (TCR) genes (7–9), etc. Although theseachievements may suggest that CTLs present in cancer patientshave the potential to cure them, researchers and clinicians havebeen facing a major obstacle in getting enough antigen-specificCTLs for therapy. Some groups have reported that tumor infil-trating lymphocytes (TIL) or tumor antigen specific CTLs can beefficiently expanded (10–13), but in general it is not so easy. Thisproblem is mainly attributable to the nature of CTLs; they easilybecome exhausted or die during cultivation.

To solve this problem, we came to the idea of utilizing inducedpluripotent stem cell (iPSC) technology for the cloning and

expansion of CTLs. When iPSCs are established from antigen-specific T cells (T-iPSC), they should inherit rearranged TCRgenes, and thus all T cells regenerated fromT-iPSCs should expressthe same TCR. Because iPSC expansion in vitro is almost unlim-ited, it should be possible to obtain asmany freshCTLs as needed.In keeping with this idea, we have recently succeeded in regen-erating MART1-specific CTLs from iPSCs originally derived fromCTLs of a melanoma patient (14). Other groups also have regen-erated viral antigen-specific T cells (15, 16), T cells expressing aninvariant TCRs (17–19), and T cells that were genetically engi-neered to express a so-called chimeric antigen receptor (CAR;ref. 20).

None of the previous studies, however, showed whether theseregenerated CTLs are as good as primary CTLs in terms of antigen-specific cytotoxic activity. Indeed, in the study using CAR-T-iPSCs,the authors stated that regenerated CTLs are phenotypically sim-ilar to gdT cells (20).

Authentic CTLs are CD4� CD8aþ CD8bþ (CD8ab hetero-dimerþ). CD8aa type T cells represent a minor population inmost lymphoid tissues but are one of the major populationsin the mucosal tissues, where they are regarded as "innatetype" T cells. Of note, the CD8aa homodimer does notfunction well as a TCR coreceptor, because it does not effi-ciently bind to MHC class I molecules (21). Although regen-erated CD8 T cells in the previous studies all seemed to be ofthe innate type, in this study we describe a novel yet simplemethod to generate CD8ab T cells and show that they areactive against leukemia.

Materials and MethodsStudy approval

This study was approved by the Institutional Review Boardof the Graduate School of Medicine, Kyoto University

1Laboratory of Immunology, Institute for Frontier Life and Medical Sciences,Kyoto University, Kyoto, Japan. 2Department of Hematology and Oncology,Graduate School of Medicine, Kyoto University, Kyoto, Japan. 3Department ofPathology and Tumor Biology, Graduate School of Medicine, Kyoto University,Kyoto, Japan. 4Laboratory for Developmental Genetics, Riken Center for Inte-grative Medical Science (IMS), Yokohama, Japan. 5Department of InternalMedicine, Division of Hematology, Rheumatology and Respiratory Medicine,Faculity of Medicine, Kagawa University, Kagawa, Japan.

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

Corresponding Author: Hiroshi Kawamoto, Institute for Frontier Life andMedical Sciences, KyotoUniversity, 53Kawahara-cho, Shogoin, Sakyo-ku, Kyoto606-8507, Japan. Phone: (81)-75-751-3815; Fax: (81)-75-751-3839; E-mail:[email protected]

doi: 10.1158/0008-5472.CAN-16-1149

�2016 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 6839

on January 14, 2021. © 2016 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst November 21, 2016; DOI: 10.1158/0008-5472.CAN-16-1149

http://cancerres.aacrjournals.org/

(approval number: G761) and abided by the tenets of theDeclaration of Helsinki. All specimens from healthy indivi-duals and patients were collected after written informedconsent was obtained.

Cell linesOp9,OP9/DLL1, and 409B2were purchased fromRIKENBRC.

THP1 was purchased from ATCC. HL60, HEL, and KG1a weremaintained in our department. OUN1 and MEG01 were giftsfrom Dr. Masaki Yasukawa (Ehime University, Ehime, Japan).K562 was a gift from Dr. Kiyotaka Kuzushima (Aichi CancerCenter, Aichi, Japan). C1R-A�24:02 was a gift from Dr. MasafumiTakiguchi (Kumamoto University, Kumamoto, Japan). Uponreceiving the cell lines, frozen stocks were prepared within onetofive passages andnew stockswere thawed frequently to keep theoriginal condition. The cell lines were passaged for less than 3

months after receipt or resuscitation. Theywere also authenticatedby morphology, growth rate, and surface phenotype, especiallyexpression of HLA class I.

Differentiation of T-iPSCs to CD8ab single positive cellsT-iPSCs were differentiated to CD4/8 double-positive (DP)

cells using the OP9 and OP9/DLL1 stromal cell coculture systemsas described (14, 22, 23), with slight modification. In brief, iPSCcolonies were dissociated using trypsin (0.25%) and collagenaseIV (1 mg/mL) and mechanically disrupted into small clumps bypipetting. About 600 iPSC clumps were collected and plated ongelatin precoated OP9 dishes filled with OP9 medium, that is,a-MEM (Invitrogen) with 20% FCS. On day 13, colonies weretreated with collagenase Type IV (50 U/mL) and trypsin-EDTA(0.05%). Cells were plated in a OP9/DLL1 semiconfluent dish onOP9mediumcontaining hIL-7 (5 ng/mL), hFlt-3L (5ng/mL), and

0

20

40

60

80

100

100 pmol/L

0 mol/L

A CB

H

E

F

D

LMP

2 T

etra

mer

Week 0 Week 4

0

20

40

60

80

100

0

20

0 mol/

L

100 p

mol/L

1 μmol/

L

1 nmol/

L

10 nm

ol/L

100 n

mol/L

40

60

80

100

0

20

40

60

80

100

68.9%

7.7%

56.3%

10 nmol/L

100 nmol/L

1 nmol/L

0 mol/L

1 nmol/L

10 nmol/L

Original CTL

Regenerated CD8αα CTL

LMP

2 T

etra

mer

CD3

CD

4

CD8α CD3C

D4

CD

8β

68.0%

12.0% 87.7%

LMP

2 T

etra

mer

K562

Ann

exin

V+

(%)

E:T Ratio

E:T Ratio E:T Ratio

LMP2 Peptide concentration

97.7%

G

Ann

exin

V+ (

%)

Ann

exin

V+ (

%)

Ann

exin

V+ (

%)

CD8α

CD8α

Figure 1.

Regeneration of CD8aa innate type T cells by the conventional method. A, Flow cytometric profiles of PBMC of a healthy volunteer before and afterstimulation with LMP2 peptide pulsed LCLs, which were used as antigen-presenting cells. LMP2-specific CTLs were detected as the CD8aþ LMP2-tetramerþ

fraction (boxed). B, In vitro cytotoxic assay of LMP2-specific CTLs from A against HLA-A�24:02þ THP1 cells pulsed with LMP2 peptide at differentconcentrations. E:T ratio, effector-to-target ratio. C, Photomicrograph of T-iPSCs; LMP2#1-1 established from the LMP2-specific CTLs. Scale bar, 500 mm. D,Flow cytometric profiles of cells generated from T-iPSCs; LMP2#1-1 at day 40 of culture. E, Flow cytometric profiles of cells generated from those shownin D after stimulation with CD3 Ab for 6 days. F, In vitro cytotoxic assay of regenerated CD8aa T cells against THP1 targets pulsed with LMP2 peptideat different concentrations. Representative of three independent experiments. G, Comparison of the cytotoxic activity between the original CTLs andthe regenerated CD8aa T cells against THP1 targets at different concentrations of LMP2 peptide. The E:T ratio was fixed at 3:1. H, In vitro cytotoxic assayof regenerated CD8aa T cells against K562 cells. Representative of three independent experiments.

Maeda et al.

Cancer Res; 76(23) December 1, 2016 Cancer Research6840




hSCF (5 ng/mL). On day 15, semiadherent cells were collectedand passage into a new dish layered with OP9/DLL1 cells. Fromthis point, passagewas done every 7 days.Onday 40,floating cellswere collected and CD4/8 DP cells were enriched by using CD4microbeads (Milteny Biotec). DP cells were stimulated with CD3Ab (15 ng/mL; OKT-3, eBioscience) in the presence of hIL-2 (100U/mL) and hIL-7 (5 ng/mL). CD8 SP cells were stimulated one to

five times by HLA-A�24:02þ lymphoblastoid cell line (LCL)pulsed with the peptide in the presence of hIL-7 (5 ng/mL) andhIL-21 (10 ng/mL).

Coculture of immature DP and double-negative cellsInduced DP and double-negative (DN) cells (day 40) from

T-iPSCs LMP2#1-1 were isolated and labeled with Cell Trace

C

D

A

B

76.9%98.8%

80.9%

0

1

2

3

Bulk DP DN

CD8αβSP

97.3%

CD3

72.5%

96.0%

64.9%

7.8%

89.8%

13.1%

87.0%

8.4%

72.4%

89.1%

88.1%

5.7%

83.3%

CD8ααSP

Control

CD3 Ab

99.9%

CD8α

CD

4

CD8α

CD

4

CD8α

CD

4

CD8α

CD

4

CD8α

CD

8β

LMP

2 T

etra

mer

CD3

CD8α

CD

8β

CD

4

LMP

2 T

etra

mer

Day 1 Day 2 Day 4Day 0

CD

8β

CD

4

Fol

d ga

inCD8 SP Gated

80.4%

Day 40 Day 46

Isol

atio

nBulk

CD8α

Cell trace

Figure 2.

Induction of CD8ab SP cells from DPcells. A, Flow cytometric profiles ofgenerated immature T cells on day 40(top left), purified CD4/8 DP cells(middle left), purified CD4/8 DN cells(bottom left), and activated T cellsderived from total cells (top right),purified CD4/8 DP cells (middle right)or purified CD4/8 DN cells (bottomright) after stimulation with CD3 Abfor 6 days. B, Fold gain of regeneratedCD8ab SP cells and CD8aa SP cellsfrom each population as described inA. Representative of threeindependent experiments. C, Flowcytometric profiles of regeneratedT cells derived from CD4/8 DP cellsstimulated by LCL pulsed with LMP2peptide. D, Time course analysis fromDP cells to CD8ab SP cells by CD3 Abstimulation. The bottom row showsthe dilution of Cell Trace Violet. Redlines andblue lines depict cellswithoutand with CD3 Ab, respectively.

Regeneration of Potent Tumor Antigen-Specific CTL from iPSC

www.aacrjournals.org Cancer Res; 76(23) December 1, 2016 6841




Day 40

DPViolet+

DNCFSE+

Mix

CD3 Ab

After 5 hoursAnnexin V/PI stain

CFSE+

CD8SP Gated

0

20

40

60

1:0 3:1 1:1 1:3

H

Ann

exin

V/P

I–/–

(%) Control

CD3 Ab

DP:DN Ratio

CD3 Ab

G

Control

DP:DN = 1:0 DP:DN = 3:1 DP:DN = 1:1 DP:DN = 1:3

41.7% 44.9% 48.2%

36.6% 14.4%

44.7%

5.8%

Ann

exin

V

PI

69.1%

19.0%

DP:DN = 1:1

Violet+ gated CFSE+ Gated

9.2%

I

00.10.20.30.40.50.6

1:0 3:1 1:1 1:3Ann

exin

V/P

I–/–

(fol

d nu

mbe

r)

DP:DN Ratio

Control

CD3 Ab

J

Granzyme A

CD8α

CD

4

DP

DN

29.6%

17.0%

0.1%

0.0%

Per

forin

0

0.5

1

1.510:0 9:1 5:5 1:9 0:10

CD8α

Cel

l tra

ce v

iole

t

CFSE

Violet+

gated

DP:DN=10:0

9:1

5:5

1:9

0:10Violet+ CFSE+

Fol

d ga

inB

C

D

0

0.5

1

1.5

2

Violet+ CFSE+

DP onlyDN onlyDP+DN MixedDP+DN Transwell

Fol

d ga

in

CD

4

CD

8β

CD

4

CD

8β

Violet+

CD8SP gated

CFSE+

Gated

N.D

.

N.D

.

N.D

.

N.D

.

DP Cells Violet+

DN Cells CFSE+

+ CD3 Ab

E

A

F

Day 40

DPViolet+

DNCFSE+

mix

CD3 Ab

Day 6FACS assay

Maeda et al.





Figure 3.DN cells kill DP cells upon stimulation. A, Scheme of the DP and DN coculture experiment for B and C. Violetþ purified CD4/8 DP cells and CFSEþ DN cells weremixed at various ratios and cultured with CD3 Ab. B, Flow cytometric profiles of generated cells from DP and DN cells as depicted in A. DP-derivedcells and DN-derived cells were detected as Cell Trace Violet and CFSE positive cells, respectively. C, Fold gain of CD8ab SP cells from each populationdescribed in B. Representative of three independent experiments. N.D., not done. D, Scheme of the transwell coculture experiment for examining theeffect of soluble factors. Violetþ purified CD4/8 DP cells and CFSEþ DN cells were cultured in the lower and upper compartments, respectively. E, Fold gainof CD8ab SP cells derived from DP cells (Violetþ) and DN cells (CFSEþ) in each culture condition. Isolated DP cells, isolated DN cells, DP and DN cells mixed,and DP cells and DN cells cocultured using transwell as depicted in D were stimulated with CD3 Ab for 6 days. Representative of three independentexperiments. F, Scheme of the cytotoxic assay of DN cells against DP cells for G. Violetþ purified CD4/8 DP cells and CFSEþ DN cells were mixed atvarious ratios, cultured with CD3 Ab for 5 hours, and stained with Annexin V and PI. G, Flow cytometric profiles of DP (Violetþ) or DN (CFSEþ) cells as shown inF. The percentages of Annexin V/PI�/�; live cells are highlighted by the boxes. H and I, Cytotoxic assay of DN cells against DP cells. Plot showspercentage (H) and fold number (I) of Annexin V/PI�/�; live cells among Violetþ cells as shown in G. Representative of three independent experiments.J, Flow cytometric analysis of intracellular granzyme A and perforin expression in DP and DN cells at day 40 of culture without TCR stimulation.

A

F

0

4

8

CXCR3ID2

CD28CCR7ID3FOXP1FASPRDM1LEF1IFNGKLRG1ZEB2KLF2SELLCD27TCF7TBX21IL2RB

TIM3ITK

GNLYGZMBCD3E

FAS

LCKGZMAPRF1LATZAP70LAG3CD8BCD28CTLA4CD4ITGA2PDCD1ITGA1LTABCL11B

CD8A

ICAM1FASLGNOTCH1TRAIL

HIF1B

SPI1RORCTOX2SOX13ID2RUNX3ETS1IKZF1TBX21BATFZBTB16EOMESRUNX1BCL6FOXO1TOXFOXO3AHRFOXP1ID3PRDM1BCL11BNFIL3GATA3GFI1HIF1ATCF3

12Value

60

Cou

nt 500

Color key and histogram

B

E

C D

CCL5IL2RG

IL2CCR3IFNGR1IL15RAIFNGIL21RIL7RCCR4CCR7CCR2CCR1CCR6CXCR6CCR5TNFIL2RACXCR4IL4RIL16CCL3CCL4CXCR3IL2RBTGFB1

DNAM1

RORCNCAM1NCR2FCGR3AKIR2DL4KIR2DS4KIR2DL1KIR2DL3KIR3DL1SOX13KIR3DL2KIR3DL3NKG2DNCR3KLRB1DAP12NKG2CNKG2EID3BCL6EGR2NCR1KLRD1ZBTB16NKG2A

F

0

4

8

CXCR3ID2

CD2DD 8CCR7ID3FOXOO PXX 1FASPRDM1LEF1EEIFNGKLRG1ZEBEE 2BBKLF2SELEE LCD2DD 7TCTT F7TBXBB 2XX 1IL2RB

TIM3ITK

GNLNN YGZMBCD3E

FAS

LCKGZMAPRF1LATZAP7077LAG3CD8BCD2DD 8CTLA4CD4ITGAGG 2PDCD1ITGAGG 1LTABCL11B

CD8A

ICAM1FASLGNOTCTT H1TRAIL

HIF1B

SPI1RORCTOXOO 2XXSOX13ID2RUNX3UUETS1IKZF1KKTBXBB 2XX 1BATFZBTBTT 16EOMESRUNX1UUBCL6FOXOO O1XXTOXFOXOO O3XXAHRFOXOO PXX 1ID3PRDM1BCL11BNFIL3GAGG TA3GFIFF 1HIF1ATCTT F3

12Value

60

Cou

nt 505050000

Color keyand histogram

DDDNANN M1

RRRORCRORCRRNNNCACC M1NNNCR2RRFFFCGR3A33KKKIRKK 2RR DL4KKKIRKK 2RR DS4KKKIRKK 2RR DL1KKKIRKK 2RR DL3KKKIRKK 3DL1SSSOX13KKKIRKK 3DL2KKKIRKK 3DL3NNNKG2KK DNNNCR3KKKLRB1DDDAP12NNNKG2KK CNNNKG2KK EIIID3BBBCL6EEEGR2NNNCR1KKKLRD1ZZBTBTT 16NNNKG2KK A22

Figure 4.

Gene expression profiles of regenerated CD8ab SP cells. A–E, Heat map comparing expression of the indicated mRNA transcripts detected by RNA sequencing ofregenerated CD8ab T cells (r-CD8ab), regenerated CD8aa T cells (r-CD8aa), peripheral blood CD8ab T cells (p-CD8ab), and peripheral blood CD8aa T cells(p-CD8aa). The FPKMvalueswere log2 transformed. The heatmaps represent different ontologygroups, including genes encoding for cytotoxicity (A), transcriptionfactors (B), NK-associated (C), cytokines and chemokines (D), and T-cell maturation markers (E). F, Unsupervised hierarchical clustering of the indicatedtranscriptomes generated by RNA sequencing. Bars above the heat maps indicate similarities between the different samples presented in each of the lanes. p-gdT,peripheral blood gd T cells.






Violet andCFSE, respectively. A total of 105 cells of theDP andDNcell mixture at the indicated ratios were cultured with CD3 Ab (15ng/mL). After 6 days, cells were analyzed by flow cytometer.

Transwell coculture of immature DP and DN cellsThe transwell system was used for the detection of soluble

mediator effects. Labeled DP and DN cells (5� 104 cells each) asabove were seeded in the lower and upper compartments, respec-tively, of a 0.4mmpore size transwell plate (Coaster), and culturedin the presence of CD3Ab (15 ng/mL). In control wells, DP orDNcells (5 � 104 cells) or their mixture (5 � 104 cells each) werecultured with CD3 Ab. After 6 days, cells were analyzed by flowcytometer.

Cytotoxic assay of DN cells against DP cellsLabeled DP and DN cells as above were cocultured at the

indicated ratio with or without CD3 Ab for 5 hours. Annexin V(BioLegend) and PI double-negative cells were considered asalive.

RNA sequencing and data analysisRegenerated LMP2-specific CD8aa T cells and CD8ab T cells

were isolatedby FACSAria III (BDBioscience). Isolated peripheralblood CD8aa T cells (TCRabþ CD4� CD8aþ CD8b�), CD8ab Tcells (TCRabþ CD4� CD8aþ CD8bþ), and gdT cells (TCRgdþ

TCRab�) were stimulated with CD3/28 stimulator (MiltenyBiotec) for 7 days according to the manufacturer's instructionsin the presence of hIL-7 (10 ng/mL) and hIL-15 (20 ng/mL). Total

RNA was isolated using an RNeasy Plus Mini Kit (QIAGEN).Libraries for RNA sequence were prepared using NEBNext UltraRNA Library Prep Kit for Illumina (New England Biolabs).Sequence alignment was performed using our in-house pipelinesas previously described (24). Fragments per kilobase of exon permillion fragments mapped (FPKM) was calculated for Refseqgenes and then log-transformed. The hierarchical clustering ofgene expression data was performed using Pearson's dissimilarityas distance measure and centroid method for linkage. For thisanalysis, we removed genes with no or low expression (meanFPKM < 2), then identified the most variant 5% of genes (n ¼527).Multi Experiment Viewer softwarewas used to generate heatmaps and hierarchical clustering. GEO accession number is GSE81975.

Statistical analysisAll data with error bars are presented as mean� SD. Difference

was assessed using paired t-test using Prism (GraphPad software).Values of P < 0.05 were considered significant.

ResultsCD8 T cells induced by the conventional method are of theinnate type, expressing a CD8aa homodimer

Latent membrane protein 2 (LMP2) is an Epstein–Barrvirus (EBV)-encoded antigen that is considered as a good CTLtarget in EBV-related tumors (25, 26). LMP2-specific CTLs, defin-ed by LMP2 tetramer staining, from a healthy volunteer were

Target- 0 mol/L100 pmol/L

mol/

Lm

ol/Lm

ol/Lm

ol/Lm

ol/Lm

ol/Lm

ol/L

1 nmol/L10 nmol/L 100 nmol/L1 µmol/L

0

20

40

60

80

100

0

20

40

60

80

100

A

0

20

40

60

80

100

mol/

Lm

ol/Lm

ol/Lm

ol/Lm

ol/L

mol/

L

0 mol/L

100 pmol/L

10 nmol/L

1 nmol/L

p-CTL LMP2#1

r-CD8αβLMP2#1-1

B

Ann

exin

V+ (

%)

Ann

exin

V+ (

%)

Ann

exin

V+ (

%)

Ann

exin

V+ (

%)

E

E:T Ratio

E:T RatioLMP2 Peptide concentration

r-CD8αα

r-CD8αβ

0

10

20

30

40

50

60

70Target- 0 mol/L100 pmol/L 1 nmol/L10 nmol/L 100 nmol/L1 µmol/L

TNFαIFNγ

IFN

γ/T

NF

α-p

rodu

cing

cel

ls (

%)

0

20

40

60

80

100p-CTL

LMP2#2

r-CD8αβLMP2#2-13

C

GF

LMP2 Peptide concentration

CD8α

CD

4

CD8α

CD

8β

CD3

LMP

2 te

tram

er

94.7%

95.1%

98.9%

IFN

γ(p

g /m

L)

101

102

103

104

105D

p-CTLLMP2#1

r-CD8αβLMP2#1-1

Figure 5.

Cytotoxic activity of regenerated CD8ab SP cells. A, In vitro cytotoxic assay of regenerated CD8ab T cells, r-CD8ab LMP2#1-1, against THP1 cells pulsed withLMP2 peptide at different concentrations. Representative of three independent experiments. B, Comparison of the cytotoxicity between the primary CTLs, p-CTLLMP2#1, and r-CD8ab LMP2#1-1, against THP1 cells pulsed with different concentrations of LMP2 peptide. The effector-to-target (E:T) ratio was fixed at 3:1. C,Comparison of IFNg production between p-CTL LMP2#1 and r-CD8ab LMP2#1-1 against THP1 cells pulsed with different concentrations of LMP2 peptide. D,Intracellular cytokine staining of r-CD8ab LMP2#1-1 against THP1 with different concentration of LMP2 peptide. Representative of three independent experiments.E, NK-like cytotoxicity of regenerated CD8aa T cells (r-CD8aa) and r-CD8ab against K562 cells. Representative of three independent experiments. F, Flowcytometric profiles of regenerated T cells derived from T-iPSCs, LMP2#2-13. G, Cytotoxic assay of p-CTL LMP2 #2 and r-CD8ab LMP2#2-13 against THP1 cellspulsed with different concentrations of LMP2 peptide. The E:T ratio was fixed at 3:1.

Maeda et al.





expanded by stimulation with antigen-presenting cells loadedwith LMP2 peptide (Fig. 1A). After several weeks of expansion,these antigen-specific CTLs efficiently killed leukemic cells loadedwith LMP2 peptide at 1 nmol/L concentration (Fig. 1B).

For establishment of iPS cells from CTLs, we used SV40 Large Tantigen (27, 28) in addition to Yamanaka factors to enhance theefficiency of reprogramming (Supplementary Fig. S1). For trans-duction of CTLs, we used Sendi virus vector, which has beenshown to be safe because transferred genes as well as viral genesare not integrated into genome (29, 30). By transducing theseCTLs with Yamanaka factors and SV40, two clones of iPSC lines,LMP2#1-1 and LMP2#2-13,were established. Sendai virus vectorscontaining Yamanaka factors and SV40were not detected in eitherof these clones, and expression profiles of various stem cell geneswere indistinguishable from those of control iPSCs (Supplemen-tary Fig. S2A and S2B). Hereafter, we will focus on the cloneLMP2#1-1 (Fig. 1C). Expression of stem cell genes in the LMP2#1-1 cell line was confirmed by cytochemistry, it had a normalkaryotype, and was pluripotent, as judged by teratoma formation(Supplementary Fig. S2C–S2E).

We then induced T cells from LMP2-T-iPSCs using theconventional method (14). As has been reported, DP and DNcells expressed TCR on Day 40 (Fig. 1D). Upon stimulationwith CD3 Ab, CD8 T cells were generated after 6 days and wereexclusively of the CD8aa type and LMP2-tetramer positive(Fig. 1E). However, as much as 100 nmol/L peptide wasrequired to induce efficient cytotoxic activity (Fig. 1F), makingthese cells �100-fold less competent than primary CTLs(Fig. 1G). However, the regenerated CTLs were able to effi-ciently kill K562 (Fig. 1H), indicating that they have highNatural Killer (NK)-like cytotoxicity.

Isolated DP cells give rise to CD8ab T cells upon stimulationWe searched then for conditions in which CD8ab T cells

could be induced. We found that when purified DP cells werestimulated with CD3 Ab, CD8SP cells were generated and theywere mostly of CD8ab type (Fig. 2A). By contrast, purified DNcells gave rise to CD8aa T cells upon activation (Fig. 2A and B).The CD8ab T cells could also be efficiently generated frompurified DP cells stimulated by autologous LCLs loaded withLMP2 peptide (Fig. 2C), indicating that the absence of DN cells,rather than the stimulation method, is critical for the inductionof CD8ab T cells.

To exclude the possibility that a preexisting small number ofCD8ab cells were preferentially expanded, purified DP cellswere labeled with Cell Trace Violet, and then stimulated

CD3

CD45RO

0 2 4 6 8

02468

10

101Fol

d E

xpan

sion

Fol

d ex

pans

ion

0

20

40

60

80

100

Ann

exin

V+ (

%)

Ann

exin

V+ (

%)

CD

8β

C

D

A B

CD8α CD3

CD

4

WT

1 te

tram

er

CD8α

84.7% 0.3% 89.2%

E

44.9%

WT

1 T

etra

mer

Week 0

CD8α

Week 6

Anti-CD3Ab

F

CD

45R

AW

T1

Tet

ram

er

Week 2 Week 6 Week 12

72.4%

8.1%

13.5%

66.9%

0 mol/L

100 pmol/L

10 pmol/L

1 nmol/L

Weeks

102

103

104

1

47.9%

27.6%

Day 46Day 40

105

Exp1

Exp2Exp3

E:T Ratio

99.8%

95.5%

CD

4

0

20

40

60

80

100

H

0

20

40

60

80

100G

0 mol/L

100 pmol/L

10 pmol/L

1 nmol/L

p-CTLWT1#3

r-CD8αβ WT1#3-3

WT1 Peptide concentrationE:T Ratio

mLmLmLCon

trol

LCL+

Peptid

e

mol/

Lm

ol/L

mol/

Lm

ol/L

mol/

Lm

ol/L

Figure 6.

Regeneration of functional WT1 specific CTLs. A, Flow cytometric profiles ofPBMC from a healthy volunteer before and after stimulation with WT1 peptidepulsed LCLs. Cells were stained with CD8a Ab and WT1-tetramer. B, In vitrocytotoxic assay of WT1-specific CTLs #3 against C1R-A�24:02 cells pulsed with

WT1 peptide at different concentrations. C, Flow cytometric profiles of cellsregenerated from T-iPSCs, WT1#3-3. The far left panel shows the profile ofpurified DP cells at day 40 and right three panels show the profiles ofregenerated T cells after 6 days' stimulationwith CD3Ab.D, Fold expansion of r-CD8abWT1#3-3 after stimulation by autologous LCLpulsedwithWT1 peptide orby immobilized CD3 Ab in 7 days. Representative of three independentexperiments. E, Fold expansion of r-CD8ab WT1#3-3 by repeated stimulation.CD8ab T cells were stimulated every week by coculturing with LCLs pulsed withWT1 peptide (100 nmol/L). Plot shows the data of three independentexperiments. Exp, experiment. F, Flow cytometric profiles of the phenotype ofCD8ab T cells repeatedly stimulated with LCLs. G, In vitro cytotoxic assay ofr-CD8ab WT1#3-3 against C1R-A�24:02 cells pulsed with WT1 peptide atdifferent concentrations. Representative of three independent experiments. H,Comparison of cytotoxicity between p-CTL WT1#3 and r-CD8ab WT1#3-3against C1R-A�24:02 cells loaded with different concentrations of WT1 peptide.The effector-to-target (E:T) ratio was fixed at 3:1.






0 25 50 75 100 125 150

0

20

40

60

80

100A

E:T Ratio E:T Ratio

THP1HL60

0

20

40

60

80

100

HLA block+

HLA blockHLA block–

HLA block–

HLA block–

HLA block–

–

HLA block+

HLA block+ HLA block+

HLA block+

B

0

20

40

60

HL60 THP1 OUN1 MEG01 K562 HEL KG1a

Effector– Effector+ HLA block

3.6x106 3.8x106 8.8x106 1.4x106 9.6x105 8.1x105 2.4x106

464% 224% 36.0% 28.6% 10.4% 139% 352%

HLA-A2402+ HLA-A2402–

Copy/µgRNA

%ABL

Ann

exin

V+ (

%)

Ann

exin

V+ (

%)

0

20

40

60

0

20

40

60

0

20

40

60

80

100C

Ann

exin

V+

(%

)

Copy/µgRNA

%ABL

13.9x106 1.9x106

339% 86.3% 200%

5.2x106

D

2x104 HL60 IP Injection

NOG Mouse

IL2 and IL7 Day 1~ three times/week

F

CTL

PBS

(Days)

Sur

viva

l (%

)

P < 0.0001

100

80

60

40

20

0

Control mice

CTL-treated mice

Day 37

Day 37

Day 53

Day 53

E#3 #4#2#1 #5

Human CD8α

Hum

an C

D33

Hum

an C

D33

G

0

0.002

0.004

0.006

BM Spleen PB

Hum

an C

D8+

(%

)

PBS/r-CD8αβ WT1#3-35x106/mouse day 1~

once/week

Figure 7.

Cytotoxicity of regeneratedWT1-specific CD8ab SP cells in vitro and in vivo.A,HLA-dependent cytotoxicity of r-CD8abWT1#3-3 against leukemia cell lines expressingHLA-A�24:02þ. HL60andTHP1,which express endogenousWT1protein,were used as target cells. Target cells, pretreatedwith orwithoutHLAclass I blockingAb,werecocultured with r-CD8ab WT1#3-3. Representative of three independent experiments. B, In vitro cytotoxic assay of r-CD8ab WT1#3-3 against various leukemiacell lines.HLA-A�24:02positiveor negative leukemiacell lineswereusedas target cells. Target cells, pretreatedwith orwithoutHLAclass I blockingAb,were coculturedwith r-CD8ab WT1#3-3. The effector-to-target (E:T) ratio was fixed at 3:1. WT1 mRNA expression in each cell line was quantified by real-time PCR and depictedbelow thegraphas absolute valueand relative value toABL.Representative of three independent experiments.C, Invitro cytotoxic assays of r-CD8abWT1#3-3againstprimary AML cells expressing WT1 protein and HLA-A�24:02. Primary leukemia cells from each patient, pretreated with or without HLA class I blocking Ab,were co-culturedwith r-CD8abWT1#3-3.WT1mRNAexpression in eachof theprimary leukemia cellswasmeasured and shownbeloweachgraph as inB.D,Treatmentscheme for the xenograft leukemia model. NOG mice were injected intraperitoneally with 2 � 104 HL60 leukemia cells. (Continued on the following page.)

Maeda et al.





(Fig. 2D). On day 2, DP cells started to downregulate CD4without cell division, and then CD8ab cells were generated onday 4 with only one or two cell divisions, indicating thatCD8SP cells were induced from DP cells.

DN cells kill DP cells upon stimulationThus, it seemed that DN cells might be exerting a suppressive

effect on the generation of CD8ab T cells. To test this possibility,isolated DP and DN cells were mixed at various ratios beforestimulation, and it was seen that DN cells at a one ninth ratio toDP cells almost completely suppressed the generation of CD8abT cells (Fig. 3A–C).

The possibility that DN cells suppress DP-SP development bysecreting soluble factors is unlikely, because virtually no sup-pression of CD8ab T-cell generation was seen in a transwellexperiment in which DP cells and DN cells were separated by asemipermeable membrane (Fig. 3D and E). We then observedthat the DP cells died as early as 5 hours after activation incultures with a mixture of DN and DP cells, and that thefrequency of live DP cells decreased along with an increase inthe DN/DP ratio (Fig. 3F–I), indicating that activated DN cellsdirectly kill DP cells. We further found that a significantproportion of DN cells expressed perforin and granzyme Aprior to activation (Fig. 3J). These results indicate that failure ofDP cells to generate CD8ab T cells after activation is attribut-able to the direct cytotoxicity of DN cells against DP cells. Theentire culture procedure and profiles of generated cells aredepicted in Supplementary Fig. S3.

Gene expression profiles of regenerated CD8ab T cellsWe analyzed gene expression profiles of regenerated CD8ab

T cells (r-CD8ab) in comparison with regenerated CD8aa T cells(r-CD8aa) and primary CD8ab and CD8aa T cells (p-CD8ab, p-CD8aa), using RNA-seq. Not only surface molecules, for exam-ple, CD3, CD8a, CD8b, and CD28 but also intracellular func-tional molecules, for example, GZMA, GZMB, PRF1, LCK, andBCL11B were commonly expressed in r-CD8ab and p-CD8ab(Fig. 4A). As to immune-checkpoint molecules, LAG3 wasexpressed in rCD8 cells but PDCD1 and HAVCR2 (Tim3) werevery low in expression. The signature transcription factors of CD8T cells including TBX21, RUNX3, GATA3, and IKZF1 were com-monly expressed in all four groups, but ZBTB16 (PLZF), a rep-resentative transcription factor for innate type T cells (31–33),waslow in both r-CD8ab and p-CD8ab compared with both types ofCD8aa cells (Fig. 4B). ReflectingNK-like cytotoxicity of r-CD8aa,NK-related genes, for example, NCAM1, NCR1, NCR2, KLRC2,andKLRC3were highly expressed in r-CD8aa. Although r-CD8abalso express some of them (KLRC2 and KLRC3), but others werenot expressed (Fig. 4C). However, r-CD8aa and r-CD8ab sharedsome common characteristics in terms of chemokine receptorexpression, such as high CXCR3 and low CCR7 expression, whichare reminiscent of effector CD8 T cells (Fig. 4D and E). In globalgene expression pattern, r-CD8ab were positioned closer to p-

CD8ab compared to r-CD8aa (Fig. 4F). Taken together, theseobservations indicate that gene expression profiles of r-CD8abresembled with that of p-CD8ab, although retaining some innatelymphocyte signature.

Regenerated CD8ab T cells exhibit antigen-specific cytotoxicitycomparable to the original CTLs

Regenerated CD8ab T cells killed antigen-loaded target cellsat a 1 nmol/L peptide concentration (Fig. 5A), and theirantigen-specific cytotoxic activity was found to be slightlyweaker than that of primary CTLs, but much stronger than thatof regenerated CD8aa T cells (Fig. 5B, see Fig. 1G for compar-ison). Regenerated CD8ab T cells showed comparable potentialto primary CTLs in producing IFNg , and produced IFNg andTNFa at more than 1 nmol/L concentration of peptide (Fig. 5Cand D). NK-like cytotoxicity against K562 cells was weaker thanregenerated CD8aa T cells (Fig. 5E). These results collectivelyindicate that this novel method works well to regeneratefunctional CD8ab T cells. Another LMP2-specific T-iPSC clone#2-13 also differentiated to CD8ab T cells (Fig. 5E) andexhibited peptide-specific cytotoxicity slightly weaker than theoriginal CTLs (Fig. 5F).

Regeneration of WT1-specific CTLsWe then decided to apply this method against acute mye-

loid leukemia (AML) cells expressing WT1 antigen, a cancer-testis antigen broadly expressed in various types of solidtumors as well as in leukemia (34, 35). We firstly inducedWT1-specific CTLs from a healthy volunteer by using WT1peptide (Fig. 6A) and measured their cytotoxic activity(Fig. 6B). We then produced iPSCs from these CTLs, and atotal of three lines were established (Supplementary Fig. S4and Supplementary Table S1). Hereafter we will show the datausing clone WT1#3-3. Regenerated cells were of the CD8abtype and WT1-tetramer positive (Fig. 6C), and the sequence ofthe TCR genes in this clone was determined (SupplementaryTable S1).

We then expanded CD8 T cells by stimulating with an autol-ogous B LCL carryingWT1 peptide, or with a CD3Ab, and a six- toeight-fold expansion was seen during 1 week in both cases (Fig.6D). We decided to use LCL for further studies because theyseemed slightly more efficient. During 6 to 8 weeks' culture, CD8T cells were expanded by �10,000 fold (Fig. 6E), changing theirsurface phenotype from a na€�ve to an effector/memory profile(Fig. 6F). Regenerated CD8 T cells showed the same antigen-specific cytotoxic activity as the original CTLs (Fig. 6G and H).

Regenerated WT1-specific CTLs have cytotoxicity againstleukemia cells

Regenerated WT1-CTLs also efficiently killed THP1 andHL60 cells, HLA-matched AML cell lines expressing endoge-nous WT1 protein (Fig. 7A). The cytotoxic activity was almostcompletely eliminated by adding a blocking Ab against HLA

(Continued.) Beginning at day 1, PBS or 5� 106 r-CD8abWT1#3-3were injected intraperitoneally into tumor-bearingmice everyweek for 4weeks. hIL-2 andhIL-7wereinjected intraperitoneally three times a week. E, Flow cytometric profiles of peripheral blood from control or CTL-treated mice. CD33 versus CD8 profilesgated on human CD45 positive cells are shown. HL60 and inoculated CTLswere detected as the human CD33þ or CD8þ fraction. Representative of three independentexperiments. F, A Kaplan–Meier curve depicting the percent survival of the experimental and control groups is shown (n ¼ 15). Accumulation of threeindependent experiments. P < 0.05 by the log-rank test. G, Percentage of resident regenerated CTLs in indicated samples from NOG mice inoculated with r-CD8abWT1#3-3 6 months before (n ¼ 7). NOG mice bearing no leukemia cells were inoculated with 5 � 106 r-CD8ab WT1#3-3 and cytokines intraperitoneally in the samemanner as shown in D. Each dot shows the data from each mouse.






class I, indicating that the killing was MHC restricted anddependent on TCR-MHC binding. Other AML cell lines OUN1and MEG01 were also killed, but the cytotoxic activity wasonly partially suppressed by HLA blocking Ab, suggesting thatthey were also killed by NK-like cytotoxicity (Fig. 7B). OtherAML lines, HEL and KG1a, which express WT1 but not expressHLA-A�24:02, were not killed, confirming that the TCR ofWT1#3-3 cells recognizes WT1 antigen presented on HLA-A�24:02. We further tested whether regenerated CTLs are ableto kill primary leukemic cells. Three samples expressing thehighest levels of endogenous WT1 antigen selected amongseveral samples from HLA-A�24:02þ patients were killed byregenerated CTLs (Fig. 7C).

Finally, we examined CTL activity in vivo in a xenograft model.NOD.Cg-Prkdcscid Il2rgtm1Sug/Jic (NOG)mice inoculated withHL60 cells followed by administration of regenerated CTLs (Fig.7D). Although leukemia cells expressing human CD33 weredetected from all the control mice, inoculated regenerated CTLswere present in peripheral blood of all the treated mice (Fig. 7E).The CTL treated mice showed significantly longer survival com-pared with control mice (Fig. 7F).

Regenerated WT1-specific CTLs are safeTo test the safety of transfused regenerated CTLs, we did

long-term observation in vivo. NOG mice bearing no leukemiccells were inoculated with regenerated CTLs in the same man-ner as in xenograft model (Fig. 1D), and observed for 6months. During observation period, no mice showed diarrheaor body weight loss. After 6 months, all mice were dissected,and mononuclear cells in bone marrow, spleen, and peripheralblood were analyzed. Small amount of regenerated CTLs weredetected from spleen and bone marrow of some mice (Fig. 7G).No mice suffered from tumor derived from regenerated CTLs.We also did not see any sign of tissue damage caused bytransferred CTLs.

DiscussionMany groups have been struggling to produce CD8ab type T

cells from T-iPSCs, but the critical method described hereturned out to be rather simple, simply include a DP cellpurification step before stimulation. Our results can be sum-marized as three points: upon TCR stimulation, (i) DP cells giverise to CD8ab T cells, (ii) DN cells immediately kill DP cells,(iii) DN cells give rise to CD8aa T cells. For the first point, thealternative ligand model of thymic selection may provide theexplanation. According to this model, DP cells positively select-ed in the cortex by peptides specifically expressed there cansurvive in the medulla because they do not encounter the samepeptide; thus, the agonist peptide can induce positive selectionin the cortex (36, 37). It may be that our culture systemreproduces such a cortical, positively selecting microenviron-ment. As to the second point, we have shown that DN cellsalready contain mature cells (Fig. 3J), although they do not killother cells unless activated further via their TCR. It seems likelythat DN cells kill DP cells by direct NK-like cytotoxicity;however, it remains to be studied what the target moleculesare in this killing. As to the third point, in our system, the TCR isexpressed at the DN stage (Supplementary Fig. S3C). Whenthese cells are activated, they may undergo a developmentalprogram toward gdT cells, because it is known that the DN cell

fate decision is based on TCR signaling strength. DN cells takethe gdT cell fate upon receiving a strong TCR signal via thegdTCR, whereas they take the abT cell fate when there is a weaksignal via the preTCR (38, 39). This model is also in line withprevious reports showing that DN cells expressing a transgenicabTCR at the DN stage come to exhibit a gdT cell-like pheno-type (40).

Regenerated CD8ab T cells showed antigen-specific cytotox-icity slightly less, but nearly comparable to primary CD8abT cells (Supplementary Table S2). This is a marked improve-ment compared to regenerated CD8aa T cells and might partlybe explained because the CD8ab heterodimer binds to MHCclass I and thus functions as a co-receptor to induce lcksignaling, whereas the CD8aa homodimer does not. Geneexpression profiles also showed that regenerated CD8ab T cellsare closer to primary CD8ab T cells, although still retainingsome phenotypic features of innate type T cells. As activatedT cells express some NK activation receptors and signal mole-cules like NKG2D, NKp30, DAP12, and so on (41–43), it is notsurprising that these regenerated CTLs cultured in vitro alsoexpress such NK-associated molecules and possess NK-likecytotoxic potential.

There is growing evidence that na€�ve T cells, stem cell mem-ory T cells (TSCM), or central memory T cells (TCM) are superioras source of adoptive T-cell transfer for tumor immunotherapy(44–47). In terms of surface markers, regenerated CD8ab Tcells expressing low CCR7 and high CXCR3 are considered aseffector phenotype T cells, although we are using IL21 to inhibitterminal differentiation (5, 48). Nevertheless, our culture meth-od has advantages that the regenerated CD8ab T cells can beexpanded more than 10,000 fold by repeated TCR stimulationwhile retaining comparable TCR-specific cytotoxic potential.Furthermore, it was reported to be hard to evaluate in vitrocultured CTLs just based on their surface markers with regard tothe point whether they are terminally differentiated or not (49).In addition, at present, it is also difficult to appropriatelyaddress this issue in vivo in xenograft model, because micro-environments in immune system of mouse xenograft model areso far from physiological situation in human. In case of murineCTLs regenerated from T-iPSCs, it was recently shown that theregenerated CTLs exhibited memory response in vivo in micebearing syngeneic tumor (50).

One may concern about the safety of regenerated CTLs istumorigenecity after administration. Although we use SV40 largeT antigen to establish T-iPSCs, we confirmed that SV40 wascompletely deleted from all the iPSCs. Also we examined thelong-term safety in transfused mice, and confirmed all the micewere free of tumor derived from regenerated CTLs.

Another concern about safety is off-target specificity. In ourmethod almost all the regenerated CD8ab T cells after expan-sion by LCL pulsed with cognate peptide retain the same TCRdetected as tetramer positive cells (Fig. 2C, 5F, 6C, 6F). To beexact, this is not the case at the DP stage, where RAG1 and RAG2genes are expressed and secondly rearrangement in TCR alphachain gene may occur. Indeed, a part of the CD8ab T cells justregenerated by TCR stimulation have lost specificity for originalepitope, resulting in the formation of negative population intetramer staining (Supplementary Fig. S3E, top right). Howev-er, these cells can be eliminated during expansion by LCLpulsed with cognate peptide (Supplementary Fig. S3E, bottomright). We actually did not see any sign of Graft-versus-Host

Maeda et al.





reaction in the mice inoculated with regenerated CTLs(Fig. 7G). In addition, it is also possible to completely ensureclonality by deleting RAG1 or RAG2 gene in T-iPSCs usinggenome editing technology.

In this method starting from 104 T-iPSCs in one 10-cmculture dish, we can eventually obtain 109 to 1010 regeneratedCD8ab T cells in one round culture, which could be sufficientfor one transfusion to a patient, thus providing a strongrationale for clinical application of this strategy. We haveestablished antigen-specific T-iPSCs from healthy volunteers,making an allogeneic approach possible. In such an allogeneicsetting, it would be preferable to produce T-iPSCs from an HLA-haplotype homozygous donor, because regenerated CTLs couldthen be used for HLA-haplotype heterozygous patients. In sucha strategy, "Off the shelf" T-cell drug would come true bygenerating CD8ab T cells in advance and preserving as frozenstocks.

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

Authors' ContributionsConception and design: T. Maeda, N. Kadowaki, H. KawamotoDevelopment of methodology: T. Maeda, S. Nagano, H. Ichise, D. Yamada,H. Koseki, T. Kitawaki, K. MasudaAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): T. Maeda, S. Nagano, H. Ichise, K. Kataoka, S. Ogawa,H. Koseki, K. Masuda

Analysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): T. Maeda, K. Kataoka, D. Yamada, K. MasudaWriting, review, and/or revision of the manuscript: T. Maeda, A. Takaori-Kondo, K. Masuda, H. KawamotoAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): T. Maeda, S. Nagano, K. Kataoka, A. Takaori-Kondo, K. MasudaStudy supervision: T. Maeda, N. Kadowaki, A. Takaori-Kondo, H. Kawamoto

AcknowledgmentsWe thank Eri Satoh and Toshika Senba for technical support; Mahito

Nakanishi and Manami Ohtaka for kindly providing Sendai virus of Yama-naka factors; Haruo Sugiyama and Fumihiro Fujiki (Osaka University),Masahiro Kawahara (Shiga University of Medical Science), and YutakaShimazu and Masaki Miyazaki (Kyoto University) for helpful discussion;and Peter Burrows (University of Alabama) for critical reading of themanuscript.

Grant SupportThis work was supported by a Project for the Development of Innovative

Research on Cancer Therapeutics (P-Direct) from the Ministry of Education,Culture, Sports, Science and Technology, Japan, by Astlym Co. Ltd, and byRegCell, Inc.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received April 27, 2016; revised August 30, 2016; accepted September 17,2016; published OnlineFirst November 21, 2016.

References1. Hodi FS,O'Day SJ,McDermott DF,Weber RW, Sosman JA,Haanen JB, et al.

Improved survivalwith ipilimumab in patients withmetastaticmelanoma.N Engl J Med 2010;363:711–23.

2. Topalian SL,Hodi FS, Brahmer JR,Gettinger SN, SmithDC,McDermottDF,et al. Safety, activity, and immune correlates of anti-PD-1 antibody incancer. N Engl J Med 2012;366:2443–54.

3. Wolchok JD, Kluger H, Callahan MK, Postow MA, Rizvi NA, Lesokhin AM,et al. Nivolumab plus ipilimumab in advanced melanoma. N Engl J Med2013;369:122–33.

4. Rosenberg SA, Yang JC, Sherry RM, Kammula US, Hughes MS, Phan GQ,et al. Durable complete responses in heavily pretreated patients withmetastatic melanoma using T-cell transfer immunotherapy. Clin CancerRes 2011;17:4550–7.

5. Chapuis AG, RagnarssonGB,NguyenHN,ChaneyCN, Pufnock JS, SchmittTM, et al. TransferredWT1-reactive CD8þ T cells canmediate antileukemicactivity and persist in post-transplant patients. Sci Transl Med 2013;5:174ra27.

6. Doubrovina E, Oflaz-Sozmen B, Prockop SE, Kernan NA, Abramson S,Teruya-Feldstein J, et al. Adoptive immunotherapy with unselected orEBV-specific T cells for biopsy-proven EBVþ lymphomas after allogeneichematopoietic cell transplantation. Blood 2012;119:2644–56.

7. Morgan RA, Dudley ME, Wunderlich JR, Hughes MS, Yang JC, Sherry RM,et al. Cancer regression in patients after transfer of genetically engineeredlymphocytes. Science 2006;314:126–9.

8. Johnson LA, Morgan RA, Dudley ME, Cassard L, Yang JC, Hughes MS, et al.Gene therapy with human and mouse T-cell receptors mediates cancerregression and targets normal tissues expressing cognate antigen. Blood2009;114:535–46.

9. Robbins PF,Morgan RA, Feldman SA, Yang JC, Sherry RM,DudleyME, et al.Tumor regression in patients with metastatic synovial cell sarcoma andmelanoma using genetically engineered lymphocytes reactive with NY-ESO-1. J Clin Oncol 2011;29:917–24.

10. Dudley ME, Wunderlich JR, Shelton TE, Even J, Rosenberg SA.Generation of tumor-infiltrating lymphocyte cultures for use in adop-

tive transfer therapy for melanoma patients. J Immunother 2003;26:332–42.

11. HoWY,NguyenHN,WolflM,Kuball J, Greenberg PD. In vitromethods forgenerating CD8þ T-cell clones for immunotherapy from the na€�ve reper-toire. J Immunol Methods 2006;310:40–52.

12. MausMV, Thomas AK, LeonardDG, AllmanD, Addya K, Schlienger K, et al.Ex vivo expansion of polyclonal and antigen-specific cytotoxic T lympho-cytes by artificial APCs expressing ligands for the T-cell receptor, CD28 and4-1BB. Nat Biotechnol 2002;20:143–8.

13. Butler MO, Lee JS, Ans�en S, Neuberg D, Hodi FS, Murray AP, et al. Long-lived antitumor CD8þ lymphocytes for adoptive therapy generated usingan artificial antigen-presenting cell. Clin Cancer Res 2007;13:1857–67.

14. Vizcardo R, Masuda K, Yamada D, Ikawa T, Shimizu K, Fujii S, et al.Regeneration of human tumor antigen-specific T cells from iPSCs derivedfrom mature CD8(þ) T cells. Cell Stem Cell 2013;12:31–6.

15. Nishimura T, Kaneko S, Kawana-Tachikawa A, Tajima Y, Goto H, Zhu D,et al. Generation of rejuvenated antigen-specific T cells by reprogram-ming to pluripotency and redifferentiation. Cell Stem Cell 2013;12:114–26.

16. Ando M, Nishimura T, Yamazaki S, Yamaguchi T, Kawana-Tachikawa A,Hayama T, et al. A safeguard system for induced pluripotent stem cell-derived rejuvenated T cell therapy. Stem Cell Reports 2015;5:597–608.

17. Wakao H, Yoshikiyo K, Koshimizu U, Furukawa T, Enomoto K,MatsunagaT, et al. Expansion of functional human mucosal-associated invariant Tcells via reprogramming to pluripotency and redifferentiation. Cell StemCell 2013;12:546–58.

18. Kitayama S, Zhang R, Liu TY, Ueda N, Iriguchi S, Yasui Y, et al. Cellularadjuvant properties, direct cytotoxicity of re-differentiated Va24 invariantNKT-like cells from human induced pluripotent stem cells. Stem CellReports 2016;6:213–27.

19. Yamada D, Iyoda T, Vizcardo R, Shimizu K, Sato Y, Endo TA, et al.Efficient regeneration of human Va24þ invariant natural killer T cellsand their anti-tumor activity in vivo. Stem Cells 2016 Jul 16. doi:10.1002/stem.2465.






20. Themeli M, Kloss CC, Ciriello G, Fedorov VD, Perna F, Gonen M, et al.Generation of tumor-targeted human T lymphocytes from inducedpluripotent stem cells for cancer therapy. Nat Biotechnol 2013;31:928–33.

21. Cheroutre H, Lambolez F. Doubting the TCR coreceptor function ofCD8alphaalpha. Immunity 2008;28:149–59.

22. Schmitt TM, de Pooter RF, Gronski MA, Cho SK, Ohashi PS, Z�u~niga-Pfl€ucker JC. Induction of T cell development and establishment of T cellcompetence from embryonic stem cells differentiated in vitro. Nat Immu-nol 2004;5:410–7.

23. Timmermans F, Velghe I, Vanwalleghem L, De Smedt M, Van Cop-pernolle S, Taghon T, et al. Generation of T cells from humanembryonic stem cell-derived hematopoietic zones. J Immunol 2009;182:6879–88.

24. Kataoka K, Nagata Y, Kitanaka A, Shiraishi Y, Shimamura T, Yasunaga J,et al. Integrated molecular analysis of adult T cell leukemia/lymphoma.Nat Genet 2015;47:1304–15.

25. Lee SP, Tierney RJ, Thomas WA, Brooks JM, Rickinson AB. Conserved CTLepitopes within EBV latentmembrane protein 2: a potential target for CTL-based tumor therapy. J Immunol 1997;158:3325–34.

26. Bollard CM, Gottschalk S, Torrano V, Diouf O, Ku S, Hazrat Y, et al.Sustained complete responses in patients with lymphoma receiving autol-ogous cytotoxic T lymphocytes targeting Epstein-Barr virus latent mem-brane proteins. J Clin Oncol 2014;32:798–808.

27. Park IH, Zhao R, West JA, Yabuuchi A, Huo H, Ince TA, et al. Reprogram-ming of human somatic cells to pluripotency with defined factors. Nature2008;451:141–6.

28. Mali P, Ye Z, Hommond HH, Yu X, Lin J, Chen G, et al. Improvedefficiency and pace of generating induced pluripotent stem cells fromhuman adult and fetal fibroblasts. Stem Cells 2008;26:1998–2005.

29. Fusaki N, BanH, NishiyamaA, Saeki K, HasegawaM. Efficient induction oftransgene-free humanpluripotent stem cells using a vector based on Sendaivirus, an RNA virus that does not integrate into the host genome. Proc JpnAcad Ser B Phys Biol Sci 2009;85:348–62.

30. Nishimura K, Sano M, Ohtaka M, Furuta B, Umemura Y, Nakajima Y, et al.Development of defective and persistent Sendai virus vector: a unique genedelivery/expression system ideal for cell reprogramming. J Biol Chem2011;286:4760–71.

31. Kovalovsky D, Uche OU, Eladad S, Hobbs RM, Yi W, Alonzo E, et al. TheBTB-zinc finger transcriptional regulator PLZF controls the development ofinvariant natural killer T cell effector functions. Nat Immunol 2008;9:1055–64.

32. Savage AK, Constantinides MG, Han J, Picard D, Martin E, Li B, et al. Thetranscription factor PLZF directs the effector program of the NKT celllineage. Immunity 2008;29:391–403.

33. Kreslavsky T, Savage AK, Hobbs R, Gounari F, Bronson R, Pereira P, et al.TCR-inducible PLZF transcription factor required for innate phenotype of asubset of gammadelta T cells with restricted TCR diversity. Proc Natl AcadSci U S A 2009;106:12453–8.

34. Inoue K, Sugiyama H, Ogawa H, Nakagawa M, Yamagami T, Miwa H,et al. WT1 as a new prognostic factor and a new marker for thedetection of minimal residual disease in acute leukemia. Blood1994;84:3071–9.

35. Cheever MA, Allison JP, Ferris AS, Finn OJ, Hastings BM, Hecht TT, et al.The prioritization of cancer antigens: a national cancer institute pilotproject for the acceleration of translational research. Clin Cancer Res2009;15:5323–37.

36. Klein L, Kyewski B, Allen PM, Hogquist KA. Positive and negative selectionof the T cell repertoire: what thymocytes see (and don't see). Nat RevImmunol 2014;14:377–91.

37. Kincaid EZ,Murata S, Tanaka K, Rock KL. Specialized proteasome subunitshave an essential role in the thymic selection of CD8(þ) T cells. NatImmunol 2016;17:938–45.

38. Hayes SM, Li L, Love PE. TCR signal strength influences alphabeta/gam-madelta lineage fate. Immunity 2005;22:583–93.

39. Haks MC, Lefebvre JM, Lauritsen JP, Carleton M, Rhodes M, Miyazaki T,et al. Attenuation of gammadeltaTCR signaling efficiently diverts thymo-cytes to the alphabeta lineage. Immunity 2005;22:595–606.

40. Terrence K, Pavlovich CP, Matechak EO, Fowlkes BJ. Premature expres-sion of T cell receptor (TCR)ab suppresses TCRgd gene rearrange-ment but permits development of gd lineage T cells. J Exp Med 2000;192:537–48.

41. Bauer S, Groh V, Wu J, Steinle A, Phillips JH, Lanier LL, et al. Activation ofNK cells and T cells by NKG2D, a receptor for stress-inducible MICA.Science 1999;285:727–9.

42. Karimi M, Cao TM, Baker JA, Verneris MR, Soares L, Negrin RS. Silencinghuman NKG2D, DAP10, and DAP12 reduces cytotoxicity of activatedCD8þ T cells and NK cells. J Immunol 2005;175:7819–28.

43. Pievani A, Borleri G, Pende D, Moretta L, Rambaldi A, Golay J, et al. Dual-functional capability of CD3þCD56þ CIK cells, a T-cell subset thatacquiresNK function and retains TCR-mediated specific cytotoxicity. Blood2011;118:3301–10.

44. Geginat J, Lanzavecchia A, Sallusto F. Proliferation and differentiationpotential of human CD8þmemory T-cell subsets in response to antigen orhomeostatic cytokines. Blood 2003;101:4260–6.

45. Berger C, Jensen MC, Lansdorp PM, Gough M, Elliott C, Riddell SR.Adoptive transfer of effector CD8þ T cells derived from central memorycells establishes persistent T cell memory in primates. J Clin Invest2008;118:294–305.

46. Gattinoni L, Lugli E, Ji Y, Pos Z, Paulos CM, Quigley MF, et al. A humanmemory T cell subset with stem cell-like properties. Nat Med 2011;17:1290–7.

47. Maus MV, Fraietta JA, Levine BL, Kalos M, Zhao Y, June CH. Adoptiveimmunotherapy for cancer or viruses. Annu Rev Immunol 2014;32:189–225.

48. Hinrichs CS, Spolski R, Paulos CM, Gattinoni L, Kerstann KW,Palmer DC, et al. IL-2 and IL-21 confer opposing differentiationprograms to CD8þ T cells for adoptive immunotherapy. Blood 2008;111:5326–33.

49. Carrasco J, GodelaineD, Van Pel A, Boon T, van der Bruggen P. CD45RA onhuman CD8 T cells is sensitive to the time elapsed since the last antigenicstimulation. Blood 2006;108:2897–905.

50. Saito H, Okita K, Chang AE, Ito F. Adoptive transfer of CD8þ T cellsgenerated from induced pluripotent stem cells triggers regressions oflarge tumors along with immunological memory. Cancer Res 2016;76:3473–83.


Maeda et al.




2016;76:6839-6850. Published OnlineFirst November 21, 2016.Cancer Res Takuya Maeda, Seiji Nagano, Hiroshi Ichise, et al. Potent Tumor Antigen-Specific Cytotoxicity

Derived iPSC Imparts− T Cells from T-cellβαRegeneration of CD8

Updated version

10.1158/0008-5472.CAN-16-1149doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2016/10/05/0008-5472.CAN-16-1149.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/76/23/6839.full#ref-list-1

This article cites 49 articles, 22 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/76/23/6839.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://cancerres.aacrjournals.org/content/76/23/6839To 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-16-1149

http://cancerres.aacrjournals.org/content/suppl/2016/10/05/0008-5472.CAN-16-1149.DC1

http://cancerres.aacrjournals.org/content/76/23/6839.full#ref-list-1

http://cancerres.aacrjournals.org/content/76/23/6839.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/76/23/6839


Documents

Regeneration of CD8ab T Cells from Derived iPSC Imparts ...Cancer Res; 76(23); 6839–50. 2016 AACR. Introduction During the last several years, cancer immunotherapy using a variety