Antibody response against NY-ESO-1 in CHP-NY-ESO-1 vaccinated patients
Article first published online: 2 FEB 2007
Copyright © 2007 Wiley-Liss, Inc.
International Journal of Cancer
Volume 120, Issue 10, pages 2178–2184, 15 May 2007
How to Cite
Kawabata, R., Wada, H., Isobe, M., Saika, T., Sato, S., Uenaka, A., Miyata, H., Yasuda, T., Doki, Y., Noguchi, Y., Kumon, H., Tsuji, K., Iwatsuki, K., Shiku, H., Ritter, G., Murphy, R., Hoffman, E., Old, L. J., Monden, M. and Nakayama, E. (2007), Antibody response against NY-ESO-1 in CHP-NY-ESO-1 vaccinated patients. Int. J. Cancer, 120: 2178–2184. doi: 10.1002/ijc.22583
- Issue published online: 23 MAR 2007
- Article first published online: 2 FEB 2007
- Manuscript Accepted: 18 DEC 2006
- Manuscript Received: 1 NOV 2006
- Ministry of Education, Culture, Sports, Science and Technology of Japan
- Cancer Research Institute, New York
- cancer vaccine;
- antibody response
NY-ESO-1 specific humoral responses are frequently observed in patients with various types of NY-ESO-1 antigen expressing tumors. In a large proportion of NY-ESO-1 antibody-positive patients of NY-ESO-1-specific CD8 T-cells can also be detected suggesting that monitoring of the NY-ESO-1 specific humoral immune response may be a relevant and more practical surrogate for estimating the overall immune response against NY-ESO-1 in clinical vaccine studies. We have immunized 9 cancer patients with full length NY-ESO-1 protein formulated with cholesterol-bearing hydrophobized pullulan (CHP-NY-ESO-1) and investigated the humoral immune responses against NY-ESO-1. Seven patients were NY-ESO-1 antibody-negative and 2 patients were positive prior to vaccination. Vaccination with CHP-NY-ESO-1 resulted in the induction or increase of NY-ESO-1 antibody responses in all 9 patients immunized. Epitope analysis revealed 5 regions in the NY-ESO-1 protein molecule that were recognized by antibodies induced after vaccination. The 5 regions were also recognized by antibodies present in nonvaccinated, NY-ESO-1 antibody-positive cancer patients. A peptide spanning amino acids 91–108 was recognized in 6 out of 9 vaccinated patients and in 8 out of 9 nonvaccinated, sero-positive patients, being the most dominant antigenic epitope in NY-ESO-1 for antibody recognition in cancer patients. In conclusion, we showed that CHP-NY-ESO-1 protein vaccination had a potent activity for inducing humoral immune responses against NY-ESO-1 antigen in cancer patients. The antigenic epitopes recognized by antibodies in the vaccinated patients were similar to those recognized in cancer patients with spontaneous humoral immunity against NY-ESO-1. © 2007 Wiley-Liss, Inc.
The NY-ESO-1 antigen was originally identified in an esophageal cancer by serological expression cloning (SEREX) using autologous patient serum.1 Expression analysis revealed that it belonged to a class of cancer/testis (CT) antigens.2 Among CT antigens, NY-ESO-1 has received particular attention as a potential target for tumor vaccines because of its strong immunogenicity.3 A NY-ESO-1-specific humoral response is observed frequently in patients with various types of NY-ESO-1-expressing tumors. NY-ESO-1 expression was observed in 20–40% of melanoma lesions, and 50% of the patients with NY-ESO-1-expressing melanoma lesions were shown to induce NY-ESO-1 antibody.4 The frequencies of NY-ESO-1 expression and sero-positivity in patients with NY-ESO-1-expressing tumors varied depending on the tumor, e.g. they were 15 and 50%, respectively, in the case of prostate cancer,5 32–33% and 13% in the case of esophageal cancer,6, 7 10–40% and <3% in the case of breast cancer1,8-10 and 40 and 30% in the case of ovarian cancer.11 Spontaneous antibody induction was rarely observed in patients with tumors expressing other CT antigens, e.g., MAGE or GAGE.
Generation of NY-ESO-1-specific CD8 T-cells was observed in PBMC from sero-positive patients, suggesting that CD4 T-cell activation with NY-ESO-1 antigen which resulted in an antibody response would also much facilitate the CD8 T-cell response.12 Recently, NY-ESO-1 whole protein has been used for a cancer vaccine in some clinical trials.13 The use of the protein instead of antigenic peptides would be expected to stimulate CD4 T-cells, and induce antibody production and a CD8 T-cell response. In this regard, monitoring of the humoral immune response seemed to be useful for estimating the overall immune response.14 In addition, several antibody epitopes have been identified in the sequence of NY-ESO-1 protein.15, 16 The NY-ESO-1 antibody from sero-positive patients reacted with those synthetic peptides as strongly as with the whole protein. The use of those peptides for ELISA to detect the antibody would be beneficial to avoid possible reactions with contaminating bacterial products that might be present if the recombinant protein was used.
Cholesterol-bearing hydrophobized pullulan (CHP) is a newly developed antigen delivery vehicle that can be used to formulate nanoparticles, including protein antigens.17 Both CD8 and CD4 T-cells are efficiently activated by DCs pulsed with a complex of CHP and NY-ESO-1 (CHP-NY-ESO-1) in vitro.18 In this study, we immunized cancer patients with CHP-NY-ESO-1 as a polyvalent vaccine. We analyzed NY-ESO-1 antibody responses against synthetic NY-ESO-1 protein and overlapping peptides in CHP-NY-ESO-1 vaccinated patients. The response was also examined in nonvaccinated, sero-positive patients with NY-ESO-1-expressing cancers.
Materials and methods
Patients and sera
Advanced cancer patients including 4 patients with esophageal cancer (E-1, -2, -3 and -4), 4 patients with prostate cancer (P-1, -2, -3 and -4) and a patient with malignant melanoma (M-1) were enrolled in the clinical trial (protocol LUD 2002–2005 of the Ludwig Institute for Cancer Research, New York, NY). Nonvaccinated esophageal and lung cancer patients were those who visited the Hospitals of Osaka and Okayama Universities during 2002–2005, respectively. Sera were obtained from vaccinated patients before vaccination and 2 or 4 weeks after each immunization, cancer patients and healthy donors with written informed consent with the permission of the ethics committees of Osaka and Okayama Universities. Sera were stored at −40°C until use.
Preparation of a complex of cholesterol-bearing hydrophobized pullulan and NY-ESO-1 protein (CHP-NY-ESO-1)
Preparation of recombinant NY-ESO-1 protein for vaccine was described elsewhere.19 The complex of cholesterol-bearing hydrophobized pullulan (CHP) and NY-ESO-1 protein (CHP-NY-ESO-1) was synthesized as described previously.20
In the clinical trial, the patients were subcutaneously administered 100 μg of NY-ESO-1 recombinant protein formulated with 2 mg of CHP 4 times at biweekly intervals. Four weeks after the last administration, safety, immune response and tumor response were evaluated. Thereafter, the vaccine was administered additionally. The 9 enrolled patients received 4–12 immunizations.
Recombinant protein and overlapping peptides
Recombinant NY-ESO-1 protein used for assays was produced by using NY-ESO-1 cDNA (nt 68–607, GenBank accession number AJ275977) cloned into the SphI-SalI sites of pQE30 vector (QIAGEN, Hilden, Germany). N-His-tagged NY-ESO-1 protein was expressed in M15 E.coli cells and purified by nickel-ion affinity chromatography under denaturing conditions.
The following series of nine 30-mer NY-ESO-1 overlapping peptides was synthesized: 30.1 (1–30), 30.2 (21–50), 30.3 (41–70), 30.4 (61–90), 30.5 (81–110), 30.6 (101–130), 30.7 (121–150), 30.8 (141–170) and 30.9 (151–180). The following series of twenty-five 18-mer overlapping peptides was also synthesized: 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) and 18.28 (163–180). A 26-mer 18.25–27(145–170) was also included. These peptides were synthesized using an Fmoc solid-phase method and an ABIMED Multiple Peptide Synthesizer (AMS422) at Okayama University (Okayama, Japan).
One hundred microliters of 1 μg/ml recombinant protein or 10 μg/ml peptide solution in coating buffer (15 mM Na2CO3, 30 mM NaHCO3, pH 9.6) was added to each well of 96-well Polysorp immunoplates (Nunc, Roskilde, Denmark) and incubated overnight at 4°C. Plates were washed with PBS and blocked with 200 μl/well of 5% FCS/PBS for 1 hr at room temperature. After washing, 100 μl of serially diluted serum was added to each well and incubated for 2 hr at room temperature. After extensive washing, goat anti-human IgG (Medical & Biological Laboratories, Nagoya, Japan) or mouse anti-human IgM, IgG1, IgG2, IgG3 or IgG4-HRP (Southern Biotechnologies, Birmingham, AL) was added to the wells for secondary antibody, and the plates were incubated for 1 h at room temperature. After washing, signals were developed with 100 μl per well of 0.03% o-phenylene diamine dihydrochloride (OPDA), 0.02% hydrogen peroxide and 0.15 M citrate buffer and absorbance at 490 nm was read by using an ELISA reader (Benchmark Microplate Reader; Bio-Rad, Hercules, CA). Ovalbumin (OVA; albumin from chicken egg white, Sigma, St. Louis, MO) was used for control protein.
Plasmid construction and transfection
NY-ESO-1 cDNA was amplified by PCR and cloned into pcDNA3.1 expression plasmid vector (Invitrogen, Carlsbad, CA). 293T cells were transfected with plasmids using Lipofectamine 2000 reagent (Invitrogen). After incubation for 48 hr, lysate of transfected cells was used for Western blot analysis.
Recombinant NY-ESO-1 protein (20 ng) or cell lysate (12.5 μg) in sample buffer (100 mM Tris-HCl, pH 8.8, 0.01% bromophenol blue, 36% glycerol, 4% SDS, 1 mM dithiothreitol) was boiled for 5 min and subjected to SDS-PAGE with 10–20% polyacrylamide BioRad Ready-Gels (Bio-Rad). After electrophoresis, the membrane (Hybond-P membrane (PVDF), Amersham Pharmacia Biotech, Buckinghamshire, UK) was blocked with 3% BSA/TBS and then incubated with patients' sera diluted 1:100 at 4°C overnight. After washing, alkaline phosphatase conjugated goat anti-human IgG (Jackson Immuno Research Laboratory, West Grove, PA) was added to the membrane. Signals were developed with a 5-bromo-4-chloro-3-indolylphosphate-nitroblue tetrazolium chromogenic substrate kit (Bio-Rad).
Prediction of antibody epitopes using computer-assisted algorithm
The hydrophobicity of NY-ESO-1 was calculated based on the Kyte-Doolittle method. The prediction program ProtScale (http://www.expasy.ch/cgi-bin/protscale.pl) was used for percentage accessible residues and average area buried. The prediction program PROFacc at The Predict Protein server web site (http://cubic.bioc.columbia.edu/predictprotein/) of the Columbia University (New York, NY) was used for solvent accessibility.
NY-ESO-1 antibody response in CHP-NY-ESO-1 vaccinated patients
The NY-ESO-1 antibody response in the CHP-NY-ESO-1 vaccinated patients was investigated by ELISA using recombinant NY-ESO-1 protein. The patients included 7 baseline sero-negative and 2 baseline sero-positive patients. Figures 1a and 1b shows the ELISA values of sera from each patient after each time point of immunization. The NY-ESO-1 antibody response increased gradually with vaccination in all patients, including 2 sero-positive patients, E-2 and P-3. An increase of antibody response was observed by the fourth immunization in all patients and as early as after the first immunization in E-2 and P-3. During immunizations 3–7, antibody response reached a plateau, and it then decreased after the 10th immunization in E-1. As shown in Figure 1c, Western blot analysis showed that sera from all the vaccinated patients were reactive with recombinant NY-ESO-1 protein as well as with the lysate of NY-ESO-1-expressing 293T cells.
Antibody response against CHP itself was observed in only E-4, but not in other patients (data not shown).
Dominant IgG1 response against NY-ESO-1
The level of the IgM and IgG subclass of the NY-ESO-1 antibody was estimated in each patient by ELISA using various second antibodies. As shown in Table I, the IgG1 subclass was dominantly induced in all patients after CHP-NY-ESO-1 vaccination. Weak IgM and IgG3 antibody response was observed in E-2, E-4, P-4 and E-2, E-4, respectively, after vaccination. Neither IgG2 nor IgG4 response was observed in any patient. A dominant IgG1 response was also observed in the 2 baseline sero-positive patients before vaccination, and in all 9 nonvaccinated, sero-positive patients with NY-ESO-1-expressing cancers investigated.
|Patient||Immunization||Serum dilution||IgM||Serum dilution||IgG subclass|
|CHP-NY-ESO-1 vaccinated patients|
|Nonvaccinated, sero-positive patients|
Antibody epitope analysis using 30- and 18-mer series of overlapping peptides
NY-ESO-1 antibody epitopes were analyzed using synthetic 30- and 18-mer series of overlapping peptides. As shown in Figure 2, peptides 30.1 (1–30), 30.2 (21–50), 30.3 (41–70), 30.5(81–110) and 30.9 (151–180) were recognized by 3, 3, 2, 6 and 1 vaccinated patients, respectively. Analysis using 18-mer series of overlapping peptides revealed that peptides 18.1 (1–18)–18.4 (19–36) which corresponded to peptide 30.1 (1–30), peptides 18.6 (31–48)–18.9 (49–66) which corresponded to 30.2 (21–50) and 30.3 (41–70), and peptides 18.16 (91–108)–18.17(97–114) which corresponded to 30.5 (81–110), were recognized by 3, 3 and 6 vaccinated patients, respectively. Peptide 18.28 (163–180), which is a C-terminal 18-mer peptide and corresponded to 30.9 (151–180), was recognized by 1 vaccinated patient. Sera from E-3, P-1 and M-1 showed a weak reaction even with an increased peptide concentration of 50 μg/ml instead of the 10 μg/ml routinely used for ELISA.
Similar antibody epitope analysis was done using sera from the 9 nonvaccinated, sero-positive esophageal and lung cancer patients. As shown in Figure 3, the antibody epitopes recognized by vaccinated patients shown in Figure 2 were also recognized by nonvaccinated, sero-positive patients with NY-ESO-1-expressing cancers investigated. Only peptides 30.4 (61–90) and 18.11 (61–78) were recognized by 2 and 1, respectively, among 9 nonvaccinated, sero-positive patients, but not 9 vaccinated patients.
Peptides frequently and strongly recognized by both vaccinated and nonvaccinated, sero-positive patients were 30.1 (1–30), 30.5 (81–110) and 18.16 (91–108).
Epitope analysis of sera from baseline sero-positive patients before and after vaccination
E-2 and P-3 were baseline sero-positive patients. To investigate whether CHP-NY-ESO-1 vaccination enhanced the induction of the antibody which had already been elicited against naturally processed epitopes derived from NY-ESO-1 protein in tumors and/or induced the generation of antibody against additional epitopes, sera obtained from those patients at baseline, and after the 2nd, 4th and 8th immunizations were tested against 30-mer peptides by ELISA. As shown in Figure 4, the antibody response against natural epitopes was enhanced and no antibody response against an additional epitope was detected.
Prediction of antibody epitopes using computer-assisted algorithm
We predicted the relative solvent accessibility of NY-ESO-1 protein using the program PROFacc at the Columbia University Predict Protein web site. As shown in Figure 5, multiple regions accessible by the antibody were identified, including the 5 regions defined in this study.
In this study, we showed that CHP-NY-ESO-1 vaccination elicited an NY-ESO-1 antibody response in all 9 patients with NY-ESO-1-expressing tumors as detected by ELISA using recombinant NY-ESO-1 protein. The patients included 7 baseline sero-negative and 2 baseline sero-positive patients. Davis et al.13 similarly reported that ISCOMATRIX/NY-ESO-1 vaccination elicited an NY-ESO-1 antibody response in patients with NY-ESO-1-expressing tumors. In their study, however, no augmentation of NY-ESO-1 antibody response was observed in 3 baseline sero-positive patients. The dose of NY-ESO-1 protein used for vaccination in their study was 100 μg, the same as that used in this study. Although the exact reason for the discrepancy is unknown, it is possible that the antibody present in the patients bound to NY-ESO-1 molecules in the ISCOMATRIX formulation and reduced the immunogenicity.13 Alternatively, it could be due to the difference of stability of the formulations. We previously showed that an antibody response against NY-ESO-1 protein was observed only in patients with tumors highly expressing NY-ESO-1 by examining NY-ESO-1 mRNA expression in tumors quantitatively by real-time RT-PCR using TaqMan probe and protein expression by IHC.7, 10 Thus, CHP-NY-ESO-1 vaccination overcame the insufficient immunogenicity of NY-ESO-1 protein naturally expressed in most tumors and elicited an antibody response. The CHP-NY-ESO-1 formulation appeared to be the most effective of any tested to date for inducing a serological response against NY-ESO-1.
A CD4 T-cell response is necessary for an IgG response.21 Gnjatic et al.12 reported multiple epitopes that bind to various HLA class II molecules in NY-ESO-1 protein. We recently showed that CHP-NY-ESO-1 was efficiently processed in the class II pathway and was also cross-presented via the class I pathway when pulsed to immature DCs in vitro.18 It appeared to be likely that the formulation of NY-ESO-1 protein in CHP-NY-ESO-1 prevented the rapid degradation of the protein by proteases, and therefore augmented NY-ESO-1 immunogenicity in vivo.17 Furthermore, the epitope analysis in 2 sero-positive patients before and after CHP-NY-ESO-1 vaccination shown in Figure 4 indicated that the antibody response against the epitopes to which antibody had already been induced before vaccination was augmented and no recognition of additional epitopes occurred following vaccination. The findings suggested that similar antigenic epitopes were recognized by immunization with the NY-ESO-1 antigen naturally present in tumors and CHP-NY-ESO-1.
Epitope analysis using overlapping 30- and 18-mer series of peptides revealed 5 regions in the NY-ESO-1 molecule recognized by the antibodies in vaccinated and nonvaccinated, sero-positive patients. Peptide 30.1 (1–30), which corresponded to peptides 18.1 (1–18)–18.4 (19–36), was recognized in 3 of 9 vaccinated patients and all 9 nonvaccinated, sero-positive patients. Valmori et al.15 and Zeng et al.16 showed that this N-terminal region contained a dominant antibody epitope recognized in various cancer patients. Our finding is consistent with theirs, and suggested a rather lower frequency of antibody response to this region in vaccinated patients than in nonvaccinated, sero-positive patients. The second epitope region consisted of peptides 30.2 (21–50) and 30.3 (41–70), and 18.6 (31–48)–18.9 (49–66). Since the responses to these peptides frequently occurred simultaneously in vaccinated and nonvaccinated, sero-positive patients, the peptide region commonly recognized would be the peptide 41–50. The third region was the peptide 18.11 (61–78). This was recognized only in EC5, with no history of autoimmune or severe infectious disease. These second and third epitope regions were firstly shown in this study. The fourth region was the peptides 30.5 (81–110) and 18.16 (91–108) recognized in 6 out of 9 vaccinated patients and 8 out of 9 nonvaccinated, sero-positive patients with strong reaction in ELISA, suggesting that it is the most dominant antigen in the antibody response. The fifth region was the peptide 163–180. The fourth and the fifth regions have also been reported by Valmori et al.15 as recognized by the antibodies from cancer patients.
We predicted the relative solvent accessibility of NY-ESO-1 protein using the program PROFacc at the Columbia University Predict Protein web site. As shown in Figure 5, the 5 regions defined in this study, peptides 1–30, 41–50, 61–78, 91–108 and 163–180, were included in the accessible regions identified by this method. The dominant region peptide 18.16 (91–108) shown as solvent accessible in this study had been shown to be located in the hydrophobic region by hydrophilicity plot analysis using the Kyte-Doolittle method by Zeng et al.16 The solvent accessibility analysis of the protein using the program PROFacc appeared to be useful to analyze epitopes.
The dominant subclass induced in the antibody response against NY-ESO-1 protein was IgG1 in both vaccinated and nonvaccinated, sero-positive patients. IgG1 in the human corresponds to IgG2a in the mouse, of which the induction results from Th1 type CD4 T-cell activation,22 providing favorable conditions for effector T-cell activation in the immune response to NY-ESO-1. NY-ESO-1 is a nuclear antigen and therefore it is unlikely that the antibody itself will have some effect on tumor cells.
In conclusion, we showed that a formulation of CHP-NY-ESO-1 protein had a potent activity for inducing NY-ESO-1 antibody at a dose of 100 μg. The regions in the NY-ESO-1 molecule recognized by the antibody in the vaccinated patients were similar to those recognized by the antibody in nonvaccinated, sero-positive patients. Moreover, 5 regions that were recognized by the antibody were identified in the NY-ESO-1 molecule. It would be important to determine if there is a relation between epitope regions recognized by B-cell and those recognized by T-cells. The cellular immune responses are now being analyzed and will be published elsewhere.
We thank Ms. T. Akimoto for excellent technical assistance and Ms. J. Mizuuchi for preparation of the manuscript.
- 17Presentation of a major histocompatibility complex class 1-binding peptide by monocyte-derived dendritic cells incorporating hydrophobized polysaccharide-truncated HER2 protein complex: implications for a polyvalent immuno-cell therapy. Blood 2002; 99: 3717–24., , , , , , , , , , , , et al.
- 18In vitro stimulation of CD8 and. CD4 T cells by dendritic cells loaded with a complex of cholesterol-bearing hydrophobized pullulan and NY-ESO-1 protein: identification of a new. HLA-DR15-binding. CD4 T-cell epitope. Clin Cancer Res 2006; 12: 1921–7., , , , , , , , , , , , et al.