Induction of CD8 T-cell responses restricted to multiple HLA class I alleles in a cancer patient by immunization with a 20-mer NY-ESO-1f (NY-ESO-1 91-110) peptide

Authors


  • This article was published online on 22 June 2012. An error was subsequently identified. This notice is included in the online and print versions to indicate that both have been corrected 21 November 2012.

Abstract

Immunogenicity of a long 20-mer NY-ESO-1f peptide vaccine was evaluated in a lung cancer patient TK-f01, immunized with the peptide with Picibanil OK-432 and Montanide ISA-51. We showed that internalization of the peptide was necessary to present CD8 T-cell epitopes on APC, contrasting with the direct presentation of the short epitope. CD8 T-cell responses restricted to all five HLA class I alleles were induced in the patient after the peptide vaccination. Clonal analysis showed that B*35:01 and B*52:01-restricted CD8 T-cell responses were the two dominant responses. The minimal epitopes recognized by A*24:02, B*35:01, B*52:01 and C*12:02-restricted CD8 T-cell clones were defined and peptide/HLA tetramers were produced. NY-ESO-1 91-101 on A*24:02, NY-ESO-1 92-102 on B*35:01, NY-ESO-1 96-104 on B*52:01 and NY-ESO-1 96-104 on C*12:02 were new epitopes first defined in this study. Identification of the A*24:02 epitope is highly relevant for studying the Japanese population because of its high expression frequency (60%). High affinity CD8 T-cells recognizing tumor cells naturally expressing the epitopes and matched HLA were induced at a significant level. The findings suggest the usefulness of a long 20-mer NY-ESO-1f peptide harboring multiple CD8 T-cell epitopes as an NY-ESO-1 vaccine. Characterization of CD8 T-cell responses in immunomonitoring using peptide/HLA tetramers revealed that multiple CD8 T-cell responses comprised the dominant response.

The NY-ESO-1 antigen was originally identified in esophageal cancer by serological expression cloning (SEREX) using autologous patient serum.1, 2 NY-ESO-1 expression is observed in a wide range of human malignancies, but the expression is restricted to germ cells in the testes in normal adult tissues.1–4 Therefore, NY-ESO-1 has emerged as a prototype of a class of cancer/testis (CT) antigens.5 The efficacy of the NY-ESO-1 antigen as a cancer vaccine has been studied extensively using various preparations, e.g., peptide, protein or DNA, etc. of the antigen with various adjuvants.6–14 These studies established the safety of the NY-ESO-1 vaccine and demonstrated its immunogenicity.

In a phase I clinical trial, we immunized cancer patients with a complex of cholesterol-bearing hydrophobized pullulan and NY-ESO-1 whole protein (CHP-NY-ESO-1) and showed that the vaccine had potent capacity to induce the NY-ESO-1 antibody in vaccinated patients.13, 14 The most dominant serological antigenic epitope was NY-ESO-1 91-108. The CHP-NY-ESO-1 vaccine also elicited CD4 and CD8 T-cell responses in immunized patients.14 Analysis of T cell responses against overlapping peptides (OLPs) spanning the NY-ESO-1 molecule revealed that two dominant NY-ESO-1 regions, regions II (73-114) and III (121-144), were recognized by either CD4 or CD8 T-cells in most patients irrespective of their HLA type. Essentially similar findings were obtained by studies using other preparations of NY-ESO-1 protein vaccine.11, 12, 15

CD8 T-cells induced by immunization with NY-ESO-1 class I short epitope peptides have been shown to be of low affinity and do not recognize naturally processed NY-ESO-1 on tumor cells.16 However, the advantage of synthetic long peptides over short peptides for use as vaccines has been reported.17 Long peptides do not bind to MHC class I molecules directly, and require antigen processing by dendritic cells to be presented. Therefore, the use of long peptides prevents the antigen peptides from direct binding to MHC class I molecules on nonprofessional antigen-presenting cells (APC), which may cause transient activation and subsequent anergy of CTLs in the absence of appropriate costimulatory signals.17–19 Based on these findings, we recently used a long peptide spanning a peptide region NY-ESO-1 91-110 (NY-ESO-1f peptide) which included the dominant serological antigenic epitope and overlapped one of the two dominant regions recognized by CD4 and CD8 T-cells for a vaccine in a clinical trial.20 Ten patients received the NY-ESO-1f peptide vaccine. The NY-ESO-1f peptide vaccine was well tolerated and elicited humoral, CD4 and CD8 T-cell responses in immunized patients.

In this study, we demonstrated that internalization of the peptide was necessary to present CD8 T-cell epitopes on APC treated with the long 20-mer NY-ESO-1f peptide. Analysis of the CD8 T-cell response in an NY-ESO-1f peptide-immunized patient revealed occurrence of responses restricted to all five HLA class I alleles defined in the patient. The frequency of A*24:02, B*35:01, B*52:01, C*03:03 and C*12:02-restricted CD8 T-cells in PBMCs was defined by clonal analysis revealing B*35:01- and B*52:01-restricted CD8 T-cell responses as dominant. By establishing clones from those HLA-restricted CD8 T-cells, new epitopes on A*24:02, B*35:01, B*52:01 and C*12:02 were defined and peptide/HLA tetramers were prepared. Clonal analysis showed that CD8 T-cells that recognize natural epitopes on tumor cells were induced in a significant proportion by immunization with the NY-ESO-1f peptide. Immunomonitoring using the tetramers revealed that multiple CD8 T-cell responses comprised the dominant response.

Material and Methods

Clinical trial

A phase I clinical trial of the NY-ESO-1f peptide vaccine was conducted to evaluate the safety, immune response and tumor response.20 Patients with advanced cancers that were refractory to standard therapy and expressed NY-ESO-1 as assessed by immunohistochemistry (IHC) were eligible. The protocol was approved by the Ethics Committee of Tokyo, Osaka and Okayama Universities in light of the Declaration of Helsinki. Written informed consent was obtained from each patient before enrolling in the study. The study was performed in compliance with Good Clinical Practice. The study was registered in the University hospital Medical Information Network Clinical Trials Registry (UMIN-CTR) Clinical Trial (Unique trial number: UMIN000001260) on July 24, 2008 (UMIN-CTRURL: http://www.umin.ac.jp/ctr/ index.htm).

Blood samples

Patient TK-f01 was a lung cancer patient who received a right middle lobectomy in October, 2004.20 As the tumor continued to grow despite chemotherapy, he was enrolled in the study in June, 2008. The patient received 12 vaccinations once every 3 weeks. Peripheral blood was drawn from patient TK-f01 with informed consent for immunological monitoring. Peripheral blood mononuclear cells (PBMCs) were isolated from heparinized blood by density gradient centrifugation using a Histopaque 1077 (Sigma-Aldrich, St. Louis, MO). CD4-, CD8- and CD19-positive cells were purified by magnetic cell sorting (Miltenyi Biotec, Bergisch Gladbach, Germany). The residual cells were kept for use as APC. The cells were stored in liquid N2 until use. HLA tying was done with PBMCs by a sequence-specific oligonucleotide probe and sequence-specific priming of genomic cDNA using a standard procedure.

Cell lines

LC99A and OU-LC-OK are lung cancer cell lines. SK-OV3 is an ovarian cancer cell line and SK-MEL37 is a melanoma cell line. These cell lines were kept by serial passage in tissue culture. EBV-B cells were generated from CD19-positive peripheral blood B cells using a culture supernatant from EBV-producing B95-8 cells. The medium used to maintain these cell lines was RPMI1640 supplemented with 10% FCS (JRM, Bioscience, Lenexa, KA), 2 mmol/l Glutamax, antibitotics, and 10 mmol/l HEPES (Invitrogen, Carlsbad, CA).

Antibodies

Anti-human CD4, anti-human CD8, anti-HLA class I and anti-HLA class II mAbs were purchased from BD Bioscience (San Jose, CA).

Peptides

The following series of 28 18-mer OLPs and a C-terminal 30-mer peptide spanning the entire NY-ESO-1 protein were used: 18.1 (1-18), 18.2 (7-24), 18.3 (13-30), 18.4 (19-36), 18.5 (25-42), 18.6 (31- 48), 18.7 (37-54), 18.8 (43-60), 18.9 (49-66), 18.10 (55-72), 18.11 (61-78), 18.12 (67-84), 18.13 (73-90), 18.14 (79-96), 18.15 (85-102), 18.16 (91-108), 18.17 (97-114), 18.18 (103-120), 18.19 (109-126), 18.20 (115-132), 18.21 (121-138), 18.22 (127-144), 18.23 (133-150), 18.24 (139-156), 18.25 (145-162), 18.26 (149-166), 18.27 (153-170), 18.28 (156-173) and 30.9 (151-180). Various N- and C-termini truncated peptides in the NY-ESO-1f peptide were also used. These peptides were synthesized using standard solid-phase methods based on N-(9-fluorenyl)-methoxycarbonyl (Fmoc) chemistry on an ABIMED Multiple Peptide Synthesizer (AMS422, ABIMED, Langenfeld, Germany) at Okayama University (Okayama, Japan). The carboxyfluorescein (FAM)-conjugated NY-ESO-1f peptide (5(6)-FAM-YLAMPFATPMEAELARRSLA) was synthesized by Operon (Tokyo, Japan).

Recombinant NY-ESO-1 protein

Recombinant NY-ESO-1 protein was prepared as described earlier.13 NY-ESO-1 cDNA was cloned into the SphI/SalI and BamHI/SphI sites of the pQE-30 vector. N-His tagged protein was purified by nickel-ion affinity chromatography under denaturing conditions.

Preparation of immature dendritic cells

Monocytes were isolated from PBMCs using anti-CD14 mAb-coated magnetic beads (Miltenyi Biotec) and cultured in AIM-V medium supplemented with 5% heat-inactivated pooled human serum, 10 ng/ml rhGM-CSF (Kyowa Hakko Kirin, Tokyo, Japan) and 10 ng/ml rhIL-4 (PeproTech) for 7 days at 37°C in a 5% CO2 atmosphere.

In vitro stimulation of bulk CD4 and CD8 T-cells

CD4 and CD8 T-cells (2 × 106/well) were cultured with a mixture of 29 NY-ESO-1 OLPs (10−6 M) in the presence of an equal number of irradiated (40 Gy), autologous CD4- and CD8-depleted cells as APC in a 24-well culture plate (BD Bioscience) for 12 days at 37°C in a 5% CO2 atmosphere. The medium was AIM-V (Invitrogen) supplemented with 5% heat-inactivated pooled human serum, 2 mM L-glutamine, 100 IU/ml penicillin, 100 μg/ml streptomycin, 10 units/ml recombinant human (rh) IL-2 (Takeda Chemical Industries, Osaka, Japan) and 10 ng/ml rhIL-7 (Peprotech, London, UK).

Cloning of CD8 T-cells

CD8 T-cells were cloned by limiting dilution after in vitro stimulation in round-bottomed 96-well plates in the presence of irradiated (40 Gy) allogeneic PBMCs freshly prepared from the healthy donor as feeder cells. The medium used was AIM-V (Invitrogen) supplemented with 5% heat-inactivated pooled human serum, 2 mM L-glutamine, 100 IU/ml penicillin, 100 μg/ml streptomycin, 20 units/ml rhIL-2 (Takeda Chemical Industries), 10 ng/ml rhIL-7 (Peprotech) and 1 μg/ml phytohemagglutinin (PHA)-L (Sigma-Aldrich).

Clonal expansion

Cloned CD8 T-cells (1 × 103) obtained by limiting dilution were expanded in a round-bottomed 96 well plate in the presence of irradiated (40 Gy) PBMCs (5 × 104) freshly prepared from allogeneic healthy donors as feeder cells. Fresh medium was added every 3 days. After culture for 14 days, the cells were harvested and kept frozen at 2 × 106/tube. The medium used was AIM-V supplemented with 5% heat-inactivated pooled human serum, 2 mM L-glutamine, 100 IU/ml penicillin, 100 μg/ml streptomycin, 20 units/ml rhIL-2, 10 ng/ml rhIL-7 and 1 μg/ml phytohemagglutinin (PHA)-L.

IFNγ capture assay

Bulk CD4 or CD8 T-cells (1 × 105) from the in vitro stimulation culture were cultured with autologous or allogeneic EBV-B cells (1 × 105) pulsed with OLPs for 4 hr. The cells were then treated with a bi-specific CD45 and IFNγ antibody (IFNγ catch reagent) (2 μl) for 5 min on ice. The cells were diluted in AIM-V medium (3 ml) and placed on a slow rotating device (Miltenyi Biotec) to allow IFNγ secretion at 37°C in a 5% CO2 atmosphere. After incubation for 45 min, the cells were washed with cold buffer and treated with PE-conjugated anti-IFNγ (detection reagent), and FITC-conjugated anti-CD4 or anti-CD8 mAb. After incubation for 10 min at 4°C, the cells were washed and analyzed by a FACS Canto II (BD Bioscience).

Tetramer construction and staining

Peptide/HLA tetramers were produced as described earlier.21, 22 NY-ESO-1 91-101/A*24:02, NY-ESO-1 92-100/B*35:01, NY-ESO-1 92-102/B*35:01, NY-ESO-1 94-104/B*35:01 and NY-ESO-1 96-106/C*12:02 tetramers were used. The HIV Env/A*24:02 tetramer was used as a control. For staining, cells were incubated with tetramer at a concentration of 20 μg/ml for 15 min at 37°C, followed by incubation with an FITC-conjugated anti-CD8 mAb (Miltenyi Biotec) on ice for 15 min and analyzed by a FACS Canto II (Becton Dickinson).

Cytotoxicity assay

Cytotoxicity was assayed by a luminescent method using the aCella-Tox kit (Cell Technology, Mountain View, CA). Effector cells were incubated with 5,000 target cells at various ratios in 96-well round bottomed culture plates for 12 hr at 37°C in a 5% CO2 atmosphere. The plate was read by a luminometer (multi-detection microplate reader, DS Pharma, Osaka, Japan).

IFNγ ELISA

CD8 T-cell clones (5 × 103) were cultured with autologous or allogeneic EBV-B cells (5 × 103) pulsed with the peptides in a 96-well round bottomed culture plate for 24 hr at 37°C in a 5% CO2 atmosphere. Culture supernatants were then collected and the amount of IFNγ was measured by sandwich ELISA. For antibody blocking experiments, each mAb (5 μg/ml) was added to the assay culture. To inhibit internalization of the antigens to DCs, cytochalasin B (10 μM) was added to the culture.

Results

Internalization of 20-mer NY-ESO-1f peptide to present CD8 T-cell epitopes on antigen presenting cells

We investigated the need for internalization of the 20-mer NY-ESO-1f peptide (NY-ESO-1 91-110: YLAMPFATPMEAELARRSLA) to present CD8 T-cell epitopes on APC. Immature dendritic cells (iDCs) were prepared from purified CD14-positive cells from a B*35:01 healthy donor by treating them with IL-4 and GM-CSF for 7 days and used as APC. As shown in Figure 1a, internalization of the FAM-conjugated NY-ESO-1f peptide was observed in approximately 10% of DCs after culture with the peptide for 12 hr. Treatment of DCs with the peptide in the presence of cytochalasin B diminished internalization. Recognition by a B*35:01-restricted CD8 T-cell clone 2H10 of DCs treated with the peptides was investigated. As shown in Figure 1b, clone 2H10 recognized DCs pulsed with the short peptide NY-ESO-1 92-100 in the presence of cytochalasin B. On the other hand, recognition of DCs pulsed with NY-ESO-1f peptide was inhibited in the presence of cytochalasin B. Only marginal recognition was observed with DCs pulsed with recombinant NY-ESO-1 protein.

Figure 1.

Internalization and processing of the long 20-mer NY-ESO-1f-peptide by APC in CD8 T-cell recognition. In (a), immature dendritic cells (iDCs) from a healthy donor (HD) NO PBMCs (A*02:07/*26:02, B*35:01/*46:01, C*08:03/*14:03) were cultured with an NY-ESO-1f peptide (10 μM) or a FAM NY-ESO-1f peptide (10 μM) for 12 hr in the presence or absence of cytochalasin B (10 μM). After culture, the internalization of the FAM NY-ESO-1f peptide was analyzed by FACS Canto II. In (b), the CD8 T-cell clone 2H10 (5 × 103) was stimulated with iDCs (5 × 103) pulsed with the short peptide (NY-ESO-1 92-100: LAMPFATPM, 9-mer; 1 μM), the NY-ESO-1f peptide (NY-ESO-1 91-110: YLAMPFATPMEAELARRSLA, 20-mer; 1 μM) or recombinant NY-ESO-1 protein (1 μM) for the indicated time in the presence or absence of cytochalasin B (10 μM). The amount of IFNγ in the culture supernatant was determined by ELISA. One representative of three independent experiments is shown.

Multiple HLA class I-restricted CD8 T-cell responses in PBMCs from patient TK-f01 immunized with the 20-mer NY-ESO-1f peptide

TK-f01 was a lung adenocarcinoma patient immunized with the NY-ESO-1f peptide (600 μg) with Picibanil OK-432 and Montanide ISA-51 12 times once every 3 weeks.20 We investigated multiple HLA class I-restricted CD8 T-cell responses in PBMCs from patient TK-f01. The patient's HLA class I was A*24:02, B*35:01, B*52:01, C*03:03 and C*12:02. Purified CD8 T-cells from the patient's PBMCs obtained at day 64 after the third vaccination were stimulated once with a mixture of 29 18-mer series of NY-ESO-1 OLPs spanning the entire protein for 12 days using CD4 and CD8-depleted PBMCs as APC. The cells were harvested and stimulated with NY-ESO-1f peptide for 4 hrs using autologous and allogeneic EBV-B cells as APC and the response was assayed by IFNγ capture assay. As shown in Figure 2, TK-f01 CD8 T-cells responded strongly to NY-ESO-1f peptide-pulsed autologous EBV-B cells. Use of various allogeneic EBV-B cells as APC showed TK-f01 CD8 T-cell responses against NY-ESO-1f peptide presented on multiple HLA class I molecules shared with patient TK-f01.

Figure 2.

Multiple HLA class I-restricted CD8 T-cell responses in patient TK-f01 immunized with the 20-mer NY-ESO-1f peptide. CD8 T-cells (2 × 106) were stimulated with a mixture of NY-ESO-1 OLPs (10−6 M) in the presence of APC for 12 days. The cells were harvested and stimulated with NY-ESO-1f peptide for 4 hr using autologous and allogeneic EBV-B cells as APC and the response was assayed by an IFNγ capture assay.

Clonal analysis of the multiple HLA class I-restricted TK-f01 CD8 T-cell responses against the NY-ESO-1f peptide

The multiple HLA class I-restricted TK-f01 CD8 T-cell responses against the NY-ESO-1f peptide were investigated by clonal analysis. Purified CD8 T-cells from the patient's PBMCs obtained at day 64 after the third vaccination were stimulated once with a mixture of NY-ESO-1 OLPs for 12 days using CD4 and CD8-depleted PBMCs as APC as above. The cells were then collected and seeded at 3 cells per well in 96-well culture plates to expand clones. After culture for 14 days, 300 clones were obtained. Each of the 300 clones were split into 7 and the responses against NY-ESO-1f peptide on autologous, OS-P06, OS-P08, HD NM, HD NO, OY-P04 and OY-P02 EBV-B cells shown in Figure 3b were investigated by IFNγ ELISA for the culture supernatant to determine restriction molecules. As shown in Table 1, the frequencies of B*35:01-, B*52:01-, A*24:02-, C*12:02- and C*03:03-restricted CD8 T-cell clones were 11.0, 9.3, 2.3, 1.3 and 0.7%, respectively. The frequencies of CD8 T-cell clones determined after clonal expansion were comparable to the results shown in Figure 2.

Figure 3.

The restriction molecule and the minimal epitope in recognition of NY-ESO-1f peptide by the CD8 T-cell clone 4E10. A CD8 T-cell clone 4E10 (5 × 103) obtained from the culture shown in Table 1 was cultured with autologous EBV-B cells pulsed with the NY-ESO-1f-peptide (1 μM). The restriction molecule was analyzed by antibody blocking (5 μg/ml) (a) and using various EBV-B cells as APC (b). The minimal epitope was determined by using various N- and C-termini truncated peptides (1 μM) (c) and by titration of the peptide concentration (d). In (e), the binding of the NY-ESO-1 91-101/A*24:02 tetramer to the clone 4E10 is shown. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Table 1. Clonal analysis of multiple HLA class I-restricted CD8 T cell responses against the NY-ESO-1f peptide
inline image

The frequency of NY-ESO-1f peptide-reactive CD8 T-cells in PBMCs restricted to each allele was calculated assuming the doubling time of the cells to be 24 hr during the culture period for 12 days as listed in Table 1.

Determination of minimal epitopes derived from the 20-mer NY-ESO-1f peptide recognized by TK-f01 CD8 T-cell clones in restriction to A*24:02, B*35:01, B*52:01 and C*12:02

We established stably proliferating TK-f01 CD8 T-cell clones: 4E10, 2H10, 8D5, 10-10U and 7B recognizing the 20-mer NY-ESO-1f peptide in restriction to A*24:02, B*35:01, B*35:01, B*52:01 and C*12:02, respectively, from the culture shown in Table 1 and determined the minimal epitopes. Figure 3 shows the confirmation of the restriction molecule and determination of a minimal epitope in recognition of the NY-ESO-1f peptide by the CD8 T-cell clone 4E10. As shown in Figure 3a, the response was blocked by anti-CD3 mAb, anti-CD8 mAb and anti-HLA class I mAb, but not anti-CD4 mAb or anti-class II mAb, confirming that the response was HLA class I restricted. As shown in Figure 3b, use of a panel of allogeneic EBV-B cells as APC to present the NY-ESO-1f peptide confirmed the response of clone 4E10 was A*24:02-restricted. As shown in Figures 3c and 3d, use of various N- and C-termini truncated peptides and titration of the peptide concentration revealed that the minimal epitope was NY-ESO-1 91-101 (11-mer). Figure 3e shows the binding of the tetramer constructed using the epitope peptide to the clone 4E10.

Similarly, as shown in Supporting Information Figures S1–S4, we confirmed B*35:01-, B*35:01-, B*52:01- and C*12:02-restricted recognition of the NY-ESO-1f peptide by clones 2H10, 8D5, 10-10U and 7B, respectively, determined minimal epitopes and successfully produced tetramers.

We investigated the cytotoxicity of CD8 T-cell clones using an aCella-Tox system to examine G3PDH release from target tumor cell lines. As shown in Figure 4a, cytotoxicity against tumor cell lines expressing NY-ESO-1 and an appropriate HLA class I allele was observed with the A*24:02-restricted clone 4E10, B*35:01-restricted clone 8D5, B*52:01-restricted clone 10-10U and C*12:02-restricted clone 7B, but not with the B*35:01-restricted clone 2H10. With those clones, no cytotoxicity was observed against tumor cell lines which lacked either NY-ESO-1 or a matched HLA class I allele.

Figure 4.

Clonal analysis of CD8 T-cells recognizing tumor cells induced by NY-ESO-1f peptide vaccination. In (a), cytotoxicity of the CD8 T-cell clones shown in Figure 3, and Supporting Information Figures S1–S4 was assayed by a luminescent method using the aCella-Tox kit. Shared HLA class I alleles in target tumors are shown in bold. In (b), purified CD8 T-cells (2 × 106) from TK-f01 PBMCs obtained at day 0 and at day 85 (after the fourth vaccination) were stimulated once with a mixture of NY-ESO-1 OLPs for 12 days in the presence of APC for 12 days. The cells were collected and seeded at 3 cells per well in 96-well culture plates to expand clones. After culture for 14 days, 290 clones were obtained from a culture at day 0 and 436 clones were obtained from a culture at day 85. Those clones were split into two groups and cultured with the NY-ESO-1f peptide (1 μM) pulsed-HD NO EBV-B cells or NY-ESO-1 positive SK-OV3 both sharing only B*35:01 with the patient to determine the B*35:01-restricted response and with NY-ESO-1f (1 μM) peptide pulsed-OS-P06 EBV-B cells or NY-ESO-1-positive LC99A both sharing A*24:02 with the patient to determine the A*24:02-restricted response. The amount of IFNγ in the culture supernatant was determined by ELISA.

Frequency of CD8 T-cells recognizing tumor cells

We then investigated the frequency of CD8 T-cells recognizing tumor cells by clonal analysis. Purified CD8 T-cells from TK-f01 PBMCs obtained at day 0 and at day 85 (after the fourth vaccination) were stimulated once with a mixture of NY-ESO-1 OLPs for 12 days using CD4 and CD8-depleted PBMCs as APC as above. The cells were collected and seeded at 3 cells per well in 96-well culture plates to expand clones. After culture for 14 days, 290 clones were obtained from a culture of CD8 T-cells at day 0 and 436 clones were obtained from a culture of CD8 T-cells at day 85. Those clones were split into two groups and the B*35:01- and A*24:02-restricted responses were investigated. As shown in Figure 4b, in CD8 T-cells obtained at day 0, no B*35:01-restricted clone reactive against NY-ESO-1f peptide was obtained. There was only one A*24:02-restricted clone reactive against the NY-ESO-1f peptide (1/145). On the other hand, in CD8 T-cells obtained at day 85, the number of B*35:01-restricted clones reactive against the NY-ESO-1f peptide was 19 (7.7%), and 6 of 19 (31.6%) clones were also reactive against the NY-ESO-1-positive and B*35:01-positive tumor SK-OV3. The number of A*24:02-restricted clones reactive against the NY-ESO-1f peptide was 12 (6.4%), and 7 of 12 (58.3%) clones were also reactive against the NY-ESO-1-positive and A*24:02-positive tumor LC99A. Recognition of tumor cells was also confirmed by an aCella-Tox system (data not shown).

Immunomonitoring of CD8 T-cell responses by IFNγ capture assay and by peptide/HLA class I tetramers in PBMCs from TK-f01 during vaccination

After initiating the vaccine, the tumor remained stable for 6 months and was classed as SD at the end of the sixth vaccination in this patient.20 However, the tumor started to grow after the eighth vaccination, consistent with an accelerated elevation in the serum CEA level.

Figure 5a shows the results of the IFNγ capture assay demonstrating the CD8 T-cell response against NY-ESO-1f peptide after stimulation with a mixture of NY-ESO-1 OLPs for 12 days. A vigorous CD8 T-cell response was observed even after the first vaccination (day 22), continuing until after the fifth vaccination (day 106). Thereafter, the response decreased suddenly. As shown in Figure 5b, the number of CD8 T-cells in the same culture positively stained with NY-ESO-1 92-100 and NY-ESO-1 94-104/B*35:01 tetramers was parallel with that of CD8 T-cells in an IFNγ capture assay. The findings indicated that CD8 T-cell responses against the peptides on B*35:01 comprised the dominant response, consistent with the clonal analysis shown in Table 1.

Figure 5.

Immunomonitoring of the CD8 T-cell response by IFNγ capture assay and by peptide/HLA class I tetramers in PBMCs from TK-f01 during vaccination. CD8 T-cells (2 × 106) were cultured with a mixture of NY-ESO-1 OLPs (1 μM) in the presence of APC for 12 days. In (a), the response was assayed by an IFNγ capture assay. In (b), staining by PE-labeled NY-ESO-1 92-100/B*35:01 and NY-ESO-1 94-104/B*35:01 tetramers is shown. An HIV Env/A*24:02 tetramer was used as control.

Discussion

In this study, we showed that internalization of the peptide was necessary for presenting CD8 T-cell epitopes derived from the long 20-mer NY-ESO-1f peptide on APC. Labeled NY-ESO-1f peptide was detected in approximately 10% of DCs after incubation with the peptide for 12 hr. No CD8 T-cell recognition of NY-ESO-1f peptide-pulsed DCs was observed after blocking internalization with cytochalasin B. Thus, the CD8 T-cell epitope was presented following internalization with a long 20-mer NY-ESO-1f peptide, contrasting with the direct presentation of the short epitope. Recognition by CD8 T-cells of NY-ESO-1f peptide-pulsed DCs, but not the recombinant NY-ESO-1 protein-pulsed DCs, indicated the occurrence of efficient cross presentation on MHC class I molecules after internalization of the 20-mer NY-ESO-1f peptide. Cross presentation of the long peptides on MHC class I molecules after internalization was previously shown with a peptide derived from HPV.23 Processing of the peptide in APC may facilitate the presentation of natural epitopes. This study showed a high frequency of CD8 T-cells recognizing natural epitopes on tumor cells (see below). It has been shown previously that immunization with the short peptide induced CD8 T-cells with low affinity incapable of recognizing natural epitopes on tumor cells.16 Use of the long peptide for immunization is beneficial to induce CD8 T-cells that recognize tumor cells.

We showed multiple HLA class I-restricted CD8 T-cell responses against a long 20-mer NY-ESO-1f peptide in the lung cancer patient TK-f01, who was immunized with the peptide with Picibanil OK-432 and Montanide ISA-51. Clonal analysis showed that the frequencies of A*24:02, B*35:01, B*52:01, C*03:03 and C*12:02-restricted CD8 T-cells were 2.3, 11.0, 9.3, 0.7 and 1.3%, respectively, in the bulk CD8 T-cell population stimulated with NY-ESO-1 OLPs, indicating that B*35:01 and B*52:01-restricted CD8 T-cell responses were the two dominant CD8 T-cell responses. The minimal epitopes recognized by A*24:02, B*35:01, B*52:01 and C*12:02, but not C*03:03-restricted CD8 T-cell clones, were defined and the peptide/HLA tetramers were produced using those epitopes. NY-ESO-1 91-101 (11-mer) on A*24:02, NY-ESO-1 92-102 (11-mer) on B*35:01, NY-ESO-1 96-104 (9-mer) on B*52:01 and NY-ESO-1 96-106 (9-mer) on C*12:02 were new epitopes first defined in this study. Particularly, identification of this A*24:02 binding epitope and successful production of a tetramer with the epitope will be highly relevant when studying Japanese populations because of its high expression frequency (60%). Interestingly, within 25 HLA class I epitopes in the NY-ESO-1 molecule defined to date,24–26 15 epitopes including the epitopes shown in this study clustered in the peptide region NY-ESO-1 89-104. These findings suggested the usefulness of a long 20-mer NY-ESO-1f peptide harboring multiple CD8 T-cell epitopes for an NY-ESO-1 vaccine and this study indeed showed that vaccination with a long 20-mer NY-ESO-1f peptide elicited multiple HLA class I-restricted CD8 T-cell responses against the respective epitopes.

We also estimated the frequency of B*35:01- and A*24:02-restricted CD8 T-cells recognizing only the peptide-pulsed DCs, or both the peptide-pulsed DCs and the tumor cells naturally expressing the epitope and the matched HLA class I. At day 0, no B*35:01-restricted CD8 T-cell clones were detected from 145 clones obtained from 960 wells in total of the bulk CD8 T-cells stimulated with NY-ESO-1 OLPs for 12 days. Only a single A*24:02-restricted CD8 T-cell clone reactive against the peptide, but not the tumor, was obtained from the same number of clones. On the other hand, at day 85 after the fourth vaccination, 19 B*35:01-restricted and 12 A*24:02-restricted CD8 T-cell clones reactive against the peptide were detected from 248 and 188 clones, respectively, obtained from 960 wells for each of the bulk CD8 T-cells stimulated as above. 31.6% (6 of 19) of B*35:01-restricted and 58.3% (7 of 12) of A*24:02-restricted CD8 T-cell clones were also reactive against the tumors naturally expressing the epitope and the matched HLA class I. Thus, a significant frequency of CD8 T-cells recognizing tumor cells naturally expressing the epitopes and matched HLA was induced. This was likely to be related to the internalization and processing of the peptide in APC.

It has been reported that vaccination with NY-ESO-1 protein with CpG and Montanide elicited detectable CD8 T-cell responses in half of the immunized patients (9/18) and that vaccine-induced CD8 T-cells mostly recognized NY-ESO-1 81-110 in restriction to either HLA-B35 or HLA-Cw3.27 B35 binding peptide epitopes 94-102 and 94-104 and Cw3 binding peptide epitopes 92-100 and 96-104 have been described.25, 28 In our previous study, we showed that an NY-ESO-1f peptide (NY-ESO-1 91-110) vaccine elicited a response to peptide 16 (NY-ESO-1 91-108) in all six patients analyzed with or without B35 and/or Cw3. The fact that the NY-ESO-1f peptide vaccine elicited CD8 T cell responses in patients with various HLA types suggests the advantage of a long peptide over the whole protein for vaccination.

Patient TK-f01 showed stable disease until the eighth vaccination and the disease exacerbated thereafter. In immunomonitoring CD8 T-cell responses following 20-mer NY-ESO-1f peptide vaccination, we observed vigorous CD8 T-cell responses, even after the first vaccination (day 22), lasting until the fifth vaccination (day 106) by IFNγ capture assay with a decrease thereafter, consistent with the clinical course. Characterization of CD8 T-cell responses using the various peptide/HLA tetramers showed that vigorous CD8 T-cell responses against the peptides restricted to B*35:01 were observed at day 64 after the third vaccination. These findings suggested that the multiple B*35:01-restricted CD8 T-cell responses comprised the dominant response.

Because Th cells help CTL induction, introduction of a Th epitope into the vaccine or physical linking of Th and CTL epitope peptides facilitated an increase in the immunogenicity of CTL vaccines.29, 30 The synthetic long peptide containing overlapping CD4 and CD8 T cell epitope sequences in the antigens is expected to generate both CD4 and CD8 T cell responses as a vaccine. In our study, the NY-ESO-1f peptide was shown to contain multiple HLA class II epitopes, as well as class I epitopes,20 and is considered to be beneficial to elicit CD8 T-cell responses efficiently.

Acknowledgements

We thank Ms. Junko Mizuuchi for preparation of the manuscript. This article is dedicated to the memory of Lloyd J. Old, M.D.

Ancillary