[CANCER RESEARCH 63, 8996–9006, December 15, 2003]

MICA/NKG2D-Mediated Immunogene Therapy of Experimental Gliomas

Manuel A. Friese,1 Michael Platten,1 Stefan Z. Lutz,3 Ulrike Naumann,1 Steffen Aulwurm,1 Felix Bischof,1

Hans-Jorg Buhring,2 Johannes Dichgans,1 Hans-Georg Rammensee,3 Alexander Steinle,3 and Michael Weller1

Departments of 1Neurology and 2Hematology, Oncology, and Immunology, and 3Institute for Cell Biology, Department of Immunology, University of Tubingen, Tubingen,Germany

ABSTRACT

The failure of conventional cancer therapy renders glioblastoma anattractive target for immunotherapy. Tumor cells expressing ligands ofthe activating immunoreceptor NKG2D stimulate tumor immunity medi-ated by natural killer (NK), �� T, and CD8� T cells. We report thathuman glioma cells express the NKG2D ligands MICA, MICB, and mem-bers of the UL16-binding protein family constitutively. However, gliomacells resist NK cell cytolysis because of high MHC class I antigen expres-sion. Plasmid-mediated or adenovirus-mediated overexpression of MICAin glioma cells enhances their sensitivity to NK and T-cell responses invitro and markedly delays the growth of s.c. and intracerebral LN-229human glioma cell xenografts in nude mice and of SMA-560 gliomas insyngeneic VMDk mice. Glioma cells forming progressive tumors afterimplantation of stably MICA-transfected human LN-229 cells lost MICAexpression, indicating a strong selection against MICA expression in vivo.Rejection of MICA-expressing SMA-560 cells in VMDk mice resulted inprotective immunity to a subsequent challenge with wild-type tumor cells.Finally, the growth of syngeneic intracerebral SMA-560 tumors is inhib-ited by peripheral vaccination with adenovirus-mediated, MICA-infectedirradiated tumor cells, and vaccination results in immune cell activation inthe NK and T-cell compartments in vivo. These data commend MICAimmunogene therapy as a novel experimental treatment for human ma-lignant gliomas.

INTRODUCTION

Glioblastoma, the most frequent malignant primary brain tumor,carries a poor prognosis, with a median survival after resection,radiotherapy, and chemotherapy of 12 months (1). The lack of effec-tive immune responses to glial tumors inside the central nervoussystem has been attributed to the immune-privileged status of thebrain conferred by the blood-brain barrier, the lack of conventionallymphatics, and the local release of immunosuppressive factors (2–4).However, both lymphocytes and macrophages infiltrate malignantgliomas, indicating the potential for lymphocyte homing and presen-tation of processed tumor antigens (2).

NKG2D is a C-type, lectin-like homodimeric receptor expressed byhuman natural killer (NK), �� T, and CD8� �� T cells (5). In themouse, NKG2D expression is found on NK cells, activated CD8� Tcells, and lipopolysaccharide-stimulated macrophages (6). Ligation ofNKG2D may be a critical activating pathway to provide tumor sur-veillance (7–9). NKG2D interacts with ligands that are not constitu-tively expressed but are inducibly expressed, including human MHCclass I chain-related A (MICA) and MICB, distant cell stress-induc-ible homologs of MHC class I (10–12). The tissue distribution of MIC

molecules is highly restricted to intestinal epithelia, but they arefrequently expressed in epithelial tumors (11, 13). A second family ofhuman NKG2D ligands (NKG2DL), designated UL16-binding pro-teins (ULBP)1, ULBP2, and ULBP3, has been characterized (14–16).Although no murine MIC molecules have been found, there are mouseorthologs of ULBP, the retinoic acid early inducible gene-1 products(RAE-1). The minor histocompatibility antigen H60 is another murineNKG2DL (6, 17–19). H60 and RAE-1 are up-regulated on tumor cellsand in response to genotoxic stress and induce tumor immunity (7–9).

Activation of NK and effector T cells is inhibited by interactions ofMHC class I antigens with certain receptors originally identified asNK cell receptors but also variably expressed on �� and �� T cells.These inhibitory NK cell receptors include isoforms of the killer cellimmunoglobulin-like receptors (KIR) and immunoglobulin-like tran-script/leukocyte immunoglobulin-like receptors (ILT/LIR) that inter-act with HLA-A, -B, -C, or -G antigens, and a heterodimer formed bythe C-type lectin CD94/NKG2A that interact with HLA-E (20–24).Because a decrease or loss of MHC class I antigen expression mayaccompany neoplastic transformation, NK cells represent a naturaldefense to sense and selectively eliminate abnormal cells (25–27).This NK-mediated cytolytic activity must be down-regulated in thebrain to prevent autoimmune damage because of low expressionlevels of MHC class I antigens on glial cells and a lack of MHC classI antigen expression on neurons (28, 29).

We show here that malignant glioma cells, in contrast to nonneo-plastic astrocytic cells, express the NKG2DL MIC, and ULBP, whichis paralleled by an increase of MHC class I expression, counteractingthe activating potential of NKG2DL. By ectopic overexpression ofMICA, engagement of NKG2D overcomes the inhibitory effect ofMHC class I expression, inducing potent glioma cell sensitization toNK cells and providing costimulation of TCR-dependent T-cell acti-vation in vitro and in vivo. We also demonstrate the feasibility ofMICA overexpression as a strategy to enhance tumor immune sur-veillance and to inhibit the growth of gliomas in human xenogeneicand murine syngeneic glioma models.

MATERIALS AND METHODS

Cell Lines and Transfectants. The human Sv-FHAS astrocytic cell linewas provided by D. Stanimirovic (Institute of Biological Sciences, NationalResearch Council of Canada, Ontario, Ottawa, Canada). The human malignantglioma cell lines were provided by Dr. N. de Tribolet (Centre HospitalierUniversitaire Vaudois, Lausanne, Switzerland). Primary glioblastoma cellswere established from freshly resected tumors, cultured in monolayers, andused between passages 4 and 9 (30). The murine glioma cell lines GL261 andSMA-560 were gifts of X. O. Breakefield (Harvard Medical School, Boston,MA) and D. D. Bigner (Duke University Medical Center, Durham, NC). Thecells were maintained in DMEM supplemented with 2 mM L-glutamine (LifeTechnologies, Inc., Paisley, United Kingdom), 10% FCS (Biochrom KG,Berlin, Germany), and penicillin (100 IU/ml)/streptomycin (100 �g/ml)(Gibco). NKL cells, kindly provided by M. J. Robertson (Indiana UniversitySchool of Medicine, Indianapolis, IN), COS-7, YAC-1, and K562 cells (Amer-ican Type Culture Collection, Rockville, MD) were grown in RPMI 1640medium supplemented with 15% FCS, 2 mM L-glutamine, 1 mM sodiumpyruvate, and penicillin (100 IU/ml)/streptomycin (100 �g/ml). Transfectionswith cDNA clones, RSV.5neo, RSV.5neo-MICA*01 (10, 11) or RSV.5neo-

Received 7/14/03; revised 10/1/03; accepted 10/9/03.Grant support: Grant Fo.01KS9602 from the Federal Ministry of Education and

Research and a grant from the Interdisciplinary Center of Clinical Research Tubingen(IZKF; to M. W.).

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

Requests for reprints: Michael Weller, Department of Neurology, University ofTubingen, Hoppe-Seyler-Strasse 3, D-72076 Tubingen, Germany. Phone: 49-7071-29-87637; Fax: 49-7071-29-5260; E-mail: michael.weller@uni-tuebingen.de; or AlexanderSteinle, Institute for Cell Biology, Department of Immunology, Auf der Morgenstelle 15,72076 D-Tubingen, Germany. Phone 49-7071-29-87648; Fax: 49-7071-5653; E-mail:alexander.steinle@uni-tuebingen.de.

8996

Research. on August 16, 2020. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

PatrMIC1 (31) were conducted using FuGENE6 (Roche, Mannheim, Ger-many) and selection with G418 (1 mg/ml). PatrMIC1 is the chimpanzeehomolog of MICA, which shows 87% homology and was used as a negativecontrol in some experiments (31). Stable, positive isolates were identified byflow cytometry and, to speed up the selection process, were sorted using aFACSvantage (Becton Dickinson, Heidelberg, Germany).

Reverse Transcription (RT)-PCR. Total RNA was prepared usingRNAeasy (Qiagen, Hilden, Germany) and transcribed according to standardprotocols. The conditions for all standard PCR were: 5 min at 95°C, 30 cyclesof 1 min at 95°C, 1 min at 65°C, 1 min at 72°C, and 10 min at 72°C, using thefollowing PCR primers: H60: sense 5�-GATGGTACAGACTCTCTAAGT-TGT-3� and antisense 5�-CAGACCCTGGTTGTCAGAATTATG-3� (554 bp);RAE-1: sense 5�-ATGGCCAAGGCAGCAGTGACCAAG-3� and antisense5�-CACATCGCAAATGCAAATGCAAATAAT-3� (748 bp); MICA: sense5�-GTGCCCCAGTCCTCCAGAGCTCAG-3� and antisense 5�-GTGGCATC-CCTGTGGTCACTCGTC-3� (635 bp); MICB: sense 5�-GGCGTCAGGAT-GGGGTATCTTTGA-3� and antisense 5�-GGCAGGAGCAGTCGTGAGTT-TGCC-3� (690 bp); ULBP1: sense 5�-CTGCAGGCCAGGATGTCTTGTGA-G-3� and antisense 5�-TGAGGGTGGTGGCCATGGCCTTGG-3� (319 bp);ULBP2: sense 5�-CTGCAGGCAAGGATGTCTTGTGAG-3� and antisense5�-TGAGGGTGGTGGCTGTGGCCCTGA-3� (327 bp); ULBP3: sense 5�-CTGCAGGTCAGGATGTCTTGTGAG-3� and antisense 5�-TGAGGGTG-GTGGCTATGGCTTTGG-3� (321 bp); �-actin: sense 5�-AGCCATGTACG-TAGCCAT-3� and antisense 5�-TTTGATGTCACGCACGATTT-3� (232 bp).

Generation of Monoclonal Antibodies. The mouse P815 mastocytomacell line was transfected using FuGENE6 with full-length cDNA of NKG2DLMICA*01, MICA*04, MICB*02, ULBP1, ULBP2, or ULBP3 in RSV.5 neo(15). Transfectants were selected with G418. BALB/c mice were immunizedeither with a mixture of MICA*01-, MICA*04-, and MICB*02-, or ULBP1-,ULBP2-, and ULBP3-expressing P815 cells. Splenocytes of immunized micewere fused with P3 � 63Ag8.653 myeloma cells. Hybridoma supernatantswere tested by indirect immunofluorescence using a FACSCalibur (BectonDickinson) for selective binding to P815-NKG2DL transfectants and to COScells transiently transfected with the various NKG2DL cDNAs. Hybridomasproducing NKG2DL-specific mAbs were subcloned twice, and immunoglobu-lins were isotyped using an isotyping kit (Roche). BAMO-1 (IgG1) andBAMO-2 (IgG2a) recognize MICA and MICB; AMO-1 (IgG1) is MICAspecific; BMO-1 (IgG1) is MICB specific; AUMO-1 (IgG1) is ULBP1 spe-cific; BUMO-2 (IgG1) ULBP2 specific; and CUMO-1 (IgM) is ULBP3specific (32).

Production of Soluble mNKG2D in Insect Cells. Recombinant solublemNKG2D lacking the NH2-terminal cytoplasmic region and transmembranedomain was produced in Sf9 insect cells (Invitrogen, Karlsruhe, Germany)using the BAC-to-BAC system (Life Technologies). The expression constructincluded an NH2-terminal FLAG/hexahistidine (15). Soluble mNKG2D waspurified from culture supernatant by affinity chromatography on Ni2�-chargedchelating Sepharose (BD PharMingen, Heidelberg, Germany) and size-exclu-sion filtration.

Monoclonal Antibodies and Flow Cytometry. Cell surface expression ofMHC class I antigens and NKG2D was assayed with the following monoclonalantibodies (mAbs): W6/32, IgG2a pan anti-HLA-A, -B, -C, -E, -G (Biozol,Echingen, Germany); anti-H-2Kb (BD PharMingen, Heidelberg, Germany);anti-H-2Db (Caltag Laboratories, Hamburg, Germany); and M585, IgG1 anti-NKG2D (kindly provided by Immunex, Seattle, WA). IgG1, IgG2a, and IgMisotype-matched antibodies were used as controls (BD PharMingen). For themNKG2D binding assays, mAb anti-FLAG M2 (5 �g/ml; Sigma) was mixedwith soluble mNKG2D (3 mg/ml) before addition to cells. COS-7 cells weretransiently transfected with neo-control, MICA*01, or PatrMIC1. MICA/B andPatrMIC1 expression levels were detected by flow cytometry using the mAbAMO-1. Interactions of human MICA*01 and chimpanzee PatrMIC1 withmouse NKG2D were analyzed by staining with flagged soluble mouseNKG2D and anti-FLAG mAb. Glioma cells were detached using Accutase(PAA, Vienna, Austria), preincubated in PBS with 2% BSA, and incubatedwith the specific antibody or matched mouse immunoglobulin isotype (10�g/ml) as a control for 30 min on ice. Specific binding was detected withphycoerythrin-conjugated goat anti-mouse IgG (Sigma). Cell surface expres-sion of CD16 (Fc�RIII) was assayed with the mAb 3G8, IgG1 anti-CD16-Cy-Chrome (BD PharMingen). Fluorescence was measured in a Becton DickinsonFACScalibur. Specific fluorescence indexes (SFIs) were calculated by dividing

mean fluorescence obtained with specific antibody by mean fluorescenceobtained with control antibody.

Adenoviral Constructs and Transduction. Replication-deficient recom-binant adenoviruses (Ad5dE1,3) that had the E1A region replaced by thefull-length MICA*01 cDNA (Ad-MICA) were constructed using the AdEasySystem, kindly provided by Bert Vogelstein (The Howard Hughes MedicalInstitute, Johns Hopkins University School of Medicine, Baltimore, MD).Recombinant adenoviral particles were produced in 293 cells (ATCC CRL-1573), and transgene expression was assessed by RT-PCR. Infection of targetcells with recombinant viruses was accomplished by exposing cells to differentconcentrations of adenovirus in PBS containing 10% glycerol for 15 min,followed by the addition of serum-containing medium for 2 days.

Purification of PBL and Isolation of NK and T Cells. Peripheral bloodmononuclear cells were derived from healthy donors by density gradientcentrifugation (Biocoll, Biochrom KG) and depletion of plastic-adherent cells.Peripheral blood lymphocytes (PBLs) were cultured on irradiated RPMI 8866feeder cells to obtain polyclonal NK cell populations. To further enrich NKcells, PBLs were sorted by immunomagnetic depletion using Dynabeads (NKCell Negative Isolation kit; Dynal, Oslo, Norway). CD3�CD56� cells wereused for cytotoxicity assays. To obtain purified T cells, fresh PBLs weredepleted of B cells and monocytes using the LymphoKwik T reagent (OneLambda, Inc., Canoga Park, CA).

Mouse Lymphocyte Isolation. Murine NK cells were prepared fromsplenocytes from CD1-deficient BALB/c nude mice or VMDk mice. NK cellswere positively selected using DX5 mAb-coupled magnetic beads with thecorresponding column system (Miltenyi Biotech, Bergisch Gladbach, Ger-many). Polyclonal mouse NK cells were cultured with mouse IL-2 (5000units/ml; Preprotech, London, United Kingdom) for at least 10 days before usein cytotoxicity assays.

Cytotoxicity Assay. Cytotoxicity was assessed in 4-h 51Cr release assays inthe absence or presence of various mAbs or soluble mNKG2D. The concen-trations for the masking experiments were 10 �g of mAb/ml and 20 �g/ml forsoluble mNKG2D. The 51Cr release assay was performed using 2000 51Cr-labeled targets/well. Effector and target cells were incubated at various E:Tratios for 4 h. Spontaneous release of 51Cr was determined by incubating thetarget cells with medium alone. Maximum release was determined by addingNP40 to a final concentration of 2%. The percentage of 51Cr release wascalculated as follows: 100 � ([experimental release � spontaneous release]/[maximum release � spontaneous release]).

T-Cell Reaction against Allogeneic Glioma Cells. To suppress LN-229glioma cell proliferation, the cells were irradiated at 100 Gy. HLA-A2-positiveT cells or peripheral blood mononuclear cells (105/well) were cocultured with104 irradiated LN-229 HLA-A2-negative glioma cells in 96-well plates intriplicates. After 4 days the cells were pulsed for 24 h with [methyl-3H]thy-midine (0.5 �Ci; Amersham, Braunschweig, Germany) and harvested with acell harvester (Tomtec, Hamden, CT). Incorporated radioactivity was deter-mined in a Wallac 1450 Microbeta Plus Liquid Scintillation Counter. For thepriming assays, nonirradiated glioma cells were incubated with peripheralblood mononuclear cells or T cells for specific priming and clonal expansion.Lymphocytes were harvested on day 5 and used as effectors in a 4-h 51Crrelease cytotoxicity assay. Mock-transfected LN-229.neo cells or LN-229 cellstransfected with MICA*01 or PatrMIC1 were used as targets.

Interleukin (IL)-2 Release Assay. T cells (1 � 106/well) isolated at day 10after tumor inoculation from spleens of vaccinated or control animals werestimulated ex vivo with irradiated SMA-560 cells (1 � 105). T-cell superna-tants from triplicate wells were collected and pooled 48 h later. IL-2 releasewas determined by ELISA (R&D Systems, Minneapolis, MN).

Mice and Animal Experiments. Athymic CD1-deficient BALB/c nudemice were purchased from Charles River Laboratories (Sulzfeld, Germany),and VMDk mice were purchased from the TSE Research Center (Berkshire,United Kingdom). Mice of 6–12 weeks of age were used in all experiments.The experiments were performed according to NIH guidelines, Guide for theCare and Use of Laboratory Animals. Groups of 4–6 mice were injected s.c.in the right flank with transfected LN-229 tumor cells or uninfected or infectedSMA-560 glioma cells (1 � 106 cells) in 0.1 ml of PBS, as indicated. Micewere examined regularly for tumor growth using a metric caliper and killedwhen tumors reached �12-mm diameter. Mice were anesthetized by i.p.injection of 7% chloral hydrate before all intracranial procedures. For intracra-nial implantation, the mice were placed in a stereotactic fixation device

8997

MICA/NKG2D-MEDIATED THERAPY OF GLIOMAS

Research. on August 16, 2020. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

(Stoelting, Wood Dale, IL), and a burr hole was drilled in the skull 2 mmlateral to the bregma. The needle of a Hamilton syringe (Hamilton, Darmstadt,Germany) was introduced to a depth of 3 mm. Five x 104 LN-229 glioma cellsor 5 � 103 SMA-560 cells in a volume of 2 �l of PBS were injected into theright striatum. The mice were observed daily and sacrificed when they devel-oped neurological symptoms.

Immunohistochemistry. Cryosections (0.7 mm) of SMA-560 gliomas,prepared at day 10 after glioma cell inoculation, were fixed in acetone andblocked with 2% normal rabbit serum and 2% BSA. Tissue sections werestained with antibodies (1:50) to CD8 (53-6.7; CD8 T cells) and CD11b(M1/70; macrophages/microglia; BD PharMingen, Heidelberg, Germany). Abiotinylated anti-rat secondary antibody (Zymed, San Francisco, CA) was usedat 1:150. Avidin-biotin complex was added, and the staining was developedwith diaminobenzidine. Murine spleens served as positive controls. Normalmurine brain served as a negative control.

Statistics. Where indicated, analysis of significance was performed usingthe two-tailed Student t test with P � 0.05 considered significant and P � 0.01considered highly significant (Excel; Microsoft, Seattle, WA). Evaluation ofsurvival patterns in mice bearing intracerebral gliomas was performed by theKaplan-Meier method (33), and Ps were evaluated by the Mantel log-rank test(34).

RESULTS

Expression of NKG2DL and MHC Class I Antigens by HumanGlioma Cells. To investigate the expression of the NKG2DL MICA,MICB, ULBP1, ULBP2, and ULBP3 by glioma cells, 12 glioma celllines, a nonneoplastic astrocyte line, Sv-FHAS, and five primary exvivo cultured glioma cells were assessed by RT-PCR (Fig. 1). MICAand ULBP1, ULBP2, and ULBP3 mRNA were expressed by almostall cell lines and most of the primary glioma cell cultures. MICB wasnot detectable in U251MG or U373MG cells and several primaryglioma cell cultures. In contrast, none of the NKG2DLs were detectedin Sv-FHAS cells. We next monitored the cell surface expression ofNKG2DL and MHC class I antigens using mAbs. All glioma cell linesexpressed MICA, ULBP2, and ULBP3. MICB was expressed by 9 of12 cell lines, whereas ULBP1 was only detectable for 5 of 12 cell

lines. All of the primary gliomas expressed MICA and ULBP2, andthe other NKG2DL at various levels (Table 1). MICB was expressedby TU132, TU140, and TU203, whereas TU113 and TU207 did not(Table 1). All cell lines and primary cultures showed high levels ofMHC class I antigen expression. The nonneoplastic Sv-FHAS cell linedid not express any NKG2DL, as determined by RT-PCR and flowcytometry. Sv-FHAS cells showed low MHC class I antigen expres-sion (Table 1).

Stable Transfection and Adenoviral Gene Transfer of MICAinto Glioma Cells Stimulate NK Cytolytic Activity. To determinewhether increased NKG2DL expression confers additional stimula-tory signals for NK cells, we generated stable MICA*01 andPatrMIC1 [chimpanzee MIC homolog (31)] transfectants of the LN-229 and Sv-FHAS cell lines (Fig. 2, A and B). PatrMIC1 was includedin this study to serve as a negative control for the nude mouse modelin vivo (see below). Enhanced MICA or PatrMIC1 expression ren-dered cells susceptible to lysis by human NKL effector cells, suggest-ing that ectopic MIC expression overcomes the inhibitory effect ofMHC class I antigens (Fig. 2, A and B, lower panels). Similar resultswere obtained with freshly isolated polyclonal NK cells (data notshown). For a gene therapeutic approach, we generated an adenoviralvector encoding MICA*01 (Ad-MICA). Infection of LN-229, primaryglioma cells (Fig. 2C), or other glioma cell lines (data not shown) withAd-MICA resulted in a marked increase in MICA expression. Ad-enoviral MICA gene transfer also promoted NK-mediated cytolysis(Fig. 2C, lower panel).

For blocking experiments with mAbs, we first ruled out that apotential lysis seen in the presence of mAbs is attributable to anti-body-dependent cellular cytotoxicity (ADCC). Unlike fresh poly-clonal NK cells isolated from PBLs, the NKL cell line used here doesnot express Fc�RIII (CD16), which mediates ADCC (Fig. 3A; Ref.35). LN-229.neo transfectants were sensitized to NK cytolysis bytreatment with the anti-MHC class I mAb W6/32 (Fig. 3B), confirm-ing that classical MHC class I antigens protect LN-229.neo cells fromNK cytolysis. Anti-MICA/B (BAMO-1) or anti-NKG2D (M585)mAb had no effect on cytolysis in LN-229.neo cells. However, thesemAbs either attenuated (anti-MICA) or blocked (anti-NKG2D) thesensitization mediated by anti-MHC class I mAb. The superior effectof anti-NKG2D compared with anti-MICA reflects the composition ofthe system of a single receptor (NKG2D) in NK cells but multipleligands (MIC and ULBP molecules) expressed by LN-229 cells (Fig.1; Table 1). Soluble mouse NKG2D (mNKG2D) attenuated the anti-MHC class I mAb-mediated sensitization as effectively as anti-MICA,suggesting that murine NKG2D interacts with some human NKG2DL.NK cytolysis of LN-229.MICA targets pretreated with anti-MICA oranti-NKG2D mAb was reduced (Fig. 3C), demonstrating that en-hanced cytolysis of LN-229.MICA cells is attributable to MICAoverexpression, overruling MHC class I antigen inhibitory signals.Similar results were obtained with LN-229.PatrMIC1 transfectants(Fig. 3D).

Costimulatory Functions of MICA-transfected Glioma Cells.Allostimulation assays were performed to analyze whether MICA/NKG2D interactions promote the induction of T-cell responses. LN-229.MICA or LN-229.PatrMIC1 cells were substantially more potentstimulators (3-fold) of T-cell proliferation than LN-229.neo cells (Fig.4A). mAb masking of MICA abrogated the augmentation of T-cellproliferation mediated by MICA but had no effect when T cells werecoincubated with LN-229.neo cells (Fig. 4B). Additional evidence forthe costimulatory potential of NKG2D for T cells was obtained byrechallenging primed T cells. T cells were primed for 4 days withLN-229.neo, LN-229.MICA, or LN-229.PatrMIC1 cells and then usedas effectors against fresh LN-229.neo, LN-229.MICA, or LN-229.PatrMIC1 cells. No target cell lysis was seen when LN-229.neo

Fig. 1. NKG2DL expression in human glioma cells. NKG2DL (MICA, MICB, ULBP1,ULBP2, ULBP3) mRNA expression was assessed by RT-PCR in human glioma cell lines(A) and primary glioma cells (B). �-actin was used as a standard.

8998

MICA/NKG2D-MEDIATED THERAPY OF GLIOMAS

Research. on August 16, 2020. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

cells, LN-229.MICA, or LN-229.PatrMIC1 cells were used as targetsfor LN-229.neo-primed T cells (Fig. 4, C–E). In contrast, LN-229.MI-CA-primed or LN-229.PatrMIC1-primed T cells efficiently killedLN-229.neo, LN-229.MICA, or LN-229.PatrMIC1 cells with compa-rable efficacy, indicating that NKG2D ligation by MICA or PatrMIC1acts as a costimulatory signal in the priming phase of T cells (Fig. 4,C–E).

MICA Expression Delays Growth of Glioma Xenografts inNude Mice. We next sought to examine the immune-stimulatorycapacity of MICA-overexpressing glioma cells in murine in vivomodels. We first tested the physical and functional engagement ofMICA*01 and PatrMIC1 by mouse NKG2D (mNKG2D). Interactionof MICA*01 and PatrMIC1, respectively, with mNKG2D, was as-sessed by using soluble mNKG2D. COS-7 cells transiently transfectedwith either MICA*01 or PatrMIC1 cDNA stained with the mAbAMO-1, but only MICA*01-transfected COS cells bound solublemNKG2D (data not shown). We concluded that mNKG2D interactswith human MICA*01 but not with PatrMIC1. PatrMIC1 differs fromMICA and MICB in several positions directly involved in NKG2Dinteraction, which may selectively interfere with binding of mNKG2D(10–12). Accordingly, NK cell cytotoxicity assays using polyclonalmouse NK cells as effector cells revealed marked lysis of humanMICA glioma cell transfectants but not of mock or PatrMIC1 trans-fectants (Fig. 5A), indicating an activation by MICA*01 of the mouseNKG2D receptor. PatrMIC1 transfectants served hereafter as negativecontrols in the ensuing nude mice studies.

The immune-stimulatory potential of MICA-transfected LN-229cells for the NK cell-mediated, anti-glioma response in vivo wasexamined in s.c. and intracerebral xenograft glioma models in nudemice that have NK cells but lack T cells. LN-229 glioma cells wereinjected s.c., and the tumor sizes were measured every 2 days. LN-229.PatrMIC1 cells grew rapidly to form compact tumors, whereasLN-229.MICA tumor growth was significantly delayed (Fig. 5B).Tumor volumes differed significantly (t test, P � 0.01) from day 17to the end of the experiment. When LN-229 cells were implantedstereotactically into the brains of nude mice, animals carrying LN-229.PatrMIC1 tumors developed neurological symptoms and had tobe sacrificed at days 20–21. In contrast, animals carrying LN-229.MICA tumors showed significantly prolonged survival (Fig. 5C;log-rank test, P � 0.01). Appropriate control experiments disclosedno difference between the proliferation of LN-229.MICA and LN-229.PatrMIC1 cell lines in vitro (Table 2). Because MICA genetransfer delayed glioma growth but did not lead to tumor rejection, we

characterized the glioma cells forming progressive tumors in nudemice. To this end, we isolated the tumor cells, which had grown s.c.,and analyzed the expression of MICA and PatrMIC1 by flow cytom-etry. These experiments yielded two findings that support a proficientrole for MICA as a potent molecule inhibiting the growth of gliomas:(a) glioma cells recovered from mice that had received LN-229.PatrMIC1 cells showed unaltered levels of high PatrMIC1 ex-pression before and after tumor inoculation, consistent with the failureof PatrMIC1 to interact with mNKG2D (Fig. 5D, left); (b) all tumorcells harvested from mice that had received LN-229.MICA*01 cellswere essentially devoid of MICA. This is probably because of tumorformation by a small non-MICA-expressing population among theLN-229.MICA cells (Fig. 5D, right). Endogenous MICA expressionin LN-229.PatrMIC1 cells was also reduced after glioma cell passag-ing in vivo: the SFI decreased from 2.6 before inoculation to 1.4 afterinoculation. MHC class I antigen expression measured by flow cy-tometry did not change after MICA transfection or in vivo passaging(data not shown).

NKG2DL Expression by Mouse Glioma Cell Lines. We evalu-ated two syngeneic mouse glioma models, SMA-560 cells in VMDkmice (spontaneous tumor) and GL261 cells in C57BL/6 mice (3-methylcholanthrene induced), for a therapeutic trial of MICA genetransfer in vivo. The mNKG2DL RAE-1� was prominently expressedby SMA-560 cells, whereas GL261 showed only a weak expression.H60 was only expressed by SMA-560 and not by GL261 cells,consistent with the report that C57BL/6 mice do not express H60 (Ref.8; Fig. 6A). Soluble mNKG2D was more prominently bound bySMA-560 than GL261 cells (Fig. 6B), indicating higher expressionlevels of mNKG2DL in SMA-560 cells. Whereas SMA-560 cellsexpressed both MHC class I antigens H-2Kb and H-2Db, which arepotential ligands for inhibitory Ly49 receptors, GL261 cells showedonly low level H-2Db expression (Fig. 6B). SMA-560 cells thusresemble the phenotype of human glioma cell lines and were used infurther animal experiments. When SMA-560 cells were infected withAd-MICA, MICA expression increased in a multiplicity of infection(MOI)-dependent manner: the SFI values were 6.6 at 100 MOI, 12.2at 300 MOI, and 22.8 at 1000 MOI (Fig. 7B). Similar to LN-229 cells,the growth of SMA-560 cells was unaffected by forced MICA ex-pression (Table 2). The functional significance of increased MICAexpression on mouse glioma cells after Ad-MICA infection wasconfirmed in cytotoxicity assays using syngeneic VMDk NK cells aseffectors (Fig. 6C). Blocking experiments with anti-MICA mAb andsoluble mNKG2D demonstrated that the increased killing of Ad-

Table 1 Flow cytometric analysis of MHC class I antigens and NKG2DL expression by human malignant glioma cell lines, non-neoplastic Sv-FHAS astrocytes, and primaryglioma cell cultures

The cells were stained with the indicated mAb or isotype-matched Ig. Expression was quantified as SFI values (mean fluorescencespecific mAb/mean fluorescenceisotyp control).

Cell lineMHC class I

(W6/32)MICA

(AMO-1)MICB

(BMO-1)MICA/B

(BAMO-1)ULBP1

(AUMO-1)ULBP2

(BUMO-2)ULBP3

(CUMO-1)

A172 145 13.0 2.3 11.2 1.8 4.1 1.8D247 113 8.2 1.5 6.5 1.0 1.6 1.4LN-18 85 6.1 3.6 7.3 1.0 2.4 1.9LN-229 215 6.9 2.0 5.2 1.1 5.8 6.7LN-308 153 4.7 1.8 5.0 1.1 3.1 2.6LN-319 112 9.7 2.2 8.9 1.3 3.7 4.4LN-428 129 5.9 1.6 6.3 1.2 3.1 2.2T98G 480 12.3 5.0 10.3 1.6 6.7 8.6U87MG 351 8.6 2.2 6.4 1.3 5.9 8.4U138MG 346 11.7 1.2 11.2 2.4 3.6 3.4U251MG 216 14.5 1.0 9.2 1.0 7.6 5.3U373MG 358 11.6 1.0 7.0 1.0 11.8 6.0Sv-FHAS 90 1.0 1.0 1.0 1.0 1.1 1.2TU113 29 4.0 1.0 3.9 1.0 1.7 1.4TU203 23 3.8 1.3 3.5 1.3 1.7 1.1TU207 38 1.4 1.1 1.4 1.1 1.7 1.1TU132 95 6.2 1.4 4.4 1.0 1.3 1.0TU140 100 9.1 2.5 7.4 1.3 1.8 1.0

8999

MICA/NKG2D-MEDIATED THERAPY OF GLIOMAS

Research. on August 16, 2020. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

Fig. 2. Ectopic MICA*01 and PatrMIC1 expression by glioma cells promote NK cell-mediated cytolysis. LN-229 (A) or Sv-FHAS (B) cells transfected with MICA*01 or PatrMIC1were assessed for MICA (mAb BAMO-1) or PatrMIC1 (mAb BAMO-2) expression (filled profiles) as compared with mock transfectants (neo-control) by flow cytometry. Open profilesshow the labeling with isotype-matched immunoglobulin. The SFI values are indicated in the upper right of each panel. C, LN-229 or TU113 primary glioma cells were infected withAd-MICA or Ad-dE1 (300 MOI) and assessed for MICA expression accordingly. In the lower panels of A–C, the cell lines were tested as targets for human NKL cells in 51Cr-releaseassays. Data are expressed as specific lysis at different E:T ratios (means; bars, SD). t test: �, P � 0.05; ��, P � 0.01.

9000

MICA/NKG2D-MEDIATED THERAPY OF GLIOMAS

Research. on August 16, 2020. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

MICA-infected SMA-560 cells was attributable to MICA expression.Attenuation of the constitutive lysis of control-infected SMA-560cells by �50% by mNKG2D probably reflects the activation byendogenous NKG2DLs such as RAE-1 and H60 (Fig. 6D). Accord-ingly, neutralization of MICA was less effective in inhibiting lysis ofAd-MICA-infected cells than mNKG2D.

MICA Expression in SMA-560 Glioma Cells Delays TumorGrowth. To analyze the immune-stimulatory capacity of MICA-transduced SMA-560 tumor cells in vivo, VMDk mice were inocu-lated s.c. with syngeneic tumor cell transductants. There was nodifference between the in vivo growth of Ad-dE1-infected and unin-

Fig. 4. Modulation of alloreactivity to glioma cells by MICA and PatrMIC1. A,irradiated LN-229.neo, LN-229.MICA, or LN-229.PatrMIC1 cells were coincubated withHLA-A2-mismatched T cells. [3H]Thymidine was added on day 4 for 16 h, and incor-poration was measured by liquid scintillation counting. B, irradiated glioma cells werepreincubated with isotype control antibody or anti-MICA (BAMO-1) and processed as inA. Data are expressed as cpm (bars, SD). t test: ��, P � 0.01, compared with LN-229.neoin A or isotype control antibodies in B. C–E, T cells were incubated with glioma cells,harvested on day 4, and used as effectors in a 4-h 51Cr-release assay at the indicated E:Tratios (means; bars, SD). t test: �, P � 0.05; ��, P � 0.01.

Fig. 3. NKG2DL and MHC class I antigen expression determines NK cytolysis. A, theNKL cell line and fresh polyclonal NK cells were assessed for CD16 (Fc�RIII) expressionby flow cytometry, which might mediate ADCC in mAb blocking experiments. Openprofiles show the labeling with isotype-matched immunoglobulin. The SFI values areindicated in the upper right of each panel. B–D, different target cell lines were pretreatedwith control immunoglobulin or anti-MICA BAMO-1 mAb, anti-PatrMIC1 BAMO-2mAb, anti-MHC class I W6/32 mAb, or soluble mNKG2D. NKL effector cells werepretreated with control immunoglobulin or anti-NKG2D M585 mAb for 30 min before usein 51Cr-release cytotoxicity assays. LN-229.neo (B), LN-229.MICA (C), or LN-229.PatrMIC1 (D) were used as targets. The specific lytic activities are given for an E:Tratio of 40:1 (means; bars, SD). t test: �, P � 0.05; ��, P � 0.01, compared with isotypecontrol immunoglobulin.

9001

MICA/NKG2D-MEDIATED THERAPY OF GLIOMAS

Research. on August 16, 2020. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

fected SMA-560 cells. Conversely, MICA*01 expression by SMA-560 glioma cells resulted in a substantial delay of tumor growth in aMOI-dependent manner (Fig. 7A). Of note, 2 of 6 animals in thegroups challenged with 300 and 1000 MOI-infected glioma cells,respectively, developed no tumor. Irrespective of the initial level ofMICA expression before inoculation, glioma cells harvested fromprogressive tumors at day 10 after inoculation were devoid of MICA(Fig. 7B). To address whether prior immunization with glioma cellsthat express MICA induces protective immunity to wild-type gliomacells, mice that had previously rejected MICA-transduced tumor cellswere rechallenged with wild-type SMA-560 cells 10 weeks after thefirst exposure in the opposite flank. Wild-type SMA-560 cells grewprogressively in naıve VMDk mice but were rejected by mice that hadbeen exposed previously to the MICA-transduced tumor cells (Fig.7C).

s.c. Vaccination with MICA-expressing Glioma Cells after Tu-mor Inoculation Delays Tumor Growth. To address whether s.c.vaccination with tumor cells expressing MICA induces protectiveimmunity to pre-established SMA-560 gliomas, mice were inoculatedintracerebrally with SMA-560 wild-type cells and then immunizedwith irradiated wild-type cells or irradiated Ad-dE1 (1000 MOI)-infected or Ad-MICA (1000 MOI)-infected SMA-560 cells at days 3and 8 (Fig. 7D). Peripheral vaccination with Ad-MICA-infected irra-diated cells resulted in a significant prolongation of survival with 2 of6 animals still alive and asymptomatic after 90 days (log-rank test,P � 0.01). NK cells isolated from splenocytes from mice vaccinatedwith Ad-MICA-transduced SMA-560 cells at day 10 showed a sub-stantially enhanced cytotoxic activity against YAC-1 target cellscompared with NK cells from animals vaccinated with control-trans-duced SMA-560 cells (Fig. 7E). Evidence for T-cell stimulation byMICA was obtained by restimulating T cells isolated at day 10 aftertumor cell inoculation. Only T cells from animals vaccinated withAd-MICA-infected SMA-560 cells released high levels of IL-2 afterrestimulation with SMA-560 wild-type cells, whereas no such effectwas seen with control-transduced tumor cells (Fig. 7F). To determinethe in situ distribution of glioma-infiltrating immune cells, we ana-lyzed brain sections at day 10 after tumor inoculation by immunohis-tochemistry (Fig. 8). Using specific mAbs for CD8� T cells (CD8)and macrophages/microglia (CD11b), we observed an increase in thenumber of infiltrating immune cells in SMA-560 gliomas after SMA-560.Ad-MICA vaccination compared with SMA-560 or SMA-560.Ad-dE1-vaccinated animals. We were not able to stain murineNK cells with the currently available antibodies (36). An antibodyrecognizing Ly-49G2 (4D11), which has been reported to stain murineNK cells in immunohistochemistry, provided no signal in our hands,probably because VMDk mice do not express Ly-49G2 (36).

DISCUSSION

This study investigates the potential of NKG2DL to promote animmune response to glioblastoma. We show that NKG2D-mediated

Fig. 5. Ectopic MICA expression delays the growth of s.c. and intracerebral humanglioma xenografts in nude mice. A, LN-229 transfectants were tested as targets forBALB/c nude mice NK cells in a 51Cr release assay (means; bars, SD). t test: ��,P � 0.01). B, the growth of s.c. LN-229.PatrMIC1 or LN-229.MICA tumors wasmonitored every 2 days (mean; bars, SD). t test: ��, P � 0.01. C, LN-229.PatrMIC1 (solidline) or LN-229.MICA*01 (broken line) cells (5 � 104) were inoculated intracerebrally inBALB/c nude mice. Survival data for five animals per group are shown, evaluated by theKaplan-Meier method (log-rank test, P � 0.01). D, at 25 days after injection of tumorcells, the mice were killed, and freshly isolated tumor cells from s.c. tumors were analyzedfor PatrMIC1 in PatrMIC1 tumors (left) or MICA in MICA tumors (right) by flowcytometry. Representative tumor isolates are shown as gray profiles, the preinoculationlevels as black profiles, and an isotype control as open profiles.

Table 2 MICA gene transfer does not modulate glioma cell growth as assessed by[3H]thymidine incorporation in vitro

[3H]Thymidine incorporation was measured after 24 h by liquid scintillation counting.Data are expressed as cpm � SD.

Proliferation (cpm)

LN-229 SMA-560

Parental 1802 � 401 3078 � 523PatrMIC1 (stable) 1766 � 431MICA (stable) 1722 � 386Ad-dE1 (300 MOI) 3153 � 500Ad-MICA (300 MOI) 3122 � 489

9002

MICA/NKG2D-MEDIATED THERAPY OF GLIOMAS

Research. on August 16, 2020. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

NK cell triggering induced by NKG2DLs expressed on glioma cells isoverruled by MHC class I antigens that engage inhibitory receptors onNK cells. Gene transfer-mediated ectopic expression of MICA onhuman and murine glioma cells permits their in vitro immune recog-nition and lysis and in vivo growth inhibition, despite a high levelMHC class I antigen expression. These observations provide a ration-ale for MICA-based immunogene therapy of gliomas.

Malignant transformation of glial cells is associated with anNKG2DL expression, i.e., of MICA/B and ULBP molecules, at thecell surface (Fig. 1; Table 1). However, the NK cell inhibitorypotential of MHC class I antigens expressed on human glioma cellsmasks the activating signals of NKG2DL. The interaction of effectorNKG2D with endogenous tumor MIC and ULBP molecules promotesNK cell-mediated tumor cell lysis in vitro when MHC class I antigensare masked (Fig. 3B). NK cell-mediated tumor cell lysis was alsoachieved when MICA expression was enhanced via plasmid transfec-tion or adenoviral gene transfer (Figs. 2 and 3C; Ref. 27). Thus,glioma cells outbalance activating ligand expression by inhibitoryMHC class I antigens and prevent NK cell lysis. This concept issupported by a highly significant correlation between MHC class Iantigen expression and NKG2DL expression on human glioma cellswhen the SFI values for the NK cell inhibitory MHC class I antigenswere correlated with a sum score of the SFI values for MICA, MICB,ULBP1, ULBP2, and ULBP3 (t test, r2 0.6523, P 0.0015).

One way of immune escape for glioma cells may be the expressionof MHC class I antigens at a level that allows sufficient binding of NKinhibitory receptors and escape from NK-mediated killing. High ex-pression of MHC class I antigens in human glioblastomas in vivo hasbeen confirmed by several studies (37, 38). MHC class I down-regulation, conversely, is a hallmark of many peripheral cancers anddecreases the susceptibility of tumor cells to lysis by cytotoxic T cells(26, 39). Possibly only those glioma cells that show an up-regulationof MHC class I antigens expression to compensate for the acquisitionof activating NKG2DL expression during malignant transformationwill survive. This selection may render tumor cells more vulnerabletoward T cell-dependent elimination through tumor-associated anti-gen detection (26, 39). Because the brain lacks specific brain-associ-ated lymphoid tissue, antigens that are introduced directly into thebrain parenchyma exclusively elicit transient innate inflammatoryimmune responses but no adaptive immunity (40–44). However, ifimmunization is elicited by injecting the identical antigen outside thebrain, the adaptive immune system becomes primed, and antigenicepitopes anywhere within the brain will be recognized as targets ofeither activated effector T cells, B cells, or antibodies (40, 42, 44).This immunological peculiarity may explain high expression levels ofMHC class I antigens on glioma cells that could effectively presenttumor-associated antigens to primed T cells. Because intracerebralT-cell priming is not found, the MHC class I antigens may only serveto suppress NK cell activation by NKG2DL.

The highly lethal nature of glioblastoma suggests that the levels ofactivating NKG2DLs expressed by glioma cells are too low to induceantitumor immunity. Thus, we concluded that immunity to gliomasmay be boosted by engineering cells expressing high levels of acti-vating NKG2DL. We noted that MICA gene transfer induces NK andT-cell responses to glioma cells in vitro (Figs. 2–4) and delays growthin a human xenograft and syngeneic mouse glioma model in vivo(Figs. 5–7). In this model, the in vivo selection for low MICA-expressing glioma cells strikingly demonstrated the NK cell activatingpotential of the NKG2D/NKG2DL system (Figs. 5D and 7B). Previ-ously, it was shown that ligation of NKG2D on NK and T cells canpromote subsequent T-cell immunity to parental tumors that lackNKG2DL, although T-cell priming was NK cell independent (8).However, a similar study did not detect such T cell-mediated memory

Fig. 6. SMA-560 glioma cells in VMDk mice as syngeneic model for MICA genetransfer. A, expression of the murine NKG2DL RAE-1� and H60 mRNA was assessed byRT-PCR. YAC-1 cells were used as a positive control. �-actin was used as a standard. B,NKG2DL and MHC class I antigen expression (H-2Kb, H-2Db) was assessed by flowcytometry. The cell lines were stained with soluble mNKG2D/anti-FLAG or the indicatedmAb (solid profiles) or with isotype-matched control antibody (open profiles). SFI valuesare indicated in the upper right corner of each panel. C, uninfected, Ad-dE1 (300MOI)-infected or Ad-MICA (300 MOI)-infected SMA-560 cells were tested as targets forNK cells isolated from VMDk mice in 51Cr release assays at 48 h (means; bars, SD). t test:��, P � 0.01. D, Ad-dE1- or Ad-MICA-infected (300 MOI) SMA-560 cells wereuntreated or pretreated with soluble mNKG2D, or control antibody or anti-MICA, andtested as targets in a 51Cr-release assay, using VMDk NK cells as effectors. Data areexpressed as specific lytic activities at an E:T ratio of 40:1 (means; bars, SD). t test: �,P � 0.05; ��, P � 0.01.

9003

MICA/NKG2D-MEDIATED THERAPY OF GLIOMAS

Research. on August 16, 2020. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

Fig. 7. MICA delays SMA-560 glioma growth in syngeneic mice. A, the growth of s.c. tumors formed by SMA-560 cells infected in vitro with Ad-dE1 (1000 MOI) or Ad-MICA(100, 300, or 1000 MOI) was monitored every 2 days. Two of 6 animals challenged with Ad-MICA-infected cells at 300 and 1000 MOI did not develop a measurable tumor. B,SMA-560 cells infected with Ad-MICA at increasing MOI were assessed for MICA expression before subcutaneous inoculation (black profiles). Open profiles show the labeling withisotype-matched immunoglobulin. Furthermore, glioma cells were freshly isolated from s.c. tumors at 10 days after inoculation and reanalyzed for MICA expression (gray profiles).C, VMDk mice that had previously rejected MICA-transduced SMA-560 cells were inoculated s.c. with wild-type SMA-560 glioma cells in the opposite flank. Primary exposurepreceded the challenge by 10 weeks (means; bars, SD). t test: ��, P � 0.01. D, 5 � 103 SMA-560 cells were inoculated intracerebrally in syngeneic VMDk mice (day 1). At days3 and 8, the animals were vaccinated s.c. with 1 � 106 irradiated Ad-dE1- or Ad-MICA-infected (1000 MOI) or uninfected SMA-560 cells. The graph shows survival data for 6animals/group, evaluated by the Kaplan-Meier method (log-rank test, P � 0.01). E and F, at day 10, splenocytes were recovered from the differently vaccinated animals, and T andNK cells were isolated. E, NK cells were used as effector cells in a 51Cr release assay using YAC-1 cells as targets (means; bars, SD). t test: �, P � 0.05; ��, P � 0.01. F, isolatedT cells were restimulated with irradiated wild-type SMA-560 cells, and IL-2 release was measured by ELISA 48 h later.

9004

MICA/NKG2D-MEDIATED THERAPY OF GLIOMAS

Research. on August 16, 2020. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

against the parental tumor in mice (7). Our data support that NKG2Dengagement costimulates human CTL responses because prior immu-nization with tumor cells expressing the NKG2DL MICA inducesprotective immunity against a challenge with wild-type glioma cells inmice that had previously rejected MICA-transduced glioma cells (Fig.7C; Ref. 8, 45). Finally, the peripheral vaccination with irradiatedAd-MICA-infected cells after the intracerebral implantation of wild-type tumor cells delayed the growth of syngeneic gliomas and pro-moted immune activation in the NK and T-cell compartments (Fig. 7,E and F, and Fig. 8). NKG2DL expression and consequent activationof NK cells and T cells may thus provide a novel therapeutic approachfor the treatment of human gliomas. If the magnitude of the effectachieved with peripheral vaccination, i.e., a median prolongation ofsurvival from 27.5 in control animals to 46 days in SMA-560.Ad-MICA-vaccinated animals, with 2 of 6 animals (33%) alive andasymptomatic after 90 days (Fig. 7D), could be transferred into thehuman situation, this would be a major advance, compared with allchemotherapy trials performed in the recent 20 years (46).

ACKNOWLEDGMENTS

We thank B. Frank, U. Obermuller, and S. Weit for excellent technicalassistance, J. Wischhusen and U. Herrlinger for helpful discussions, and F.Duffner and E. H. Grote for surgical glioma specimens.

REFERENCES

1. Surawicz, T. S., Davis, F., Freels, S., Laws, E. R., Jr., and Menck, H. R. Brain tumorsurvival: results from the National Cancer Data Base. J. Neurooncol., 40: 151–160,1998.

2. Weller, M., and Fontana, A. The failure of current immunotherapy for malignantglioma. Tumor-derived TGF-�, T-cell apoptosis, and the immune privilege of thebrain. Brain Res. Brain Res. Rev., 21: 128–151, 1995.

3. Saas, P., Walker, P. R., Hahne, M., Quiquerez, A. L., Schnuriger, V., Perrin, G.,French, L., Van Meir, E. G., de Tribolet, N., Tschopp, J., and Dietrich, P. Y. Fasligand expression by astrocytoma in vivo: maintaining immune privilege in the brain?J. Clin. Investig., 99: 1173–1178, 1997.

4. Miller, D. W. Immunobiology of the blood-brain barrier. J. Neurovirol., 5: 570–578,1999.

5. Bauer, S., Groh, V., Wu, J., Steinle, A., Phillips, J. H., Lanier, L. L., and Spies, T.Activation of NK cells and T cells by NKG2D, a receptor for stress-inducible MICA.Science (Wash. DC), 285: 727–729, 1999.

6. Diefenbach, A., Jamieson, A. M., Liu, S. D., Shastri, N., and Raulet, D. H. Ligandsfor the murine NKG2D receptor: expression by tumor cells and activation of NK cellsand macrophages. Nat. Immunol., 1: 119–126, 2000.

7. Cerwenka, A., Baron, J. L., and Lanier, L. L. Ectopic expression of retinoic acid earlyinducible-1 gene (RAE-1) permits natural killer cell-mediated rejection of a MHCclass I-bearing tumor in vivo. Proc. Natl. Acad. Sci. USA, 98: 11521–11526, 2001.

8. Diefenbach, A., Jensen, E. R., Jamieson, A. M., and Raulet, D. H. Rae1 and H60ligands of the NKG2D receptor stimulate tumour immunity. Nature (Lond.), 413:165–171, 2001.

9. Girardi, M., Oppenheim, D. E., Steele, C. R., Lewis, J. M., Glusac, E., Filler, R.,Hobby, P., Sutton, B., Tigelaar, R. E., and Hayday, A. C. Regulation of cutaneousmalignancy by �� T cells. Science (Wash. DC), 294: 605–609, 2001.

10. Bahram, S., Bresnahan, M., Geraghty, D. E., and Spies, T. A second lineage ofmammalian major histocompatibility complex class I genes. Proc. Natl. Acad. Sci.USA, 91: 6259–6263, 1994.

11. Groh, V., Bahram, S., Bauer, S., Herman, A., Beauchamp, M., and Spies, T. Cellstress-regulated human major histocompatibility complex class I gene expressed ingastrointestinal epithelium. Proc. Natl. Acad. Sci. USA, 93: 12445–12450, 1996.

12. Li, P., Morris, D. L., Willcox, B. E., Steinle, A., Spies, T., and Strong, R. K. Complexstructure of the activating immunoreceptor NKG2D and its MHC class I-like ligandMICA. Nat. Immunol., 2: 443–451, 2001.

13. Groh, V., Rhinehart, R., Secrist, H., Bauer, S., Grabstein, K. H., and Spies, T. Broadtumor-associated expression and recognition by tumor-derived �� T cells of MICAand MICB. Proc. Natl. Acad. Sci. USA, 96: 6879–6884, 1999.

14. Cosman, D., Mullberg, J., Sutherland, C. L., Chin, W., Armitage, R., Fanslow, W.,Kubin, M., and Chalupny, N. J. ULBPs, novel MHC class I-related molecules, bind

Fig. 8. Ad-MICA gene transfer promotes im-mune cell infiltration in the SMA-560 syngeneicmouse glioma model. VMDk mice were inoculatedintracerebrally with SMA-560 wild-type cells andvaccinated s.c. with SMA-560 (A–C), SMA-560.Ad-dE1 (D–F), or SMA-560.Ad-MICA (G–I)at days 3 and 8 after inoculation. Cryosections weregenerated from brains at day 10 after tumor inoc-ulation and stained with mAb directed againstmacrophages/microglia (A, D, and G) or CD8� Tcells (B, E, H), or with isotype-matched controlmAb (C, F, I). Spleen sections from untreatedanimals were used as positive controls (J, K, and L)Magnification for all sections was �20. M, quan-tification of positively stained cells per visual fieldin glioma sections in �20 magnification. Eachcolumn represents the average number (bars, SD; ttest: ��, P � 0.01) of positively stained cells fromat least three different brain sections.

9005

MICA/NKG2D-MEDIATED THERAPY OF GLIOMAS

Research. on August 16, 2020. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

to CMV glycoprotein UL16 and stimulate NK cytotoxicity through the NKG2Dreceptor. Immunity, 14: 123–133, 2001.

15. Steinle, A., Li, P., Morris, D. L., Groh, V., Lanier, L. L., Strong, R. K., and Spies, T.Interactions of human NKG2D with its ligands MICA, MICB, and homologs of themouse RAE-1 protein family. Immunogenetics, 53: 279–287, 2001.

16. Sutherland, C. L., Chalupny, N. J., Schooley, K., Van den Bos, T., Kubin, M., andCosman, D. UL16-binding proteins, novel MHC class I-related proteins, bind toNKG2D and activate multiple signaling pathways in primary NK cells. J. Immunol.,168: 671–679, 2002.

17. Nomura, M., Zou, Z., Joh, T., Takihara, Y., Matsuda, Y., and Shimada, K. Genomicstructures and characterization of Rae1 family members encoding GPI-anchored cellsurface proteins and expressed predominantly in embryonic mouse brain. J. Biochem.(Tokyo), 120: 987–995, 1996.

18. Malarkannan, S., Shih, P. P., Eden, P. A., Horng, T., Zuberi, A. R., Christianson, G.,Roopenian, D., and Shastri, N. The molecular and functional characterization of adominant minor H antigen, H60. J. Immunol., 161: 3501–3509, 1998.

19. Cerwenka, A., Bakker, A. B., McClanahan, T., Wagner, J., Wu, J., Phillips, J. H., andLanier, L. L. Retinoic acid early inducible genes define a ligand family for theactivating NKG2D receptor in mice. Immunity, 12: 721–727, 2000.

20. Brooks, A. G., Posch, P. E., Scorzelli, C. J., Borrego, F., and Coligan, J. E. NKG2Acomplexed with CD94 defines a novel inhibitory natural killer cell receptor. J. Exp.Med., 185: 795–800, 1997.

21. Carena, I., Shamshiev, A., Donda, A., Colonna, M., and Libero, G. D. Majorhistocompatibility complex class I molecules modulate activation threshold and earlysignaling of T cell antigen receptor-�/� stimulated by nonpeptidic ligands. J. Exp.Med., 186: 1769–1774, 1997.

22. Colonna, M., Navarro, F., Bellon, T., Llano, M., Garcia, P., Samaridis, J., Angman,L., Cella, M., and Lopez-Botet, M. A common inhibitory receptor for major histo-compatibility complex class I molecules on human lymphoid and myelomonocyticcells. J. Exp. Med., 186: 1809–1818, 1997.

23. Braud, V. M., Allan, D. S., O’Callaghan, C. A., Soderstrom, K., D’Andrea, A., Ogg,G. S., Lazetic, S., Young, N. T., Bell, J. I., Phillips, J. H., Lanier, L. L., andMcMichael, A. J. HLA-E binds to natural killer cell receptors CD94/NKG2A, B andC. Nature (Lond.), 391: 795–799, 1998.

24. Lanier, L. L. NK cell receptors. Annu. Rev. Immunol., 16: 359–393, 1998.25. Ljunggren, H. G., and Karre, K. In search of the “missing self”: MHC molecules and

NK cell recognition. Immunol. Today, 11: 237–244, 1990.26. Garrido, F., Ruiz-Cabello, F., Cabrera, T., Perez-Villar, J. J., Lopez-Botet, M.,

Duggan-Keen, M., and Stern, P. L. Implications for immunosurveillance of alteredHLA class I phenotypes in human tumours. Immunol. Today, 18: 89–95, 1997.

27. Pende, D., Cantoni, C., Rivera, P., Vitale, M., Castriconi, R., Marcenaro, S., Nanni,M., Biassoni, R., Bottino, C., Moretta, A., and Moretta, L. Role of NKG2D in tumorcell lysis mediated by human NK cells: cooperation with natural cytotoxicity recep-tors and capability of recognizing tumors of nonepithelial origin. Eur. J. Immunol.,31: 1076–1086, 2001.

28. Joly, E., Mucke, L., and Oldstone, M. B. Viral persistence in neurons explained bylack of major histocompatibility class I expression. Science (Wash. DC), 253:1283–1285, 1991.

29. Neumann, H., Cavalie, A., Jenne, D. E., and Wekerle, H. Induction of MHC class Igenes in neurons. Science (Wash. DC), 269: 549–552, 1995.

30. Rieger, J., Wick, W., and Weller, M. Human malignant glioma cells express sema-phorins and their receptors, neuropilins and plexins. Glia, 42: 379–389, 2003.

31. Steinle, A., Groh, V., and Spies, T. Diversification, expression, and �� T cellrecognition of evolutionarily distant members of the MIC family of major histocom-

patibility complex class I-related molecules. Proc. Natl. Acad. Sci. USA, 95: 12510–12515, 1998.

32. Welte, S. A., Sinzger, C., Lutz, S. Z., Singh-Jasuja, H., Sampaio, K. L., Eknigk, U.,Rammensee, H. G., and Steinle, A. Selective intracellular retention of virally inducedNKG2D ligands by the human cytomegalovirus UL16 glycoprotein. Eur J. Immunol.,33: 194–203, 2003.

33. Kaplan, E. L., and Meyer, P. Non-parametric estimation from incomplete observa-tions. J. Am. Stat. Assoc., 53: 457–481, 1958.

34. Mantel, N. Evaluation of survival data and two new rank order statistics arising in itsconsideration. Cancer Chemother. Rep., 50: 163–170, 1966.

35. Robertson, M. J., Cochran, K. J., Cameron, C., Le, J. M., Tantravahi, R., and Ritz,J. Characterization of a cell line, NKL, derived from an aggressive human naturalkiller cell leukemia. Exp. Hematol., 24: 406–415, 1996.

36. Dokun, A. O., Chu, D. T., Yang, L., Bendelac, A. S., and Yokoyama, W. M. Analysisof in situ NK cell responses during viral infection. J. Immunol., 167: 5286–5293,2001.

37. Saito, T., Tanaka, R., Yoshida, S., Washiyama, K., and Kumanishi, T. Immunohis-tochemical analysis of tumor-infiltrating lymphocytes and major histocompatibilityantigens in human gliomas and metastatic brain tumors. Surg. Neurol., 29: 435–442,1988.

38. Klein, B., Loven, D., Lurie, H., Rakowsky, E., Nyska, A., Levin, I., and Klein, T. Theeffect of irradiation on expression of HLA class I antigens in human brain tumors inculture. J. Neurosurg., 80: 1074–1077, 1994.

39. Hicklin, D. J., Marincola, F. M., and Ferrone, S. HLA class I antigen downregulationin human cancers: T-cell immunotherapy revives an old story. Mol. Med. Today, 5:178–186, 1999.

40. Knopf, P. M., Harling-Berg, C. J., Cserr, H. F., Basu, D., Sirulnick, E. J., Nolan, S. C.,Park, J. T., Keir, G., Thompson, E. J., and Hickey, W. F. Antigen-dependentintrathecal antibody synthesis in the normal rat brain: tissue entry and local retentionof antigen-specific B cells. J. Immunol., 161: 692–701, 1998.

41. Perry, V. H. A revised view of the central nervous system microenvironment andmajor histocompatibility complex class II antigen presentation. J. Neuroimmunol.,90: 113–121, 1998.

42. During, M. J., Symes, C. W., Lawlor, P. A., Lin, J., Dunning, J., Fitzsimons, H. L.,Poulsen, D., Leone, P., Xu, R., Dicker, B. L., Lipski, J., and Young, D. An oralvaccine against NMDAR1 with efficacy in experimental stroke and epilepsy. Science(Wash. DC), 287: 1453–1460, 2000.

43. Thomas, C. E., Schiedner, G., Kochanek, S., Castro, M. G., and Lowenstein, P. R.Peripheral infection with adenovirus causes unexpected long-term brain inflammationin animals injected intracranially with first-generation, but not with high-capacity,adenovirus vectors: toward realistic long-term neurological gene therapy for chronicdiseases. Proc. Natl. Acad. Sci. USA, 97: 7482–7487, 2000.

44. Flugel, A., Berkowicz, T., Ritter, T., Labeur, M., Jenne, D. E., Li, Z., Ellwart, J. W.,Willem, M., Lassmann, H., and Wekerle, H. Migratory activity and functionalchanges of green fluorescent effector cells before and during experimental autoim-mune encephalomyelitis. Immunity, 14: 547–560, 2001.

45. Groh, V., Rhinehart, R., Randolph-Habecker, J., Topp, M. S., Riddell, S. R., andSpies, T. Costimulation of CD8�� T cells by NKG2D via engagement by MICinduced on virus-infected cells. Nat. Immunol., 2: 255–260, 2001.

46. Stewart, L. A. Chemotherapy in adult high-grade glioma: a systematic review andmeta-analysis of individual patient data from 12 randomised trials. Lancet, 359:1011–1018, 2002.

9006

MICA/NKG2D-MEDIATED THERAPY OF GLIOMAS

Research. on August 16, 2020. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from

2003;63:8996-9006. Cancer Res   Manuel A. Friese, Michael Platten, Stefan Z. Lutz, et al.   Experimental GliomasMICA/NKG2D-Mediated Immunogene Therapy of

  Updated version

  http://cancerres.aacrjournals.org/content/63/24/8996

Access the most recent version of this article at:

   

   

  Cited articles

  http://cancerres.aacrjournals.org/content/63/24/8996.full#ref-list-1

This article cites 46 articles, 18 of which you can access for free at:

  Citing articles

  http://cancerres.aacrjournals.org/content/63/24/8996.full#related-urls

This article has been cited by 27 HighWire-hosted articles. Access the articles at:

   

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

  SubscriptionsReprints and

  .pubs@aacr.orgDepartment at

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

  Permissions

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

.http://cancerres.aacrjournals.org/content/63/24/8996To request permission to re-use all or part of this article, use this link

Research. on August 16, 2020. © 2003 American Association for Cancercancerres.aacrjournals.org Downloaded from