Autoreactive, Cytotoxic T Lymphocytes Specific for Peptides Derived From

Vol. 10, 1047–1056, February 1, 2004 Clinical Cancer Research 1047

Autoreactive, Cytotoxic T Lymphocytes Specific for Peptides Derived from Normal B-Cell Differentiation Antigens in Healthy Individuals and Patients with B-Cell Malignancies

Matthias Grube,1 Katayoun Rezvani,1 avidity would be necessary for T-cell immunotherapeutic Adrian Wiestner,2 Hiroshi Fujiwara,1 approaches using the peptides studied. Giuseppe Sconocchia,1 Jan J. Melenhorst,1 1 3 INTRODUCTION Nancy Hensel, Gerald E. Marti, ϩ 2 2 CD8 CTLs specifically recognize peptides derived from Larry W. Kwak, Wyndham Wilson, and endogenous proteins presented by class I HLA molecules on the 1 John A. Barrett surface of malignant cells. T-cell-mediated immune reactions 1National Heart Lung Blood Institute, 2National Cancer Institute, against autologous tumor cells have been described in a variety 3 NIH, and Food and Drug Administration, Bethesda, Maryland of malignant diseases (1–4). Several recent studies have identified tumor-reactive CD4ϩ and CD8ϩ T-lymphocytes in non- Hodgkin lymphoma and chronic lymphocytic leukemia (CLL; ABSTRACT Refs. 5–8). Furthermore, immunoglobulin idiotype vaccination Purpose: To investigate potential immunotherapeutic in myeloma and non-Hodgkin lymphoma can induce immune strategies in B lymphocytic malignancies we looked for responses to malignant B cells (9–12). However normal tissue- CTLs recognizing CD19 and CD20 epitopes. specific self-antigens, which are candidates for immunotherapy Experimental Design: Three CD19 and CD20 peptides in solid tumors (13), have not been widely explored in B-cell binding to HLA-A*0201 were identified and used to detect malignancies (14, 15). T cells reactive to self-antigens are peptide specific CTLs by a quantitative real-time PCR to mainly eliminated by negative thymic selection (16–19). How- ␥ measure IFN- mRNA expression in 23 healthy individuals ever, the process is incomplete, leaving behind low-avidity T and 28 patients (18 chronic lymphocytic leukemia (CLL), 7 cells autoreactive to self-antigens (20, 21). The therapeutic follicular lymphoma, 2 acute lymphocytic leukemia, and 1 potential of monoclonal antibodies targeting the B-cell differ- large cell lymphoma). Peptide-specific CTLs were expanded entiation epitopes CD19 and CD20 makes these molecules of in culture with CD40-activated B cells to test lytic activity in particular interest to study as potential T-cell epitopes (22–26). three patients. T cells, genetically modified to recognize CD19 or CD20 mol- ؉ Results: In healthy individuals, CD8 T-cell responses ecules through antibody ligands can efficiently eliminate their were detected in one to CD1974–82, in three to CD20127–135, respective B-cell target (27–30). Peptides derived from CD20 and three to CD20188–196. Seven of 27 patients (6 with CLL) have been used to generate CTLs to treat B-cell malignancies in ؉ had CD8 T cells recognizing CD1974–82. Seven patients mice (31). CD19 and CD20 play an important role in the responded to CD20127–135 and three to CD20188–196. All development, differentiation, and activation of B cells. CD19 is were CLL patients. CD1974–82-specific CTLs from three a signal-transduction molecule and CD20 is part of a multimeric patients were expanded over 4 weeks. These cells were HLA- receptor complex regulating cell cycle progression and B-cell A*0201 specific and lytic for peptide-loaded antigen- differentiation. Both are expressed by B-progenitor cells, persist presenting cells but not to malignant or unpulsed B cells. during all stages of B-cell maturation, and are lost on terminal Conclusions: CTLs that recognize CD19 and CD20 differentiation to plasma cells (32–34). CD19 and CD20 are epitopes exist in healthy individuals and may be increased largely restricted to normal and neoplastic B cells (although in CLL patients. They are of low avidity and require high CD19 can be also found on dendritic cells). The majority of doses of peptide for activation. Strategies to increase T-cell B-lineage malignancies express CD19 and CD20 (35, 36). These attributes make these molecules good candidates for T-cell immunotherapeutic research. In this study we set out to determine whether cytotoxic CD8ϩ T-cell responses specific for peptides derived from CD19 Received 8/21/03; revised 10/20/03; accepted 10/22/03. and CD20 antigens occur in healthy individuals and patients Grant support: The Dr. Mildred-Scheel-Stiftung Deutsche Krebshilfe Foundation (to M. G.). with B-cell malignancies. We then used a sensitive quantitative The costs of publication of this article were defrayed in part by the real-time (RT) PCR (qPCR) technique to search for the antici- payment of page charges. This article must therefore be hereby marked pated low frequencies of circulating T cells recognizing peptides advertisement in accordance with 18 U.S.C. Section 1734 solely to derived from these molecules. Our results show that low-avidity indicate this fact. Requests for reprints: John Barrett, NIH, 9000 Rockville Pike, Build- T cells recognizing both CD19 and CD20 circulate in normal ing 10, Room 7C 103, Hematology Branch, Bethesda, Maryland 20892. individuals and in patients with B-cell malignancies with a Phone: (301) 402-3296; Fax: (301) 435-8655; E-mail: [email protected]. notable increase in CD19-specific T cells in patients with CLL.

MATERIALS AND METHODS zation assay using the T2 cell line as described previously (41, Patients and Donors. After informed consent, cells from 42). T2 cells were pulsed with 100 ␮M peptide and 5 ␮g/ml ␤2 HLA-A2-positive patients with B-cell malignancies [18 CLL, 7 microglobulin (Sigma, St. Louis, MO) for 18 h at 26°C. CMV follicular lymphoma, 2 acute lymphocytic leukemia, 1 large cell (pp65) and gp100 peptides were used as positive controls and lymphoma (non-Hodgkin lymphoma)], which were enrolled in the background expression of HLA-A2 was determined using clinical trials approved by the institutional review board and DMSO as a negative control. The level of stabilized HLA-A2 on healthy individuals (n ϭ 23) were obtained from leucopheresis the surface of T2 was determined by using FITC-conjugated products. Peripheral blood mononuclear cells were separated mouse antihuman monoclonal antibody (BB7.2; BD Bio- using Ficoll-Hypaque gradient-density (Organon Teknika Co., sciences PharMingen, San Diego, CA). The fluorescence index Durham, NC) and were subsequently frozen in RPMI 1640 (FI) determined by fluorescence-activated cell sorting analysis ϭ complete medium (CM; 25 mM HEPES buffer, 2 mML-gluta- for each peptide was calculated by the following formula: FI mine, 100 units/ml penicillin, and 100 ␮g/ml streptomycin; Life (mean channel of fluorescence T2 ϩ peptide)/(mean channel of Technologies, Inc., Gaithersburg, MD) containing 20% heat- fluorescence T2 without peptide). Peptides were considered to inactivated FCS and 10% DMSO according to the standard stabilize HLA-A2 molecules with high affinity, if the FI was protocols. Before use, cells were thawed, washed, and resus- Ն1.5 and low affinity if the FI was between 1.1 and 1.49 at a pended in CM supplemented with 10% human AB serum (HS) peptide concentration of 100 ␮M. All of the assays were re- and rested overnight. peated a minimum of three times, and the results are given as HLA-Typing. High-resolution HLA-class I genotyping means of replicate experiments. ؉ ϩ was performed by sequence-specific PCR using genomic DNA CD8 T-Cell Selection. CD8 T cells were isolated (HLA-Laboratory, Department of Transfusion Medicine, War- from the peripheral blood mononuclear cells of patients and ren G. Magnusson Center, NIH, Bethesda, MD). healthy individuals by using a CD8-positive isolation kit (Dynal, Cell Lines. T2 cells (American Type Culture Collection, Oslo, Norway). After detachment of the immunomagnetic beads ϩ Manassas, VA) are a HLA-A*0201-positive hybrid human cell by using DetachaBead (Dynal) a purity of at last 98% CD8 T line. They expresses very low levels of cell surface HLA-A 2.1 cells was determined by flow cytometry. ؉ and are unable to present endogenous antigens because of the CD8 T Cell in Vitro Stimulation. To determine ϩ lack of most of the MHC class II region including the known peptide-specific CD8 T-cell reactivity, we measured the IFN-␥ ϩ TAP (transporter proteins for antigenic peptide) and proteasome mRNA gene expression by CD8 T cells stimulated with can- genes (37). C1R-A2 cells are a MHC-class I-defective LCL cell didate peptides. Multiple experiments to optimize assay condi- ϩ line, that expresses a transfected genomic clone of HLA-A2.1 tions were performed with CD8 T cells obtained from HLA- (38, 39) and were kindly provided by the laboratory of Dr. A*0201-positive, CMV-seropositive donors, stimulated with Jeffrey Schlom (National Cancer Institute, NIH, Bethesda, MD). HLA-A*0201-associated CMV peptide (pp65)-pulsed C1R-A2 6 ϩ The cells were maintained in CM supplemented with 10% FCS. cells. After purification, 1 ϫ 10 CD8 T cells were plated in a Peptide Predictions. The protein sequence from CD19 96-well flat-bottomed plate in 200 ␮l of CM supplemented with was reviewed for 9-mer peptides that could potentially bind to 10% HS and incubated overnight at 37°C, 5% CO2, to minimize MHC class-I molecules using the computer-based prediction background expression of IFN-␥ mRNA expression due to ϩ analysis of H. G. Rammensee, (University of Tu¨bingen, Tu¨bin- lymphocyte manipulation. CD8 T cells were then stimulated in gen, Germany; Ref. 40).4 In this analysis, peptides are ranked vitro with test peptides using an adapted protocol from previous according to a score that takes the presence of primary and studies (43). Briefly C1R-A2 cells were washed three times in secondary MHC-binding anchor residues into account. The serum-free CM and incubated with test peptide at 50 ␮g/ml in analysis was performed for MHC-class-I allele HLA*0201 be- CM at 37°C, 5% CO2, for 2 h. The peptide-loaded cells were cause of the prevalence of this allele in the Caucasian popu- then irradiated with 7500 cGy, washed once, suspended in CM ϩ lation. containing HS, and added to the isolated CD8 T cell at a 1:1 ϩ Peptide Synthesis. All of the peptides used in this ratio. As controls, CD8 T cells were either incubated with study were synthesized by Biosynthesis (Lewisville, TX) to a unloaded C1R-A2 cells (negative control), or with C1R-A2 cells minimum of 95% purity as measured by high-performance and 5 ␮g/ml staphylococcus enterotoxin B (Sigma-Aldrich, St. liquid chromatography. The following peptides were used: Louis, MO; positive control). After3hofcoincubation at 37°C, 5% CO , cells were harvested for RNA-isolation and cDNA CD1974– 82 (HMRPLASWL), CD20127–135 (AISGMILSI), 2 synthesis. Additional negative controls included HLA-A*0201- CD20188–196 (SLFLGILSV), Cytomegalovirus (CMV) pep- negative individuals, CMV seronegative HLA-A*0201-positive tide495–503 (NLVPMVATV) derived from the immunodomi- nant pp65 protein and the synthetically modified gp100 pep- individuals and C1R-A2 cells pulsed with gp100 (209–2M) as tide 209–217(2M; IMDQVPFSV) as a negative control. irrelevant peptide. Peptides were dissolved in DMSO at a concentration of 5 RNA Extraction and cDNA Synthesis. Total RNA was mg/ml, further diluted in PBS, and stored at Ϫ20°C. isolated from test samples using the RNeasy Mini kit (Quiagen, MHC Stabilization Assay. Peptides were tested for their Valencia, CA) and stored at Ϫ80°C. For cDNA synthesis, 1 ␮g ability to bind to HLA-A2 molecules in a MHC-class I stabili- of total RNA was reverse transcribed into DNA with Advantage RT-for-PCR Kit (BD Biosciences, Clontech, Palo Alto, CA) and stored at Ϫ20°C. qPCR. Measurement of IFN-␥ mRNA gene expression 4 Internet address: http://www.uni-tuebingen.de/uni/kxi/. was performed using an ABI Prism 7900 Sequence Detection

system (Perkin-Elmer, Foster City, CA) as described previously patients were stimulated with autologous, irradiated (7500 cGy), (44, 45). The feasibility of this approach for the analysis of peptide-pulsed CD40-B-cells. Briefly, CD40-B-cells were antigen-specific T-cell responses, both in peripheral blood lym- washed three times in serum-free CM and were incubated for phocytes and in tumor tissues, has been validated previously 2 h with test peptide at 50 ␮g/ml in the presence of 5 ␮g/ml (46). Primers for IFN-␥, CD8, and Taqman Probes (Custom human ␤2 microglobulin (Sigma, St. Louis, MO) in CM at 37°C

Oligonucleotide Factory, Foster City, CA) were designed to in 5% CO2. CD40-B-cells were then irradiated (7500 cGy), span exon-exon junctions to prevent transcription of genomic washed once, and added to the isolated CD8ϩ T cell at a ratio of DNA. To create a standard curve, the cDNA was generated by T:CD40-B of 4:1 in CM containing 10% HS and recombinant reverse transcription using the same technique used for the human IL-7 (10 ng/ml; Peprotech). After 7 days in culture, a preparation of test cDNA. IFN-␥ and CD8 cDNA was amplified second stimulation was performed, and the following day, 20 by PCR using the same primers designed for the RT-PCR, IU/ml recombinant human IL-2 (Biosource International, Ca- purified and quantified by UV spectrophotometry. The number marillo, CA) was added. After 14 days in culture, a third of cDNA copies was calculated using the molecular weight of stimulation was performed, followed on day 15 by the addition each gene amplicon. Serial dilutions of the amplified gene at of recombinant human IL-2. After a total of four stimulations, known concentrations were tested by RT-PCR. qPCRs of cDNA the peptide-stimulated T cells were obtained and tested for specimens, cDNA standards, and water as negative control were peptide-specific cytotoxicity. conducted in a total volume of 25 ␮l with Taqman Master mix Flow Cytometric Analysis. CD8ϩ T cells, obtained after (Perkin-Elmer), 400 nM primers, and 200 nM probe. Primer primary isolation, were stimulated with peptide-loaded antigen- sequences were as follows: IFN-␥ (forward), 5Ј-AGCTCTG- presenting cells (APCs) and were stained for intracellular IFN-␥ CATCGTTTTGGGTT; IFN-␥ (reverse), 5Ј-GTTCCATTATC- production. Briefly, C1R-A2 cells were washed three times in CGCTACATCTGAA; IFN-␥ (probe), FAM-TCTTGGCTGT- serum-free CM and were incubated with test peptide at 50 Ј ␮ TACTGCCAGGACCCA-TAMRA; CD8 (forward), 5 -CCC- g/ml for2hinCMat37°C and 5% CO2. C1R-A2-cells were TGAGCAACTCCATCATGT; CD8 (reverse), 5Ј-GTGGGCT- then washed once, irradiated (7500 cGy), and suspended with TCGCTGGCA; and CD8 (probe), FAM-TCAGCCACTTCGT- 1 ϫ 106 CD8ϩ T cells at a 1:1 ratio in CM containing 10% HS GCCGGTCTTC-TAMRA. The thermal cycler parameters were and 10 ␮g/ml Brefeldin A (Sigma). Cells were coincubated for

2 min at 50°C, 10 min at 95°C, and 40 cycles each of 95°C for 16hat37°C and 5% CO2. Unpulsed C1R-A2 cells and C1R-A2 15 s and 60°C for 1 min. Standard curve extrapolation of copy cells pulsed with gp100 peptide (irrelevant peptide) were used number was performed for both IFN-␥ and CD8. The calculated as negative controls. Positive controls were performed by stim- number of copies of IFN-␥ mRNA in each sample was normal- ulating CD8ϩ T cells with 5 ␮g/ml staphylococcus enterotoxin ized to the number of copies of CD8 mRNA by dividing the B (Sigma). CD8ϩ T cells were then stained by incubation with number of copies of IFN-␥ transcripts by the number of copies phycoerythrin-conjugated monoclonal antibodies (Becton Dick- of CD8 transcripts. All of the PCR assays were performed in inson, San Jose, CA). The intracellular staining for IFN-␥ was duplicates and reported as the mean. A 2-fold difference in gene performed after adding FACS Lysing Solution and FACS Per- expression was found to be within the discrimination ability of meabilization Solution (Becton Dickinson) by using FITC- the assay. conjugated monoclonal antibodies (Becton Dickinson). Data Generation of CD40-Activated B Cells. CD40 acti- acquisition was performed on FACSCalibur and was analyzed vated B cells (CD40-B-cells) were generated from CD8- using CellQuest Software (Becton Dickinson). depleted peripheral blood mononuclear cells, as described pre- Cytotoxicity Assay. To determine peptide specific lysis, viously (47). In brief, human CD40L-transfected murine we used a semi-automated mini-cytotoxicity assay, as described fibroblasts (LTK-CD40L) were lethally irradiated (7500 cGy), previously (43, 48). Effector cells were diluted to different were subsequently plated on 6-well plates (BD Biosciences, concentrations and plated in 40-␮l, 60-well Terasaki trays (Rob- Franklin Lakes, NJ) in CM supplemented with 10% FCS, and bins Scientific, Sunnyvale, CA) with six replicates per dilution. ϫ 6 were cultured overnight at 37°C, 5% CO2. After washing the Target cells at a concentration of 2 10 cells/ml were stained plates twice with PBS, we cocultured the peripheral blood with 10 ␮g/ml Calcein-AM (CAM; Molecular Probes Inc., mononuclear cells with fibroblasts at 2 ϫ 106 cells/ml in the Eugene, OR) for 60 min at 37°C. After washing three times in presence of interleukin (IL)-4 (4 ng/ml; Peprotech, Rocky Hill, CM, cells were resuspended at 105 cells/ml in CM supplemented NJ) and Cyclosporin A 0.7 ␮g/ml in Iscove‘s modified Dulbec- with 10% HS. Target cells (103)in10␮l of medium were added co’s medium (Invitrogen), supplemented with 10% HS, 50 to each well containing effector cells. Wells with target cells ␮g/ml transferrin (Boehringer Mannheim, Indianapolis, IN), 5 alone and medium alone were used for maximum and minimum ␮g/ml insulin (Sigma Chemical Co., St. Louis, MO) and peni- fluorescence emission, respectively. After 4-h incubation at ␮ cillin/streptomycin at 37°C, 5% CO2. Cultured cells were trans- 37°Cin5%CO2, 5 l of FluoroQuench (EB, Stain-Quench ferred to new plates with freshly prepared, irradiated fibroblasts Reagent; One Lambda Inc., Canoga Park, CA) was added to every 3–5 days. each well and the trays were centrifuged for 1 min at 60 ϫ g Before use, CD40-B cells were Ficoll-density centrifugated before measurement of fluorescence using an automated followed by washing twice with PBS to remove nonviable cells Lambda Fluoroscan (One Lambda Inc.). A decrease in the including remaining fibroblasts. fluorescence emission is proportional to the degree of target cell Expansion of Peptide-Specific CD8؉ T Cells Using lysis once the released dye is quenched by the hemoglobulin Peptide-Pulsed CD40-B-Cells. Purified (Ͼ98%) CD8ϩ T contained in the FluoroQuench reagent. The percentage of lysis cells (CD8ϩ T cell Isolation Kit; Miltenyi, Auburn, CA) from was calculated as follows.

% lysis Table 2 CD8ϩ T-cell reactivities to peptides in healthy individuals Stimulation index (SI)b Mean experimental emission Ϫ mean minimum Healthy ϭ ͩ1 Ϫ ͪ a Mean maximum Ϫ mean minimum individuals CD1974–82 CD20127–135 CD20188–196 1 n.d.c n.d. 1.0 ϫ 100 2 n.d. n.d. 1.9 3 1.3 n.d. 0.9 MHC restriction of lytic activity was tested by blockade of 4 1.3 n.d. 1.0 HLA-A2 using monoclonal antibody BB7.2 and the correspond- 5 1.6 n.d. 1.7 ing isotype control (Becton Dickinson). 6 1.2 n.d. 1.4 Statistical Analysis. To determine specific response to 7 0.5 0.6 0.8 ϩ 8 1.5 n.d. 1.2 stimulation, mRNA for IFN-␥ from CD8 T cells stimulated 9 0.8 n.d. 1.6 with test-peptide versus unpulsed APC (background) was de- 10 1.1 0.6 0.6 tected by qPCR. The IFN-␥ mRNA copy number was first 11 2.6 1.6 1.2 corrected for CD8 mRNA. A cutoff value of 2.0 for the ratio of 12 0.1 0.2 0.1 13 1.2 2.4 0.6 IFN-␥ mRNA obtained from CD8ϩ T cells stimulated with ϩ 14 1.9 1.1 2.2 relevant test peptides to that obtained from CD8 T cells 15 1.5 3.5 1.5 stimulated with unpulsed APC was considered to be evidence of 16 0.9 2.0 1.1 epitope specificity. The cutoff value was derived by analyzing 17 0.6 0.7 0.1 IFN-␥ mRNA transcripts detectable in CD8ϩ T cells both from 18 1.2 n.d. 0.6 19 0.1 0.3 0.3 healthy donors and from patients stimulated with gp100 (209- 20 1.2 1.5 1.3 ϩ 2M; irrelevant peptide) to background. Analysis of these CD8 21 1.9 1.9 2.0 T cells identified a mean ratio of 1.0 (range, 0.7–1.3) with 95 22 1.1 1.0 1.1 and 99% confidence intervals of 1.0 Ϯ 0.89 and 1.0 Ϯ 1.16, 23 0.6 1.8 2.3 respectively, a SE of 0.06, and a SD of 0.20. The cutoff ratio a Corresponding numbers of healthy individuals are maintained (stimulation index) was estimated by adding the mean to three throughout this article. b Stimulation index (SI) determined by the ratio of IFN-␥ mRNA SDs, which equaled 1.6. To minimize the possibility of falsely ϩ ϩ copy number obtained from CD8 T cells stimulated with relevant test considering CD8 T cells immunoreactive, we accepted a peptides to that obtained from CD8ϩ T cells stimulated with unpulsed 2-fold increase in stimulated:unstimulated IFN-␥ transcript ratio antigen-presenting cells. A cutoff value of 2.0 was considered to be as evidence of epitope-specific reactivity. evidence of epitope specificity. SI displayed in underlined bold repre- Fisher’s exact test was calculated to determine, whether sent positive reactivities. c n.d., not done. there was a statistically significant difference in T-cell response to test peptides between normal healthy individuals and patients. Statistical significance was achieved, when P Ͻ 0.05. low-affinity binding and a FI greater that 1.5 high-affinity RESULTS binding. As shown in Table 1, T2 cells pulsed with positive control CMV (pp65) or gp100 peptide showed a FI of 2.3 or 2.2 Selection and Binding Activity of Peptides to HLA- respectively. Among the CD19- and CD20-derived peptides, A*0201 Molecules. Using the SYFPETHI computer-based pulsing with CD20 led to the highest FI (1.6), whereas prediction analysis we selected candidate 9-mer peptides that 127–135 the other peptides showed a FI of 1.3 and 1.4 (CD19 and could potentially bind to MHC-class I molecules (40). The 74–82 CD20 ) corresponding to a lower-affinity binding to the peptides CD19 , CD20 , and CD20 were syn- 188–196 74–82 127–135 188–196 HLA-A*0201 molecules. thesized and their binding to HLA-A*0201 molecules evaluated Identification of CD19- and CD20-Peptide-Specific in the T2 cell assay. A FI between 1.11 and 1.49 was considered ؉ CD8 T-Cell Reactivity in Healthy Individuals and Patients with B-Cell Malignancies. To determine whether CD8ϩ T cells recognizing CD19 and CD20 exist in healthy individuals ␥ Table 1 Peptides and predicted binding to HLA-A*0201 and patients with B-cell malignancies, we looked for IFN- mRNA production in peptide-stimulated CD8ϩ T cells using Rammensee qPCR. We analyzed 23 healthy donors and 28 patients with Peptide Positiona Sequenceb scorec FId B-cell malignancies (18 CLL, 7 follicular lymphoma, 2 acute e CMV (pp65) 495–503 NLVPMVATV 30 2.3 lymphocytic leukemia and 1 large cell lymphoma). As controls, gp100 (209-2M) 209–217 IMDQVPFSV 22 2.2 CD8ϩ T cells were stimulated either with CMV(pp65)-peptide CD1974–82 74–82 HMRPLASWL 21 1.3 CD20127–135 127–135 AISGMILSI 27 1.6 or with staphylococcus enterotoxin B (positive controls) and CD20188–196 188–196 SLFLGILSV 32 1.4 gp100 (209–2M) peptide (negative control). A positive re- 4 a Peptide position within entire amino acid sequence. sponse was defined as Ն100 IFN-␥ mRNA copies/10 CD8 b Peptide sequence with single letter abbreviations for amino acids. copies and a stimulation index (SI) of Ն2, where SI ϭ IFN-␥ c Rammensee score of predicted peptide binding in arbitrary units mRNA copies/104 CD8 copies in peptide-pulsed T2 cell cul- (40). d tures/unpulsed cultures. FI, fluorescence index of peptide binding determined by MHC- ϩ stabilization assay. CD8 T-cell responses to CD1974–82 were detected in 1 of e CMV, cytomegalovirus. 21 healthy individuals with a SI of 2.6 (Table 2; Fig. 1A).

CD20188–196 (donors 14, 21, and 23) with SI between 2.2 and 2.3 (Table 2; Fig. 1, B and C). None of the healthy donors had a response to more than one peptide or to gp100 peptide. Clinical data for the patients with B-cell malignancies are ϩ shown in Table 3. A CD8 T-cell response to CD1974–82 was observed in 7 of 27 patients with B-cell malignancies (SI range, 2.1–7.1; Table 4; Fig. 1A). Positive T-cell responses after stimulation with CD20-derived peptides were detected in eight patients. Seven of 24 patients had a positive response to stimula-

tion with CD20127–135 (SI range, 2.3–3.0). Three of 26 patients showed a response to CD20188–196 (SI range, 2.3–3.7; Table 4; Fig. 1, B and C). In two patients (patients 3 and 16), there was

a positive response to both CD1974–82 and CD20127–135 peptides. Two patients showed a positive response to all three peptides (patients 5 and 8). No patient responded to stimulation with gp100 peptide. It was notable that six of seven positive CD8ϩ T-cell

responses against CD1974–82 occurred in CLL-patients. A ϩ CD1974–82 CD8 T-cell response was significantly more frequent in the CLL group compared with normal donors (Fig. 1A; Table 5).

Similarly, positive CD20127–135 and CD20188–196 re-

Table 3 Patients: clinical properties Age Yr after Diagnosis Patient Sex (yr) Stagea Treatmentb diagnosis CLLc 1 M 64 I CHOP, Flu, CVP 10 2 F 54 I none 1 3 F 56 I none 3 4 F 80 0 none 3 5 F 62 0 none 4 6 F 51 0 none 5 7 M 58 0 none 8 8 F 57 0 none 4 9 M 57 0 none 2 10 M 70 I none 6 11 F 72 I CHOP 2 12–18d FL 19 F 60 III none 0 Fig. 1 CD8ϩ T cell response to stimulation with the HLA-A*0201- 20 M 54 III none 0 restricted peptides. Selected T cells from healthy individuals (n ϭ 23, 21 F 40 IV none 0 E), chronic lymphocytic leukemia (CLL) patients (n ϭ 18, F) and 22 M 60 IV none 0 patients with follicular lymphoma, large cell non-Hodgkin-lymphoma or 23 M 32 IV none 0 acute lymphocytic leukemia (non-CLL, n ϭ 10, ‚) were incubated for 24 M 34 IV none 0 3 h with unpulsed antigen-presenting cells (APCs) or APCs pulsed with 25 M 60 III none 0 peptides. Peptides used were CD1974–82 (A), CD20127–135 (B) and NHL 26 F 31 III CHOP 1 CD20188–196 (C). Values represent the stimulation index (SI) of single experiments where SI is determined by the ratio of IFN-␥ mRNA copy ALL 27 M 44 CR Hyper CVAD 0 number obtained from CD8ϩ T cells stimulated with relevant test peptides to that obtained from CD8ϩ T cells stimulated with unpulsed 28 F 21 CR Daunorubicin/ 3 APCs. A cutoff value of 2.0 was considered to be evidence of epitope Vincristin specificity. p, statistical difference of T-cell reactivity between CLL- a Rai stage (CLL); Ann Arbor classification (FL, NHL); CR: com- patients and, respectively, non-CLL-patients and normal individuals plete remission. (p ϭ n.s., not significant). b Treatment: none, not on active treatment at the time of analysis; CHOP, cyclophosphamide/vincristine/Adriamycin/prednisone; Flu, flu- darabine; CVP, cyclophosphamide/vincristine/prednisone; CVAD, cyclophosphamide/vincristine/cytarabine/doxorubicin. c Positive T-cell responses after stimulation with CD20-derived CLL, chronic lymphocytic leukemia; FL, follicular lymphoma; NHL, non-Hodgkin lymphoma (large cell); ALL, acute lymphocytic peptides were detected in 6 of 23 healthy individuals. Three leukemia. d 6 responded to stimulation with CD20127–135 (donors 13, 15, and Patients with CLL and leucocytes Ͼ20 ϫ 10 /ml, treatment 16) with SI between 2.0 and 3.5 and three to stimulation with conditions not known.

sponses occurred only in the CLL patient group. However, the Table 5 Summary of CD8ϩ T-cell reactivities in healthy individuals, frequency of positive responses did not differ significantly from chronic lymphocytic leukemia (CLL) patients, and patients with other that found in normal individuals. Overall the CD19 peptide B-cell malignancies induced the greatest CD8ϩ T-cell reactivity [SI, 3.8 compared Stimulation index Ն 2a with 3.2 (CD20188–196) and 2.6 (CD20127–135)]. CD19 CD20 CD20 Peptide specificity determined by qPCR was confirmed 74–82 127–135 188–196 Healthy individuals 1/21 (5%)b 3/14 (21%) 3/23 (13%) previously using intracellular cytokine assay and tetramer anal- CLL 6/17 (35%) 7/15 (47%) 3/16 (19%) ysis (49). In six healthy donors, IFN-␥ mRNA expression Non-CLLc 1/10 (10%) 0/9 (0%) 0/10 (0%) (qPCR) after CMV pp65 stimulation was compared with 495–503 a Stimulation index determined by the ratio of IFN-␥ mRNA copy intracellular IFN-␥ production and tetramer assay. There was number obtained from CD8ϩ T-cells stimulated with relevant test pep- strong correlation between qPCR, intracellular cytokine assay, tides to that obtained from CD8ϩ T-cells stimulated with unpulsed and tetramer assay (R2 ϭ 0.92 and 0.78, respectively). The antigen presenting cells. A cutoff value of 2.0 was considered to be qPCR assay was the most sensitive, detecting 1/105 CMV pp65- evidence of epitope specificity. b Positive CD8ϩ T-cell reactivities out of total number of tested specific CD8ϩ T cells (data not shown). Good correlation be- ϩ individuals, percentage of reactivities. tween the analysis of CD8 T-cell reactivity by flow cytometry c Follicular lymphoma, non-Hodgkin lymphoma (large cell), and (detection of intracellular IFN-␥ expression) and by qPCR is acute lymphocytic leukemia.

shown in Figs. 2 and 3 after stimulation with CD1974–82 in patient 1 (representative experiment).

In Vitro Expansion of Peptide-Specific CD8؉ T

ϩ Cells with Recombinant Peptide. To determine whether Table 4 CD8 T-cell reactivities to peptides in patients with B-cell malignancies CD1974–82-specific T cells could be expanded in vitro, B cells

b stimulated with CD40L and loaded with CD1974–82 were used Stimulation index (SI) ϩ to weekly restimulate autologous CD8 T cells in media con- a Patients CD1974–82 CD20127–135 CD20188–196 taining 10 units/ml IL-2 and 10 ng/ml IL-7. Three patients CLLc 1 7.1 n.d.c 0.6 (patient 1, 8, and 26), in whom we were able to detect antigen- 2 1.2 0.9 0.9 specific T cells, were studied. The cytotoxicity of the expanded 3 2.5 2.5 n.d. T cells was then tested in a Calcein-AM release assay. Cyto- 4 1.6 2.5 0.8 toxicity against CD19 -pulsed T2 cells ranged between 24 5 2.1 3.0 3.5 74–82 6 0.2 0.7 0.2 and 48% at E:T-ratios between 30:1 and 70:1 in patients 1, 8, 7 1.6 2.9 1.4 and 26 (Fig. 4, A–C). Negative controls (T2 cells pulsed with 8 6.0 2.3 3.7 gp100 peptide or K562 cells) showed no lysis. Lysis of the 9 1.4 1.4 0.9 T2-loaded target was partial after treatment with a monoclonal 10 0.8 1.6 1.7 11 0.6 2.7 0.8 antibody against HLA-A2 confirming that the CTLs were 12 1.4 1.5 1.5 HLA-A2 restricted (Fig. 5). However, there was no significant 13 1.4 n.d. 1.4 lysis of CD40-activated autologous B cells (Fig. 4) nor lysis of 14 n.d. n.d. 1.6 unstimulated B cells or allogeneic HLA-A2-positive EBV-LCL 15 2.4 0.9 n.d. (data not shown). These observations suggest that, despite pep- 16 4.3 2.3 0.9 ϩ 17 0.8 1.1 1 tide specificity of the CD8 cells, the expression of peptide on 18 0.3 1.1 2.3 the unpulsed B cells was too low to activate lytic activity. FL 19 1.4 0.8 1.4 20 1.6 1.9 1.1 DISCUSSION 21 1.1 0.8 1.3 Recent studies in non-Hodgkin lymphoma of the B-cell 22 1.2 1.1 1.2 type indicate the presence of autologous CD4ϩ and CD8ϩ 23 0.9 0.7 0.8 24 1.3 1.4 1.4 T-lymphocytes reactive against malignant B cells (5–8, 50). 25 1 0.9 0.8 Because malignant B cells can act as APCs (51), they could induce T-cell responses directly or indirectly via professional NHL 26 2.3 n.d. 1.4 APCs. Recently it has been shown, that dendritic cells induce ALL 27 0.9 1.8 1.1 proliferation and antitumor-responses against B-cell lymphomas 28 1.5 1.0 1.2 and CLL (52–55). Here we set out to determine, whether the a Corresponding numbers of patients are maintained throughout proteins CD19 and CD20, abundantly expressed by many ma- this article. lignant B cells, could induce such cytotoxic CD8ϩ T-cell re- b Stimulation index (SI) determined by the ratio of IFN-␥ mRNA copy number obtained from CD8ϩ T cells stimulated with relevant test sponses to B-cell leukemias and lymphomas, in patients with peptides to that obtained from CD8ϩ T cells stimulated with unpulsed B-cell malignancies and in normal individuals. antigen-presenting cells. A cutoff value of 2.0 was considered to be We studied the B-cell differentiation antigens CD19 and evidence of epitope specificity. SI displayed in underlined bold repre- CD20, because these molecules play a critical role in the devel- sent positive reactivities. c CLL, chronic lymphocytic leukemia; FL, follicular lymphoma; opment, differentiation, and activation of B cells. They are NHL, non-Hodgkin lymphoma (large cell); ALL, acute lymphocytic tissue-restricted to B-lineage cells and are expressed in the leukemia; n.d., not done. majority of B-cell malignancies (35, 36). We selected 9-mer

peptide sequences (CD1974–82, CD20127–135, and CD20188–196) on the basis of HLA-A2 binding as candidates for the detection and the expansion of antigen-specific CTLs. Interestingly, despite relatively weak binding of these peptides to MHC molecules, all three were immunogenic. Using a qPCR assay that has 10-fold greater sensitivity than tetramer analysis (49), CD8ϩ T cells recognizing CD1974–82, CD20127–135, and CD20188–196 were detected in 5, 21, and 13%, respectively, of normal individuals. The presence of self-reactive T cells, that have escaped thymic deletion in the periphery, does not necessarily imply functional autoimmunity: In the thymus, T cells with a high binding affinity to self-MHC/self-peptide complexes are eliminated from the T-cell repertoire by clonal deletion (21, 56, 57). ϩ However, T cells with low-avidity for self-antigens escape Fig. 3 Analysis of CD8 T-cell response against CD1974–82 by quan- clonal deletion and enter the post-thymic compartment (58–60). titative real-time PCR (qPCR) in patient 1. Data are shown as absolute ␥ ϩ We assume, that the CD19- and CD20-specific T cells described numbers of IFN- mRNA copies after 3-h stimulation of CD8 selected T cells with unpulsed (left) or peptide-pulsed (right) antigen-presenting here were of low avidity, because T-cell responses to peptide cells (C1R-A2). Error bars, the SD.

were detected only after pulsing the APCs with high concentrations of peptide (50␮g/ml). (In preliminary experiments, peptide doses of 0.1 or 1␮g/ml never elicited T-cell responses).

T cells reactive against CD1974–82 were also detected in patients with B-cell malignancies but predominantly in CLL (35

versus 5%). T-cell responses to CD1974–82 were significantly more frequent and greater in amplitude than in normal individ-

uals. CD20127–135 but not CD20188–196 recognizing T cells were also found twice as frequently in the CLL group compared with the normal individuals (47 versus 21%), suggesting that peptide

CD20188–196 was less immunogenic than CD1974–82 and CD20127–135. Although T-cell responses to CD19 and CD20 were commonest in CLL, insufficient numbers of patients with other B-cell malignancies were tested to determine whether there was a statistically greater probability of CLL patients developing responses to B-cell antigens. The increased reactivity to self-peptides in CLL compared with normals could be due to either direct or indirect T-cell stimulation from CLL or dendritic cells presenting quantitatively larger amounts of CD19 and CD20 because of the characteristic increase in B cell production in CLL. However, because CLL cells only weakly express critical adhesion and costimulatory molecules (61–63), CD19 and CD20 peptide epitopes are more likely to be presented by professional APCs. Although the expression of CD19 and CD20 on CLL cells is somewhat decreased, the massive leucocytosis in CLL could counteract any decreased antigenic potential from these molecules (64, 65). It is possible that CLL patients had the best CD19 and CD20 responses because of a higher tumor load than patients with lymphomas. Unlike high- avidity T cells, which can be eliminated from the repertoire by high concentrations of antigen, stimulation of low-avidity T cells by these self-antigens could elicit increased frequencies of Fig. 2 Flow cytometric analysis of CD8ϩ T-cell response to stimula- persisting non-autoreactive T cells. However, T cells analyzed tion with CD1974–82 in patient 1. Peptide-reactive T cells were identi- in this study were derived from peripheral blood. Additional fied by intracellular IFN-␥ expression (IC IFN-G) after stimulation with studies to look for T cells recognizing CD19 and CD20 epitopes peptide-pulsed antigen-presenting cells. A and B, the CD8/IFN-␥ profile in lymph node biopsies would be worthwhile. of CD8ϩ T cells after coincubation with unpulsed (A) and peptide- pulsed (B) antigen-presenting cells (C1R-A2). Frequencies of IFN-␥- Because of the significant increase in CD19 responses in positive T cells are shown as percentages of the total number of CD8ϩ CLL, we studied T-cell responses to this peptide further in three T cells. PE, phycoerythrin. patients (2 CLL and 1 large cell lymphoma). After 4 weeks of

Fig. 5 Inhibition of CD1974–82-specific target cell lysis by anti- HLA-A2 monoclonal antibody (mAb). The cytotoxic activity of ϩ CD1974–82-specific CD8 T cells (patient 1) against peptide-pulsed antigen-presenting cells (T2 cells ᭜) occurs in a MHC-restricted manner and is inhibited by the addition of anti-HLA-A2 mAb (Œ). (‚) indicates lytic activity after addition of a control (isotype)-antibody (Ab).

␮ stimulation with 50 g/ml recombinant CD1974–82, peptide- specific, MHC-restricted cytotoxicity of peptide-pulsed APCs (T2 cells) occurred in all three cases. However these T-cell lines were not cytotoxic to unpulsed autologous or allogeneic B cells whether or not they were activated by CD40L. This suggested that the peptides studied were not naturally presented, or that antigen density is below the threshold required to cause a cytotoxic response in the CD8ϩ T cells. Activation of B cells by CD40L has recently been shown to effectively up-regulate expression of costimulatory molecules in normal and malignant B cells including CLL cells, making it unlikely that the CLL targets failed to deliver costimulatory signals (61–63). Further- more, CD40-activated CLL B cells were capable of expanding T cells in culture. In conclusion, these studies indicate that T-cell responses to peptides derived from CD19 and CD20 molecules are common in both normal individuals and patients with CLL but involve only low-avidity T cells, requiring high doses of peptide to stimulate them. Our findings suggest that the T-cell repertoire is thus shaped in CLL by the elimination of any high-avidity CLL-specific T cells and the expansion of low-avidity T cells, recognizing common and highly expressed molecules such as CD19 and CD20 during the course of CLL. This suggests a form of T-cell repertoire shaping by CLL similar to that described for proteinase 3 antigen PR1 in patients with chronic myelogenous leukemia (50). Development of T-cell immunotherapy approaches to CLL using CD19 or CD20 peptide antigens would, therefore, require strategies to increase T-cell avidity to generate effective B-cell-specific cytotoxic T cells. Because nonmalig- Fig. 4 Cytotoxicity assay to show lytic activity of CD1974–82-specific CD8ϩ T cells after 4 weeks expansion. CD8ϩ T cells from patient 1 (A), nant B cells would also be recognized by CD19- and CD20- patient 8 (B), and patient 26 (C) were weekly stimulated using CD40- specific T cells, it would be necessary to limit the in vivo activated, peptide-loaded autologous B cells. Cytotoxic responses to T2 persistence of specific T cells to avoid continuous B-cell deple- cells pulsed with CD19 (᭜), gp100-peptide (f), and unpulsed T2 74–82 tion using, for example, IL-2-dependent T cells, or T cells cells (᭛) are shown. K562 cells (Œ) were used as control to rule out natural killer cell activity. To analyze lytic activity against B cells, coexpressing a suicide gene such as the herpes simplex virus autologous CD40-activated B cells (Ⅺ) were used as target cells. thymidine kinase (HSV-TK).

ACKNOWLEDGMENTS 18. Sha, W. C., Nelson, C. A., Newberry, R. D., Kranz, D. M., Russell, We thank Dr. Ernst Holler for his support and encouragement and J. H., and Loh, D. Y. Positive and negative selection of an antigen receptor on T cells in transgenic mice. Nature (Lond), 336: 73–76, 1988. Dr. Mathias Oelke for his support and helpful discussions. 19. Pircher, H., Burki, K., Lang, R., Hengartner, H., and Zinkernagel, R. M. Tolerance induction in double specific T-cell receptor transgenic REFERENCES mice varies with antigen. Nature (Lond), 342: 559–561, 1989. 1. Rosenberg, S. A. A new era for cancer immunotherapy based on the 20. Amsen, D., and Kruisbeek, A. M. Thymocyte selection: not by TCR genes that encode cancer antigens. Immunity, 10: 281–287, 1999. alone. Immunol. Rev., 165: 209–229, 1998. 2. Brandle, D., Brasseur, F., Weynants, P., Boon, T., and Van den, E. B. 21. Stockinger, B. T lymphocyte tolerance: from thymic deletion to A mutated HLA-A2 molecule recognized by autologous cytotoxic T peripheral control mechanisms. Adv. Immunol., 71: 229–265, 1999. lymphocytes on a human renal cell carcinoma. J. Exp. Med., 183: 2501–2508, 1996. 22. Vitetta, E. S., Thorpe, P. E., and Uhr, J. W. Immunotoxins: magic bullets or misguided missiles? Immunol. Today, 14: 252–259, 1993. 3. Gueguen, M., Patard, J. J., Gaugler, B., Brasseur, F., Renauld, J. C., Van Cangh, P. J., Boon, T., and Van den Eynde, B. J. An antigen 23. Grossbard, M. L., Press, O. W., Appelbaum, F. R., Bernstein, I. D., recognized by autologous CTLs on a human bladder carcinoma. J. Im- and Nadler, L. M. Monoclonal antibody-based therapies of leukemia munol., 160: 6188–6194, 1998. and lymphoma. Blood, 80: 863–878, 1992. 4. Yi, Q., Eriksson, I., He, W., Holm, G., Mellstedt, H., and Osterborg, 24. Waldmann, T. A. Monoclonal antibodies in diagnosis and therapy. A. Idiotype-specific T lymphocytes in monoclonal gammopathies: evi- Science (Wash. DC), 252: 1657–1662, 1991. dence for the presence of CD4ϩ and CD8ϩ subsets. Br. J. Haematol., 25. Maloney, D. G., Brown, S., Czerwinski, D. K., Liles, T. M., Hart, 96: 338–345, 1997. S. M., Miller, R. A., and Levy, R. Monoclonal anti-idiotype antibody 5. Rezvany, M. R., Jeddi-Tehrani, M., Rabbani, H., Lewin, N., Avila- therapy of B-cell lymphoma: the addition of a short course of chemo- Carino, J., Osterborg, A., Wigzell, H., and Mellstedt, H. Autologous T therapy does not interfere with the antitumor effect nor prevent the lymphocytes may specifically recognize leukaemic B cells in patients emergence of idiotype-negative variant cells. Blood, 80: 1502–1510, with chronic lymphocytic leukaemia. Br. J. Haematol., 111: 608–617, 1992. 2000. 26. Dillman, R. O. Antibodies as cytotoxic therapy. J. Clin. Oncol., 12: 6. Gitelson, E., Hammond, C., Mena, J., Lorenzo, M., Buckstein, R., 1497–1515, 1994. Berinstein, N. L., Imrie, K., and Spaner, D. E. Chronic lymphocytic 27. Cooper, L. J., Topp, M. S., Serrano, L. M., Gonzalez, S., Chang, leukemia-reactive T cells during disease progression and after autolo- W. C., Naranjo, A., Wright, C., Popplewell, L., Raubitschek, A., For- gous tumor cell vaccines. Clin. Cancer Res., 9: 1656–1665, 2003. man, S. J., and Jensen, M. C. T-cell clones can be rendered specific for 7. Shinohara, N., Watanabe, M., Sachs, D. H., and Hozumi, N. Killing CD19: toward the selective augmentation of the graft-versus-B-lineage of antigen-reactive B cells by class II-restricted, soluble antigen-specific leukemia effect. Blood, 101: 1637–1644, 2003. CD8ϩ cytolytic T lymphocytes. Nature (Lond), 336: 481–484, 1988. 28. Roessig, C., Scherer, S. P., Baer, A., Vormoor, J., Rooney, C. M., 8. Osterborg, A., Yi, Q., Bergenbrant, S., Holm, G., Lefvert, A. K., and Brenner, M. K., and Juergens, H. Targeting CD19 with genetically Mellstedt, H. Idiotype-specific T cells in multiple myeloma stage I: an modified EBV-specific human T lymphocytes. Ann. Hematol., 81 evaluation by four different functional tests. Br. J. Haematol., 89: (Suppl. 2): S42–S43, 2002. 110–116, 1995. 29. Brentjens, R. J., Latouche, J. B., Santos, E., Marti, F., Gong, M. C., 9. Rasmussen, T., Hansson, L., Osterborg, A., Johnsen, H. E., and Lyddane, C., King, P. D., Larson, S., Weiss, M., Riviere, I., and Mellstedt, H. Idiotype vaccination in multiple myeloma induced a Sadelain, M. Eradication of systemic B-cell tumors by genetically reduction of circulating clonal tumor B cells. Blood, 101: 4607–4610, targeted human T lymphocytes co-stimulated by CD80 and interleukin- 2003. 15. Nat. Med., 9: 279–286, 2003. 10. Li, Y., Bendandi, M., Deng, Y., Dunbar, C., Munshi, N., Jagannath, 30. Jensen, M., Cooper, L., Wu, A., Forman, S., and Raubitschek, A. S., Kwak, L. W., and Lyerly, H. K. Tumor-specific recognition of Engineered CD20-specific primary human cytotoxic T lymphocytes for human myeloma cells by idiotype-induced CD8(ϩ) T cells. Blood, 96: targeting B-cell malignancy. Cytotherapy, 5: 131–138, 2003. 2828–2833, 2000. 31. Roberts, W. K., Livingston, P. O., Agus, D. B., Pinilla-Ibarz, J., 11. Ruffini, P. A., Neelapu, S. S., Kwak, L. W., and Biragyn, A. Zelenetz, A., and Scheinberg, D. A. Vaccination with CD20 peptides Idiotypic vaccination for B-cell malignancies as a model for therapeutic induces a biologically active, specific immune response in mice. Blood, cancer vaccines: from prototype protein to second generation vaccines. 99: 3748–3755, 2002. Haematologica, 87: 989–1001, 2002. 32. Nadler, L. M., Anderson, K. C., Marti, G., Bates, M., Park, E., 12. Bendandi, M., Gocke, C. D., Kobrin, C. B., Benko, F. A., Sternas, Daley, J. F., and Schlossman, S. F. B4, a human B lymphocyte- L. A., Pennington, R., Watson, T. M., Reynolds, C. W., Gause, B. L., associated antigen expressed on normal, mitogen-activated, and malig- Duffey, P. L., Jaffe, E. S., Creekmore, S. P., Longo, D. L., and Kwak, nant B lymphocytes. J. Immunol., 131: 244–250, 1983. L. W. Complete molecular remissions induced by patient-specific vac- 33. Nadler, L. M., Korsmeyer, S. J., Anderson, K. C., Boyd, A. W., cination plus granulocyte-monocyte colony-stimulating factor against Slaughenhoupt, B., Park, E., Jensen, J., Coral, F., Mayer, R. J., Sallan, lymphoma. Nat. Med., 5: 1171–1177, 1999. S. E., et al. B cell origin of non-T cell acute lymphoblastic leukemia. A 13. Coulie, P. G. Human tumour antigens recognized by T cells: new model for discrete stages of neoplastic and normal pre-B cell differen- perspectives for anti-cancer vaccines? Mol. Med. Today, 3: 261–268, tiation. J. Clin. Investig., 74: 332–340, 1984. 1997. 34. Tedder, T. F., and Engel, P. CD20: a regulator of cell-cycle pro- 14. De Visser, K. E., Schumacher, T. N., and Kruisbeek, A. M. CD8ϩ gression of B lymphocytes. Immunol. Today, 15: 450–454, 1994. T cell tolerance and cancer immunotherapy. J. Immunother., 26: 1–11, 35. Campos, L., Guyotat, D., Larese, A., Mazet, L., Bourgeot, J. P., 2003. Ehrsam, A., and Fiere, D. Expression of CD 19 antigen on acute 15. Rosenberg, S. A. Progress in human tumour immunology and monoblastic leukemia cells at diagnosis and after TPA-induced differ- immunotherapy. Nature (Lond), 411: 380–384, 2001. entiation. Leuk. Res., 12: 369–372, 1988. 16. Kappler, J. W., Roehm, N., and Marrack, P. T cell tolerance by 36. Tedder, T. F., and Schlossman, S. F. Phosphorylation of the B1 clonal elimination in the thymus. Cell, 49: 273–280, 1987. (CD20) molecule by normal and malignant human B lymphocytes. 17. Kisielow, P., Bluthmann, H., Staerz, U. D., Steinmetz, M., and von J. Biol. Chem., 263: 10009–10015, 1988. Boehmer, H. Tolerance in T-cell-receptor transgenic mice involves 37. Zweerink, H. J., Gammon, M. C., Utz, U., Sauma, S. Y., Harrer, T., deletion of nonmature CD4ϩ8ϩ thymocytes. Nature (Lond), 333: 742– Hawkins, J. C., Johnson, R. P., Sirotina, A., Hermes, J. D., Walker, 746, 1988. B. D., et al. Presentation of endogenous peptides to MHC class I-

restricted cytotoxic T lymphocytes in transport deletion mutant T2. 51. Hathcock, K. S., Laszlo, G., Pucillo, C., Linsley, P., and Hodes, cells. J. Immunol., 150: 1763–1771, 1993. R. J. Comparative analysis of B7-1 and B7-2 costimulatory ligands: 38. Storkus, W. J., Howell, D. N., Salter, R. D., Dawson, J. R., and expression and function. J. Exp. Med., 180: 631–640, 1994. Cresswell, P. NK susceptibility varies inversely with target cell class I 52. Banchereau, J., and Steinman, R. M. Dendritic cells and the control HLA antigen expression. J. Immunol., 138: 1657–1659, 1987. of immunity. Nature (Lond), 392: 245–252, 1998. 39. Hogan, K. T., Shimojo, N., Walk, S. F., Engelhard, V. H., Maloy, 53. Kwak, L. W., Campbell, M. J., Czerwinski, D. K., Hart, S., Miller, ␣ W. L., Coligan, J. E., and Biddison, W. E. Mutations in the 2 helix of R. A., and Levy, R. Induction of immune responses in patients with HLA-A2 affect presentation but do not inhibit binding of influenza virus B-cell lymphoma against the surface-immunoglobulin idiotype ex- matrix peptide. J. Exp. Med., 168: 725–736, 1988. pressed by their tumors. N. Engl. J. Med., 327: 1209–1215, 1992. 40. Rammensee, H., Bachmann, J., Emmerich, N. P., Bachor, O. A., 54. Hsu, F. J., Benike, C., Fagnoni, F., Liles, T. M., Czerwinski, D., and Stevanovic, S. SYFPEITHI: database for MHC ligands and peptide Taidi, B., Engleman, E. G., and Levy, R. Vaccination of patients with motifs. Immunogenetics, 50: 213–219, 1999. B-cell lymphoma using autologous antigen-pulsed dendritic cells. Nat. 41. McIntyre, C. A., Rees, R. C., Platts, K. E., Cooke, C. J., Smith, Med., 2: 52–58, 1996. M. O., Mulcahy, K. A., and Murray, A. K. Identification of peptide epitopes of MAGE-1, -2, -3 that demonstrate HLA-A3-specific binding. 55. Krackhardt, A. M., Harig, S., Witzens, M., Broderick, R., Barrett, Cancer Immunol. Immunother., 42: 246–250, 1996. P., and Gribben, J. G. T-cell responses against chronic lymphocytic leukemia cells: implications for immunotherapy. Blood, 100: 167–173, 42. McArdle, S. E., Rees, R. C., Mulcahy, K. A., Saba, J., McIntyre, C. A., and Murray, A. K. Induction of human cytotoxic T lymphocytes 2002. that preferentially recognise tumour cells bearing a conformational p53 56. Kruisbeek, A. M., and Amsen, D. Mechanisms underlying T-cell mutant. Cancer Immunol. Immunother., 49: 417–425, 2000. tolerance. Curr. Opin. Immunol., 8: 233–244, 1996. 43. Molldrem, J., Dermime, S., Parker, K., Jiang, Y. Z., Mavroudis, D., 57. Sebzda, E., Mariathasan, S., Ohteki, T., Jones, R., Bachmann, Hensel, N., Fukushima, P., and Barrett, A. J. Targeted T-cell therapy for M. F., and Ohashi, P. S. Selection of the T cell repertoire. Annu. Rev. human leukemia: cytotoxic T lymphocytes specific for a peptide derived Immunol., 17: 829–874, 1999. from proteinase 3 preferentially lyse human myeloid leukemia cells. 58. De Visser, K. E., Cordaro, T. A., Kioussis, D., Haanen, J. B., Blood, 88: 2450–2457, 1996. Schumacher, T. N., and Kruisbeek, A. M. Tracing and characterization 44. Heid, C. A., Stevens, J., Livak, K. J., and Williams, P. M. Real time of the low-avidity self-specific T cell repertoire. Eur. J. Immunol., 30: quantitative PCR. Genome Res., 6: 986–994, 1996. 1458–1468, 2000. 45. Kruse, N., Pette, M., Toyka, K., and Rieckmann, P. Quantification 59. Miller, J. F., and Morahan, G. Peripheral T cell tolerance. Annu. of cytokine mRNA expression by RT PCR in samples of previously Rev. Immunol., 10: 51–69, 1992. frozen blood. J. Immunol. Methods, 210: 195–203, 1997. 60. Targoni, O. S., and Lehmann, P. V. Endogenous myelin basic 46. Kammula, U. S., Lee, K. H., Riker, A. I., Wang, E., Ohnmacht, protein inactivates the high avidity T cell repertoire. J. Exp. Med., 187: G. A., Rosenberg, S. A., and Marincola, F. M. Functional analysis of 2055–2063, 1998. antigen-specific T lymphocytes by serial measurement of gene expres- 61. Ranheim, E. A., and Kipps, T. J. Activated T cells induce expression in peripheral blood mononuclear cells and tumor specimens. J. Im- sion of B7/BB1 on normal or leukemic B cells through a CD40- munol., 163: 6867–6875, 1999. dependent signal. J. Exp. Med., 177: 925–935, 1993. 47. Schultze, J. L., Michalak, S., Seamon, M. J., Dranoff, G., Jung, K., 62. Yellin, M. J., Sinning, J., Covey, L. R., Sherman, W., Lee, J. J., Daley, J., Delgado, J. C., Gribben, J. G., and Nadler, L. M. CD40- Glickman-Nir, E., Sippel, K. C., Rogers, J., Cleary, A. M., Parker, M., activated human B cells: an alternative source of highly efficient antigen presenting cells to generate autologous antigen-specific T cells for et al. T lymphocyte T cell-B cell-activating molecule/CD40-L mole- adoptive immunotherapy. J. Clin. Investig., 100: 2757–2765, 1997. cules induce normal B cells or chronic lymphocytic leukemia B cells to express CD80 (B7/BB-1) and enhance their costimulatory activity. 48. Jiang, Y. Z., Mavroudis, D., Dermime, S., Hensel, N., Couriel, D., J. Immunol., 153: 666–674, 1994. Molldrem, J., and Barrett, A. J. Alloreactive CD4ϩ T lymphocytes can exert cytotoxicity to chronic myeloid leukaemia cells processing and 63. Van den Hove, L. E., Van Gool, S. W., Vandenberghe, P., Bakkus, presenting exogenous antigen. Br. J. Haematol., 93: 606–612, 1996. M., Thielemans, K., Boogaerts, M. A., and Ceuppens, J. L. CD40 triggering of chronic lymphocytic leukemia B cells results in efficient 49. Rezvani, K., Grube, M., Brenchley, J. M., Sconocchia, G., Fuji- alloantigen presentation and cytotoxic T lymphocyte induction by up- wara, H., Price, D. A., Gostick, E., Yamada, K., Melenhorst, J., Childs, R., Hensel, N., Douek, D. C., and Barrett, A. J. Functional leukemia- regulation of CD80 and CD86 costimulatory molecules. Leukemia (Bal- associated antigen-specific memory CD8ϩ T cells exist in healthy timore), 11: 572–580, 1997. individuals and in patients with chronic myelogenous leukemia before 64. Ginaldi, L., De Martinis, M., Matutes, E., Farahat, N., Morilla, R., and after stem cell transplantation. Blood, 102: 2892–2900, 2003. and Catovsky, D. Levels of expression of CD19 and CD20 in chronic B 50. Molldrem, J. J., Lee, P. P., Kant, S., Wieder, E., Jiang, W., Lu, S., cell leukaemias. J. Clin. Pathol., 51: 364–369, 1998. Wang, C., and Davis, M. M. Chronic myelogenous leukemia shapes host 65. Cabezudo, E., Carrara, P., Morilla, R., and Matutes, E. Quantitative immunity by selective deletion of high-avidity leukemia-specific T analysis of CD79b, CD5 and CD19 in mature B-cell lymphoprolifera- cells. J. Clin. Investig., 111: 639–647, 2003. tive disorders. Haematologica, 84: 413–418, 1999.

Downloaded from clincancerres.aacrjournals.org on September 26, 2021. © 2004 American Association for Cancer Research. Autoreactive, Cytotoxic T Lymphocytes Specific for Peptides Derived from Normal B-Cell Differentiation Antigens in Healthy Individuals and Patients with B-Cell Malignancies

Matthias Grube, Katayoun Rezvani, Adrian Wiestner, et al.

Clin Cancer Res 2004;10:1047-1056.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/10/3/1047

Cited articles This article cites 62 articles, 28 of which you can access for free at: http://clincancerres.aacrjournals.org/content/10/3/1047.full#ref-list-1

Citing articles This article has been cited by 5 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/10/3/1047.full#related-urls

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/10/3/1047. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.