ex vivo expanded adaptive nk cells effectively kill ...€¦ · 4 hours, blasts were stained with...

Research Article

Ex Vivo Expanded Adaptive NK Cells EffectivelyKill Primary Acute Lymphoblastic Leukemia CellsLisa L. Liu1, Vivien B�eziat2,3, Vincent Y.S. Oei4,5, Aline Pfefferle1, Marie Schaffer1,S€oren Lehmann6,7, Eva Hellstr€om-Lindberg7, Stefan S€oderh€all8, Mats Heyman8,Dan Grand�er9, and Karl-Johan Malmberg1,4,5

Abstract

Manipulation of human natural killer (NK) cell repertoirespromises more effective strategies for NK cell–based cancerimmunotherapy. A subset of highly differentiated NK cells,termed adaptive NK cells, expands naturally in vivo in responseto human cytomegalovirus (HCMV) infection, carries uniquerepertoires of inhibitory killer cell immunoglobulin-like recep-tors (KIR), and displays strong cytotoxicity against tumor cells.Here, we established a robust and scalable protocol for ex vivogeneration and expansion of adaptive NK cells for cell therapyagainst pediatric acute lymphoblastic leukemia (ALL). Cultureof polyclonal NK cells together with feeder cells expressingHLA-E, the ligand for the activating NKG2C receptor, led toselective expansion of adaptive NK cells with enhanced allor-

eactivity against HLA-mismatched targets. The ex vivo expandedadaptive NK cells gradually obtained a more differentiatedphenotype and were specific and highly efficient killers ofallogeneic pediatric T- and precursor B-cell acute lymphoblasticleukemia (ALL) blasts, previously shown to be refractory tokilling by autologous NK cells and the NK-cell line NK92currently in clinical testing. Selective expansion of NK cellsthat express one single inhibitory KIR for self-HLA class I wouldallow exploitation of the full potential of NK-cell alloreactivityin cancer immunotherapy. In summary, our data suggest thatadaptive NK cells may hold utility for therapy of refractory ALL,either as a bridge to transplant or for patients that lack stem celldonors. Cancer Immunol Res; 5(8); 654–65. �2017 AACR.

IntroductionNatural killer (NK) cells are innate lymphoid cells charac-

terized by a potent intrinsic cytotoxic potential and an ability tokill virus-infected and transformed cells (1–4). In humans, thefunctional potential of NK cells is regulated through expressionof germline-encoded activating and inhibitory receptors (5).Activating receptors bind to stress-induced ligands, which arehighly expressed on tumor cells (6, 7). Inhibitory receptors,including killer cell immunoglobulin-like receptors (KIR) and

CD94-NKG2A heterodimers, bind to groups of HLA class Ialleles and the nonclassical HLA class I molecule HLA-E,respectively (8, 9). The variegated and random distribution ofKIR in the NK-cell compartment endows the innate immunesystem with a broad and highly diverse range of responsivenesswithout receptor rearrangement (10, 11). The functional poten-tial of a given NK cell is quantitatively tuned by the net inputobtained through interactions with ligands for activating recep-tors and self-MHC molecules (12, 13). This functional calibra-tion process is termed NK-cell education and allows the cell tosense a sudden decrease of normal HLA expression as a result ofvirus infection or tumor transformation (14).

In the context of allogeneic adoptive NK-cell therapy or stemcell transplantation (SCT), such reactivity due to a lack ofrecognized MHC-encoded inhibitory ligands is triggered whenNK cells are transferred across HLA barriers (15). There is com-pelling evidence that NK cells play an important role as anti-leukemic effector cells in T-cell–depleted haploidentical, KIRligand–mismatched, SCT for patients with acute myeloid leuke-mia (AML; refs. 16–18). NK cell–mediated graft-versus-leukemia(GVL) effects have been less well documented in the context ofacute lymphoblastic leukemia (ALL), in particular in adult ALL(19, 20). Being derived from lymphoid progenitors, ALL blastsexpress more HLA class I than myeloid cells, potentially con-tributing to its increased resistance to NK-cell killing (21). Othersuggested mechanisms of resistance include insufficient bindingto LFA-1 and lower expression of ligands for activating NK-cellreceptors, including MICA/B on target cells (15, 19).

The KIR repertoire in any given donor is determined by KIRgene content, copy number variation, and a stochastic epigeneticregulation at the promoter level (22, 23). Thus, the frequency of

1Center for Infectious Medicine, Department of Medicine Huddinge, KarolinskaInstitutet, Stockholm, Sweden. 2Human Genetics of Infectious Diseases Labo-ratory, INSERM U1163, Imagine Institute, Paris, France. 3Paris Descartes Univer-sity, Imagine Institute, Paris, France. 4Institute for Cancer Research, OsloUniversity Hospital, Oslo, Norway. 5The KG Jebsen Center for Cancer Immuno-therapy, Institute of Clinical Medicine, University of Oslo, Oslo, Norway. 6Depart-ment of Medical Sciences, Uppsala University, Uppsala, Sweden. 7Department ofMedicine, Center for Hematology and Regenerative Medicine, Karolinska Insti-tutet, Karolinska University Hospital Huddinge, Stockholm, Sweden. 8Depart-ment of Women's and Children's Health & the Pediatric Cancer Unit, KarolinskaInstitutet and Karolinska University Hospital, Stockholm, Sweden. 9Departmentof Oncology and Pathology, Cancer Centre Karolinska, Karolinska Institutet,Stockholm, Sweden.

Note: Supplementary data for this article are available at Cancer ImmunologyResearch Online (http://cancerimmunolres.aacrjournals.org/).

Corresponding Author: Karl-Johan Malmberg, Oslo University Hospital,Ullernchausseen 70, Oslo 0130, Norway. Phone: 474-539-0926; E-mail:[email protected]

doi: 10.1158/2326-6066.CIR-16-0296

�2017 American Association for Cancer Research.

CancerImmunologyResearch

Cancer Immunol Res; 5(8) August 2017654

on September 13, 2020. © 2017 American Association for Cancer Research. cancerimmunolres.aacrjournals.org Downloaded from

Published OnlineFirst June 21, 2017; DOI: 10.1158/2326-6066.CIR-16-0296

http://crossmark.crossref.org/dialog/?doi=10.1158/2326-6066.CIR-16-0296&domain=pdf&date_stamp=2017-7-14

http://cancerimmunolres.aacrjournals.org/

alloreactive NK cells varies, ranging from a few percent to 60% to70% of the NK-cell population in exceptional cases (24). There-fore, expressionofHLA class I on acute leukemia blasts remains animportant limiting factor for the efficacy of NK-cell therapydespite the use of allogeneic NK cells. ALL cells are resistant totheNK92 cell line (19).NK92 lacksKIRs, but expressesNKG2A, aninhibitory receptor that binds to HLA-E, which may contribute tothe NK-cell resistance. However, pediatric ALL cells express littleHLA-E (25). Thus, it remains unclear whether NK-cell recognitionof ALL cells is functionally regulated through activating andinhibitory input delivered throughNKG2CandNKG2A receptors,respectively, binding to their shared ligand, HLA-E.

Although several GMP-compliant NK-cell expansion protocolsyield large numbers of NK cells for therapy, it has proven difficultto selectively expand specific subsets of NK cells expressing asingle self-specific KIR, which would allow more effective target-ing of allogeneic leukemia cells (26, 27). Insights into the naturalimmune response to human cytomegalovirus (HCMV), mani-fested as expansion and differentiation of highly cytotoxic NKcells with unique and predictable KIR expression profiles, mayhold keys to the design of more selective in vitro expansionprotocols. Numerous reports have shown in vivo expansion of socalled "adaptive" NKG2CþNK cells expressing a single self-KIR inHCMV seropositive individuals (28–30). HCMV reactivation andthe associated expansion of adaptive NK cells have been linked toa lower relapse rate in AML patients after cord blood allogeneicSCT, a type of transplant in which one or more HLA mismatchesare common (31). This suggests that spontaneously occurringadaptive NK cells may have beneficial effects mediated througheradication of residual leukemic blasts. In vitro expansion ofNKG2Cþ adaptive NK cells can be achieved by coculturing na€�veNK cells with CMV-infected fibroblasts or HLA-E transfected celllines (28, 32), but the robustness and efficacy of such expansionprotocols remains to be determined. Herein, we have examinedthe use of a feeder-based platform for selective expansion anddifferentiation of educated NK cells aiming to target primary ALLblasts.

Materials and MethodsHuman participants and cells

This study was conducted in accordance with the Declarationof Helsinki and was approved by the Regional Ethical ReviewBoard in Stockholm, Sweden. Healthy blood donors were used asa source for NK-cell expansion experiments. For all donors,peripheral bloodmononuclear cells (PBMC)were separated frombuffy coats by density gravity centrifugation (Ficoll–Hypaque; GEHealthcare) andwere cryopreserved in 10%DMSOand90%heat-inactivated FBS for later use. Bone marrow samples from patientswith pediatric ALL and adult Myelodysplastic Syndromes (MDS)were processed like the healthy PBMCs. When required, genomicDNA was isolated from 200 mL of whole blood using DNeasyBlood and Tissue Kit (Qiagen).

Expansion of NK cellsThawed PBMCs from healthy blood donors were subjected to

NK-cell negative selection (NKCell Isolation Kit,Miltenyi Biotec).NK cells were then cocultured with irradiated (40 Gy) 721.221.AEH cells (33) in completemedium (RPMI, glutamine, 10%FCS)in the presence of IL15 (final concentration 100 ng/mL, R&DSystems) at a final concentration of 5 � 106/mL in a 96-well

U-bottom plate (E:T ratio 10:1). Cells were labeled with CellTraceviolet (1 mmol/L, Life Technologies) prior to coculture.

Flow cytometry–based cytotoxic assayIsolated NK cells, either rested overnight or expanded after

14 days, were labeled with CellTrace violet and used as effectorcells at a final concentration of 1.25 � 106/mL. Blasts werethawed and added at a final concentration of 2.5 � 105, andcaspase-3 substrate was added at the beginning of the assay. After4 hours, blasts were stained with the live/dead cell marker aqua(DCM) and relevant extracellular receptors. The percentage ofcytotoxic activity was calculated using the following formula:percent specific killing ¼ (C3þ – spontaneous C3þ) þ(C3þDCMþ � spontaneous C3þDCMþ)/[live targets þ (C3þ

� spontaneus C3þ) þ (C3þDCMþ � spontaneous C3þDCMþ)].

KIR and KIR ligand typingKIR ligands were determined using the KIR HLA Ligand Kit

(Olerup SSP; Qiagen) for the detection of, HLA-C1, HLA-C2, andfour HLA-Bw4 motifs. KIR genotyping was performed using ahigh-throughput technology called quantitative KIR automatedtyping (qKAT; ref. 34).

HCMV serologyCMV serology was determined using an ELISA-based assay on

plasma obtained during sample preparation. Purified nuclearCMV antigen (AD 169) was used, and the cut-off level forseropositivity was an absorbance of �0.2 at a dilution of 1/100.

Antibodies, tetramers, and flow cytometry analysisStainings were performed using an appropriate mix of the

following antibodies (clone names are given in brackets):CD14-V500 (M5E2), CD19-V500 (HIB19), CD3-PE-Cy5.5 orPE.Cy5 (UCHT1), CD56-ECD or PE-Cy5.5 (N901), CD57 puri-fied (TB01) or BV605 (NK-1), anti-mouse-IgM-EF650 (II/41),FCER1g-FITC (rabbit polyclonal), CD7-PE.Cy7 (8H8.1), ILT2-PE(HP-F1), CD161-BV605 (HP-3G10), NKG2A-APC or APC.AF750or PE-Cy7 (Z199), NKG2C-PE (FAB138P, REA205), DNAM-1-PE.vio770 (Dx11), NKG2D-BV711 (1D11), CD16-AF700 or BV785(3G8), 2B4-PE (C1.7), NKp46-BV786 (9E2), CD2-PB (TS1/8),KIR2DL3-FITC (180701), KIR2DL1-APC (143211), KIR3DL1-AF700 (Dx9), KIR2DS4-QD585 (179315), KIR3DL2-biotin(Dx31), KIR2DL2/L3/S2-PE.Cy5.5 (GL183), KIR2DL1/S1-PE.Cy7(EB6), CD10 (HI10a), granzyme A (CB9), granzyme B (GB11),perforin (B-D48), and TRAIL (RIK-1). Dead cells were labeledusing live/dead aqua (Life Technologies). Biotin-conjugated anti-bodies were visualized using streptavidin-Qdot 585 or 605 (LifeTechnologies). After staining, cells were fixed and permeabilizedusing a Fixation/Permeabilization Kit (eBioscience) prior to intra-cellular staining. Samples were acquired using an LSR-Fortessa18-color flow cytometer (Becton Dickinson), and data wereanalyzed with FlowJo software version 9 (Tree Star, Inc.). TheBD-LSR Fortessa instrument was equipped with a 100 mW405 nm laser, a 100 mW 488 nm laser, a 50 mW 561 nm laser,and a 40 mW 639 nm laser.

NK-cell functional assayIsolated NK cells were rested overnight in complete medium

and distributed at a final concentration of 2.5 � 105 cells/mLin a 96-well U-bottom plate. All target cells were added to NK

Adaptive NK Cells Recognize Primary ALL

www.aacrjournals.org Cancer Immunol Res; 5(8) August 2017 655




cells at a final concentration of 2.5 � 105 cells/mL. For con-ventional functional assays, ALL blasts, PHA blasts, 721.221.wt,or 721.221.AEH cells were used as target cells. The cells wereincubated for 6 hours (37�C, 5% CO2) after addition ofmonensin (GolgiStop, BD Biosciences, 1/1,500 final concen-tration), brefeldin A (GolgiPlug, BD Biosciences, 1/1,000 finalconcentration), and CD107a-BV421 (H4A3, 1/100 final con-centration). After incubation, cells were washed and stainedfor extracellular receptors, permeabilized (fixation/perme-abilization buffer, eBioscience) and stained for intracellularTNF-APC (MAb11) and IFNg-AF700 (B27) prior to analysiswith flow cytometry.

Statistical analysisFor comparisons of independent groups, the Student t test or

the Mann–Whitney test was performed. For comparisons ofmatched groups, the paired Student t test or the Wilcoxonmatched test was performed depending on the sample size.For comparisons of qualitative variables, the Fisher exact t testwas performed. In the relevant figures, n.s. indicates not signif-icant; �� indicates P < 0.001; �� indicates P < 0.01; and � indicatesP < 0.05. Analyses were performed using GraphPad software.

ResultsSelective expansion of educated NKG2Cþ NK cells expressinga single KIR

Here, we set out to establish a platform for selective ex-pansion of NKG2Cþ NK cells expressing a single self-specificKIR (self-KIR). NK cells were cocultured together with 721.221cells transfected with HLA-E (221.AEH) in the presence ofIL15 for 14 days and stained for inhibitory KIRs, NKG2A, andNKG2C. As previously reported (28), a profound skewingtoward NKG2Cþ NK cells expressing self-KIR was seen after14 days in culture (Fig. 1A). Stratification of NK-cell subsetsduring proliferation revealed that CD57þ NK cells dominateamong nondividing NK cells (generation 0), whereas self-KIRþNKG2Cþ NK cells dominated in the fraction of NK cellsreaching later generations (Fig. 1B). To explore the robust-ness of this platform, we generated adaptive NK cells from14 HCMV seropositive donors, with or without a preexistingNKG2CþCD57þ adaptive NK-cell populations. The 2-weekculture protocol successfully expanded NKG2Cþ NK cells fromall donors tested. The skewing of the KIR repertoire with a biastoward expression of self-KIRs was noted both at the bulk NK-cell level and in the NKG2CþNKG2A� NK-cell subset. Indonors with preexisting expansions of adaptive NK cells, theskewing of the repertoire was further enhanced (Fig. 1C). Thiswas also observed in C1/C2 donors where the expanded phe-notype was determined by the preexisting CMV-inducedimprint in the KIR repertoire (Supplementary Fig. S1).Although it cannot be excluded that the enrichment of theadaptive NK cells was partly a consequence of selective survival,the protocol also led to a 2.4-fold expansion of the adaptiveNK-cell subset (Fig. 1D). Culture of adaptive NK cells in IL15alone or in combination with 221.wt cells did not induce asimilar skewing toward an end product enriched for NKG2A�

NK cells (Supplementary Fig. S2). These results demonstratethe feasibility and robustness of the expansion of self-KIRþ

NKG2Cþ NK cells from HCMV-seropositive donors using HLA-E–overexpressing feeder cells.

In vitro expansion differentiates cells toward an adaptivephenotype

We monitored the phenotypic changes and granular loadin the NK cells pre- and post in vitro expansion and foundthat both conventional and adaptive NK cells displayedlower expression of CD7, Siglec-7, and PLZF and upregulationof CD2 and DNAM-1 (Fig. 2). Whereas the frequency ofCD57þ NK cells decreased somewhat, due to preferentialproliferation of CD57– NKG2Cþ single-KIRþ NK cells(Fig. 1B; ref. 28), the expression intensity of CD57 increasedover the course of the 14-day culture. Loss of FceR1g and Sykare additional hallmarks of adaptive NK cells (35, 36). Incontrast to the surface receptors, these adaptor molecules wereeither stable or normalized during the 14-day culture. Expres-sion of the inhibitory receptor LILRB1 did not show anyconsistent change.

Adaptive NK cells can express slightly more CD16 andperform better in antibody-dependent cytotoxicity assaysassays (37). However, activation of NK cells with target cellsand/or cytokines leads to loss of CD16 through a metallo-protease called ADAM17 (38). Indeed, the feeder-basedexpansion protocol led to a relative loss in CD16 expressiondown to approximately 50% positivity, in line with previouslyreported data (Fig. 2). Notably, CD16 was more stable inadaptive NK cells, leading to a relatively bigger differencebetween adaptive and conventional NK cells after coculturewith HLA-E–expressing feeder cells. In addition to theobserved changes in surface receptors and adaptor molecules,the levels of granzyme A and perforin increased or remainedstable in all NK-cell subsets (Supplementary Fig. S3). Togeth-er, these data suggest that the ligation of NKG2C under theinfluence of IL15 leads to a gradual shift toward an adaptivephenotype.

Cytotoxicity and specificity of in vitro generated adaptive NKcells

The dynamic change toward a more adaptive phenotypeprompted us to test the specificity and functional potential ofin vitro generated NK cells. To this end, expanded NK cellsexpressing either KIR2DL3 or KIR2DL1, derived from HLA-C1/C1 (C1/C1) or HLA-C2/C2 (C2/C2) donors, respectively, weretested against C1/C1þ and C2/C2þ PHA blasts in a flowcytometry–based cytotoxicity assay. Target cell death was mea-sured with caspase-3 (C3) activity and a live/dead cell marker(DCM). The expanded NK cells displayed effective and specifickilling of mismatched PHA blasts, reaching nearly 100%killing of C1/C1 blasts by KIR2DL1 single-positive NK cells(Fig. 3B). Despite the 2-week culture in IL15, negligible killingwas seen in the KIR-HLA–matched setting (Fig. 3A and B). Toevaluate the impact of the selective expansion protocol, thecytotoxicity of the expanded NK cells was compared withresting NK cells before expansion. All NK-cell expansionsdisplayed significantly more potent cytotoxicity againstHLA-C–mismatched PHA blasts on a per cell basis comparedto the resting NK cells day 0 (Fig. 3B). In the presence ofHLA class I–blocking antibodies, expanded NK cells werealso able to kill PHA blasts in KIR-HLA–matched settings. Incontrast, HLA class I blockade had no effect on the killing ofHLA-mismatched PHA blasts, suggesting that the repro-grammed NK-cell cultures displayed a maximized alloreactiveresponse (Fig. 3C).

Liu et al.

Cancer Immunol Res; 5(8) August 2017 Cancer Immunology Research656




NKG

2A

NKG2C

#177 (C1/C1)Day 0 Day 14

A

KIR2DL1

KIR

2DL3

NKG

2A

NKG2C

KIR

2DL3

KIR2DL1

#54 (C2/C2)Day 0 Day 14

C

Day 0 Day 14 Day 0 Day 14Day 0 Day 14

% N

KG2C

+ NKG

2A–

% S

elf-s

peci

fic K

IR

% S

elf-s

peci

fic K

IR in

N

KG2C

+ NKG

2A–

NK

cells

0

20

40

60

80

100

0

20

40

60

80

100

0

20

40

60

80

100*** *** *

B

D

0

2

4

6

8

Fold

exp

ansi

on

Conv. Adaptive

*

Gen 0

Gen 1

Gen 2

Gen 3Gen 4+

CD

57

NK

G2C

KIR

2DL1

KIR

2DL3

CD

57

NKG2C

NKG2C+ NK cells (C2/C2)Bulk NK cells

CellTrace violet

Gen 0 Gen 1 Gen 2 Gen 3 Gen 4+

Figure 1.

Selective expansion of educated NKG2Cþ NK cells expressing a single KIR. A, Representative FACS plot showing frequency of NKG2C, NKG2A, KIR2DL1,and KIR2DL3 of one C1/C1 and C2/C2 donor after 14-day expansion. B, Representative FACS plots showing cell division of the indicated subsetmeasured through the dilution of CellTrace violet. C, Compiled graph showing frequency of NKG2CþNKG2A� NK cells and NK cells with self-KIR in bulkCD56dim NK cells and self-KIRþ NK cells within the NKG2CþNKG2A� compartment in 14 HCMV seropositive donors before and after expansion. D, Foldexpansion of the adaptive NK-cell subset versus NK cells with conventional phenotypes. Summary of 7 experiments and 8 donors.






Adaptive NK cells efficiently recognize primary pediatricALL blasts.

Previous studies have reported that ALL blasts are relativelyresistant to lysis by allogeneic NK cells and the NK cell lineNK92 (19, 39). Therefore, KIR2DL1- and KIR2DL3 single-positive, in vitro expanded adaptive NK cells were tested againsta panel of blasts from 23 pediatric, primary ALL bone marrowsamples of three different subgroups; T-cell ALL and B-cellprecursor cells with either high hyperdiploidy 47-50 chromo-

somes (Heh) or (t12;21) (Table 1). All samples were taken atthe time of diagnosis and were >80% blasts. We established aFACS-based killing assay using an individualized gating strat-egy for each ALL sample, based on the immune phenotypingdeveloped by the Nordic society of pediatric hematology andoncology (Fig. 4A). Expanded NK cells displayed high killingactivity against HLA-mismatched primary ALL blasts (Fig. 4Band C). The average killing was approximately 70% at aneffector to target ratio of 5:1 for ALL blasts in the mismatched

% C

D16

0

20

40

60

80

100 ** **

0

50

100

Bulk Conventional Adaptive%

2B

4

0

50

100

% C

D2

Bulk Conventional Adaptive

0

20

40

60

80

100

% C

D7

Bulk Conventional Adaptive

0

20

40

60

80

100

% F

cεR

Iγ

0

20

40

60

80

100

% P

LZF

0

20

40

60

80

100

% S

yk

0

20

40

60

80

100

% C

D16

1

0

20

40

60

80

100

% S

igle

c-7

* **

*** *** *** *** *

***

*

Day 0

Day 0

Day 0

Day 0

Day 7

Day 7

Day 14

Day 14

Day 14

Day 14

Day 0

Day 14

Day 0

Day 14

Day 0

Day 0

Day 14

Day 14

Day 0

Day 14

Day 0

Day 14

Day 14

0

20

40

60

80

100

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

Day 7

Day 14

Day 0

% D

NA

M-1

* nsns

0

2,000

4,000

6,000

8,000

10,000

12,000

0

20

40

60

80

100

% C

D57

MFI

CD

57

0

20

40

60

80

100%

LILR

B1

*** *****

* ****

Figure 2.

In vitro expansion is associated with cellular differentiation toward an adaptive phenotype. Compiled graph from three experiments (n between 5 and 10)showing the expression of the indicated surface receptors and adaptor molecules on bulk, conventional, and adaptive NK cells at days 0, 7, and 14.

Liu et al.





setting, independent of disease subgroup (Fig. 4B and C).A similar efficiency was noted when selectively expanded NKcells were tested against primary MDS and AML blasts (Sup-plementary Fig. S4), suggesting the possibility of a potentiallybroader clinical implementation of in vitro expanded adaptiveNK cells.

ALL blasts from pediatric patients express little HLA-E (25).Thus, it remained an open question whether NKG2C contrib-uted to the recognition of the mismatched targets. In linewith previous reports, we observed lower expression of HLA-Eon B-ALL blasts compared with healthy controls (Fig. 4D). Toaddress the relative contribution of the NKG2C–HLA-E inter-action, degranulation was monitored in NKG2CþNKG2A–

and NKG2C–NKG2Aþ NK-cell subsets following stimulationwith ALL blasts and 221.AEH cells in the presence or absenceof anti-CD94 blockade (Fig. 4F). Although predictable recog-nition patterns were observed for NKG2A–NKG2Cþ NK cellsand NKG2AþNKG2C– NK cells against 221.AEH cells overex-pressing HLA-E (Fig. 4E), no significant difference in degranu-

lation was seen with or without anti-CD94 mAb in eitherNKG2A–NKG2Cþ or NKG2AþNKG2C– NK cells when stimu-lated with ALL blasts (Fig. 4F). Taken together, these resultssuggest that although NKG2C is functional on the in vitroexpanded adaptive NK cells, its ligation is not required for thepotent killing of ALL blasts in an HLA-mismatched setting.

Adaptive NK cells display an unleashed allogeneic responseThe finding that NKG2C appeared redundant for the ability

of the expanded adaptive NK cells to kill ALL blasts led us toexamine the impact of the size of the alloreactive subset in moredetail. To this end, we generated parallel cultures and moni-tored the ability of resting, short-term IL15 stimulated andexpanded NK cells to kill mismatched ALL blasts (Fig. 5A). Thekilling of primary ALL blasts correlated with the generation of alarge alloreactive subset (Fig. 5B). To further explore the impactof allorecognition and the intrinsic cytotoxic potential of adap-tive NK cells, we sorted NKG2CþNKG2A–KIR2DL1þ NK cellsfrom C2/C2 donors to 100% purity (Supplementary Fig. S5)

2DL1+(C2/C2)

2DL3+(C1/C1)

NK CellsNK Cells

PHA Blasts (C1/C1) PHA Blasts (C2/C2)

2DL1+(C2/C2)

2DL3+(C1/C1)

A

0

20

40

60

80

100%

Spec

ific

killi

ng

0

20

40

60

80

100

Day 0 Day 140

20

40

60

80

1002DL3+ NK Cells

Day 0 Day 140

20

40

60

80

1002DL1+ NK Cells

B

0

20

40

60

80

100

Day 0 Day 14

2DL1+ NK Cells

PHA BlastsC2/C2

PHA BlastsC1/C1

0

20

40

60

80

100

Day 0 Day 14

C

%Sp

ecifi

c ki

lling

%Sp

ecifi

c ki

lling

%Sp

ecifi

c ki

lling

%Sp

ecifi

c ki

lling

%Sp

ecifi

c ki

lling

Caspase-3Caspase-3DCMLive cells

2DL3+ NK Cells

0

50

100

% S

peci

fic k

illin

g

PHA BlastsC2/C2

PHA BlastsC1/C1

2DL1+ NK Cells

HLA Class I blockingIsotype control

*

ns

PHA BlastsC1/C1

PHA BlastsC2/C2

* *

*** *

Figure 3.

In vitro expanded adaptive NK cells are highly cytotoxic and specifically recognize HLA-mismatched targets. A, Representative graph showing specificlysis of PHA blasts derived from C1/C1 or C2/C2 donors using KIR2DL1þ or KIR2DL3þ expanded NK cells. B, Compiled graph of specific lysis usingKIR2DL3þ and KIR2DL1þ NK cells expanded from different donors (n ¼ 7, n ¼ 3, respectively) against PHA blasts derived from C1/C1 or C2/C2 donors.C, Specific lysis of C1/C1 or C2/C2 PHA blasts in a FACS-based killing assay with KIR2DL1þ NK cells in the presence or absence of HLA class I–blockingantibodies. The figure shows a representative experiment out of three. Error bars represent the SD between three donors.






and compared their ability to kill HLA-mismatched PHA blastswith unsorted, expanded adaptive NK cells from the samedonor. Notably, the sorted adaptive NK cells were highly potentagainst mismatched C1/C1þ PHA blasts, although expandedNK cells showed slightly higher killing against the same targets(Fig. 5C–D). These results show that adaptive NK cells, bothresting and expanded, represent a unique subset with enhancedalloreactive capacity due to the near complete lack of inhibitoryMHC-binding receptors when transferred across HLA-C bar-riers. The selective expansion and repertoire skewing usingHLA-E–expressing feeder cells leads to further potentiation ofthe adaptive NK cells and holds promise for the next-generationNK cell–based cancer immunotherapy.

DiscussionThe survival rate for pediatric ALL currently exceeds 90%, but

the remaining patients who relapse remain a major clinicalchallenge requiring effective treatment modalities. AllogeneicSCT has become the established procedure for treating childrenwith high-risk and relapsed leukemia. Over the past decade,accumulating evidence indicates that NK cells may contributeto the GVL effect associated with allogeneic SCT. However, thecontribution of NK cells in the GVL effect against ALL appearslimited, possibly reflecting the poor ability of heterogeneousNK-cell populations to kill ALL blasts in vitro (39, 40). A largenumber of clinical trials are currently exploring the potential ofresting and cytokine-activated NK cells as a therapeutic modal-ity against various cancer types. Most of the existing protocolsinvolve transfer of highly diverse populations of NK cells.Although currently available clinical NK-cell products containcells with the desired alloreactive specificity, it is obvious thatfurther refinement of the procedures hold promise to improvethe efficacy of NK cell–based immunotherapy (26). Indeed, thesize of the alloreactive NK-cell subset and induction of molec-ular remission and prolonged disease-free survival are corre-

lated (41). In particular, it represents an exciting possibilityto selectively expand and/or guide differentiation of adaptiveNK cells showing high cytolytic potential and unique KIRspecificities (42–44). In this study, we have shown the robust-ness of a preclinical platform for selective expansion of adap-tive NKG2Cþ NK cells and its potential in targeting primaryALL blasts.

The protocol developed in this study yielded NK cells with adifferentiated phenotype displaying profound skewing towardexpression of a single self-KIR. Consequently, the expandedadaptive NK cells displayed an elevated specificity for allogeneictargets lacking the cognate HLA ligand, demonstrating the main-tained signaling potential of inhibitory KIRs, despite the 2-weekculture in IL15. This is an important and unique feature of thisparticular strategy for expanding adaptive NK cells, becauseother ex vivo expansion protocols have been associated withloss of self-tolerance by overcoming KIR inhibition (45–48). It iswell documented that the proliferative capacity of the CD57þ

adaptive NK-cell subset is limited (49). This raises the questionwhether the observed enrichment of adaptive NK cells is merelya consequence of selective survival. We have previously shownthat a proportion of CMV-seropositive patients with TAP defi-ciency also displays expansion of NKG2Cþ NK cells with phe-notypic hallmarks of adaptive NK cells (50). However, theseexpansions had polyclonal KIR repertoires, suggesting that theselective skewing in MHC-sufficient individuals depends on aselective advantage of inhibition through self-specific KIRs.This aligns with earlier studies in the mouse showing preferen-tial proliferation and expression of Bcl-2 in NK cells expressingself-MHC Ly49þ in the mouse (51, 52). Notably, the currentprotocol yielded a small, but significant increase in absolutenumbers of adaptive NK cells, reflecting their expansion incombination with differentiation from less differentiatedNKG2Cþ precursors, as previously suggested (28). NKG2C isunique among activating NK-cell receptors, and is the only NK-cell receptor in addition to CD16, that can activate resting NKcells (53, 54). Thus, we propose that the enrichment of adaptiveNK cells under the current protocol is a combined effect ofsurvival, NKG2C-driven expansion, and differentiation fromearlier precursors.

The most appealing setting for clinical implementation ofthe current protocol is to transfer ex vivo expanded adaptive NKcells across HLA-C barriers to patients who are refractory toconventional treatment modalities. In such protocols, cellsgenerated from a C1/C1 donor could be transferred to patientswith C2/C2 genotypes and vice versa. Notably, C1/C2 donorscould also be used pending on their CMV imprinted repertoire.In C1/C2 donors with a preexisting expansion of KIR2DL1þ

adaptive NK cells, the current protocol expanded the samephenotype, possibly extending the number of possible donorsfor C1/C1 patients, which may be of clinical relevance giventhat the C2/C2 genotype is less common. Although not inves-tigated here, it is possible to expand adaptive NK cells fromCMV-na€�ve repertoires, albeit the robustness and efficiency ismuch lower (28).

Although no off-target effects have been noted in more than200 patients that have so far received polyclonal conventionalallogeneic NK cells, it is possible that the highly potent adaptiveNK cells, lacking all inhibitory receptors to recipient HLA class I,may cause tissuedamage from targetingmismatchednormal cells.It will be essential to design phase I/II studies with transfer of low

Table 1. Patient characteristics

Sex Age of diagnosis Diagnosis HLA-C typing

M 7 B cell (t12;21) C2/C2M 1 B cell (t12;21) C1/C2M 5 B cell (t12;21) C1/C2M 4 B cell (t12;21) C1/C1F 11 B cell (t12;21) C1/C2F 7 B cell (t12;21) C1/C2F 3 B cell (t12;21) C1/C2M 6 B cell (t12;21) C1/C2F 2 B cell (t12;21) C1/C2M 4 B cell (t12;21) C2/C2F 8 B cell (t12;21) C1/C1F 4 B cell (Heh) C1/C2F 5 B cell (Heh) C1/C1M 3 B cell (Heh) C1/C1M 7 B cell (Heh) C2/C2M 5 B cell (Heh) C2/C2M 16 B cell (Heh) C1/C1F 1 T cell C1/C2F 3 T cell C1/C2M 7 T cell C1/C1M 12 T cell C1/C1M 2 T cell C1/C1F 5 T cell C1/C2

Abbreviations: F, female; Heh, high hyperdiploidy; M, male.

Liu et al.





C1/C1 C1/C2 C2/C20

20

40

60

80

100

%Sp

ecifi

c ki

lling

2DL1+ NK cellsT-ALLB-ALL (t12;21)B-ALL (Heh)

SS

C

SS

C

CD

10

Cas

pase

-3

FSC Cell Trace CD19 Dead cell marker

ALL Blasts only

NK Cells + ALL blasts(matched)

NK Cells + ALL blasts(mismatched)

0

20

40

60

80

100

2DL1+(C2/C2)

2DL3+(C1/C1)

NK Cells

ALL 3140 (C1/C1)

2DL1+(C2/C2)

2DL3+(C1/C1)

NK Cells

ALL 3084 (C2/C2)

0

20

40

60

80

100

2DL1+(C2/C2)

2DL3+(C1/C1)

NK Cells

ALL 3856 (C1/C2)

Caspase-3 DCM

Caspase-3

Live cells

B C

0

20

40

60

80

100

%Sp

ecifi

c ki

lling

%Sp

ecifi

c ki

lling

%Sp

ecifi

c ki

lling

A

B cells

ALL bl

asts

0

100

200

300

MFI

HLA

-E

0

1,000

2,000

3,000

MFI

HLA

-I

*

B cells

ALL bl

asts

ns

FMOALL221.AEH

HLA-E

% o

f Max

D

******

E

F

%C

D10

7a+

NK

cel

ls

+aCD94

ALL blasts 221.AEH

20

40

60

80

0

+aCD94

**** NKG2A+NKG2C-NKG2A-NKG2C+

**

*

***

Figure 4.

Primary pediatric ALL blasts are susceptible to selectively expanded NK cells.A, Example of gating strategy for a pre-B-ALL sample using FACS-based killing assay.B, Representative graphs showing frequency of specific cytolysis using expanded KIR2DL1þ or KIR2DL3þ NK cells against three different ALL patients beingC1/C1, C1/C2, or C2/C2. C, Compiled graph showing frequency of specific lysis using KIR2DL1þ NK cells against three different HLA subgroups of primary pediatricALL blasts being C1/C1, C1/C2, or C2/C2. D, MFI of HLA-E and HLA class I on healthy B cells and B-ALL blasts. Data generated in five experiments on 23 patients(E) Representative graph showing HLA-E level on ALL blasts compared with 721.221.AEH cells. F, Frequency of CD107aþ NKG2C�NKG2Aþ and NKG2CþNKG2A�

NK cells stimulated with primary ALL blasts or 221.AEH with or without blocking antibody to CD94. Shown is a representative experiment out of three.






cell doses of expandedNKcells using gradually increasing doses ofpreconditioning with chemotherapy as a means to indirectlytitrate the length of persistence of this highly cytotoxic subset invivo. Paralleling the clinical testing of new chimeric antigenreceptors by means of mRNA electroporation, any off-targeteffects will be transient in the nonconditioned host, as theallogeneic NK cells will be rapidly rejected by recipient T cells.

Another attractive and potentially safer clinical indication forexpanded adaptive NK cells is for patients with tumors thatdisplay downregulation or loss of HLA class I (55, 56). Resistanceto checkpoint inhibition therapy with anti-PD-1 for patients withmelanoma is associated with point mutations causing decreasedantigen presentation and in one case the complete loss of HLAclass I (57). If this observationwere common acrossmore tumors,

ALL blasts (C2/C2)

Resting IL15 221.wt 221.AEH0

20

40

60

80

100

**

***

ns

0 20 40 60 80 1000

20

40

60

80

100 Resting NK cellsNK cells + 221.wtNK cells + 221.AEH

B%

Spe

cific

kill

ing

% S

peci

fic k

illin

g

Alloreactive subset in CD56dim NK cells (%)

2DL3+ NK Expansions

D

A

Caspase-3Caspase-3 DCMLive cells

PHA Blasts (C1/C1)

Expanded Sorted0

20

40

60

80

100

% S

peci

fic k

illin

g

KIR2DL1+ NK cells(C2/C2)

PHA Blasts (C2/C2)

Expanded Sorted0

20

40

60

80

100

% S

peci

fic k

illin

g

KIR2DL1+ NK cells(C2/C2)

69.8

92.1 3.3

0.93.7

70.8 4.7

4.120.4NKG2C KIR2DL3

KIR2DL3

KIR

2DL1

KIR

2DL1

NK

G2A

99.7

0.0

5.494.6

0.0

KIR2DL3

KIR

2DL1

NKG2C

NK

G2A

Expanded (C2/C2) Sorted (C2/C2)C Figure 5.

Unleashed alloreactive response byresting and expanded adaptive NKcells. A, Killing of mismatched ALLblasts by resting NK cells, overnight-activated IL15-stimulated NK cells, orcells expanded by the indicated feedercells in IL15 for 14 days. Data arecompiled from three experiments andinclude n ¼ 4 to 8 donors for thedifferent conditions. B, Graph showscorrelation between the size of thealloreactive subset and the killing ofmismatched ALL blasts. Data arecompiled from three experiments and4 to 8 donors. C, Expression of NKG2Cand KIR on expanded NK cells at day 14(left) and sorted adaptive NK cellsfrom the same donor. D, Specific lysisof C1/C1 or C2/C2 PHA blasts byexpanded and sorted NK cells in aFACS-based killing assay. C and D,Data are representative of twoindependent sorts and cytotoxicityassays.

Liu et al.





NK cells may find their therapeutic niche as a combination withcheckpoint inhibition or a rescue therapy following relapse. Infact, total or partial loss of HLA class I has been reported atdiagnosis in numerous cancer types (55, 58, 59). Previous obser-vations show a correlation betweenHLA-I expression andNK-cellsusceptibility in pediatric B-ALL (20). Extending those results, wefound that adaptive NK cells efficiently lyse HLA-matched targetsin the presence of HLA class I blockade, mimicking the selectiveloss of HLA class I in HLA-matched tumor cells. The fact that invitro expanded NK cells retained their sensitivity to cognate KIR–HLA interactions despite culture in IL15 suggests that theywill notbe toxic to normal tissues when transferred to HLA-matchedpatients. Furthermore, targeting patients whose tumors displayedlow or absent HLA class I will facilitate treatment of patients whoare C1/C2, the most common KIR ligand genotype.

Reusing and colleagues reported low expression of HLA-E onprimary ALL blasts (25). This raised the question whetherNKG2C was involved in the recognition of the ALL blasts.Antibody-blocking experiments suggested a redundant role forNKG2C in the activation of the in vitro expanded adaptive NKcells. NK-cell recognition of ALL blasts depends on DNAM-1(40). Adaptive NK cells highly express DNAM-1 and CD2(28, 53, 60), of which CD2 acts in synergy with both CD16and NKG2C (53). Both these receptors are stable or inducedduring the in vitro expansion and may contribute to the efficientkilling of ALL cells.

Although our expansion protocol resulted in highly specificcytotoxicity against all targets tested, the frequency of cells thatmobilized CD107a to the cell surface was relatively low, rangingbetween 5%and 30%. A previous report showed a subpopulationof NK cells, termed "serial killers" that can deliver multiple lytichits in a row before exhaustion (61). It is tempting to speculatethat the some of the expanded adaptiveNK cells are "serial killers"and display different killing kinetics that give rise to the discrep-ancy between high specific cytolysis and measured CD107amobilization.

In conclusion, we have demonstrated the robustness of aculture platform yielding expansion and differentiation of adap-tive NK cells with high cytotoxic potential and with predictablespecificity. The feeder cell–based expansion platform should bescalable into a GMP-compliant process, having immediate impli-cations for harnessing adaptive NK cells in cancer immunother-

apy. These results may also guide future attempts to developinduced pluripotent stem cell–derived off-the-shelf adaptiveNK cells (26).

Disclosure of Potential Conflicts of InterestV. Beziat has ownership interest (includingpatents) in PatentWO2014037422

A1. K.-J.Malmberg is a visiting professor at Karolinska Institute, reports receiving acommercial research grant from, has ownership interest (including patents) in,and is a consultant/advisory board member for Fate Therapeutics. No potentialconflicts of interest were disclosed by the other authors.

Authors' ContributionsConception and design: L.L. Liu, V. B�eziat, S. S€oderh€all, K.-J. MalmbergDevelopment of methodology: L.L. Liu, V. B�eziat, K.-J. MalmbergAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): L.L. Liu, V.Y.S. Oei, A. Pfefferle, M. Schaffer,S. Lehmann, E. Hellstr€om-Lindberg, S. S€oderh€all, M. Heyman, D. Grand�erAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): L.L. Liu, V.Y.S. Oei, A. Pfefferle, M. Schaffer,D. Grand�er, K.-J. MalmbergWriting, review, and/or revision of the manuscript: L.L. Liu, V.Y.S. Oei,A. Pfefferle, S. Lehmann, S. S€oderh€all, M. Heyman, D. Grand�er, K.-J. MalmbergAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): V.Y.S. OeiStudy supervision: K.-J. MalmbergOther (Patient clinical subgroup identification during project planning):S. S€oderh€all

Grant SupportThis work was supported by grants from the Swedish Research Council, the

Swedish Children's Cancer Society, the Swedish Cancer Society, the TobiasFoundation, the Karolinska Institutet, the Wenner-Gren Foundation, the Nor-wegian Cancer Society, the Norwegian Research Council, the South-EasternNorway Regional Health Authority, the European Commission Horizon 2020Programme 692180-STREAMH2020-TWINN-2015, grant from National Sci-ence Center, Poland 2014/13/N/NZ6/02081 (toM. Schaffer), and Stiftelsen KGJebsen. V. B�eziat was supported by the French National Research Agency (ANR;grant no. NKIR-ANR-13-PDOC-0025-01).

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 October 27, 2016; revised April 13, 2017; accepted June 14, 2017;published OnlineFirst June 21, 2017.

References1. Kiessling R, Klein E, Pross H, Wigzell H. "Natural" killer cells in the mouse.

II. Cytotoxic cells with specificity for mouse Moloney leukemia cells.Characteristics of the killer cell. Eur J Immunol 1975;5:117–21.

2. Kiessling R, Klein E, Wigzell H. "Natural" killer cells in the mouse. I.Cytotoxic cells with specificity for mouse Moloney leukemia cells.Specificity and distribution according to genotype. Eur J Immunol1975;5:112–7.

3. Herberman RB, Nunn ME, Holden HT, Lavrin DH. Natural cytotoxicreactivity of mouse lymphoid cells against syngeneic and allogeneictumors. II. Characterization of effector cells. Int J Cancer 1975;16:230–9.

4. HerbermanRB,NunnME, LavrinDH.Natural cytotoxic reactivity ofmouselymphoid cells against syngeneic acid allogeneic tumors. I. Distribution ofreactivity and specificity. Int J Cancer 1975;16:216–29.

5. Lanier LL. NK cell recognition. Annu Rev Immunol 2005;23:225–74.6. Carlsten M, Malmberg KJ, Ljunggren HG. Natural killer cell-mediated

lysis of freshly isolated human tumor cells. Int J Cancer 2009;124:757–62.

7. Gonzalez S, Groh V, Spies T. Immunobiology of human NKG2D and itsligands. Curr Top Microbiol Immunol 2006;298:121–38.

8. Moretta A, Bottino C, Pende D, Tripodi G, Tambussi G, Viale O, et al.Identification of four subsets of human CD3-CD16þ natural killer (NK)cells by the expression of clonally distributed functional surfacemolecules:correlation between subset assignment of NK clones and ability tomediatespecific alloantigen recognition. J Exp Med 1990;172:1589–98.

9. Braud VM, Allan DS, O'Callaghan CA, Soderstrom K, D'Andrea A, Ogg GS,et al. HLA-E binds to natural killer cell receptors CD94/NKG2A, B and C.Nature 1998;391:795–9.

10. Uhrberg M, Valiante NM, Shum BP, Shilling HG, Lienert-Weidenbach K,Corliss B, et al. Human diversity in killer cell inhibitory receptor genes.Immunity 1997;7:753–63.

11. Cichocki F, Miller JS, Anderson SK. Killer immunoglobulin-like receptortranscriptional regulation: a fascinating dance of multiple promoters.J Innate Immunity 2011;3:242–8.

12. Elliott JM, Yokoyama WM. Unifying concepts of MHC-dependent naturalkiller cell education. Trends Immunol 2011;32:364–72.






13. Chen S, Yang M, Du J, Li D, Li Z, Cai C, et al. The self-specific activationreceptor SLAM family is critical for NK cell education. Immunity2016;45:292–304.

14. Brodin P, Hoglund P. Beyond licensing and disarming: a quantitative viewon NK-cell education. Eur J Immunol 2008;38:2934–7.

15. Ruggeri L, CapanniM, Casucci M, Volpi I, Tosti A, Perruccio K, et al. Role ofnatural killer cell alloreactivity in HLA-mismatched hematopoietic stemcell transplantation. Blood 1999;94:333–9.

16. Ruggeri L, Aversa F, Martelli MF, Velardi A. Allogeneic hematopoietictransplantation andnatural killer cell recognition ofmissing self. ImmunolRev 2006;214:202–18.

17. Ruggeri L, Capanni M, Urbani E, Perruccio K, ShlomchikWD, Tosti A, et al.Effectiveness of donor natural killer cell alloreactivity in mismatchedhematopoietic transplants. Science 2002;295:2097–100.

18. Miller JS, Soignier Y, Panoskaltsis-Mortari A, McNearney SA, Yun GH,Fautsch SK, et al. Successful adoptive transfer and in vivo expansion ofhuman haploidentical NK cells in patients with cancer. Blood 2005;105:3051–7.

19. Romanski A, Bug G, Becker S, Kampfmann M, Seifried E, Hoelzer D, et al.Mechanisms of resistance to natural killer cell-mediated cytotoxicity inacute lymphoblastic leukemia. Exp Hematol 2005;33:344–52.

20. Pfeiffer M, Schumm M, Feuchtinger T, Dietz K, Handgretinger R, Lang P.Intensity of HLA class I expression and KIR-mismatch determine NK-cellmediated lysis of leukaemic blasts from children with acute lymphaticleukaemia. Br J Haematol 2007;138:97–100.

21. Godal R, Bachanova V, Gleason M, McCullar V, Yun GH, Cooley S, et al.Natural killer cell killing of acute myelogenous leukemia and acutelymphoblastic leukemia blasts by killer cell immunoglobulin-like recep-tor-negativenatural killer cells afterNKG2AandLIR-1blockade. Biol BloodMarrow Transplant 2010;16:612–21.

22. Yawata M, Yawata N, Draghi M, Partheniou F, Little AM, Parham P. MHCclass I-specific inhibitory receptors and their ligands structure diversehuman NK-cell repertoires toward a balance of missing self-response.Blood 2008;112:2369–80.

23. Li H, Pascal V, Martin MP, Carrington M, Anderson SK. Genetic control ofvariegated KIR gene expression: polymorphisms of the bi-directionalKIR3DL1 promoter are associated with distinct frequencies of gene expres-sion. PLoS Genet 2008;4:e1000254.

24. Fauriat C, Andersson S, Bjorklund AT, Carlsten M, Schaffer M, BjorkstromNK, et al. Estimation of the size of the alloreactiveNK cell repertoire: studiesin individuals homozygous for the group A KIR haplotype. J Immunol2008;181:6010–9.

25. Reusing SB, Manser AR, Enczmann J, Mulder A, Claas FH, Carrington M,et al. Selective downregulation of HLA-C and HLA-E in childhood acutelymphoblastic leukaemia. Br J Haematol 2016;174:477–80.

26. Liu LL, Pfefferle A, Yi Sheng VO, Bjorklund AT, Beziat V, Goodridge JP, et al.Harnessing adaptive natural killer cells in cancer immunotherapy.Mol Oncol 2015;9:1904–17.

27. PendeD,Marcenaro S, FalcoM,Martini S, BernardoME,MontagnaD, et al.Anti-leukemia activity of alloreactive NK cells in KIR ligand-mismatchedhaploidentical HSCT for pediatric patients: evaluation of the functionalrole of activating KIR and redefinition of inhibitory KIR specificity. Blood2009;113:3119–29.

28. Beziat V, Liu LL,Malmberg JA, IvarssonMA, Sohlberg E, BjorklundAT, et al.NK cell responses to cytomegalovirus infection lead to stable imprints inthe human KIR repertoire and involve activating KIRs. Blood 2013;121:2678–88.

29. GumaM,AnguloA, Vilches C,Gomez-LozanoN,MalatsN, Lopez-BotetM.Imprint of human cytomegalovirus infection on the NK cell receptorrepertoire. Blood 2004;104:3664–71.

30. Lopez-Verges S, Milush JM, Schwartz BS, Pando MJ, Jarjoura J, York VA,et al. Expansion of a unique CD57þNKG2Chi natural killer cell subsetduring acute human cytomegalovirus infection. Proc Natl Acad Sci U S A2011;108:14725–32.

31. Cichocki F, Cooley S, Davis Z, DeFor TE, Schlums H, Zhang B, et al.CD56dimCD57þNKG2Cþ NK cell expansion is associated withreduced leukemia relapse after reduced intensity HCT. Leukemia 2016;30:456–63.

32. GumaM,BudtM, SaezA, Brckalo T,HengelH,AnguloA, et al. ExpansionofCD94/NKG2Cþ NK cells in response to human cytomegalovirus-infectedfibroblasts. Blood 2006;107:3624–31.

33. Lee N, Goodlett DR, Ishitani A, Marquardt H, Geraghty DE. HLA-E surfaceexpression depends on binding of TAP-dependent peptides derived fromcertain HLA class I signal sequences. J Immunol 1998;160:4951–60.

34. Jiang W, Johnson C, Jayaraman J, Simecek N, Noble J, Moffatt MF, et al.Copy number variation leads to considerable diversity for B but not Ahaplotypes of the human KIR genes encoding NK cell receptors. GenomeRes 2012;22:1845–54.

35. Hwang I, Zhang T, Scott JM, KimAR, Lee T, Kakarla T, et al. Identification ofhuman NK cells that are deficient for signaling adaptor FcRgamma andspecialized for antibody-dependent immune functions. Int Immunol2012;24:793–802.

36. Lee J, Zhang T, Hwang I, Kim A, Nitschke L, Kim M, et al. Epigeneticmodification and antibody-dependent expansion of memory-like NKcells in human cytomegalovirus-infected individuals. Immunity 2015;42:431–42.

37. Lopez-Verges S, Milush JM, Pandey S, York VA, Arakawa-Hoyt J, PircherH, et al. CD57 defines a functionally distinct population of mature NKcells in the human CD56dimCD16þ NK-cell subset. Blood 2010;116:3865–74.

38. Romee R, Foley B, Lenvik T, Wang Y, Zhang B, Ankarlo D, et al. NK cellCD16 surface expression and function is regulated by a disintegrin andmetalloprotease-17 (ADAM17). Blood 2013;121:3599–608.

39. Archimbaud E, Thomas X, Campos L, Magaud JP, Fiere D, Dore JF.Susceptibility of adult acute lymphoblastic leukemia blasts to lysis bylymphokine-activated killer cells. Leukemia 1991;5:967–71.

40. Torelli GF, Peragine N, Raponi S, Pagliara D, De Propris MS, Vitale A, et al.Recognition of adult and pediatric acute lymphoblastic leukemia blastsby natural killer cells. Haematologica 2014;99:1248–54.

41. Curti A, Ruggeri L, Parisi S, Bontadini A, Dan E, Motta MR, et al. Larger sizeof donor alloreactive NK cell repertoire correlates with better responseto NK cell immunotherapy in elderly acute myeloid leukemia patients.Clin Cancer Res 2016;22:1914–21.

42. Foley B, Cooley S, Verneris MR, Pitt M, Curtsinger J, Luo X, et al. Cyto-megalovirus reactivation after allogeneic transplantation promotes a last-ing increase in educated NKG2Cþ natural killer cells with potent function.Blood 2012;119:2665–74.

43. Della Chiesa M, Falco M, Bertaina A, Muccio L, Alicata C, Frassoni F, et al.Human cytomegalovirus infection promotes rapid maturation of NK cellsexpressing activating killer Ig-like receptor in patients transplanted withNKG2C-/- umbilical cord blood. J Immunol 2014;192:1471–9.

44. Sun JC, Beilke JN, Lanier LL. Adaptive immune features of natural killercells. Nature 2009;457:557–61.

45. Alici E, Sutlu T, Bjorkstrand B, Gilljam M, Stellan B, Nahi H, et al.Autologous anti-tumor activity by NK cells expanded from myelomapatients using GMP-compliant components. Blood 2008;111:3155–62.

46. Berg M, Lundqvist A, McCoy P Jr., Samsel L, Fan Y, Tawab A, et al. Clinical-grade ex vivo-expanded human natural killer cells up-regulate activatingreceptors and death receptor ligands and have enhanced cytolytic activityagainst tumor cells. Cytotherapy 2009;11:341–55.

47. Fei F, Lim M, George AA, Kirzner J, Lee D, Seeger R, et al. Cytotoxicity ofCD56-positive lymphocytes against autologous B-cell precursor acutelymphoblastic leukemia cells. Leukemia 2015;29:788–97.

48. Imai C, Iwamoto S, Campana D. Genetic modification of primary naturalkiller cells overcomes inhibitory signals and induces specific killing ofleukemic cells. Blood 2005;106:376–83.

49. Beziat V, Descours B, Parizot C, Debre P, Vieillard V. NK cell terminaldifferentiation: correlated stepwise decrease of NKG2A and acquisition ofKIRs. PLoS One 2010;5:e11966.

50. Beziat V, Sleiman M, Goodridge J, Kaarbo M, liu l, Rollag H, et al.Polyclonal Expansion of NKG2Cþ NK Cells in TAP-deficient Patients.Front Immunol 2015;6:507.

51. Brodin P, Lakshmikanth T, Johansson S, Karre K, Hoglund P. Thestrength of inhibitory input during education quantitatively tunes thefunctional responsiveness of individual natural killer cells. Blood 2009;113:2434–41.

52. Kim S, Poursine-Laurent J, Truscott SM, Lybarger L, Song YJ, Yang L, et al.Licensing of natural killer cells by host major histocompatibility complexclass I molecules. Nature 2005;436:709–13.

53. Liu LL, Landskron J, Ask EH, Enqvist M, Sohlberg E, Traherne JA, et al.Critical role of CD2 co-stimulation in adaptive natural killer cell responsesrevealed in NKG2C-deficient humans. Cell Rep 2016;15:1088–99.

Liu et al.





54. Luetke-Eversloh M, Hammer Q, Durek P, Nordstrom K, Gasparoni G,Pink M, et al. Human cytomegalovirus drives epigenetic imprinting ofthe IFNG locus in NKG2Chi natural killer cells. PLoS Pathogens 2014;10:e1004441.

55. Masuda K, Hiraki A, Fujii N,Watanabe T, TanakaM,Matsue K, et al. Loss ordown-regulation of HLA class I expression at the allelic level in freshlyisolated leukemic blasts. Cancer Sci 2007;98:102–8.

56. Malmberg K, Sohlberg E, Goodridge JP, Ljunggren H. Immune selectionduring tumor checkpoint inhibition therapy paves way for NK-cell "miss-ing self" recognition. Immunogenetics 2017. DOI :10.1007/s00251-017-1011-9, IMMU-D-17-00101.0.

57. Zaretsky JM, Garcia-Diaz A, Shin DS, Escuin-Ordinas H, Hugo W, Hu-Lieskovan S, et al. Mutations associated with acquired resistance to PD-1blockade in melanoma. N Engl J Med 2016;375:819–29.

58. Wang ZG, Cao Y, D'Urso C, Houghton AN, Ferrone S. Molecular abnor-malities in the expression of HLA class-I antigens by melanoma cells. Int JCancer Suppl 1991;6:101–5.

59. Garrido F, Ruiz-Cabello F, Cabrera T, P�erez-Villar JJ, L�opez-Botet M,Duggan-Keen M, et al. Implications for immunosurveillance of alteredHLA class I phenotypes in human tumours. Immunol Today 1997;18:89–95.

60. Enqvist M, Ask EH, Forslund E, Carlsten M, Abrahamsen G, Beziat V, et al.Coordinated Expression of DNAM-1 and LFA-1 in Educated NK Cells.J Immunol 2015;194:4518–27.

61. Vanherberghen B, Olofsson PE, Forslund E, Sternberg-Simon M, Khor-shidi MA, Pacouret S, et al. Classification of human natural killer cellsbased on migration behavior and cytotoxic response. Blood 2013;121:1326–34.






2017;5:654-665. Published OnlineFirst June 21, 2017.Cancer Immunol Res Lisa L. Liu, Vivien Béziat, Vincent Y.S. Oei, et al. Lymphoblastic Leukemia Cells

Expanded Adaptive NK Cells Effectively Kill Primary AcuteEx Vivo

Updated version

10.1158/2326-6066.CIR-16-0296doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerimmunolres.aacrjournals.org/content/suppl/2017/07/12/2326-6066.CIR-16-0296.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerimmunolres.aacrjournals.org/content/5/8/654.full#ref-list-1

This article cites 60 articles, 24 of which you can access for free at:

Citing articles

http://cancerimmunolres.aacrjournals.org/content/5/8/654.full#related-urls

This article has been cited by 3 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

Permissions

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

.http://cancerimmunolres.aacrjournals.org/content/5/8/654To request permission to re-use all or part of this article, use this link



http://cancerimmunolres.aacrjournals.org/lookup/doi/10.1158/2326-6066.CIR-16-0296

http://cancerimmunolres.aacrjournals.org/content/suppl/2017/07/12/2326-6066.CIR-16-0296.DC1

http://cancerimmunolres.aacrjournals.org/content/5/8/654.full#ref-list-1

http://cancerimmunolres.aacrjournals.org/content/5/8/654.full#related-urls

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

mailto:[email protected]

http://cancerimmunolres.aacrjournals.org/content/5/8/654


ex vivo expanded adaptive nk cells effectively kill ...€¦ · 4 hours, blasts were stained with...

Documents