Cancer Research Cancer Epigenetics Telomeres
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Maia, S.
Right arrow Articles by Cardoso, A. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Maia, S.
Right arrow Articles by Cardoso, A. A.
[Cancer Research 65, 10050-10058, November 1, 2005]
© 2005 American Association for Cancer Research

Immunology

Gene Expression Profiling Identifies BAX-{delta} as a Novel Tumor Antigen in Acute Lymphoblastic Leukemia

Sara Maia1,5, W. Nicholas Haining2, Sascha Ansén1, Zhinan Xia1, Scott A. Armstrong2,4, Nilufer P. Seth3, Paolo Ghia1, Monique L. den Boer6, Rob Pieters6, Stephen E. Sallan2, Lee M. Nadler1 and Angelo A. Cardoso1

Departments of 1 Medical Oncology, 2 Pediatric Oncology, and 3 Cancer Immunology and AIDS, Dana-Farber Cancer Institute, Harvard Medical School; 4 Division of Hematology/Oncology, Children's Hospital, Boston, Massachusetts; 5 Unit of Tumor Biology, Institute of Molecular Medicine, University of Lisbon, Portugal; and 6 Department of Pediatric Oncology/Hematology, Sophia Children's Hospital, Erasmus University Medical Center, Rotterdam, the Netherlands

Requests for reprints: Angelo A. Cardoso, Department of Medical Oncology, Dana-Farber Cancer Institute, Room D-540B, 44 Binney Street, Boston, MA 02115. Phone: 617-632-6706; Fax: 617-632-4369; E-mail: cardoso{at}mbcrr.harvard.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The identification of new tumor-associated antigens (TAA) is critical for the development of effective immunotherapeutic strategies, particularly in diseases like B-cell acute lymphoblastic leukemia (B-ALL), where few target epitopes are known. To accelerate the identification of novel TAA in B-ALL, we used a combination of expression profiling and reverse immunology. We compared gene expression profiles of primary B-ALL cells with their normal counterparts, B-cell precursors. Genes differentially expressed by B-ALL cells included many previously identified as TAA in other malignancies. Within this set of overexpressed genes, we focused on those that may be functionally important to the cancer cell. The apoptosis-related molecule, BAX, was highly correlated with the ALL class distinction. Therefore, we evaluated BAX and its isoforms as potential TAA. Peptides from the isoform BAX-{delta} bound with high affinity to HLA-A*0201 and HLA-DR1. CD8+ CTLs specific for BAX-{delta} epitopes or their heteroclitic peptides could be expanded from normal donors. BAX-{delta}–specific T cells lysed peptide-pulsed targets and BAX-{delta}–expressing leukemia cells in a MHC-restricted fashion. Moreover, primary B-ALL cells were recognized by BAX-{delta}–specific CTL, indicating that this antigen is naturally processed and presented by tumor cells. This study suggests that (a) BAX-{delta} may serve as a widely expressed TAA in B-ALL and (b) gene expression profiling can be a generalizable tool to identify immunologic targets for cancer immunotherapy.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Current therapies for acute lymphoblastic leukemia (ALL) fail to cure most adults and many children with the disease, and new treatment approaches are necessary. T-cell–based immunotherapy offers great potential as a specific and potent strategy for the treatment of human cancer. Several studies have shown that T-cell responses specific for distinct tumor antigens can be primed or expanded in the cancer-bearing host (1–4), suggesting that the immune response against tumors can be manipulated in humans for therapeutic benefit. We and others have shown that patients with ALL possess leukemia-specific cytotoxic effector cells in their T-cell repertoire (5–9). This suggests that immunotherapy directed at tumor antigens expressed by ALL cells may offer a new therapeutic strategy for the disease. However, a limiting step in the development of tumor vaccines or adoptive immunotherapy is the identification of tumor antigens expressed by ALL.

In contrast to other malignancies, few tumor-associated antigens (TAA) have been identified in ALL. These include the oncogenic fusion protein TEL/AML-1 (6), the minor histocompatibility antigens HA-1, HA-2, and HB-1 (8, 10), Wilms' tumor (WT1; ref. 11), and HER-2/neu (12). The clinical potential of these antigens is hampered by their limited expression in ALL patients or by their restricted presentation by nonprevalent HLA alleles. Therefore, the identification of novel, broadly expressed TAA in leukemia cells is essential for the effective translation of immunotherapy strategies in ALL. Ideally, these antigens should be preferentially or uniquely expressed by malignant cells, play essential roles on tumor cell survival or proliferation, or be broadly expressed on tumors, and their reactive T-cell repertoire should be spared from negative selection or functional inactivation (13).

We reasoned that gene expression profiling might allow the identification of a large number of genes that are overexpressed in ALL cells with respect their normal counterparts, B-cell precursors (BCP). This set of genes would fulfill an essential requirement of tumor antigens (i.e., their overexpression in cancer cells compared with normal tissues) and serve as a pool of target antigens from which to identify those capable of eliciting a tumor-specific T-cell response. We report here the use of gene expression profiling to identify BAX-{delta} as a novel candidate tumor antigen in ALL.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Primary cells and cell lines. Bone marrow and peripheral blood from ALL patients (diagnostic; ≥90% blasts) or healthy donors were collected according to institutional guidelines and with appropriate informed consent. ALL patients had a median age of 5.3 years (range, 0.1-17.0) and a median WBC count of 37.7 x 109 cells/L (range, 2.0-1,175.0). Mononuclear cells were isolated on Ficoll-Paque (Amersham, Piscataway, NJ) and HLA-A2*0201 individuals were screened using an anti-HLA-A2 monoclonal antibody (mAb; PharMingen, San Diego, CA). BCPs (CD19+ Ig{kappa}/Ig{lambda}–) were selected from bone marrow of pediatric healthy donors using fluorescence-activated cell sorting. T2 cells and the ALL cell lines 207, Reh, SUP-B15, NALM-6, and RS4;11 were obtained from American Type Culture Collection (Manassas, VA) or DSZM (Braunschweig, Germany).

Gene expression profiling. Total RNA was obtained using TRIZOL (Invitrogen, Carlsbad, CA) and used for subsequent production of biotinylated antisense RNA as described (14). Labeled RNA was hybridized to Affymetrix U133A (Affymetrix, Santa Clara, CA) oligonucleotide arrays (16 hours at 45°C), which were then washed and stained with streptavidin-PE (Molecular Probes, Eugene, OR). Signal amplification using biotinylated anti-streptavidin mAb (3 µg/mL; Vector Laboratories, Burlingame, CA) was followed by a second staining with streptavidin-PE. Normal goat IgG was used as a blocking agent. Scans were carried out on Affymetrix scanners, and expression values were calculated using Affymetrix GeneChip software and normalized based on a linear scaling method (15).

Reverse transcription-PCR and sequencing. Total RNA was obtained using TRIZOL and quantified by spectrophotometry, and cDNA was prepared by reverse transcription at 42°C in a mix containing Improm II (Promega, Madison, WI) and pd(N) hexamers (Amersham). PCR amplification of BAX isoforms was done from 100 ng cDNA using the primers: 5'-GAATTCGCCGCCACCATGGACGGGTCCGGGGAGCAG-3' (exon 1) and 5'-CTTGACGAATTCTCATCAGCCCATCTTCTTCCAGAT-3' (exon 6). As control, amplification of the housekeeping gene PBGD was done. Specific bands were extracted from agarose gels, purified using a QIAquick Gel Extraction kit (Qiagen, Valencia, CA), and analyzed in an automatic sequencer (Applied Biosystems, Foster City, CA).

Immunoblotting. Cells were washed in ice-cold DPBS and lysed in NP40 lysis buffer, and lysates were quantified by Bradford colorimetric assay (Bio-Rad, Hercules, CA). Equal amounts of protein (50 µg) were analyzed in a 15% SDS-PAGE gel and transferred to nitrocellulose membranes (Schleider & Schuell, Keene, NH). After blocking with 5% milk-TBST (TBS-0.1% Tween), immunoblotting was done using an antibody specific for BAX (N-20; Santa Cruz Biotechnology, Santa Cruz, CA) or actin (C11; Sigma-Aldrich, St. Louis, MO) followed by incubation with horseradish peroxidase–conjugated anti-rabbit antibody (Promega) and developed using the Renaissance Chemiluminescence kit (NEN, Boston, MA).

Peptides and HLA-A*0201 and HLA-DR binding assay. The BIMAS (16) and SYFPEITHI (17) algorithms were employed to predict high-affinity binding to HLA-A*0201, whereas the latter was used for predicting HLA-DR binding. All peptides were synthesized by New England Peptide (Gardner, MA).

Binding and stabilization of HLA-A*0201 was evaluated using the T2-binding assay. Briefly, T2 cells were pulsed (10 µg/mL peptide; 16 hours at 37°C) in serum-free medium. Cells were washed, stained with FITC-conjugated anti-HLA-A2 mAb (PharMingen), and analyzed in a Cytomics FC500 cytometer (Beckman-Coulter, Fullerton, CA) using the FlowJo software (Tree Star, Ashland, OR). Mean fluorescence intensities (MFI) were measured for unpulsed and pulsed cells, and HLA-A*0201 binding was quantified as: fluorescence index = [(MFI with peptide / MFI without peptide) – 1]. Binding of the high-affinity peptide HIV RT-pol476 was used as reference.

Binding to HLA-DR1, HLA-DR2, and HLA-DR4 alleles was determined using a competition assay (18), measuring the ability of BAX peptides to dislodge the biotinylated, high-affinity HLA-DR binder peptide influenza HA-307-319. Test peptides (30 nmol/L-30 µmol/L) were incubated overnight at 37°C with 1.7 µmol/L synthetic HLA-DR and 30 nmol/L biotinylated HA-307 peptide. Following washing, the peptide/HLA-DR mix was added to an ELISA plate precoated with anti-pan HLA-DR mAb (L243 clone; 200 ng/mL; overnight at 4°C) and incubated for 2 hours at room temperature. Plates were washed and incubated with Europium-streptavidin–labeled solution (1 hour at room temperature), and enhancement solution was added before plate reading. Fluorescence was measured in a Victor2 fluorescence reader (Perkin-Elmer, Boston, MA), and HLA-DR binding was quantified as the concentration of BAX-{delta} peptide necessary to reduce HA-307 binding to HLA-DR by 50% (IC50). All reactions were done in duplicate.

Generation of peptide-specific CTLs. Dendritic cells and activated B cells were used as antigen-presenting cells (APC) as described (4, 19). Dendritic cells were prepared from plastic-adherent peripheral blood mononuclear cells cultured for 7 days with granulocyte macrophage colony-stimulating factor (50 ng/mL; Genzyme, Cambridge, MA) and interleukin (IL)-4 (20 ng/mL; R&D, Minneapolis, MN). Activated B cells were prepared by culture of mononuclear cells on CD40L-expressing fibroblasts supplemented with IL-4 (2 ng/mL) and cyclosporin A (2.8 µg/mL; Novartis, Basel, Switzerland). For generation of peptide-specific T cells, peptide-pulsed dendritic cells (10 µg/mL; 4 hours at 37°C) were irradiated at 3.2 Gy and mixed to autologous T cells (T cells/dendritic cells ratio, 20:1) and cocultured for 7 days in RPMI plus 10% human AB serum (Valley Biomedical, Winchester, VI; RPMI-HS10) and IL-7 (10 ng/mL; R&D). For restimulation/amplification, on day 7 and weekly thereafter, T cells were rechallenged with irradiated, peptide-pulsed (1 µg/mL), activated B cells (T cells/B cells ratio, 5:1). Cytotoxicity and quantification of IFN-{gamma}-secreting cells were assessed 5 to 7 days after the third restimulation or when indicated.

Detection of tetramer-binding T cells. Recombinant tetrameric HLA-A*0201/peptide complexes with {delta}-25Y1 and {delta}-26Y1 were synthesized by a standard approach and conjugated with streptavidin-PE. CTL were incubated with tetramer for 30 minutes, washed in 1% bovine serum albumin-DPBS, and incubated with CD3-PC5 and CD8-PC7 mAbs (Beckman-Coulter) for 30 minutes. Alternatively, cells were stained with CD8-PC5 and Annexin V-FITC mAbs (PharMingen) to exclude nonviable cells. Samples were acquired in a Cytomics FC500 cytometer and analyzed using the FlowJo software.

Cytotoxic assay. Cell-mediated toxicity was assessed using a standard 51Cr release assay (5). Target cells (pulsed T2 cells or ALL cells) were incubated with 51Cr (75 µCi; 90 minutes at 37°C), washed four times, and plated at 5 x 103 per well in RPMI-HS10. T cells were added at increasing ratios and incubated for 4 hours at 37°C, supernatants harvested into Skatron filters (Skatron, Sterling, VA), and radioactivity was measured in a {gamma}-scintillation counter (Wallac Wizard, Perkin-Elmer). Specific lysis was determined as: Specific lysis (%) = [(Experimental 51Cr release – Spontaneous 51Cr release) / (Maximum 51Cr release – Spontaneous 51Cr release) x 100]. Maximum release was determined by addition of 2% Triton X-100 (Sigma-Aldrich). All experiments were done in triplicate.

Quantification of peptide-specific T cells. The ELISPOT assay was used to detect peptide-specific IFN-{gamma}-producing T cells. ELISPOT IP plates (Millipore, Billerica, MA) were coated with 1-D1K mAb (5 µg/mL; Mabtech, Nacka, Sweden; overnight at 4°C). After removal of unbound antibody, primary leukemia cells (5 x 104/well) or T2 cells (1 x 104/well; unpulsed or pulsed) were added followed by CTL (2 x 105/well). Cocultures were carried for 48 hours at 37°C in RPMI-HS10. After extensive washing, wells were incubated with biotin-labeled anti-IFN-{gamma} detection mAb (1 µg/mL; clone 7B6-1; Mabtech; 3 hours at room temperature), washed, and incubated with SA-ALP-PQ (1:1,000; Mabtech; 1.5 hours at room temperature). After washing, alkaline phosphatase substrate solution (Bio-Rad) was added. IFN-{gamma} spots were digitally imaged and quantified using an Immunospot Analyzer (Cellular Technology, Cleveland, OH). All experiments were done in duplicate. As positive control, CTL were stimulated with 10 µg/mL phytohemagglutinin (Sigma-Aldrich), whereas CTL alone or cocultured with RT-pol476-pulsed T2 cells were used as negative controls.

Statistical analysis. Statistical analysis was done using the two-tailed nonparametric Mann-Whitney test, and differences were considered statistically significant when P ≤ 0.05. This analysis was done using GraphPad Prism version4.0 (GraphPad Software, San Diego, CA).

Expression array analysis. Genes were correlated with the particular class descriptions as described (14, 15). Signal-to-noise statistic (µ0 – µ1) / ({sigma}0 + {sigma}1) was used, where µ and {sigma} represent the mean and SD of expression, respectively, for each class. The data were filtered so that only genes with an expression level of >20 and <16,000 and whose expression varied at least 3-fold between classes were included in the analysis. We carried out 100 permutations of the samples to determine whether correlations were greater than would be expected by chance with a 99% confidence.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Gene expression profiling identifies known tumor-associated antigen in acute lymphoblastic leukemia samples. To identify potential leukemia-associated antigens, we compared the transcriptome of primary B-cell ALL (B-ALL) specimens with that of normal BCP by using gene expression arrays. In this initial comparison, we wished to limit heterogeneity among ALL samples. Therefore, we restricted the initial screening to diagnostic specimens collected from children with t(12;21) rearrangement (TEL/AML-1; n = 20), because (a) gene expression profiling data suggest that TEL/AML-1 genotype is a relative homogenous group within ALL (20) and (b) TEL/AML-1 represents the most frequent recognized translocation in childhood ALL (21). Because B-ALL cells share phenotypic properties of normal BCP (pro/pre–B cells and immature B cells), the latter were used as the normal leukemia counterpart. Therefore, BCP from the bone marrow of four healthy children were sorted and used as the control population (Fig. 1A).



View larger version (50K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Identification of genes differentially expressed in B-ALL in comparison with BCP. A, sorting strategy for normal BCP. The contour plot is gated on CD19+ cells from the bone marrow of a representative healthy donor. The cells encompassed by the purple gate correspond to pro/pre–B cells; those in the green gate are immature B cells. B, heat map of the 75 probes that are most highly correlated with ALL versus BCP. Each column represents a leukemia or BCP sample, and each row represents an individual gene. Expression levels are normalized for each gene, where the mean is 0, expression levels greater than the mean are shown in red and levels less than the mean are in blue. Increasing distance from the mean is represented by increasing color intensity. Symbols over BCP samples indicate cell type (purple, pro/pre–B cells; green, immature B cells). Gene symbols and gene accession numbers or DNA sequence names are labeled on the right. C, expression in ALL and BCP of apoptosis genes selected as highly correlated with the ALL class distinction. Relative expression levels of BIRC7/ML-IAP, BNIP3L, and BAX in ALL and BCP samples are shown. Each point represents an individual sample, and the red bar is the median value for each class. The expression values are raw data obtained from Affymetrix GeneChip analysis after the arrays are scaled as described in Materials and Methods.

 
We first determined whether among the 22,277 probes tested we could identify genes whose expression pattern correlated with the leukemia sample class.7 Genes were sorted based on their degree of correlation with ALL/BCP class distinction, and the statistical significance of the observed difference in expression was determined using permutation analysis (Fig. 1B). For the 28 samples tested, ~4,000 probes were overexpressed in ALL cells with respect to their normal counterparts, showing that primary ALL cells have a markedly different gene expression profile when compared with their normal counterparts.

Analysis of the genes differentially expressed between ALL cells and BCP revealed several genes that have been identified as overexpressed or predominantly expressed in cancer cells. These include genes involved in tumorigenesis, such as BIRC7/ML-IAP, FSCN1, PTP4A3, CD99, RNF11, FGFR1, SDC-2, etc. (see Supplementary Table S1). Because we hypothesized that the set of genes overexpressed by ALL cells would be enriched for TAA, we examined our data set for genes encoding antigens known to elicit tumor-specific immune responses. We did not find overexpression of WT1, HER-2/neu, HB-1, or HA-1 (HA-2 is not probed in this array), which were reported previously as serving as TAA in ALL. However, genes identified or validated previously as TAA in other tumor types (e.g., BIRC7/ML-IAP, MSRA, CD63, EZH2/ENX1, D2S448/MG50, and MDM2; refs. 22–24) were relatively overexpressed in ALL compared with normal BCP (Fig. 1B; Supplementary Table S2).

A potential limitation of gene expression profile as tool for discovery of new TAA was that the list of genes differentially expressed is very large. Therefore, we narrowed our search to genes involved in pathways known to be biologically important to tumor cells (13). We chose to focus on the apoptotic pathway for three reasons: (a) regulation of apoptosis is deranged in many tumor types, including ALL, and may be central to the leukemic phenotype (25–27); (b) critical mediators or regulators of apoptosis have been identified previously as TAA (22, 28, 29); and (c) BIRC7/ML-IAP, an antiapoptotic molecule described previously as a potent TAA in melanoma (22), was markedly overexpressed in our data set (Fig. 1B). We therefore surveyed the gene list for other members of the apoptotic pathway. We observed that BAX and BNIP3L were among those with a higher relative expression in ALL cells (Fig. 1C). After BIRC7/ML-IAP, BAX is the apoptosis-related gene showing highest correlation with the ALL class distinction (Fig. 1C) and therefore was selected as candidate TAA for further studies.

BAX-{delta} is expressed by primary leukemia B cells. The microarray analysis showed significant degree of overexpression of BAX in ALL samples with a TEL/AML-1 translocation. For BAX to be most clinically useful as a TAA, it should be expressed by ALL cells with other genetic abnormalities. Expression of BAX was evaluated in an unselected cohort of 55 B-ALL patients to determine whether this gene is overexpressed in ALL patients with distinct genotypes. Reverse transcription-PCR (RT-PCR) was done using primers designed to amplify the full-length BAX mRNA (BAX-{alpha}) and its isoforms BAX-{delta}, BAX-{varepsilon}, BAX-{sigma}, and BAX-{zeta}. Transcripts for BAX-{alpha} were detected in all cases tested, constituting the prevalent form. BAX-{delta}, which lacks the death domain encoding exon 3, was detected in all patients tested (n = 51; Fig. 2A). Most patients also expressed BAX-{sigma} and BAX-{zeta}, whereas BAX-{varepsilon} mRNA was never detected. BAX-{delta} was undetectable in BCP derived from healthy donors studied (Fig. 2A; data not shown). Nucleotide sequencing confirmed that the PCR product indeed corresponded to the BAX-{delta} isoform. It also showed that these BAX-{delta} transcripts did not contain mutations (i.e., none affecting the exon 2-exon 4 junction; n = 12 patients; data not shown). BAX-{delta} mRNA was not detected in normal human cells, including blood mononuclear cells, bone marrow endothelial cells, or bone marrow stroma (data not shown). Expression by Western blot analysis using an antibody recognizing the NH2 terminus of BAX showed that BAX-{delta} protein was detected in all patient specimens (n = 20 patients) as well as in all B-ALL cell lines (n = 5) tested (Fig. 2B; data not shown).



View larger version (28K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Primary B-ALL cells express BAX-{delta}. A, electrophoretic analysis of RT-PCR products from primary B-ALL and normal BCP cells. Arrows, BAX-{delta} and PBGD housekeeping cDNA amplification, respectively. B, Western blot for BAX-{delta} protein expression in primary ALL and B-ALL cell lines using an antibody recognizing the NH2-terminal portion of BAX. BAX-{delta} is identified as a 15-kDa band. The control is bone marrow endothelium derived from a normal donor.

 
BAX-{delta} peptide epitopes bind HLA-A*0201 and HLA-DR. Because BAX-{delta} is widely expressed in B-ALL and was not detected in normal BCP, we reasoned that it would allow the most specific targeting of ALL cells. We evaluated the peptide sequence corresponding to exon 2-exon 4 junction for optimal HLA-A*0201 anchor motifs and identified three BAX-{delta} native peptides that may be capable of binding—the nonamers {delta}-24 and {delta}-26 and the decamer {delta}-25. Using the PY1 heteroclitic peptide approach (30), we also selected the variants {delta}-25Y1 and {delta}-26Y1 (Fig. 3). Initial analyses using the HLA-binding algorithms BIMAS (16) and SYFPEITHI (17) predicted stable binding to HLA*A-0201 for the peptides {delta}-25 and {delta}-26 but not for {delta}-24. The BIMAS analysis predicted a markedly increased binding for the heteroclitic peptides {delta}-25Y1 and {delta}-26Y1 in comparison with their native epitopes. We used the standard T2 assay to evaluate the ability of BAX-{delta} peptides to bind and stabilize HLA*A-0201 (4, 31) using the immunodominant peptide RT-pol476 of HIV-1 as reference. As shown in Fig. 3, {delta}-25, {delta}-26, and their heteroclitic peptides stably bind to HLA*A-0201 with high affinity, comparable with that observed for the RT-pol476 epitope. In all experiments done (n = 4), the heteroclitic peptides showed an increased binding stability. The peptide {delta}-24 showed no binding to HLA-A*0201 and thereby was excluded from the functional studies.



View larger version (13K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. BAX-{delta}-derived peptides bind to human HLA-A*0201. Amino acid sequence of selected peptides encompassing the exon 2-exon 4 (in bold) junction of BAX-{delta} are shown along with their binding affinity to HLA-A*0201 as assessed using a standard T2 assay. Results are expressed as fluorescence index (see Materials and Methods). The high-affinity RT-pol476 peptide was used as reference.

 
Because optimal T-cell responses to antigens require T-cell help generally elicited by MHC II–bound immunogenic epitopes derived from the same antigen, we also screened BAX-{delta} for peptides that may potentially bind to HLA-DR alleles. Again, we focused on the sequence encompassing the BAX-{delta} exon 2-exon 4 junction. Using the SYFPEITHI algorithm for predicting peptide affinity to HLA-DR molecules (17), we selected three peptides—the 26-mer {delta}-16 and the 15-mers {delta}-20 and {delta}-22. To determine the binding of these BAX-{delta} peptides to HLA-DR1 (DRB1*0101), HLA-DR2 (DRB1*1501), and HLA-DR4 (DRB1*0401) molecules, we employed a novel binding assay, in which the peptides of interest compete for HLA-DR binding with the dominant 13-mer influenza HA-307 peptide (18). Peptides {delta}-16 and {delta}-20 strongly bound to HLA-DR1 as shown by their efficacy in displacing the high-affinity peptide HA-307 (Table 1). Peptide {delta}-16 also showed moderate binding to HLA-DR2 and HLA-DR4, whereas peptide {delta}-22 did not significantly bind to any of these alleles (Table 1).


View this table:
[in this window]
[in a new window]

  		Table 1. BAX-{delta}-derived peptides bind to human HLA-DR alleles

 
Generation of CTL responses specific for BAX-{delta} peptides. Given that the generation of epitope-specific CTL is essential for an effective antitumor T-cell–mediated immune response, we concentrated our efforts on defining the immunogenicity of the BAX-{delta} peptides binding to HLA-A*0201. First, we evaluated the ability of these peptides to prime in vitro–specific CTL responses in HLA-A*0201+ donors (5-6 donors/peptide tested). Epitope-specific T cells were generated and amplified using peptide-pulsed dendritic cells and/or CD40-activated B cells from the same donor as described (see Materials and Methods; refs. 4, 19). Epitope-reactive T cells could be expanded for >35 days from all donors (n = 6). To detect peptide-specific CTL expanded in this culture system, we produced A*0201 peptide tetramers for the high-affinity binders {delta}-25Y1 and {delta}-26Y1. Flow cytometry analysis was done to detect T cells specific for these BAX-{delta} peptides. After 3 weeks in culture, we were able to detect CD8+ T cells that bind to tetramer complexes containing {delta}-25Y1 (0.59-1.09%) and {delta}-26Y1 (0.89-3.19%; Fig. 4A), showing that T cells expressing TCR recognizing BAX-{delta} epitopes can be expanded ex vivo.



View larger version (36K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. CTL specific for BAX-{delta} peptides can be expanded and lyse target cells pulsed with cognate peptides. CTLs were generated from the peripheral blood of HLA-A*0201 healthy donors. A, BAX-{delta}-specific CD8+ T cells were identified using tetramers for {delta}-25Y1 and {delta}-26Y1 peptides as depicted for a representative case (of three individual experiments). B, CTL activity was evaluated against T2 cells pulsed with native peptides {delta}-25 ({blacksquare}) and {delta}-26 ({bullet}), T2 cells pulsed with heteroclitic peptides {delta}-25Y1 ({square}) and {delta}-26Y1 ({circ}), or control T2 cells pulsed with RT-pol476 ({triangleup}). Results are shown for CTL at day 28 of culture for a representative donor (of six donors used). At 30:1 ratio, the lysates (mean ± SE, n = 6) of targets pulsed with the cognate versus irrelevant peptides were 55.5 ± 4.8 versus 18.9 ± 2.5 for CTL-{delta}-25, 51.8 ± 5.2 versus 12.6 ± 4.0 for CTL-{delta}-25Y1, 54.4 ± 5.6 versus 16.5 ± 1.4 for CTL-{delta}-26, and 53.3 ± 7.8 versus 18.4 ± 2.7 for CTL-{delta}-26Y1. C, CTL generated against BAX-{delta} heteroclitic peptides ({square} and {circ}) also lysed T2 cells pulsed with their corresponding BAX-{delta} native peptides ({blacksquare} and {bullet}) but not control T2 cells pulsed with RT-pol476 ({triangleup}). Results are shown for CTL at day 28 of culture for a representative donor (of six donors used). At 30:1 ratio, the lysates (mean ± SE, n = 6) of targets pulsed with the heteroclitic versus native peptides were 51.8 ± 5.2 versus 55.8 ± 5.1 for CTL-{delta}-25Y1 and 53.3 ± 7.8 versus 36.1 ± 4.2 for CTL-{delta}-26Y1. D, CTL generated against peptides {delta}-25Y1 or {delta}-26Y1 secrete IFN-{gamma} in response to T2 cells pulsed with their cognate epitopes (black columns) but not to T2 cells pulsed with irrelevant peptide (white columns). All experiments were done in triplicates, and the mean number of spots per peptide was determined for each condition. Results are shown for CTL at day 28 of culture for a representative donor (of four donors used). IFN-{gamma} production (mean ± SE, n = 4) in response to targets pulsed with cognate versus irrelevant peptide was 41.3 ± 10.1 versus 16.9 ± 2.6 for CTL-{delta}-25Y1 and 43.7 ± 2.9 versus 14.0 ± 3.1 for CTL-{delta}-26Y1, respectively.

 
To evaluate the cytotoxic activity of T cells reactive to BAX-{delta} peptide-pulsed APC, 51Cr release assays were done using as targets T2 cells pulsed with the respective peptides. As shown in Fig. 4B for a representative donor, expanded T cells lysed T2 cells pulsed with their corresponding cognate peptides but not with the RT-pol476 peptide. Because T2 cells pulsed with these different epitopes express equivalent levels of HLA-A*0201 molecules, this further highlights the specificity of these CTLs for the BAX-{delta} peptides. No marked differences were observed in the cytolytic activity of CTL specific for the native or the heteroclitic peptides (Fig. 4B). Importantly, CTL generated using the heteroclitic peptides {delta}-25Y1 and {delta}-26Y1 effectively lysed target cells pulsed with the corresponding native peptides (Fig. 4C), validating the use of the heteroclitic peptides for the generation and expansion of BAX-{delta}-specific T cells.

To confirm these findings, ELISPOT assays were done using peptide-pulsed T2 cells as stimulators and CTL as responders (days 28-35 of culture system). T-cell lines raised against native and heteroclitic peptides produced IFN-{gamma} when stimulated with targets pulsed with native and heteroclitic peptides (Fig. 4D), showing their immunogenicity. These studies indicate that T cells specific for BAX-{delta} peptides are present in the repertoire of normal donors.

BAX-{delta} peptides {delta}-25 and {delta}-26 are naturally processed by human leukemia cells. To determine whether BAX-{delta} peptides are naturally processed and presented by B-ALL cells, we evaluated the ability of epitope-specific CTL to lyse two BAX-{delta}+ leukemia cell lines: NALM-6 (HLA-A*0201+) and RS4;11 (HLA-A*0201–). As shown in Fig. 5A, CTL specific for {delta}-25Y1 and {delta}-26Y1 lysed NALM-6 cells (40:1 ratio; n = 4; 29.9 ± 2.6 and 24.9 ± 4.6, respectively) but not RS4;11 cells (40:1 ratio; n = 4; 6.6 ± 1.7 and 3.2 ± 0.9, respectively), indicating that both BAX-{delta} epitopes are naturally processed by these leukemia cells. This experiment also showed that this cytolytic response was allele restricted, because no significant lysis of HLA-A*0201– leukemia cells was observed (Fig. 5A) despite their expression of BAX-{delta}. Importantly, CTL specific for {delta}-25Y1 or {delta}-26Y1 did not lyse activated autologous B cells (data not shown). The ability of leukemia cells to be recognized by BAX-{delta}-reactive T cells was confirmed using an ELISPOT assay (Fig. 5B), with NALM-6 cells triggering IFN-{gamma} secretion by resting CTL specific for {delta}-25Y1 and {delta}-26Y1 (Fig. 5B; n = 4; 47.8 ± 5.3 and 48.6 ± 3.3 spots for NALM-6 and 5.0 ± 1.7 and 4.3 ± 1.2 spots for RS4;11, respectively). B cells from HLA-A*0201+ healthy donors, which do not express BAX-{delta}, did not stimulate IFN-{gamma} secretion by these CTL (data not shown).



View larger version (27K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. BAX-{delta} epitopes are naturally processed and presented by leukemia cells in a HLA-A*0201-restricted manner. A, CTL specific for {delta}-25Y1 or {delta}-26Y1 lysed HLA-A*0201+ NALM-6 cells (x) but not HLA-A*0201– RS4;11 cells ({diamond}). Results are shown for CTL at day 28 of culture for a representative donor (of four donors used). B, BAX-{delta}-specific CD8+ T cells produce IFN-{gamma} in response to HLA-A*0201+; BAX-{delta}+ NALM-6 cells (black columns) but not to RS4;11 cells (BAX-{delta}+; HLA-A*0201–; white columns). Each experiment was done in triplicate, and the mean number of spots per peptide calculated for each condition. Results are shown for CTL at day 28 of culture for a representative donor (of four donors used). C, CTL-mediated lysis of peptide-pulsed target cells ({blacksquare}, T2/{delta}-25; {square}, T2/{delta}-25Y1; {bullet}, T2/{delta}-26; {circ}, T2/{delta}-26Y1) was assessed in the absence or presence of an excess of BAX-{delta}+, HLA-A*0201+ cells (NALM-6; x) using a competition assay. Results are shown for a representative experiment (of two donors used). At 30:1 ratio, the lysates (mean ± SE, n = 2) in the presence versus the absence of cold target were 2.95 ± 0.96 versus 57.1 ± 5.2 for CTL-{delta}-25, 4.7 ± 1.7 versus 56.4 ± 5.3 CTL-{delta}-25Y1, 18.2 ± 1.6 versus 60.5 ± 5.4 for CTL-{delta}-26, and 8.9 ± 2.9 versus 50.7 ± 3.1 for CTL-{delta}-26Y1.

 
To further show natural processing of BAX-{delta} epitopes by leukemia cells, we did a competition assay using leukemia cells (NALM-6 cells; HLA-A*0201+ BAX-{delta}+) as cold targets. Cytotoxic assays were done using T2 cells pulsed with BAX-{delta} peptides as radioactive targets in the presence of an excess (50:1 ratio) of leukemia cells. These leukemia cells abrogated or significantly inhibited the cytotoxic activity of BAX-{delta}-specific CTL against the T2 cells pulsed with their corresponding cognate peptides (Fig. 5C). In contrast, cold target RS4;11 cells (HLA-A*0201– BAX-{delta}+) did not affect the lysis of T2 cells pulsed with {delta}-25Y1 or {delta}-26Y1 by their respective CTL (data not shown). Taken together, these studies show that the BAX-{delta} peptides {delta}-25 and {delta}-26 are naturally processed by leukemia cells and are recognized by CTL and that the T-cell responses to BAX-{delta} peptides are HLA restricted.

Primary leukemia cells are recognized by BAX-{delta}–specific T cells. To determine the immunogenicity of BAX-{delta} peptides in B-ALL, we assessed whether primary leukemia cells were capable of stimulating BAX-{delta}-specific T cells using IFN-{gamma} production as a readout. ELISPOT assays were done using primary ALL cells from 14 HLA-A*0201+ patients as targets and BAX-{delta}-specific T-cell lines from normal donors (n = 5). As shown in Fig. 6, primary leukemia cells stimulated IFN-{gamma} secretion from CTL reactive to both {delta}-25Y1 and {delta}-26Y1 epitopes of BAX-{delta}. Statistical analysis showed that this leukemia-induced secretion of IFN-{gamma} was significantly higher than that stimulated by T2 cells pulsed with control peptide RT-pol476 (P = 0.02 for CTL-{delta}-25Y1 and P = 0.01 for CTL-{delta}-26Y1; Fig. 6). It is unclear why the responses induced by primary cells are lower than those observed with NALM-6 cell line (Fig. 5B), although this may be due to increase apoptosis of primary tumor cells as reported in other studies (8). These observations show that primary leukemia cells naturally process and display BAX-{delta} immunogenic epitopes, generating functional responses by epitope-specific T cells.



View larger version (12K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6. CTL specific for BAX-{delta} peptides recognize primary ALL cells. Leukemia cells from the peripheral blood and bone marrow of HLA-A*0201+ ALL patients (n = 14) were evaluated for their ability to stimulate IFN-{gamma} secretion in CTL generated against {delta}-25Y1 and {delta}-26Y1 peptides (n = 5 donors). ELISPOT assay was done as described in Materials and Methods. RT-pol476-pulsed T2 cells or CTL alone were used as controls. Each circle indicates the mean number of IFN-{gamma}-spots for each individual patient tested.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In this study, we show that gene expression profiling can be used as a tool for the identification of candidate tumor antigens and have used it, coupled with reverse immunology, to identify BAX-{delta} as a widely expressed leukemia-associated tumor antigen in pediatric B-ALL. Peptide sequences from the exon 2-exon 4 fusion region of BAX-{delta} were identified that (a) bind with high affinity to MHC I and II molecules, (b) are naturally displayed by HLA-A*0201 alleles, and (c) elicit specific CTL responses. Importantly, BAX-{delta}-expressing primary leukemia cells trigger responses from peptide-specific T cells.

Epitope deduction and functional validation of candidate antigens has proven to be a successful strategy to identify novel TAA (32). However, a limitation of this approach is that the selection of the candidate antigens is not based on a comprehensive analysis of the molecular or biological signature of the tumors but rather from limited expression studies or from a priori hypotheses. Strategies to increase the number of biologically relevant genes overexpressed in tumors from which to select candidate antigens are therefore needed. Gene expression arrays have been widely used in hematologic malignancies as an approach to biological discovery by defining the genes expressed at high levels in cancer cells (33). Here, we show that this approach can also be used as a screening tool to identify tumor antigens based on their overexpression. The use of this strategy to identify new TAA differentially expressed in leukemia stem/progenitors cells will be of great interest.

The list of genes expressed at higher levels in ALL cells compared with their normal counterparts includes tumor antigens, such as BIRC7/ML-IAP, which have already been shown to elicit tumor-specific immunity in other diseases (22). This list also features at least one novel antigen (BAX-{delta}) capable of eliciting tumor-specific T cells in vitro. To our knowledge, this is the first study in which the identification of a novel tumor antigen results from the selection of the candidate antigens through the genome-wide expression profiling of tumor samples in comparison with their normal tissue counterparts. Future studies will determine the presence of additional immunogenic epitopes encoded by genes differentially expressed by B-ALL cells and will establish their value as potential leukemia TAA.

BAX-{delta} is an isoform resulting from RNA alternative splicing of BAX, a proapoptotic molecule of the BCL-2 family (34). In BAX-{delta}, the in-frame fusion of exon 2 and exon 4 results in a molecule lacking the BH3 death domain predicted to be unable to dimerize with BCL-2 family members (35). It is not known whether BAX-{delta} affects the balance between proapoptotic and antiapoptotic BCL-2 family molecules, and its functional relevance in ALL remains to be defined. It has been suggested that BAX-{delta} or other inactive, death domain–lacking isoforms may affect cell survival by disrupting the delicate balance between antiapoptotic and proapoptotic effectors as shown for p53 dysfunction or BCL-2 overexpression (36). The mechanisms responsible for the expression of BAX-{delta} in leukemia cells are unknown, but its lack of expression in normal BCP suggests an association with the transformation process.

Although we identified BAX by comparing the gene expression profile of TEL/AML-1 with normal BCP, we subsequently validated BAX-{delta} differential expression in ALL patients with a range of genetic abnormalities. This broad expression is an attractive feature of BAX-{delta} as a putative TAA and contrasts to some previously identified ALL TAA, which have a more limited expression pattern. For instance, TEL/AML-1, the product of the t(12;21) translocation, is expressed in ~25% of the B-cell pediatric ALL patients and is not present in adult leukemia (21). The minor histocompatibility antigen HB-1 is found in most B-ALL (10), but tumor-specific CTLs recognize only the HB-1 H allele that is present in 23% of Caucasians (10). Another potential TAA, WT1, was reported to be overexpressed in most ALL (>60%; ref. 37), but its putative use as TAA in this cancer has not been established (11). HER-2/neu, a TAA well studied and validated in solid tumors, is only expressed in 15% of pediatric ALL patients (12). Target antigens well characterized in other cancers (such as hTERT, MUC1, and PRAME) have been detected in ALL (38–40), but it is not known whether leukemia cells can present epitopes derived from these molecules or be targeted by specific CTL for these antigens. The advantage of gene expression as an approach to identify candidate antigens is that the signal-to-noise algorithms used to create the class distinction between ALL and BCP prioritize those genes that are most uniformly overexpressed in the ALL class (14). It is not surprising, therefore, that PCR experiments confirmed that BAX and its isoforms were consistently overexpressed in pediatric ALL samples compared with BCP, a critical feature of optimal tumor antigens. Screening of a large group of distinct tumor cell lines (n = 25) for BAX-{delta} expression showed that all expressed this isoform (Supplementary Table S3). We are currently investigating whether BAX-{delta} is overexpressed by other tumor types and could therefore function as a TAA in cancers other than ALL.

How should novel tumor antigens like BAX-{delta} be translated into more efficient immunotherapies in ALL? Although antileukemia CTL can be elicited in vitro by modified leukemia cells or dendritic cell–based strategies (5, 9), the rapid progression of the disease and its associated lymphopenia may preclude the clinical use of autologous tumor cell vaccination in relapsed ALL (41). An alternative strategy could be the ex vivo amplification of antileukemia T cells and their adoptive transfer into patients. We and others have shown that the T-cell repertoire of most ALL patients contains T cells reactive to their tumor, suggesting that ex vivo T-cell expansion may be feasible (5–7, 9, 42). Moreover, techniques to efficiently expand therapeutically meaningful numbers of antigen-specific T cells are being developed. However, a major limitation in the development of such strategies is the identification of new antigens that are overexpressed by ALL cells. Adoptive transfer of ex vivo expanded specific T cells has proven to be an efficacious approach in EBV-associated malignancies, cytomegalovirus, and melanoma (43–45). In EBV lymphoproliferative disorders, adoptive immunotherapy induced complete remission in patients with overt disease and prevented disease progression (46). Importantly, it has been shown that adoptively transferred tumor-specific T cells expanded in vivo, contributed to the cancer host's T-cell memory pool, migrated to tumor sites, and effectively targeted malignant cells (43, 47). Future studies will be required to determine whether sufficient numbers of T cells specific for BAX-{delta} can be generated ex vivo.

A potential benefit of the identification of new target tumor antigens is that it increases the pool of immunogenic peptides that can be employed to recruit a larger, more potent repertoire of T cells targeting the tumor. Increasing numbers of studies are assessing polyepitope vaccination strategies as means to augment the efficacy and diversity of antitumor immunity. An important argument for this approach is that responses to multiple epitopes should reduce or prevent immune escape (i.e., the selection of nonimmunogenic tumor variants that would evade T-cell immunity; ref. 48). Several studies support the concept that vaccination to multiple immunogenic epitopes results in more robust immune responses (49–51). Our studies have shown that several hundred genes are overexpressed in ALL with respect to BCP. Although not all of these will prove to be valid tumor antigens, it represents a large number of theoretical targets with which to increase the pool of candidate antigens. Moreover, it may be possible to tailor the immunotherapy approach to the patients' HLA and antigenic profile, aiming at the recruitment or expansion of T cells of different specificities to maximize immunotherapeutic efficacy.

In conclusion, we used genome-wide expression profiling and reverse immunology to identify BAX-{delta} as a novel, widely expressed, TAA in B-ALL. We found that primary leukemia cells naturally process and present BAX-{delta} immunogenic epitopes and stimulate responses from peptide-specific T cells, supporting BAX-{delta} as an immunologic target for T-cell immunity in ALL. The availability of comprehensive databases of gene expression profiles of human cancers and their normal tissue counterparts should allow the systematic use of this approach. Thus, gene expression profiling may be a generalizable tool for target identification in tumor immunotherapy.


   Acknowledgments
 
Grant support: NIH grants P01-CA68484 (L.M. Nadler, S.E. Sallan, S.A. Armstrong, and A.A. Cardoso) and K08HL72750 (W.N. Haining), Fundação para Ciência e Tecnologia-Portugal grant SAU-13240 (A.A. Cardoso), Career Development Award and American Society of Clinical Oncology (W.N. Haining), Fundação para Ciência e Tecnologia-Portugal scholarship (S. Maia), and Dr. Mildred Scheel Stiftung (Deutsche Krebshilfe), Germany scholarship (S. Ansén).

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

We thank Drs. Renata Stripecke and Ana Limon for the critical reading of the article, Dr. Kay Wucherpfennig for providing the MHC II competition binding assay, and Virginia M. Dalton for gathering the information concerning the ALL samples.


   Footnotes
 
Note: P. Ghia is currently at the Department of Oncology, Universitá Vita-Salute San Raffaele, Milan, Italy.

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

7 http://research.dfci.harvard.edu/haining. Back

Received 5/ 8/05. Revised 8/ 8/05. Accepted 8/16/05.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Yee C, Gilbert MJ, Riddell SR, et al. Isolation of tyrosinase-specific CD8+ and CD4+ T cell clones from the peripheral blood of melanoma patients following in vitro stimulation with recombinant vaccinia virus. J Immunol 1996;157:4079–86.[Abstract]
  2. Wang RF, Johnston SL, Zeng G, Topalian SL, Schwartzentruber DJ, Rosenberg SA. A breast and melanoma-shared tumor antigen: T cell responses to antigenic peptides translated from different open reading frames. J Immunol 1998;161:3598–606.
  3. Bonnet D, Warren EH, Greenberg PD, Dick JE, Riddell SR. CD8(+) minor histocompatibility antigen-specific cytotoxic T lymphocyte clones eliminate human acute myeloid leukemia stem cells. Proc Natl Acad Sci U S A 1999;96:8639–44.[Abstract/Free Full Text]
  4. Scardino A, Alves P, Gross DA, et al. Identification of HER-2/neu immunogenic epitopes presented by renal cell carcinoma and other human epithelial tumors. Eur J Immunol 2001;31:3261–70.[CrossRef][Medline]
  5. Cardoso AA, Seamon MJ, Afonso HM, et al. Ex-vivo generation of anti-pre-B leukemia specific autologous cytolytic T cells. Blood 1997;90:549–61.[Abstract/Free Full Text]
  6. Yotnda P, Garcia F, Peuchmaur M, et al. Cytotoxic T cell response against the chimeric ETV6-AML1 protein in childhood acute lymphoblastic leukemia. J Clin Invest 1998;102:455–62.[Medline]
  7. Cardoso AA, Veiga JP, Ghia P, et al. Adoptive T-cell therapy for B-cell acute lymphoblastic leukemia: preclinical studies. Blood 1999;94:3531–40.[Abstract/Free Full Text]
  8. Mutis T, Verdijk R, Schrama E, et al. Feasibility of immunotherapy of relapsed leukemia with ex vivo-generated cytotoxic T lymphocytes specific for hematopoietic system-restricted minor histocompatibility antigens. Blood 1999;93:2336–41.[Abstract/Free Full Text]
  9. Montagna D, Maccario R, Locatelli F, et al. Ex vivo priming for long-term maintenance of antileukemia human cytotoxic T cells suggests a general procedure for adoptive immunotherapy. Blood 2001;98:3359–66.[Abstract/Free Full Text]
  10. Dolstra H, Fredrix H, Maas F, et al. A human minor histocompatibility antigen specific for B cell acute lymphoblastic leukemia. J Exp Med 1999;189:301–8.[Abstract/Free Full Text]
  11. Ohminami H, Yasukawa M, Fujita S. HLA class I-restricted lysis of leukemia cells by a CD8(+) cytotoxic T-lymphocyte clone specific for WT1 peptide. Blood 2000;95:286–93.[Abstract/Free Full Text]
  12. Muller MR, Grunebach F, Kayser K, et al. Expression of HER-2/neu on acute lymphoblastic leukemias: implications for the development of immunotherapeutic approaches. Clin Cancer Res 2003;9:3448–53.[Abstract/Free Full Text]
  13. Ho WY, Blattman JN, Dossett ML, Yee C, Greenberg PD. Adoptive immunotherapy: engineering T cell responses as biologic weapons for tumor mass destruction. Cancer Cell 2003;3:431–7.[CrossRef][Medline]
  14. Golub TR, Slonim DK, Tamayo P, et al. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 1999;286:531–7.[Abstract/Free Full Text]
  15. Armstrong SA, Staunton JE, Silverman LB, et al. MLL translocations specify a distinct gene expression profile that distinguishes a unique leukemia. Nat Genet 2002;30:41–7.[CrossRef][Medline]
  16. Parker KC, Bednarek MA, Coligan JE. Scheme for ranking potential HLA-A2 binding peptides based on independent binding of individual peptide side-chains. J Immunol 1994;152:163–75.[Abstract]
  17. Rammensee H, Bachmann J, Emmerich NP, Bachor OA, Stevanovic S. SYFPEITHI: database for MHC ligands and peptide motifs. Immunogenetics 1999;50:213–9.[CrossRef][Medline]
  18. Day CL, Seth NP, Lucas M, et al. Ex vivo analysis of human memory CD4 T cells specific for hepatitis C virus using MHC class II tetramers. J Clin Invest 2003;112:831–42.[CrossRef][Medline]
  19. Schultze JL, Michalak S, Seamon MJ, et al. CD40-activated human B cells: an alternative source of highly efficient antigen presenting cells to generate autologous antigen-specific T cells for adoptive immunotherapy. J Clin Invest 1997;100:2757–65.[Medline]
  20. Yeoh EJ, Ross ME, Shurtleff SA, et al. Classification, subtype discovery, and prediction of outcome in pediatric acute lymphoblastic leukemia by gene expression profiling. Cancer Cell 2002;1:133–43.[CrossRef][Medline]
  21. Shurtleff SA, Buijs A, Behm FG, et al. TEL/AML1 fusion resulting from a cryptic t(12;21) is the most common genetic lesion in pediatric ALL and defines a subgroup of patients with an excellent prognosis. Leukemia 1995;9:1985–9.[Medline]
  22. Schmollinger JC, Vonderheide RH, Hoar KM, et al. Melanoma inhibitor of apoptosis protein (ML-IAP) is a target for immune-mediated tumor destruction. Proc Natl Acad Sci U S A 2003;100:3398–403.[Abstract/Free Full Text]
  23. Bilsborough J, Van Pel A, Uyttenhove C, Boon T, Van den Eynde BJ. Identification of a second major tumor-specific antigen recognized by CTLs on mouse mastocytoma P815. J Immunol 1999;162:3534–40.[Abstract/Free Full Text]
  24. Li J, Li W, Liang S, et al. Recombinant CD63/ME491/neuroglandular/NKI/C-3 antigen inhibits growth of established tumors in transgenic mice. J Immunol 2003;171:2922–9.[Abstract/Free Full Text]
  25. Sanchez-Garcia I, Grutz G. Tumorigenic activity of the BCR-ABL oncogenes is mediated by BCL2. Proc Natl Acad Sci U S A 1995;92:5287–91.[Abstract/Free Full Text]
  26. Smith KS, Rhee JW, Cleary ML. Transformation of bone marrow B-cell progenitors by E2a-Hlf requires coexpression of Bcl-2. Mol Cell Biol 2002;22:7678–87.[Abstract/Free Full Text]
  27. Swanson PJ, Kuslak SL, Fang W, et al. Fatal acute lymphoblastic leukemia in mice transgenic for B cell-restricted bcl-xL and c-myc. J Immunol 2004;172:6684–91.[Abstract/Free Full Text]
  28. Schmitz M, Diestelkoetter P, Weigle B, et al. Generation of survivin-specific CD8+ T effector cells by dendritic cells pulsed with protein or selected peptides. Cancer Res 2000;60:4845–9.[Abstract/Free Full Text]
  29. Andersen MH, Svane IM, Kvistborg P, et al. Immunogenicity of Bcl-2 in patients with cancer. Blood 2005;105:728–34.[Abstract/Free Full Text]
  30. Tourdot S, Scardino A, Saloustrou E, et al. A general strategy to enhance immunogenicity of low-affinity HLA-A2. 1-associated peptides: implication in the identification of cryptic tumor epitopes. Eur J Immunol 2000;30:3411–21.[CrossRef][Medline]
  31. Zweerink HJ, Gammon MC, Utz U, et al. Presentation of endogenous peptides to MHC class I-restricted cytotoxic T lymphocytes in transport deletion mutant T2 cells. J Immunol 1993;150:1763–71.[Abstract]
  32. Schultze JL, Vonderheide RH. From cancer genomics to cancer immunotherapy: toward second-generation tumor antigens. Trends Immunol 2001;22:516–23.[CrossRef][Medline]
  33. Ebert BL, Golub TR. Genomic approaches to hematologic malignancies. Blood 2004;104:923–32.[Abstract/Free Full Text]
  34. Apte SS, Mattei MG, Olsen BR. Mapping of the human BAX gene to chromosome 19q13.3-q13.4 and isolation of a novel alternatively spliced transcript, BAX {delta}. Genomics 1995;26:592–4.[CrossRef][Medline]
  35. Zha H, Aime-Sempe C, Sato T, Reed JC. Proapoptotic protein Bax heterodimerizes with Bcl-2 and homodimerizes with Bax via a novel domain (BH3) distinct from BH1 and BH2. J Biol Chem 1996;271:7440–4.[Abstract/Free Full Text]
  36. McKenna SL, Cotter TG. Functional aspects of apoptosis in hematopoiesis and consequences of failure. Adv Cancer Res 1997;71:121–64.[Medline]
  37. Rosenfeld C, Cheever MA, Gaiger A. WT1 in acute leukemia, chronic myelogenous leukemia and myelodysplastic syndrome: therapeutic potential of WT1 targeted therapies. Leukemia 2003;17:1301–12.[CrossRef][Medline]
  38. Kim NW, Piatyszek MA, Prowse KR, et al. Specific association of human telomerase activity with immortal cells and cancer. Science 1994;266:2011–5.[Abstract/Free Full Text]
  39. Brossart P, Schneider A, Dill P, et al. The epithelial tumor antigen MUC1 is expressed in hematological malignancies and is recognized by MUC1-specific cytotoxic T-lymphocytes. Cancer Res 2001;61:6846–50.[Abstract/Free Full Text]
  40. Steinbach D, Viehmann S, Zintl F, Gruhn B. PRAME gene expression in childhood acute lymphoblastic leukemia. Cancer Genet Cytogenet 2002;138:89–91.[CrossRef][Medline]
  41. Haining WN, Cardoso AA, Keczkemethy HL, et al. Failure to define window of time for autologous tumor vaccination in patients with newly diagnosed or relapsed acute lymphoblastic leukemia. Exp Hematol 2005;3:286–94.
  42. Todisco E, Gaipa G, Biagi E, et al. CD40 ligand-stimulated B cell precursor leukemic cells elicit interferon-{gamma} production by autologous bone marrow T cells in childhood acute lymphoblastic leukemia. Leukemia 2002;16:2046–54.[CrossRef][Medline]
  43. Bollard CM, Aguilar L, Straathof KC, et al. Cytotoxic T lymphocyte therapy for Epstein-Barr virus+ Hodgkin's disease. J Exp Med 2004;200:1623–33.[Abstract/Free Full Text]
  44. Walter EA, Greenberg PD, Gilbert MJ, et al. Reconstitution of cellular immunity against cytomegalovirus in recipients of allogeneic bone marrow by transfer of T-cell clones from the donor. N Engl J Med 1995;333:1038–44.[Abstract/Free Full Text]
  45. Robbins PF, Dudley ME, Wunderlich J, et al. Cutting edge: persistence of transferred lymphocyte clonotypes correlates with cancer regression in patients receiving cell transfer therapy. J Immunol 2004;173:7125–30.[Abstract/Free Full Text]
  46. Rooney CM, Smith CA, Ng CY, et al. Use of gene-modified virus-specific T lymphocytes to control Epstein-Barr-virus-related lymphoproliferation. Lancet 1995;345:9–13.[CrossRef][Medline]
  47. Dudley ME, Wunderlich JR, Robbins PF, et al. Cancer regression and autoimmunity in patients after clonal repopulation with antitumor lymphocytes. Science 2002;298:850–4.[Abstract/Free Full Text]
  48. Smyth MJ, Godfrey DI, Trapani JA. A fresh look at tumor immunosurveillance and immunotherapy. Nat Immunol 2001;2:293–9.[CrossRef][Medline]
  49. Toes RE, Hoeben RC, van der Voort EI, et al. Protective anti-tumor immunity induced by vaccination with recombinant adenoviruses encoding multiple tumor-associated cytotoxic T lymphocyte epitopes in a string-of-beads fashion. Proc Natl Acad Sci U S A 1997;94:14660–5.[Abstract/Free Full Text]
  50. Smith SG, Patel PM, Porte J, Selby PJ, Jackson AM. Human dendritic cells genetically engineered to express a melanoma polyepitope DNA vaccine induce multiple cytotoxic T-cell responses. Clin Cancer Res 2001;7:4253–61.[Abstract/Free Full Text]
  51. Dutoit V, Taub RN, Papadopoulos KP, et al. Multiepitope CD8(+) T cell response to a NY-ESO-1 peptide vaccine results in imprecise tumor targeting. J Clin Invest 2002;110:1813–22.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Maia, S.
Right arrow Articles by Cardoso, A. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Maia, S.
Right arrow Articles by Cardoso, A. A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online