• Open Access

Identification of promiscuous HLA-DR-restricted CD4+ T-cell epitopes on the cancer-testis antigen HCA587

Authors


To whom correspondence should be addressed.
E-mail: yinyanhui@bjmu.edu.cn

Abstract

The cancer testis antigen HCA587 is an attractive candidate for T cell-based immunotherapy because it is overexpressed in a wide spectrum of malignant tumors but not normal tissues, except testis. Several CTL epitopes derived from HCA587 have been described. Our aim was to identify helper T lymphocyte epitopes of HCA587 for the optimization of T cell-based immunotherapies against HCA587-expressing tumors. Candidate helper T lymphocyte epitopes for HCA587 were predicted using the SYFPEITHI algorithm and were tested for their ability to induce helper T lymphocyte responses by in vitro peptide vaccination of CD4+ T lymphocytes from healthy individuals and hepatocellular carcinoma patients. Four CD4+ T-cell epitopes for HCA587 (p43–57, p145–159, p186–200 and p249–263) were identified. Among them, the p43–57 epitope was shown to be naturally processed and presented by HCA587-expressing tumor cells as well as autologous dendritic cells pulsed with whole-protein HCA587. Notably, this epitope behaved as a promiscuous T-cell epitope as it stimulated T cells in the context of more than one HLA class II allele. Thus, p43–57 is the first HCA587-derived major histocompatibility complex class II-restricted epitope to fulfil all prerequisites for use as a peptide vaccine in patients with HCA587-expressing tumors. (Cancer Sci 2011; 102: 1455–1461)

The importance of T cell-mediated antitumor immunity has been demonstrated in both animal models and human cancer therapy.(1,2) Activation of antitumor T-cell immunity relies on the recognition of tumor-associated antigens that bear immunogenic T-cell epitopes expressed on tumor cells. Most attempts have focused on the activation of tumor-specific cytotoxic T lymphocytes (CTL) as they can directly kill tumor cells. Human clinical trials using molecularly defined major histocompatibility complex (MHC) class I-restricted tumor antigens have indicated that antigen-specific T-cell responses are readily detected in patients after vaccination, but overall immune responses are weak and transient.(3–5) One of the possible explanations is that these immunotherapies have ignored the role that CD4+ T helper lymphocytes play in the generation and persistence of CD8+ T-cell responses.(6,7)

Increasing evidence from both human and animal studies indicates that CD4+ T cells play a central role in initiating and maintaining host immune responses against cancer.(8–10) Even in the absence of CTL, tumor regression can be mediated by direct and indirect killing mechanisms.(11–13) Thus, an optimal vaccination might require the participation of both CD4+ and CD8+ T cells to generate a strong and long-lasting antitumor response.

Until now, many tumor-specific MHC class I-restricted epitopes have been identified; however, only a very small number of tumor-specific helper T-cell epitopes have been characterized.(14) The lack of MHC class II-restricted helper T epitopes is a major hurdle in the use of antigen-specific CD4+ T cells in cancer vaccines. Thus, identification of MHC class II-restricted helper T peptides is important for the development of effective cancer vaccine.

HCA587 was cloned by serologic analysis of recombinant cDNA expression libraries (SEREX) from the hepatocellular carcinoma-derived cDNA libraries by our group.(15) The sequence of the HCA587 gene was identical with that of MAGE-C216. HCA587 is a cancer-testis antigen with expression in normal tissues limited to the testis, but shows a high level of expression in various types of tumors.(16,17) Our previous study demonstrated that specific CD4+ and CD8+ T-cell responses could be readily induced by in vitro stimulation of peripheral blood mononuclear cells (PBMC) with recombinant HCA587 protein.(18) Lurquin et al. recently reported that a melanoma patient who showed tumor regression after vaccination with MAGE-3.A1 and MAGE-1.A1 peptides has a high number of antitumor CTL in the blood and metastases, and remarkably, the most frequent antitumor CTL is directed against MAGE-C2 (HCA587) expressed on the tumor cells.(19,20) Because of its strong immunogenecity and frequent expression in a wide spectrum of malignant tumors, HCA587 is an attractive target for immunotherapy. To date, eight MHC class I-restricted epitopes from HCA587 have been described,(21–24) but no MHC class II-restricted epitopes have been identified. In the present study, we report the results of the analysis of the HCA587 antigen for MHC class II-restricted epitopes by a combined approach of computational identification of candidate T-cell epitopes, followed by biological validation analysis of the predicated sequences forming naturally processed epitopes recognized by CD4+ T cells.

We identified four HCA587-derived epitopes capable of inducing CD4+ T lymphocyte responses in a MHC class II-restricted fashion. Among them, the p43–57-reactive CD4+ T lymphocytes were effective in recognizing the naturally processed HCA587 protein on either the tumor cells directly or antigen-presenting cells (APC) pulsed with whole-protein HCA587. Our results indicate that p43–57 fulfils the criteria for a widely applicable HLA-DR-restricted peptide vaccine, and will be used in combination with previously described CTL epitopes to enhance the efficacy of vaccines directed against HCA587-expressing cancers.

Materials and Methods

T-cell epitope prediction and peptide synthesis.  The SYFPEITHI algorithm was used to predict HCA587-derived peptides with a high probability to bind to the six DRB1 subtypes (*0101, *0301, *0401, *0701, *1101 and *1501), and 15 pentadecamer peptides were selected for synthesis based on their ability to bind to at least two of these six HLA-DRB1 molecules (Table 1). The 15 peptides were grouped into two pools: HCA587 sequences comprising residues 43–57, 52–66, 145–159, 171–185, 186–200, 202–216, 227–241 and 249–263 formed pool 1 and 263–277, 283–297, 296–310, 315–329, 326–340, 353–367 and 359–373 formed pool 2. A mix of seven pentadecamers derived from pp65 of the human cytomegalovirus (CMV) (p32, p117, p243, p269, p299, p510 and p524) was used as a positive control.(25) Peptides were synthesized by GL Biochem Ltd (Shanghai, China). The purity (>85%) and identity of peptides were assessed by reverse-phase high-performance liquid chromatography and mass spectrometry, respectively. All peptides were dissolved completely in a mixture of water and dimethyl sulfoxide (DMSO).

Table 1.   Peptides derived from the HCA587 antigen selected for the present study
 SequencesBinding probability to HLA-DRB1
*0101*0301*0401*0701*1101*1501
  1. The scores of the concerned peptides obtained by epitope prediction using the SYFPEITHI database are indicated. Scores of 20 or above are considered as predicting a high probability of binding to the respective HLA-DR subtype.

Pool 1 residues
 P43–57S S T L Y L V F S P S S F S T291126161818
 P52–66P S S F S T S S S L I L G G P241822321122
 P145–159V A E L V E F L L L K Y E A E10231414726
 P171–185V I K Y K D Y F P V I L K R A251016181124
 P186–200R E F M E L L F G L A L I E V332120161224
 P202–216P D H F C V F A N T V G L T D111422161826
 P227–241S L L I I I L S V I F I K G N312120281320
 P249–263I W E V L N A V G V Y A G R E241320181424
Pool 2 residues
 P263–277E H F V Y G E P R E L L T K V162626161514
 P283–297Y L E Y R E V P H S S P P Y Y2632210304
 P296–310Y Y E F L W G P R A H S E S I181822102412
 P315–329L E F L A K L N N T V P S S F271620141918
 P326–340P S S F P S W Y K D A L K D V182416181822
 P353–367D A T V M A S E S L S V M S S23122030624
 P359–373S E S L S V M S S N V S F S E281220162018

Subjects and cell lines.  Peripheral blood mononuclear cells were obtained from 12 healthy subjects and four patients with hepatocellular carcinoma. All samples were collected with written consent and were approved by the Hospital Ethics Review Committee. The melanoma cell line SK-MEL-37 expressed HCA587 as determined by western blot (data not shown).

Epstein–Barr virus (EBV)-transformed lymphoblastoid cell lines (LCL) were produced from PBMC of HLA-typed healthy subjects as described previously with minor modification.(26) Briefly, PBMC (5 × 106) were incubated in 2.5 mL complete RPMI 1640 medium (Gibco, NY, USA) supplemented with 2.5 mL EBV containing B95-8 culture supernatant at 37°C and 5% CO2. Cells were supplemented with fresh complete RPMI 1640 if required. After 2–3 weeks, the EBV-transformed LCL appeared as clusters.

Generation of dentritic cells (DC).  Dendritic cells were generated from human PBMC as described previously with minor modifications.(27) Briefly, PBMC were obtained by Ficoll density gradient centrifugation, and monocytes were isolated by plastic adherence and cultured in medium supplemented with GM-CSF (50 ng/mL; R&D Systems, Minneapolis, MN, USA) and interleukin (IL)-4 (30 ng/mL; R&D Systems). Maturation was induced on day 6 by adding lipopolysaccharide (LPS, 1 μg/mL; Sigma, St Louis, MO, USA). On day 7, the immature and mature DC were characterized by flow cytometry using FITC-conjugated or phycoerythrin (PE)-conjugated anti-human monoclonal antibodies specific for CD14, HLA-DR, CD40, CD80, CD86, CD83 and isotype-matched controls. Stained cells were analyzed by FACSCalibur flow cytometer (BD Biosciences, San Jose, CA, USA) and CellQuest software (BD Biosciences). Mature DC had the phenotype CD14, CD83+, CD40++, CD80+++, CD86+++ and HLA-DR+++ (data not shown).

Generation of peptide-specific CD4+ T cells.  Synthetic peptides corresponding to pool 1 and pool 2 were used to stimulate the PBMC from the different donors. Briefly, part of the separated PBMC were suspended in serum-free RPMI 1640 (Gibco) supplemented with 2 mM l-glutamine, 100 U/mL penicillin and 50 μg/mL streptomycin and were divided into three fractions. Two of these (with equal numbers of PBMC) were pulsed with different peptide pools (2 μg/mL for each peptide) and the third fraction was pulsed with a mix of seven peptides derived from the pp65 antigen of the human CMV. Pulsing was performed for 120 min at room temperature. After washing, the pulsed cells were suspended in RPMI 1640 supplemented with 5% heat-inactivated human AB serum and were distributed at a density of 2 × 106 PBMC/mL into the wells of a 48-well plate. The cultured cells were restimulated weekly using peptide-pulsed and irradiated (30 Gy) autologous PBMC as APC. Twenty-four hours after each restimulation, 10 U/mL recombinant human IL-2 (R&D Systems) was added. After 3–4 cycles of stimulation, the ratio of CD4+ T was examined by FACS and the CD4+ T-cell responses were evaluated.

To generate peptide-specific long-term T-cell lines, PBMC were suspended in RPMI 1640 and cultured with individual peptides (10 μg/mL). The cells were repetitively restimulated at 7–14-day intervals with peptide-pulsed and irradiated (30 Gy) autologous PBMC as APC, and 10 U/mL recombinant human IL-2 was added 24 h later after each restimulation. After 4–6 rounds of restimulation, the CD4+ T-cell lines that produced peptide-specific interferon-gamma (IFN-γ) were selected and expanded.

Enzyme-linked immunospot (ELISPOT) assay.  IFN-γ ELISPOT assay was performed in nitrocellulose-bottomed 96-well plates (MAHAN 45; Millipore, Bedford, MA, USA). Wells were precoated with anti-IFN-γ capture antibody (clone 1-D1K) at the dilution recommended in the supplier’s instructions (Mabtech, Nacka, Sweden). After washing and blocking of the plates, CD4+ T cells were seeded in culture medium supplemented with irradiated autologous or HLA-DR-matched PBMC for 20 h incubation at 37°C. Plates were then washed, developed with chromogenic alkaline phosphatase substrate (5-bromo-4-chloro-3-indolylphosphateynitro-blue tetrazolium; Sigma, MO, USA) after a two-step incubation with biotin-labeled antihuman IFN-γ antibody (monoclonal antibody 7-B6-1-Biotin; 2 μg/mL; Mabtech) and streptavidin-alkaline phosphatase, and the number of spot-forming cells (SFC) per well was counted using the ELISPOT reader (IMMUNOSPOT ANALYZER; CTL Analyzers LLC, Cleveland, OH, USA).

CD4+ T cell stimulation assay.  The production of IFN-γ and IL-4 by peptide-specific CD4+ T cells was tested by ELISA kits (Biolegend, San Diego, CA, USA) following the supplier’s instructions. The detection limit for IFN-γ and IL-4 was 12.5 and 1.1 pg/mL, respectively.

Blocking of T-cell responses.  To prove the DR restriction of the respective T-cell responses, the T-cell receptor/MHC class II interaction was blocked using antibodies against HLA-DR (clone L243) or HLA-A,B,C (clone W6/32; both from Biolegend). Peptide-pulsed APC were incubated with the appropriate antibody (10 μg/mL) for 30 min on ice and then transferred to the test plate.

HLA-DR typing.  HLA-DR subtyping of PBMC and LCL was performed by high-resolution sequence-based typing (SBT) using sequence-specific primer–polymerase chain reaction (SSP-PCR), according to protocols described by Sayer et al.(28)

Statistical analysis.  The results were analyzed with the Student’s unpaired t-test. All tests were two-tailed, and P < 0.05 was considered significant.

Results

Identification of HCA587-derived CD4+ T-cell-stimulating peptides.  The SYFPEITHI algorithm predicted 15 peptides derived from HCA587 with high binding probability to six different HLA-DRB1 subtypes (Table 1). To streamline the screening of the peptides stimulating CD4+ T-cell responses, pool 1 and pool 2 were used separately for in vitro stimulation of the PBMC from six healthy donors. After 7 days, the activated cells were expanded in the presence of IL-2-containing medium and propagated by weekly restimulation with irradiated peptide-pulsed autologous PBMC as APC. After three rounds of peptide restimulation, the bulk T-cell cultures were tested using IFN-γ ELISPOT assay for reactivity against autologous PBMC that had been pulsed with each peptide forming the pool. As can be seen from Figure 1, PBMC from donor 2 significantly recognized p43–57 and p249–263, while PBMC from donor 6 developed a specific response against p145–159 and p186–200. No significant T-cell response was observed in the six donors after stimulation with seven additional HCA587-derived peptides forming the second pool. The HLA-DR haplotypes of six donors (donor 1 to donor 6) were determined and are shown in Table 2.

Figure 1.

 T-cell responses against the HCA587-derived peptides. Autologous peripheral blood mononuclear cells (PBMC) (1 × 105/well) were used as antigen-presenting cells (APC) to stimulate bulk T cells of six donors (donor 1 to donor 6). (a) T cells (5 × 104 cells/well) from donor 2 showed a significant response against peptide p43–57 and p249–263 after stimulation for 21 days with PBMC loaded with the first pool of eight HCA587-derived peptides. (b) T cells (2 × 104 cells/well) from donor 6 exhibited a significant response against peptide p145–159 and p186–200. A pool of seven peptides derived from the pp65 antigen of the human CMV was used as a positive control. The data are the means of triplicate determinations ± SD. *P < 0.05; **0.001 < P < 0.05 (determined by unpaired, two-tailed Student’s t-test). IFN, interferon.

Table 2.   HLA-DR haplotypes of donors and patients
 HLA-DRB1 alleles
Donor 107011101
Donor 209011202
Donor 308010901
Donor 414030901
Donor 501010405
Donor 601021501
Donor 709011202
Donor 813020901
Donor 903011502
Donor 1011011501
Donor 1107011301
Donor 1214011405
Patient 101010403
Patient 211011305
Patient 312021202
Patient 404061202

HLA-DR restriction of HCA587-reactive CD4+ T cells.  To obtain peptide-specific long-term T cell lines, PBMC from six healthy subjects and four patients with hepatocellular carcinoma were stimulated with autologous peptide-pulsed PBMC. The HLA-DR haplotypes of six donors (donor 7 to donor 12) and four patients are shown in Table 2. After 4–6 rounds of stimulations with a single peptide, a total of six T-cell lines were obtained from four donors and one patient. Three T-cell lines derived from donor 9, donor 10 and patient 3 were reactive with p43–57, and the other three T-cell lines derived from donor 7, donor 9 and donor 12 were specific for p145–159, p186–200 and p249–263 respectively. All the T-cell lines secreted IFN-γ (Fig. 2) but not IL-4 (data not shown) after stimulation with the corresponding peptides, indicating that these peptide-specific T cells belonged to the Th1 subset.

Figure 2.

 HLA-DR restriction analysis of HCA587-reactive T-cell lines. (a) p43–57-specific T-cell lines derived from donor 9 (D9), donor 10 (D10) and patient 3 (P3). (b) p145–159-specific T cell line from donor 7 (D7). (c) p186–200-specific T cell line from donor 9 (D9). (d) p249–263-specific T cell line from donor 12 (D12). The HLA-DR restriction of the peptide-specific T-cell lines was demonstrated by a significant blocking of the T-cell response to the peptide by incubating the antigen-presenting cells (APC) with anti-HLA-DR antibody (clone L243). Treatment of peptide-pulsed autologous peripheral blood mononuclear cells (PBMC) with anti-HLA-ABC antibody (clone W6/32) had no effect. These experiments were done using irradiated peptide-pulsed autologous PBMC or dentritic cells as APC. The data are the means of triplicate determinations ± SD.

The CD4+ T cells generated by the helper epitope p43–57 were further analyzed for Foxp3 expression and the results showed that approximately 17% of CD4+ T cells and 11% of IFN-γ-producing CD4+ T cells are Foxp3 positive following stimulation with peptide for 24 h, while no CD4+ T cells express Foxp3 in their resting state (data not shown).

In order to ascertain that T-cell responses to each peptide are HLA-DR dependent, HLA blocking experiments were performed where APC presentation of individual peptides to CD4+ T cells was assessed in the presence of either anti-HLA-DR- or anti-HLA-A,B,C-specific Ab known to specifically block antigen recognition. CD4+ T-cell stimulations induced by each peptide were significantly abrogated in the presence of an anti-HLA-DR-blocking antibody (Fig. 2). These results suggested that HCA587-derived peptides p43–57, p145–159, p186–200 and p249–263 were presented to the T cells in a HLA-DR-restricted manner.

Natural processing and presentation of the HCA587-derived peptides.  To verify whether the recognized HCA587 sequence contains naturally processed epitope(s), peptide-specific T cell lines were challenged with autologous dentritic cells pulsed with whole-protein HCA587 and human IgM as a negative control. Figure 3a shows the specific response of the p43–57 reactive T-cell line from donor 9 against the HCA587 protein processed and presented by autologous dentritic cells. Reactivity against the processed epitope derived from the protein could be blocked by preincubation of the APC with anti-DR antibody, proving once more the HLA-DR restriction of the presented epitope after processing of the HCA587 protein by dentritic cells. Neither the p145–159-specific T-cell line (from donor 7) nor the p186–200-specific T-cell line (from donor 9) was able to recognize the HCA587 protein presented by autologous dentritic cells (data not shown). Recognition of the native HCA587 protein by the p249–263-specific T-cell line (from donor 12) was not tested because the peptide-specific CD4+ T cells did not expand in culture enough to allow further characterization.

Figure 3.

 Natural processing of the HCA587-derived epitope p43–57 by dendritic cells and tumor cells. (a) p43–57-specific T-cell line from donor 9 showed a concentration-dependent response against autologous dentritic cells pulsed with whole-protein HCA587 (5, 10, 15 μg/mL), while the control protein (human IgM 15 μg/mL) remained unrecognized. The T-cell response against the naturally processed epitope p43–57 could be blocked by anti-human HLA-DR antibody, but not anti-human HLA-ABC antibody. (b) p43–57-specific T-cell line from donor 9 recognized HCA587-expressing tumor cell SK-MEL-37, which shared the HLA-DRB1*0301 subtype with responding T cells. The data are the means of triplicate determinations ± SD.

It was critical to assess whether p43–57 was produced by HCA587-expressing tumor cells via the MHC class II antigen-processing pathway. To select the appropriate tumor cells for the p43–57-specific T-cell line recognition assay, we assessed expression of the HCA587 protein and cell surface HLA-DR molecules in several tumor cell lines. The melanoma cell line SK-MEL-37 expressing HCA587 and HLA-DR molecules (HLA-DRB1*0101/0301) was used as APC. To enhance the expression of MHC molecules on the cell surface, the cells were cultured with IFN-γ (500 U/mL) for 48 h before the assays. The data presented in Figure 3b show that the p43–57-specific T-cell line was capable of recognizing antigen directly on tumor cells.

Dissection of DR restriction with high resolution.  To determine the HLA-DR subtype restricting the T-cell response against p43–57, allogeneic APC sharing the same HLA-DR haplotype with the responding cells were used in the T-cell stimulation assay. For use as allogeneic APC, four LCL (LCL8, HLA-DRB1*0701/1101; LCL13, HLA-DRB1*0406/1202; LCL22, HLA-DRB1*0102/1501; LCL26, HLA-DRB1*0901/1202) expressing one of the HLA-DRB1 alleles of the donors (donor 9, HLA-DRB1*0301/1502; donor 10, HLA-DRB1*1101/1501; patient 3, HLA-DRB1 *1202/1202) were established and pulsed with 2 μg/mL peptide p43–57. As shown in Figure 4, p43–57 was recognized in association with HLA-DRB1*1501 by donor 10, HLA-DRB1*1202 by patient 3 and HLA-DRB1*1502 by donor 9. In addition, the p43–57-specific T-cell line from donor 9 also responded to HCA587-expressing SK-MEL-37 melanoma cells (HLA-DRB1*0301/0101), as shown in Figure 3b, suggesting that p43–57 could be recognized in association with HLA-DRB1*0301. These results indicate that p43–57 was a promiscuous epitope presented in association with HLA-DRB1*1501, HLA-DRB1*1502, HLA-DRB1*1202 and HLA-DRB1*0301.

Figure 4.

 Subtype-specific dissection of the HLA-DRB1 restriction of p43–57-reactive T-cell lines. p43–57-reactive CD4+ T-cell lines (1 × 104/well) were co-cultured with the HLA-DRB1 genotype partially overlapping allogeneic lymphoblastoid cell lines (LCL) pulsed with p43–57 (2 μg/mL) or no peptide, and the IFN-γ in the culture supernatant was measured by ELISA assay 24 h later. (a) Allogeneic LCL8 (DRB1*0701 and DRB1*1101) and LCL22 (DRB1*0102 and DRB1*1501) were used as antigen-presenting cells (APC) for donor 10 (D10) (DRB1*1101 and DRB1*1501). p43–57-specific T cells responded to LCL22 but not to LCL8, suggesting that p43–57 was recognized in association with HLA-DRB1*1501 by donor 10. (b) Allogeneic LCL26 (DRB1*0901 and DRB1*1202) and LCL13 (DRB1*0406 and DRB1*1202) were used as APC for patient 3 (P3) (DRB1*1202 and DRB1*1202), indicating that p43–57 was recognized in association with HLA-DRB1*1202 by patient 3. (c) Allogeneic LCL22 was used as APC for donor 9 (D9) (DRB1*0301 and DRB1*1502), demonstrating that p43–57 was recognized in association with HLA-DRB1*1502 by donor 9. The data are the means of triplicate determinations ± SD. ***P < 0.001 (determined by unpaired, two-tailed Student’s t-test). IFN, interferon.

Owing to lack of HLA-DRB1*1502-expressing LCL, HLA-DRB1*1502 restriction of the CD4+ T-cell-mediated response to p43–57 was identified by use of HLA-DRB1*1501-expressing LCL22 as APC. HLA-DRB1*1502 is nearly identical to HLA-DRB1*1501 in the primary amino acid sequence, with only a glycine for valine substitution at position 86 that contributes to the p1 pocket involved in peptide binding (IMGT/HLA database of the EMBL-European Bioinformatics Institute). Our results indicate that there were no obvious differences for p43–57 in restriction by HLA -DRB1*1502 versus HLA -DRB1*1501.

Discussion

A prerequisite for immunotherapy of cancer is the existence of molecules that are either exclusively or preferentially expressed by malignant cells and that have the capacity to induce a specific immune response in the tumor-bearing host. Cancer-testis (CT) antigen, as a new category of antigen, has emerged to be a unique group of antigen that fulfils this requirement and has become a major focus for the development of vaccine-based clinical trials in recent years.(19,20,29,30) Up to now, a total of 110 CT genes or gene families have been entered into the CT database established recently by the Ludwig Institute for Cancer Research (http://www.cta.lncc.br),(31) of which only several CT antigens have been shown to elicit coordinated humoral and cell-mediated responses,(15,18–20,32) and HCA587 is one of the most immunogenic CT antigens. Because HCA587 is an attractive target for cancer vaccine, significant efforts were first devoted to the identification of CTL epitopes.(21–24) Here, we combined computational prediction for promiscuous MHC class II epitopes with validation by biological assays in vitro to study the CD4+ T-cell epitopes of HCA587.

The SYFPEITHI algorithm was applied for an in silico screening of the entire sequence of the HCA587 antigen. Fifteen pentadecamer peptides were predicted to have a high binding probability to the HLA-DRB1 subtypes *0101, *0301, *0401, *0701, *1101 and *1501. Because the SYFPEITHI score does not cover all known HLA-DR subtypes, we selected healthy subjects and patients with unknown HLA status for the identification of CD4+ T-cell-stimulating epitopes in the present study. In addition, to mimic what occurs in vivo where different processed peptides compete for MHC class II binding and T cell receptor recognition, we used pools of peptides for CD4+ T-cell stimulation. This strategy should favor the expansion of CD4+ T cells specific for peptides with higher affinity. By this approach, four epitopes of HCA587 (p43–57, p145–159, p186–200 and p249–263) were shown to induce CD4+ T-cell responses, and blocking experiments with an anti-HLA-DR antibody demonstrated that the four epitopes were recognized by CD4+ T cells in association with HLA-DR alleles.

As we know, the in silico prediction of the potential MHC class II ligands from any protein sequence was based on the identification of appropriately positioned anchor residues specific for a given MHC class II allele,(33) which allowed for the identification of peptides with high binding affinity. However, many such peptides failed to be presented by tumor cells or APC. Therefore, it is very important to assess whether the peptide-reactive CD4+ T-cell lines have the capacity to recognize the tumor antigen derived from tumor cells or processed by APC.(14) Of the four epitopes identified in the present study, only p43–57 was shown to be naturally processed and presented by DC and the tumor cell line SK-MEL-37, while p145–159- and p186–200-specific CD4+ T cells did not recognize the HCA587 protein after processing by autologous DC. The natural processing and presentation of p249–263 were not evaluated since the p249–263-specific T-cell line did not expand well in culture.

Dissection of the p43–57-induced T-cell response demonstrated that this epitope was presented in association with HLA-DRB1 subtypes *1501, *1502, *1202 and *0301, which is a characteristic of promiscuous helper T lymphocyte epitopes that are highly desired for clinical use because of their increased population coverage.(34–36) In addition, our preliminary results also indicate that the peptide-specific CD4+ T cells under stimulation with peptide could increase the cytotoxic activity of natural killer (NK) cells (data not shown).

p43–57 is the first HCA587-derived epitope that fulfils all requirements for a helper T lymphocyte-stimulating vaccine. The application of such a peptide in combination with CTL epitopes might help the induction of a strong antitumor response and improve the clinical efficacy of vaccines. Moreover, the identification of CD4 epitopes will help in the study of the number and function of HCA587-specific CD4+ T cells and monitoring of the immune responses in HCA587-expressing cancer patients before and after vaccination.

Acknowledgments

Grant support: The National 863 High Technology Program of China (2006AA02Z486); Beijing Municipal Government Foundation for Natural Sciences (7071006).

Disclosure Statement

There are no financial conflicts of interest.

Ancillary