Defining MHC class II T helper epitopes for WT1 tumor antigen

Abstract

The product of Wilms‘ tumor gene 1 (WT1) is overexpressed in diverse human tumors, including leukemia, lung and breast cancer, and is often recognized by antibodies in the sera of patients with leukemia. Since WT1 encodes MHC class I-restricted peptides recognized by cytotoxic T lymphocytes (CTL), WT1 has been considered as a promising tumor-associated antigen (TAA) for developing anticancer immunotherapy. In order to carry out an effective peptide-based cancer immunotherapy, MHC class II-restricted epitope peptides that elicit anti-tumor CD4⁺ helper T lymphocytes (HTL) will be needed. In this study, we analyzed HTL responses against WT1 antigen using HTL lines elicited by in vitro immunization of human lymphocytes with synthetic peptides predicted to serve as HTL epitopes derived from the sequence of WT1. Two peptides, WT1_124–138 and WT1_247–261, were shown to induce peptide-specific HTL, which were restricted by frequently expressed HLA class II alleles. Here, we also demonstrate that both peptides-reactive HTL lines were capable of recognizing naturally processed antigens presented by dendritic cells pulsed with tumor lysates or directly by WT1+ tumor cells that express MHC class II molecules. Interestingly, the two WT1 HTL epitopes described here are closely situated to known MHC class I-restricted CTL epitopes, raising the possibility of stimulating CTL and HTL responses using a relatively small synthetic peptide vaccine. Because HTL responses to TAA are known to be important for promoting long-lasting anti-tumor CTL responses, the newly described WT1 T-helper epitopes could provide a useful tool for designing powerful vaccines against WT1-expressing tumors.

Introduction

In the past decade, numerous tumor-associated antigens (TAA) derived from melanoma and various types of epithelial solid tumors have been described [1]. Identification of peptide epitopes of TAA recognized by tumor-specific CD8⁺ cytotoxic T lymphocytes (CTL) and CD4⁺ helper T lymphocytes (HTL) has led to the development of T-cell-based immunotherapy for peptide vaccines to treat and prevent relapsing cancers. The Wilms’ tumor gene 1 (WT1) product seems to be an ideal TAA since many tumor types including leukemias, lung, breast, ovary, thyroid, and colorectal cancers overexpress the WT1 protein and in some cases this expression predicts poor prognosis [2–8]. Furthermore, WT1 is only found in low levels in normal tissues such as kidney, splenic capsule, testis, and granulose cells of the ovary [9]. Recently, CTL responses specific for WT1 restricted by human leukocyte antigen (HLA),namely, HLA-A0201 and HLA-A2402 were reported [10–13] such studies have dealt with in-vitro-induced CTL killing of WT1+ leukemia cells or lung cancer cells . Moreover, in an in vivo mouse model, adoptively transferred human WT1-specific CTL were shown to inhibit growth of subcutaneously inoculated leukemia cells [14]. On the basis of these findings, a phase I clinical study of cancer peptide vaccine trial targeting WT1 in patients with leukemia, myelodysplasia (MDS) and lung and breast cancer was recently performed [15]. The study presented some anecdotal immunological responses showing HLA-A24-restricted CTL responses in vaccinated patients accompanied by tumor regression and reduction of serum carcino embryonic antigen (CEA) levels in the absence of autoimmunity. In view of these findings, WT1 appears to be a promising candidate for cancer immunotherapy against WT1+ cancers.

Because it is now generally accepted that tumor-specific HTL, in addition to CTL, play an essential role in tumor rejection in both animal models and humans [16, 17], identifying MHC class II-restricted HTL epitopes from various TAA including WT1 is an important task for improving the T-cell-based cancer immunotherapy. Because it has been observed that anti-WT1 IgG antibodies are present in some patients with leukemia, it is highly likely that CD4 HTL responses (which are necessary for IgG class switching) may already exist in some cancer patients [18, 19]. Moreover, it has been shown that two peptide sequences derived from WT1, peptide WT1_12e (modified peptide sequence derived from chromosomal jumping) and WT1_331–345 were able to induce peptide-specific HTL lines from DR4+ donors and some of the peptide-reactive HTL could recognize CML cells directly [20, 21].

In the present study, we have studied two WT1-peptide epitopes (WT1_124–138 and WT1_247–261) that are recognized by HTL in the context of the HLA-DR53 allele, which is expressed by a large proportion of the population. These WT1 HTL epitopes could be used in combination with the previously described CTL epitopes to increase the immunological responses of WT1-peptide vaccine than using only MHC class I-restricted CTL epitope peptide in clinical setting.

Methods

Cell lines

Epstein Barr Virus (EBV)-lymphoblastoid cell lines (EBV-LCL) were produced from PBMC of HLA-typed volunteers using culture supernatant from EBV-producing B95-8 cell lines. Mouse fibroblast cell lines (L-cells) transfected and expressing individual human MHC class II molecules were kindly provided by Dr. Robert W. Karr (Pfizer Global R&D, New London, CT, USA) and Dr. Takehiko Sasazuki (Tokyo, Japan). The myeloid leukemia cell lines K562, KU812, the Jurkat T-cell lymphoma, and the prostate tumor cells LNCaP and PC3 were all obtained from American Type Culture Collection (Manassas, VA, USA) and were maintained in tissue culture as recommended by the supplier. The gastric cancer cell line MKN45 was supplied by the Japanese Cancer Research Resources Bank (Tokyo, Japan). The myeloid leukemia cell lines KT1 and OUN1 [12] were kindly provided by Dr. Masaki Yasukawa (Ehime University, Ehime, Japan). The melanoma cell line 624mel was kindly provided by Dr. Steven A. Rosenberg (National Cancer Institute, NIH, Bethesda, MD, USA).

Synthetic peptides

Potential HLA-DR promiscuous CD4⁺ T-cell epitopes were selected from the amino acid sequence of the WT1 using the algorithm tables for three HLA-DR alleles (DRB1*0101, DRB1*0401, and DRB1*0701) described by Southwood et al. [22]. Peptides that displayed high algorithm scores were synthesized and purified as described. The purity (>95%) and identity of peptides were determined by HPLC and mass spectrometry, respectively.

In Vitro induction of antigen-specific HTL lines with synthetic peptides

The procedure selected for the generation of tumor antigen-reactive HTL by use of peptide-stimulated peripheral blood mononuclear cells (PBMC) has been described in detail elsewhere [23]. Briefly, dendritic cells (DC) were produced in cell culture from adherent monocytes that were cultured for 7 days at 37°C in a humidified CO₂(5%) incubator in the presence of 50 ng/ml Granulocyte Macrophage-Colony Stimulating Factor (GM-CSF) and 1000 IU/ml IL-4. Peptide-pulsed DC (3 μg/ml for 2 hr at room temperature) were irradiated (4,200 rad) and cocultured with autologous purified CD4⁺ T cells (using antibody-coated magnetic microbeads from Miltenyi Biotech, Auburn CA, USA) in 96 round-bottom-well culture plates. After 7 days, the CD4⁺ T cells were restimulated with peptide-pulsed irradiated autologous PBMC and 2 days later, human rIL-2 was added at a final concentration of 10 IU/ml. One week later, the T cells were tested for their cytokine release to peptide as described below. Those cultures exhibiting an antigen-specific lymphokine production to peptide (at least 2.5-fold over the background) were expanded in 24- or 48-well plates by weekly restimulation with peptides and irradiated autologous PBMC. Complete culture medium for all procedures consisted of AIM-V (Invitrogen/GIBCO) medium supplemented with 3% human male AB serum (Sigma). Informed consent for blood donation was obtained from all volunteers before blood sampling.

Measurement of antigen-specific responses

CD4⁺ T cells (3×10⁴ /well) were mixed with irradiated antigen presenting cells (APC) in the presence of various formulations of antigen (peptides, tumor lysates, or γ-irradiated tumor cells), in 96-well culture plates. APC consisted of either PBMC (1×10⁵ /well), HLA-DR-expressing L-cells (3×10⁴ /well), DC (5×10³ /well), or tumor cells (3×10⁴ /well). Tumor cell lines, except EBV-LCL, were treated with the interferoon gamma (IFN-γ (500 U/ml for 48 h) to induce MHC antigen expression before assay. The expression of HLA-DR molecules on tumor cells was evaluated by flow cytometry using anti-HLA-DR (L243) monoclonal antibody (mAb) conjugated with fluorescein isothiocyanate. Tumor cell lysates were prepared by three freeze-thaw cycles of 1×10⁸ tumor cells resuspended in 1 ml of serum-free RPMI-1640 medium. Lysates were used as a source of antigen at 5×10⁵ cell equivalents per ml. Culture supernatants were collected after 48 h for measuring antigen-induced lymphokine (IFN-γ or GM-CSF) production by the HTL using ELISA kits (BD Pharmingen, San Diego, CA, USA). To determine MHC restriction molecules involved in antigen presentation, blocking of the antigen-induced proliferative response was investigated by adding anti-HLA-DR mAb L243 (IgG2a, prepared from supernatants of the hybridoma HB-55 obtained from the ATCC) or anti-HLA-A, B, C mAb W6/32 (IgG2a, ATCC). All antibodies were used at a final concentration of 10 μg/ml throughout the 48-h incubation period. All assessments of ELISA were carried out at least in triplicate and results correspond to the mean values with the standard deviation (SD) of the mean.

Reverse transcription (RT)-PCR analysis

The presence of WT1 transcripts in tumor cells was assessed by RT-PCR [10]. A set of specific primers for WT1 was selected that spans several exon sequences, allowing the distinction of DNA from RNA. The WT1 primers used were 5′-GGC ATC TGA GAC CAG TGA GAA-3′ (sense primer) and 5′-GAG AGT CAG ACT TGA AAG CAG T-3′ (antisense primer), which yielded a 482-bp product using cDNAs in PCR amplification samples by performed for 30 cycles that were derived from the purified RNA samples.

Western blot analysis

Five million tumor cells were washed in phospate buffered saline (PBS) and lysed in Laemmly buffer. The cell lysate was subjected to electrophoresis in a 4–12% NuPage bis-Tris SDS-PAGE gel (Invitrogen) under reducing conditions and then transferred to Immobilon-P (Millipore, Bedford, MA, USA) membrane. The membrane was then blocked in PBS containing 0.01% Tween 20 and 5% nonfat dry milk for 1 h at room temperature and incubated first with mouse anti-human WT1 F-6 mAb (1:200 in blocker; Santa Cruz Biotech, Santa Cruz, CA, USA) overnight at 4°C and then with mouse anti-actin mAb (1:500 in blocker, Santa Cruz, CA, USA) for 30 min at room temperature. After washing, the membrane was incubated with horseradish peroxidase-labeled goat anti-mouse IgG and subjected to the enhanced chemiluminescence assay using ECL detection system (Amersham).

Results

Prediction and selection of potential HTL epitopes for WT1

Because our set goal was to identify potent MHC class II-restricted HTL epitopes for WT1, we first examined the amino acid sequence of this TAA for the presence of peptides containing binding motifs for HLA-DR1, -DR4, and -DR7 using the computer-based MHC-peptide binding algorithm developed by Southwood and collaborators [22]. This approach has been very successful in our hands, allowing the identification of several significant HTL epitopes from various TAA such as tissue-specific, tumor marker, melanoma antigen, or viral protein including HER2/ neu, gp100, MAGE-A3, EBNA2, HTLV-1 envelope protein, CEA, TARP, and PSMA [23–31]. Interestingly, in some circumstances T-cell responses induced by peptides predicted by this approach were restricted by MHC class II alleles other than HLA-DR1, -DR4 and -DR7, such as -DR9, -DR15, -DR16, -DR52, -DR53, -DQ2, and -DQ6. For the present analysis, a total of 13 peptide sequences from WT1 were identified as probable binders to HLA-DR1, -DR4, and -DR7, and thus were potentially promiscuous MHC class II-restricted T cell epitopes (data not shown). When examining the position that these peptides occupy within the WT1 molecules, it became evident that several of these peptides were proximal to or overlapped with recently described MHC class I-restricted CTL epitopes. Specifically, WT1_247–261 is proximal to the HLA-A24-restricted CTL epitope, WT1_235–243 and WT1_124–138 overlaps with the HLA-A2-restricted CTL epitope, WT1_126–134[10–13]. Since HTL play an essential role in the generation and maintenance of CTL anti-tumor responses, the design of effective vaccines should contain peptide epitopes to stimulate both CTL and HTL [32]. Thus, ideally one could use as a vaccine a single polypeptide containing epitopes for both CTL and HTL, which are naturally proximal to each other. Therefore, we selected the two candidate peptides WT1_124–138 and WT1_247–261 to determine their ability to trigger HTL responses in vitro using PBMC from normal volunteers.

Assessment of the expression of WT1 in tumor cells

Next, we assessed the expression levels of WT1 in various tumor cell lines to be used as targets for HTL recognition assays. As shown in Fig. 1a, RT-PCR assays demonstrated that several myeloid leukemia cell lines, the Jurkat T-cell lymphoma, several EBV-LCL, a gastric cancer cell line (MKN45), and some prostate tumor cell lines (LNCaP, PC3) expressed WT1 mRNA. On the other hand, the melanoma cell line 624mel did not appear to express WT1 mRNA, allowing this cell line to serve as a negative control. These results were confirmed by Western blot analysis where the WT1 protein was found to be expressed by all the tumor cell lines except for the 624mel melanoma (Fig. 1b). These results suggest that not only leukemia cell lines but also EBV-LCL and prostate tumor cell lines can be used as targets for recognition assay by WT1-peptide-reactive HTL.

Fig. 1.

Expression of WT1 in human tumor cell lines. a RT-PCR using WT1-specific primers was performed to analyze the production of WT1 mRNA. The amplified WT1 product is 482-bp long. All cell lines except melanoma 624-mel cells tested were WT1-positive. Glyceraldehyde 3-phosphate dehydrogenase GAPDH was used as the housekeeping gene control for RT-PCR and the product is 1.2 kbp long. b Western blot analysis was performed using a WT1-specific mAb performed as described in “Methods”. WT1 protein could be detected in indicated human tumor cell lines except melanoma cell line 624-mel. The expression of the housekeeping actin protein was used as a control. The WT1 protein measures approximately 52 kDa and the actin protein approximately 42 kDa.

T-cell responses to peptide epitopes from WT1

The two selected peptide sequences derived from WT1 as described above were synthesized and then tested for their ability to stimulate CD4⁺ T cells obtained from three healthy, MHC-typed individuals in primary in vitro cultures using peptide-pulsed autologous DC as APC. Both peptides were effective in inducing proliferative T-cell responses in at least three blood donors. In the case of the HLA-DR4, -DR9, -DR53 individual, peptide WT1_247–261 was able to stimulate a vigorous antigen-specific T-cell response (Fig. 2). Moreover, this response was shown to be restricted by HLA-DR53 molecules by the capacity of DR53-transfected L-cells to present peptide to the T cells and by the ability of anti-DR mAb, but not anti-class I mAb, to block the T-cell response (Figs. 2a and 2b). Furthermore, as shown in Fig. 2C, the HTL responded to peptide WT1_247–261 presented by autologous PBMC in a dose-dependent manner, where the half-maximal stimulation dose of peptide WT1_247–261 was ∼1μg/ml.

Fig. 2.

HLA-DR53-restricted T-cell responses to the peptide WT1_247–261. An HTL line was selected from an HLA-DR4, -DR9, -DR53 normal individual by weekly stimulation of CD4⁺ T cells with peptide as described in “Methods”. a When autologous PBMC were used as APC, the T-cell responses to the peptide WT1_247–261 were inhibited by anti-HLA-DR mAb (L243 at 10 μg/ml) but not by anti-HLA class I mAb (W6/32 at 10 μg/ml). b When mouse L-cell fibroblasts transfected with HLA-DR genes were used as APC, the HTL recognized the peptide WT1_247–261 in the context of HLA-DR53 molecules. c Peptide dose-response curve was performed to estimate the overall avidity of the HTL for peptide WT1_247–261 when autologous PBMC were used as APC. Values shown are triplicate determinations; error bars, SD. Columns and symbols without error bars had SD<10% the value of the mean.

Two separate T-cell lines reactive with peptide WT1_124–138 isolated from two normal individuals were studied in detail for their antigen specificity and MHC restriction pattern. The data presented in Figs. 3a, 3c indicate that both of the T-cell lines recognized this peptide presented by autologous APC and that antibodies to HLA-DR but not to HLA class I inhibited this reactivity. Interestingly, studies performed to determine the HLA-restriction elements for these T-cell lines indicated that in one case (the HLA-DR9, -DR14, -DR53 individual), the response to peptide WT1_124–138 was restricted by HLA-DR53 (Fig. 3b), whereas in the other case (HLA-DR1, -DR15 individual), the response to peptide WT1_124–138 was restricted by HLA-DR15 (Fig. 3d). The peptide dose–response curves revealed that the affinity of the DR15-restricted HTL for antigen (Fig. 4b) was ∼1000-fold higher than the DR53-restricted HTL (Fig. 4a).

Fig. 3.

HTL responses to the peptide WT1_124–138. HTL lines were selected from an HLA-DR9, -DR14, -DR53 normal individual (a, b) and from a HLA-DR1, -DR15 normal individual (c, d) by weekly stimulation of CD4⁺ T cells with peptide as described in “Methods”. When autologous PBMC were used as APC, the HTL responses to the peptide WT1_124–138 were inhibited by anti-HLA-DR mAb L243 at 10 μg/ml but not anti-HLA class I mAb W6/32 at 10 μg/ml (a, c). When mouse L-cell fibroblasts transfected with HLA-DR genes were used as APC, (b) one of the HTL recognized the peptide WT1_124–138 in the context of HLA-DR53 molecules while the other HTL recognized the peptide in the context of HLA-DR15 (d). Values shown are triplicate determinations; error bars, SD. Columns and symbols without error bars had SD<10% the value of the mean

Fig. 4.

Peptide dose-response curves were performed to estimate the overall avidity of the HLA-DR53- (a) and the HLA-DR15-restricted (B) HTL for the peptide WT1_124–138 with autologous PBMC used as APC. Values shown are triplicate determinations; error bars, SD. Symbols without error bars had SD<10% the value of the mean

Recognition of naturally processed WT1 antigens on tumor MHC Class II molecules by WT1-peptide-reactive HTL

The data presented above demonstrated that the peptides WT1_124–138 and WT1_247–261 were indeed capable of inducing CD4⁺ T-cell responses. However, in order to assume these TAA, peptide-epitopes will be useful for the design of cancer immunotherapy, these peptides need to be processed endogenously or exogenously by APC and/or WT1+ tumor cells that express MHC class II molecules. Thus, we examined the ability of peptide-induced HTL to directly recognize tumor cells that express both WT1 protein and cell surface HLA-DR molecules. We assumed that EBV-LCL should be good targets for recognition assays by peptide-reactive HTL because these cells express the WT1 protein and high levels of MHC class II molecules (without requiring IFN-γ treatment to enhance surface MHC expression). Therefore, we first used EBV-LCL as targets to assess direct recognition by the peptide-reactive HTL lines.

As shown in Fig. 5a, the HLA-DR53-restricted WT1_124–138-specific HTL was as effective in recognizing the autologous EBV-LCL (LCL-TN) as compared to the peptide pulsed (10 μg/ml) autologous PBMC. Moreover, this HTL line also was capable of reacting with another DR53-positive EBV-LCL (Ky, Wa), which expresses WT1, but failed to recognize DR53-negative lines, EBV-LCL (Sa), and Jurkat (Fig. 5b). Most importantly, the recognition of WT1-positive tumor cells by the HTL line was inhibited by anti-HLA-DR antibodies (Fig. 5b), confirming that this interaction requires the presentation of peptide by MHC class II molecules.

Fig. 5.

Anti-tumor reactivity of the HLA-DR53-restricted, peptide WT1_124–138- and WT1_247–261-specific HTL lines. a Peptide WT1_124–138 and c WT1_247–261-specific HTL (3×10⁴ /well) were tested at various numbers of APC for their capacity to recognize the WT1-positive autologous EBV-LCL, peptide-pulsed (10 μg/ml) autologous PBMC, and autologous PBMC (W/O peptide). The peptide-reactive HTL lines secreted IFN-γ as the result of recognizing antigen in the HTL-DR53⁺ EBV-LCL presenting naturally processed epitope. Production of IFN-γ was inhibited by anti-DR antibodies and could not be detected using DR53-negative EBV-LCL (EBV-Sa) or DR-negative Jurkat T cell lymphoma (b, d). Values shown are triplicate determinations; error bars, SD. Symbols without error bars had SD<10% the value of the mean

In similar experiments, the HLA-DR53-restricted, specific for peptide WT1_247–261 HTL also had the capacity to produce lymphokine when stimulated directly by WT1 positive, HLA-DR53 positive EBC-LCL, and this recognition was inhibited by anti-HLA-DR antibodies (Figs. 5c, 5d). These results indicate that the WT1-peptide epitopes associate with HLA-DR53 molecules of the WT1-positive tumor cells through endogenous processing.

We next attempted to assess whether the peptide-specific HTL had the capacity of recognizing other WT1+ tumor cells such as leukemia cells or prostate carcinomas. To carry out these experiments, tumor cells were cultured for 48 hr prior to the assay with IFN-γ to enhance the expression of MHC class II molecules on the cell surface. As shown in Fig. 6a, we were able to induce HLA-DR expression in KU812 leukemia cells and PC3 prostate tumor cells, both bearing the DR53 allele. We observed that both HLA-DR53-restricted WT1_124–138 and WT1_247–261-specific HTL were capable of recognizing the KU812 and PC3 tumor cells directly but nor WT1+, HLA-DR negative KT1 leukemia cells and WT1-negative, DR53+ 624 melanoma cells (Figs. 6b, 6c). Furthermore, the direct recognition of the tumor cells by both HTL lines was inhibited by anti-HLA-DR mAb but not by anti-class I mAb. These results indicate that WT1-expressing tumor cells are capable of directly presenting the T-cell epitopes when HLA-DR molecules are present on the cell surface.

Fig. 6.

Direct recognition of WT1 on DR+ tumor cells by HTL. a Cell surface expression of HLA-DR molecules on tumor cells. Tumor cell lines (KT1, KU812 and PC3) were cultured with IFN-γ (500 U/ml) for 48 hr to enhance MHC antigen expression. The expression of HLA-DR molecules on tumor cells was evaluated by flow cytometry using anti-HLA-DR (L243) mAb conjugated with fluorescein isothiocyanate (grey line open histograms). The staining with the isotype negative control is shown in the black-filled histograms. The HLA-DR53-restricted, WT1_124–138(b) and WT1_247–261-reactive HTL (c) lines produced IFN-γ as the results of recognizing antigen on HLA-DR53+, WT1+ KU812, or PC3 tumor cells presenting naturally processed epitopes. Values shown are triplicate determinations; error bars, SD. Columns and symbols without error bars had SD<10% the value of the mean

Recognition of naturally processed WT1 antigen by professional APC

We proceeded to assess whether APC would be able to take up the WT1 antigen derived from dead tumor cells (i.e., freeze/thaw cell lysates) expressing WT1 and appropriately process this antigen to stimulate WT1-peptide-specific HTL. The results presented in Figure 7 demonstrate that the both DR53-restricted WT1-peptide-reactive HTL were efficient in recognizing tumor cell lysates from WT1-positive tumors (K562, EBV-Ky, LNCaP, and PC3) but not control tumor lysates (624mel), when presented by autologous DC. In addition, the capacity of the HTL to recognize naturally processed WT1 antigen in the form of tumor lysates (presented by DC) was effectively blocked by anti-HLA-DR antibodies confirming that the epitopes were presented via MHC class II molecules. In summary, the overall results indicate that the T-cell epitopes represented by peptide WT1_124–138 and WT1_247–261 are not only processed by APC from tumor cell lysates but are also expressed on MHC class II molecules on WT1- positive tumor cells.

Fig. 7.

Recognition of naturally processed antigen by the HLA-DR53-restricted WT1_124–138-specific HTL line (a) and the HLA-DR53-restricted, WT1_247–261-specific HTL line (b). When autologous DC were used as APC, the peptide-reactive HTL were able to respond to cell lysates derived from various tumor cells expressing WT1 but not to WT1-negative 624 mel melanoma. The recognition of tumor lysate via autologous DC was inhibited by anti-HLA-DR mAb L243 at 10 μg/ml. Values shown are triplicate determinations; error bars, SD. Columns and symbols without error bars had SD<10% the value of the mean

Lastly, a similar analysis was performed using the peptide WT1_124–138-reactive, DR15-restricted HTL lines, but in this case, T-cell responses could not be detected by recognition assay using exogenously or endogenously processed tumor antigen (data not shown). Thus, it is possible that this epitope may not be properly produced (processed) and presented by HLA-DR15 molecules on the cell surface of APC.

Discussion

The identification of TAA recognized by CTL at the molecular level has formed the basis for the design of T-cell based immunotherapeutic cancer vaccines. Actually, a growing number of phase I and II clinical trials, which are mainly targeted at melanoma antigens, are being performed and reported during the last decade [33]. WT1 seems to be one of the most ideal TAA for cancer immunotherapy since there is accumulating evidence that WT1 is overexpressed in most leukemias and various epithelial solid tumors [2–6]. Thus, several groups have been identified the potent immunogenic peptides of WT1 for HLA-A2 and HLA-A24 that could elicit efficient CTL that exhibited cytotoxic activity against WT1-positive tumor cells in vitro [10–13]. Recently, Sugiyama et al. have performed a phase I clinical study using HLA-A24-restricted WT1 9-mer peptide WT1_235–243 in patients with leukemia, MDS, and lung cancer and breast cancer and they demonstrated that WT1-peptide vaccine could reduce tumor size or decrease serum level of tumor markers (CEA) in some patients [34]. Although CTL are the most effective immune elements for eradicating tumor cells, HTL also play a critical role in the generation and maintenance of CTL anti-tumor responses. Therefore, we believe that a vaccine should contain not only CTL epitopes but also HTL epitopes to develop an effective anti-tumor immune response. Actually, Pawelec and collaborator were the first group to induce T-helper cells specific for WT1 epitopes [20, 21]. Although the restriction elements by their HTL lines remained somewhat unclear but appeared to be via HLA-DR4, the WT1-peptide-reactive HTL lines clearly recognized CML cells directly [20, 21]. In our study, we describe two strong HTL epitopes for WT1 (WT1_124–138 and WT1_247–261), which were restricted by HLA-DR53 (DRB4*0101) molecules. It is interesting to note that the HTL epitope described by Pawelec et al [20, 21] may be presented by HLA-DR4 and -DR53 raising the possibility that this epitope represents a true promiscuous MHC class II binder. The HLA-DR53 molecule is expressed by a large proportion of the population (∼50%) owing to its linked co-expression with the HLA-DR4 (23.6% Caucasian, 6.1% Black, 40.4% Japanese, 29.8% Hispanic), HLA-DR7 (26.2% Caucasian, 11.9% Black, 1.0% Japanese, 16.6% Hispanic), and HLA-DR9 (3.6% Caucasian, 4.7% Black, 24.5% Japanese, 6.7% Hispanic) alleles [22]. Other peptide epitopes presented by HLA-DR53 have been described for various TAA such as gp100, Her2/neu, CEA, PSMA, and NY-ESO-1, which are widely expressed in melanoma, prostate, and epithelial solid tumors [23, 26, 27, 29]. Therefore, we believe that HLA-DR53-restricted TAA peptides would be quite useful for designing peptide-based vaccines for patients with various tumors. Furthermore, it is significant that the two HTL epitopes, WT1_124–138 and WT1_247–261 are found in close proximity to two CTL epitopes: WT1_126–134 (HLA-A2-restricted) and WT1_235–243 (HLA-A24-restricted). Thus, it would be possible to stimulate both CTL and HTL, utilizing a relatively short peptide.

Although the WT1 gene has been originally isolated as a gene responsible for Wilms’ tumor, it has been reported that WT1 gene acts on oncogenesis in various cancers and highly expressed in both leukemias and solid tumors [35–37]. Interestingly, we observed that expression of WT1 have been found in not only leukemia cell lines but also EBV-LCL examined by both RT-PCR and Western blots. Previously, Ragona et al. also detected WT1-specific transcripts by RT-PCR in EBV-LCL but protein expression was not assessed [38, 39]. Our results indicated that peptide WT1_124–138 and WT1_247–261-reactive HTL had the ability of directly recognizing EBV-LCL, indicating that WT1-derived peptides are presented by the endogenous MHC class II molecules of EBV-LCL. Recently, O’Reily et al. indicated that WT1-peptide-loaded EBV-LCL could significantly elicit WT1-peptide-specific T cells in vitro rather than using DC as APC and the frequencies of the generated WT1-peptide-specific T cells were almost comparable with those generated against highly immunogenic EBV peptides such as EBNA3C, BZLF-1, BMLF1, and LMP2 [14]. This information would support our results that EBV-LCL efficiently enable the WT1-peptide-specific HTL to stimulate without exogenously adding the peptide. Thus, our described WT1 HTL epitopes could be effective at eliciting potent T-helper responses against EBV-associated malignancies.

Direct evidence that leukemia cells present bcr-abl fusion TAA peptides bound to MHC molecules on leukemia cells has been shown by mass-spectrometric analysis of endogenous peptides eluted from MHC molecules of leukemia cells [40]. In the present study, we show using DR53-restricted HTL lines that peptides WT1_124–138 and WT1_247–261 are naturally associated with HLA-DR53 molecules in WT1-expressing tumor cells (PC3 and KU812 and EBV-LCL). Thus, these data indicates that some WT1-expressing tumor cells are capable of processing and presenting the antigenic peptides on their own MHC class II molecules.

Although WT1 has also found to be expressed on various epithelial tumors, including lung, colon and breast cancers, these non-hematopoietic tumors usually would express little MHC class II molecules in vivo unless these molecules become upregulated by the presence of local IFNγ. It is noteworthy that WT1-peptide-reactive HTL efficiently recognized the native antigen, given in the form of lysates from WT1-expressing tumor cells upon exogenously processing and presentation by autologous DC. These results suggest that tumor-specific HTL that recognize DC or monocytes that have taken up dead or apoptotic tumor cells could enhance CTL activity or exhibit direct effector function toward the tumor cells.