Clinical Cancer Research Prevention Award Advances in Breast Cancer
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
Clinical Cancer Research 13, 7199, December 1, 2007. doi: 10.1158/1078-0432.CCR-07-0682
© 2007 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Couch, M. E.
Right arrow Articles by Sidransky, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Couch, M. E.
Right arrow Articles by Sidransky, D.

Cancer Therapy: Preclinical

Alteration of Cellular and Humoral Immunity by Mutant p53 Protein and Processed Mutant Peptide in Head and Neck Cancer

Marion E. Couch1, Robert L. Ferris2, Joseph A. Brennan3, Wayne M. Koch4, Elizabeth M. Jaffee5, Michael S. Leibowitz2, Gerald T. Nepom6, Henry A. Erlich7 and David Sidransky4,5

Authors' Affiliations: 1 Department of Otolaryngology–Head and Neck Surgery, University of North Carolina, G0412 Neurosciences Hospital, Chapel Hill, North Carolina; 2 Departments of Otolaryngology and of Immunology, University of Pittsburgh, Pittsburgh, Pennsylvania; 3 Department of Otolaryngology–Head and Neck Surgery, Wilford Hall Medical Center, Lackland Air Force Base, San Antonio, Texas; 4 Department of Otolaryngology–Head and Neck Surgery and 5 Department of Oncology, Medicine, and Molecular Biology and Genetics, Johns Hopkins Hospital, Baltimore, Maryland; 6 Virginia Mason Research Center and the University of Washington, Benaroya Research Institute, Seattle, Washington; and 7 Roche Molecular Systems, Department of Human Genetics, Alameda, California

Requests for reprints: Robert L. Ferris, Departments of Otolaryngology and Immunology, 5117 Center Avenue, Room 2.26b, Pittsburgh, PA 15213. Phone: 412-623-7738; Fax: 412-623-7768; E-mail: ferrisrl{at}upmc.edu or David Sidransky, Department of Otolaryngology–Head and Neck Surgery, 6252 JHOC, Baltimore, MD 21287. E-mail: dsidrans{at}jhmi.edu.


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 References
 
Purpose: To determine if serologic recognition of p53 mutations at the protein level depends upon the ability of mutant p53 to express new peptide epitopes that bind to human leukocyte antigen (HLA) class II molecules, we used anti-p53 antibody production as a marker for HLA class II–restricted T-cell involvement in head and neck cancer.

Experimental Design: An anti-p53 antibody response was correlated with specific p53 mutations and the patients' HLA class II alleles and haplotypes. HLA binding studies and in vitro stimulation (IVS) of peripheral blood mononuclear cells were done using a mutant versus wild-type HLA-DQ7-binding p53 peptide.

Results: Certain HLA-DQ and HLA-DR alleles were frequently present in p53 seropositive patients who produced serum anti-p53 antibodies. Selected mutated p53 peptides fit published allele-specific HLA class II binding motifs for the HLA-DQ7 or HLA-DR1 molecules. Moreover, a mutant p53 peptide bound with a 10-fold greater affinity than the wild-type p53 peptide to HLA-DQ7 molecules. IVS of CD4+ T cells from seven healthy HLA-DQ7+ donors using this mutant p53 peptide (p53220C) was associated with a partial T helper type 2 phenotype compared with IVS using the wild-type p53210-223 peptide.

Conclusions: Our results support the hypothesis that mutated p53 neoantigens can bind to specific HLA class II molecules, leading to a break in tolerance. This may lead to skewing of the CD4+ T lymphocyte response toward a tumor-permissive T helper type 2 profile in head and neck cancer patients, as manifested by seropositivity for p53.

The immune system seems to recognize at least six categories of tumor-associated antigens: differentiation antigens, reactivated embryonic gene products, oncogene or suppressor gene products, mutated gene products, viral gene products, and idiotypic epitopes (1, 2). Tumor suppressor gene products, with mutations that may lead to new epitope formation, comprise another source of tumor antigens that are not on normal tissues. A member of this category of tumor antigens, the oncoprotein p53, may represent an attractive target for immunotherapy strategies (3) because this polypeptide chain of 393 amino acids is mutated in over 50% of all human cancers.

Mutated p53 peptides have been shown to bind to specific HLA class I molecules and generate CTLs capable of lysing head and neck squamous cell carcinoma cells containing p53 mutations in an HLA class I–restricted, tumor peptide–specific manner (4–7). There is also significant evidence that HLA class II–restricted antitumor responses are able to effectively contribute to antitumor immunity (8–10). Tumor burden and p53 seropositivity have been associated with a tumor permissive T helper type 2 (Th2) skewing of the CD4+ T-cell response in cancer patients (11) as manifested by low secretion of IFN-{gamma} and high secretion of interleukin-4 (IL-4), IL-5, and IL-13 (12, 13), However, the precise role of mutated tumor antigenic peptides in mediating such an effect on antitumor antigen immunity has not been sufficiently clarified.

Antibody production directed against mutated p53 has been found in the sera of patients with a wide variety of tumors, including breast, pancreas, lung, colon, and head and neck (14–18). The generation of these anti-p53 antibodies has previously been attributed to the presence of increased circulating levels of mutated p53 oncoprotein (19), binding of heat shock proteins to the mutated proteins (20, 21), and mutations in certain exons (22). Tumor antigen seropositivity is also thought to be reflective of a Th2 skewing of the CD4+ T-cell response in cancer patients. Thus, we used anti-p53 antibody production as evidence for HLA class II–restricted Th cell activation and investigated the relationship between T helper profiles in healthy donor peripheral blood mononuclear cells (PBMC) stimulated with a mutant p53210-223 peptide (Y->C at position 220). In this way, a form of immune escape mediated by p53 mutation may lead to presentation of a tumor antigen capable of shaping a tumor permissive immune response via HLA class II binding. Deleterious effects on antitumor immune responses may result from mutant tumor peptide specific, HLA class II antigen–restricted stimulation in cancer patients.


   Materials and Methods
Top
 Abstract
 Materials and Methods
Results
 Discussion
 References
 
Sera. Serum was collected from each of 43 head and neck cancer patients during the perioperative period (within 3 months preoperatively or postoperatively). Blood was allowed to clot and then was spun for 15 min at 2,500 rpm in a chilled centrifuge. Serum was then decanted into a cryostat tube that was stored in a –80°C freezer.

Sequence analysis. Tumors from each of the 43 patients were immediately frozen in liquid nitrogen and stored in a –80°C freezer. To remove normal tissue and inflammatory cells, the frozen specimens were microdissected, and the DNA was isolated as previously described (23). Exons 5 through 9 of the p53 gene, contained in a 1.8-kb fragment, were amplified using the PCR, then cloned, and sequenced. All reactions were repeated to confirm the results.

Immunoprecipitation. Both mutant (248, 143) and wild-type (SN3) p53 plasmids (the kind gift of B. Vogelstein) were cloned into Bluescript plasmid vectors downstream from the T7 promoter site. After growth in Escherichia coli, the DNA was extracted using the boiling lysis method then linearized with restriction enzymes for in vitro transcription using T7 RNA polymerase (Riboprobe kit, Promega Biotech). The presence of RNA was confirmed with a MOPS minigel before in vitro translation in a rabbit reticulocyte lysate system (Promega). The protein generated was labeled with m7G(5')ppp(5')G (Pharmacia). Patient serum (5 µL) was mixed with the translation product (5-10 µL) in radioimmunoprecipitation assay buffer–bovine serum albumin buffer, vortexed, and incubated for 2 h at room temperature. Staph Protein A-Sepharose (Pharmacia; 1.5 mg/25 µL radioimmunoprecipitation assay buffer–bovine serum albumin) was then added, and this slurry was incubated on a shaker for 30 min at room temperature. After centrifugation, the supernatant was discarded, and the Sepharose beads were extensively washed in sequential buffer solutions (radioimmunoprecipitation assay buffer 1 mol/L urea, radioimmunoprecipitation assay buffer 1 mol/L NaCl, and radioimmunoprecipitation assay buffers). Sample buffer was added, and the Sepharose beads were then loaded onto a SDS-PAGE gel and run at 150 V. The gels were fixed, dried, and treated with Amplify (Amersham). The autoradiographs were exposed to film overnight in –80°C freezer, after which the film was developed. A murine monoclonal antibody (mAb) to I-CAM was used as a negative control, and a murine mAb NCL-p53-1801 (Novacastra Labortories Ltd., Newcastle upon Tyne, United Kingdom), which reacts with both mutant and wild-type p53, served as a positive control. All assays were repeated to confirm the presence of antibodies.

Western blotting. Various evolutionarily conserved domains of the p53 protein were used to create p53 fragments that were expressed as glutathione S-transferase fusion proteins in E. coli. Five of these constructs, which encompassed different combinations of these domains, and an unrelated protein, Oct 1 pou, were used in this study. The plasmid pET11GTK, containing the glutathione S-transferase gene and DNA encoding portions of p53, was transformed in E. coli BL21 (DE3) cells as previously described (24). Cultures were inoculated from frozen stock and grown at 25°C in M9ZB culture medium supplemented with 100 µg/mL ampicillin in 0.4 mmol/L isopropyl-β-D-thiogalactopyranoside for induction of the protein. The cultures were centrifuged and resuspended in ice-cold lysis buffer as previously described. The fusion proteins were extracted by treating the cultures with lysozyme (100 µ/mL), multiple freeze-thaw cycles, and then sonication. The extracts were centrifuged and then aliquoted for storage in –80°C freezer. These proteins were bound to glutathione-Sepharose beads (Pharmacia), and 100 µL of this slurry were then incubated with 10 µL of patient sera. After incubation for 1 h at 4°C, the beads were washed thrice in HNN buffer and then loaded onto an SDS-PAGE gel. The gel was transferred to a polyvinylidene difluoride membrane using the Milliblot-SDE Transfer system. Secondary antibody consisting of goat anti-human peroxidase was incubated with the membrane (1:5,000 in 5% Blotto) for 1 h. The unrelated protein, Oct 1 pou, was used as a negative control and sera with anti-p53 served as a positive control. Enhanced chemiluminescence solution (Amersham) was used to develop the chemiluminescence on the membrane. Film was placed over the membrane and exposed for 1 min and then developed.

Haplotype determination. DNA (1-10 µL) isolated from the tumors was PCR-amplified for the following class II loci: DRB1, DRB2, DQB1, DQB2. The DNA was then probed with horseradish peroxidase–labeled oligonucleotide as previously described (25).

Competitive binding assay to HLA-DQ7. To investigate the binding of synthetic mutant and wild-type p53 peptides to the purified HLA-DQ7 (DQ3.1; DQA1*0301, DQB1*0301) molecule, a competitive binding assay was done as previously described (26). p53 peptides were used only if the patient possessed the HLA-DQ7 genotype. The peptide AYK functioned as a positive control, as it binds well to the purified DQ7 molecule with a known IC50 of <0.1 µmol/L. The negative control, peptide 34P, bound with an IC50 of >10 µmol/L.

In vitro stimulation of HLA-DQ7+ healthy donor PBMC. CD4+ T-cell cultures were generated from HLA-DQ7+ healthy donors identified by DNA-based typing at the University of Pittsburgh Immunogenetics facility, using PBMC as described (7, 27) with the following modifications. Monocytes were separated with a Percoll (Pharmacia) gradient, selected by plastic adherence for 2 h at 37°C, and cultured for 6 days in 10% human antibody serum (Mediatech) AIM-V with granulocyte macrophage colony–stimulating factor (100 ng/mL) and IL-4 (100 ng/mL). The resulting immature dendritic cells (DC) were matured for 24 h at 37°C in a medium supplemented with IFN-{alpha} (1000 units/mL), IFN-{gamma} (1000 units/mL), IL-1β (25 ng/mL), and tumor necrosis factor-{alpha} (50 ng/mL). Wild-type p53210-223 or mutant p53220C peptide was added at 1 µmol/L for the final 1 h of incubation. CD4+ PBMC obtained by negative selection using an antibody-coated MACS bead (Miltenyi Biotec) cocktail, including CD8, CD14, CD16, CD19, CD36, CD56, CD123, TCR {gamma}/{delta}, and glycophorin A. The isolated CD4+ T cells (>97% purity) were resuspended in AIM-V medium supplemented with 2% human antibody serum and 7% fetal bovine serum.

Multiplexed ELISA (Luminex) assays. Twenty-four hours post restimulation of PBMC, supernatants were collected and analyzed using Luminex and the human cytokine Ten-plex panel kit (Invitrogen, Carlsbad, CA), which measured IL-1β, IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, granulocyte macrophage colony-stimulating factor, tumor necrosis factor-{alpha}, and IFN-{gamma}. LabMAP technology (Luminex) combines the principle of a sandwich immunoassay with the fluorescent bead–based technology, allowing multiplexed analysis of up to 100 different analytes in a single microtiter well (Biosource, Inc.). The differentially dyed polystyrene microspheres serve as a solid support for this "sandwich" immunoassay. Usual interassay variability within the replicates is 3.5% to 7%, and intraassay variability is 11% to 15%, as validated with appropriate ELISA with an 89% to 98% correlation (28, 29).

ELISPOT assays. IFN-{gamma} or IL-5 ELISPOT assays were done in triplicate as described previously (7, 13). In all the experiments, autologous PBMC pulsed with an irrelevant peptide or with the p53 peptide (wild type or mutant) were used, respectively. Phorbol 12-myristate 13-acetate/ionomycin-treated CTL were used as a control for CTL secretion of IFN-{gamma}. Plates were read using a Carl Zeiss VISION ELISPOT reader. Background was considered as the number of spots secreted by CD4+ T cells alone. T-cell reactivity, as measured by the ELISPOT assay, was considered positive if the number of spots in test wells was significantly higher than that in background wells when using a two-tailed permutation test at {alpha} ≤ 0.05. Precursor frequency of unstimulated cells was 10 to 15 cells per well for IFN-{gamma} and negligible (0-2 cells per well) for IL-5. To determine the HLA class II antigen restriction of the CTL reactivity, target cells were incubated with either HLA-A, HLA-B, HLA-C–specific mAb W6/32 (30) or HLA-DR, HLA-DP, HLA-DQ antigen–specific mAb (LGII-612.14; ref. 31) or without any mAb. The precursor frequency of unstimulated cells for IFN-{gamma} was 10 to 15 cells per well, and IL-5 was negligible (0-2 cells per well). Each mAb was used at the final concentration of 10 µg/mL and was incubated for 30 min at 37°C before the addition of the effector CTL.


   Results
Top
 Abstract
 Materials and Methods
 Results
Discussion
 References
 
Correlation of antibody production with p53 mutations. To determine whether the mutated p53 protein could express new epitopes that would be predicted to bind the patient's HLA class II molecules, the p53 gene was sequenced (exons 5-9) in a series of 43 tumors from patients with head and neck squamous cell carcinoma. This revealed that 15 patients had mutations, mostly missense mutations, although one patient had a frameshift mutation (patient 62; Table 1 ). In our patient population, the mutations were present in multiple locations, including exons 6, 7, and 8.


View this table:
[in this window]
[in a new window]

  		Table 1. Correlation of patient's anti-p53 antibody production with p53 mutations and immunodominant epitopes

 
The ability of these patients to generate anti-p53 antibodies served as a marker for class II CD4+ Th2 helper lymphocyte activation. Immunoprecipitation and Western blot assays allowed identification of the p53 mutations that were associated with an immune response. These techniques did not cause denaturation of the p53 protein before the addition of the patient's sera. Both wild-type and mutant forms of p53 were reacted with the serum in the immunoprecipitation assay. The reactive serum determined by immunoprecipitation had antibodies that reacted with both mutant and wild-type forms of p53 (Fig. 1 ). Seven patients (of 43; 16%) had circulating anti-p53 antibodies as detected by the Western blot technique (Table 2 ). Of the patients with p53 mutations, approximately half (7 of 15 or 47%) had detectable antibody to p53 (Table 1).


Figure 1
View larger version (32K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 1. Patient's anti-p53 antibodies in immunoprecipitation assays react with both mutant (248) and wild-type (SN3) forms of the p53 protein. As a positive control, a murine monoclonal antibody to p53 was reacted with the wild-type p53 (lane a). Monoclonal antibody to I-CAM was reacted with mutant p53 as a negative control (lane b). Patient's sera without anti-p53 did not bind the labeled p53 (lanes c-f); however, sera from patients with p53 missense mutation bound both mutant and wild-type p53 (lanes g-h), respectively.

 

View this table:
[in this window]
[in a new window]

  		Table 2. Correlation of patients' antibody production and p53 mutations with class II haplotypes

 
Determination of immunodominant p53 epitopes. The evolutionarily conserved domains of the p53 protein were used as fusion proteins bound to glutathione-Sepharose beads in a sensitive Western blot assay. This technique allowed identification of possible immunodominant epitopes on the p53 molecule because the sera reacted with fragments of the p53 molecule. Sera from seven patients bound to various p53 constructs. No immunodominant epitopes could be localized in this patient population because antibodies to p53-mutated oncoproteins reacted with all domains of the p53 molecule, not just the amino and/or carboxy terminus. As seen in Fig. 2 and Table 1, internal domains (2C, 23), as well as more terminal domains (N5, NC), bound p53 antibody.


Figure 2
View larger version (57K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 2. Western blot assay using fragments of p53 as fusion proteins. No immunodominant regions were found. Examples of two patient's reactions. A, patient 296 reacted with the 35 and N5 fusion proteins. B, patient 309 had antibodies that recognized NC and N5 proteins. Oct 1 pou was an unrelated protein that served as a negative control.

 
Class II alleles and anti-p53 antibody production. The possible association between class II HLA major histocompatibility antigen expression and the immune response to p53 oncoproteins was assessed by genotyping 43 patients with head and neck squamous cell carcinoma (Table 3 ). The HLA-DQ, HLA-DR, and HLA-DP alleles and haplotypes were determined. The DPB1 alleles did not seem to have any association with antibody production. However, many of the patients who produced antibodies were found to have the DQB1*0301 (three of seven patients), DRB1*0701 (three of seven patients), DQB1*0501 (three of seven patients), and DRB1*0101 (two of seven patients) alleles. Patients with p53 mutations who did not produce antibodies frequently had the DQB1*0201 (four of eight patients) and the DRB1*0701 (four of eight patients) alleles.


View this table:
[in this window]
[in a new window]

  		Table 3. Competitive binding assay of p53 mutant peptides to DQ7 (DQA1*0301, DQB1*0301, DQB 3.1)

 
Matching of p53 mutation sequences to HLA class II binding motifs. Several studies (32) have elucidated the detailed allele-specific peptide binding motifs for some of the class II molecules. Therefore, a search was undertaken for the best possible fit of mutant p53 proteins with these published class II ligand motifs. In Falk's study, the allele DQB1*0301 ligand motif was derived from pool sequencing natural peptides of the DQA1*0501-DQB1*0301 molecule. In this particular motif, there are three important hydrophobic or aromatic anchor residues at positions 1, 5, and 7. In addition, peptide ligands usually possess charged and polar residues in clusters, as well as proline residues at positions 2, 11, 12, 13, and 14 (19). Because every patient with a known class II haplotype had their tumor DNA sequenced for p53 gene mutations, the mutant amino acid residue sequences (14-mers) could be matched with some of the allele-specific class II ligand motifs for best possible fit. All the mutant p53 peptides that fit this class II allele binding motif were also the mutant p53 molecules associated with antibody production (Fig. 3A ). The converse was also true.


Figure 3
View larger version (13K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 3. A, HLA-DQ7 (DQB1*0301) binding motif (1) and p53 mutant peptide sequences. The mutated amino acid residues are in italics and underlined. Amino acids that match predicted anchor residues are in bold letters. If the peptides were predicted to bind two or more anchor residues, antibody production was predicted. B, HLA-DR1 (DRB1*0101) binding motif and p53 mutant peptide sequences. The mutated amino acid residues are in italics and underlined. Amino acids that match predicted anchor residues are in bold letters. All patients had the DRB1*0101-DQB1*0501 haplotype. In patients wherein more than one anchor residue was predicted to bind, antibody production was seen (patients 66 and 296). Conversely, p53 serum antibodies were not seen in the patient where less than two anchor residues bound the motif (patient 245).

 
Three of the patients with known p53 mutations possessed the same DRB1*0101-DQB1*0501 haplotype. The peptide sequences containing the mutations were matched via the best possible fit to the allele-specific HLA-DR1 (DRB1*0101) binding motif published by Falk et al. (32). Anchor residues were located at positions 1, 4, and 9 on this binding motif. In Fig. 3B, the mutated p53 peptides (from patients 66 and 296) that matched two of three residues on the DRB1*0101 binding motif were associated with antibody production. The peptide (patient 245) that was predicted to bind only one of three anchor residues was not associated with anti-p53 production.

Competitive binding of p53 mutant peptides to HLADQ7 (DQA1*0301, DQB1*0301, DQ 3.1). Two patients (patients 40 and 168) with the same DQA1*0301-DQB1*0301 haplotype had tumors with a p53 missense mutation (tyr->cys) in codon 220 (exon 6). Their mutated p53 peptide (14-mer) was sequenced and used in a competitive binding assay using the DQ7 molecule DQ3.1. It would be predicted that binding of the synthetic mutated p53 peptides with the DQ7 molecule would be seen if class II presentation was involved in determining antibody production. Not only was this the case but also the mutated peptide sequence (NTFRHSVVVPCEPP) bound to the DQ7 molecule with higher affinity than the wild-type p53 peptide (NTFRHSVVVPYEPP; Fig. 3A). The substitution of a cysteine for a tyrosine in the mutated p53 peptide resulted in a 10-fold greater binding to the DQ7 molecule as shown by an IC50 of 0.58 µmol/L. The wild-type peptide had an IC50 of 4.5 µmol/L. In addition, patient 69 possessed the DQB1*0301 allele (DRB1*1101-DQB1*0301) but did not produce anti-p53 antibody. The mutated p53 peptide (MNRRPNLTIITLED) was also sequenced then used in the same competitive binding assay. As predicted by this model, the p53 mutated peptide (at codon 251) and the wild-type peptide (MNRRPILTIITLED) did not function as a ligand for this class II molecule. IC50 values were both >10 µmol/L. Patient 226 (DRB1*0407-DQB1*0301) produced anti-p53 antibodies and possessed a p53 mutation at codon 272. When the mutated peptide sequence (14-mer) was used in the competitive binding assay to the DQ7 molecule, no binding was found by either the mutated (NSFEERVCACPGRD) or wild-type p53 peptide (NSFEVRVCACPGRD). In this case, the substitution of a glutamic acid residue for a glycine residue did not alter binding to the DQ7 molecule.

HLA-DQ7–binding mutant p53220C peptide stimulation induces a partial Th2 cytokine phenotype in healthy donor CD4+ PBMC. The cytokine phenotype induced by stimulation of CD4+ T cells with either the wild-type (p53210-223) or mutant (p53220C) peptide was compared in ELISPOT assays. Healthy donor PBMCs were used because tolerance to p53 is similarly induced in the thymus of healthy donors and cancer patients, and the HLA-DQ7+ head and neck squamous cell carcinoma patients were deceased and PBMC were unavailable for these studies. Thus, CD4+ T cells from seven healthy, HLA-DQ7+ donors were magnetic-bead isolated, and in vitro stimulation (IVS) was done for 1 week using the wild-type p53210-223 or mutant p53220C peptide. To ensure that results were not skewed by the induction conditions, autologous DC or PBMC stimulator cells were used. IFN-{gamma} or IL-5 ELISPOT assays were then done, as shown in Fig. 4A and B , using autologous PBMC target stimulators pulsed with either p53 peptides or irrelevant surface peptides, respectively. As shown in Fig. 4A, significantly lower IFN-{gamma} release was observed in CD4+ T cells stimulated with the mutant p53220C peptide compared with the same CD4+ PBMC stimulated after one round of IVS using DC pulsed with the wild-type p53210-223 peptide (P ≤ 0.05, one-tailed permutation test). To determine whether this effect was influenced by the induction conditions, IVS was also done using autologous HLA-DQ7+ PBMC as stimulators, pulsed with either wild-type (p53210-223) or mutant p53220C peptides. These experiments yielded a similar pattern of cytokine release as IVS using DC stimulators, but lower overall spots were observed using PBMC (not shown). IL-5 ELISPOT assays using CD4+ T cells stimulated with the mutant (p53220C) peptide yielded HLA Class II–restricted IL-5 spots (P ≤ 0.05). The precursor frequency of unstimulated cells for IFN-{gamma} was 10 to 15 cells per well, and IL-5 was negligible (0-2 cells per well). These results were observed in separate HLA-DQ7+ PBMC IVS cultures repeated at least twice for each donor tested (n = 7).


Figure 4
View larger version (21K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 4. IFN-{gamma} and IL-5 ELISPOT assays to measure CD4+ T-cell responses in healthy HLA-DQ7+ donor PBMC after stimulation with wild-type p53210-223 or mutant p53220C peptides. A, after 1 wk of IVS using autologous, mature DC pulsed with the wild-type p53210-223 or p53220C peptide, respectively. HLA-DQ7+ healthy CD4+ T cells were assayed using IFN-{gamma} ELISPOT assays. Stimulator cells in these assays were autologous, mature day 7 DC loaded with the same wild-type p53210-223 or p53220C peptides. Pretreatment with either HLA-A, HLA-B. HLA-C–specific mAb W6/32 (30), HLA-DR–specific mAb (L243), or HLA-DR, HLA-DP, HLA-DQ antigen–specific mAb (LGII-612.14) or without any mAb was done to block the recognition of CD4+ T cells as described in Materials and Methods. Columns, mean spots; bars, SD. A representative of experiments repeated at least twice with four separate HLA-DQ7+ healthy donor PBMC (P < 0.05). B, after 1 wk of IVS using autologous, mature DC pulsed with the wild-type p53210-223 or p53220C peptide, respectively. HLA-DQ7+ healthy CD4+ T cells were assayed using IL-5 ELISPOT assays. Stimulator cells in these assays were autologous, mature day 7 DC loaded with the same wild-type p53210-223 or p53220C peptides. Pretreatment with either HLA-A, HLA-B, HLA-C–specific mAb W6/32 (30), HLA-DR–specific mAb, (L243) or HLA-DR, HLA-DP, HLA-DQ antigen–specific mAb (LGII-612.14) or without any mAb was done to block the recognition of CD4+ T cells as described in Materials and Methods. Columns, mean spots; bars, SD. A representative of experiments repeated twice with four separate HLA-DQ7+ healthy donor PBMC (P < 0.05).

 
Multiplex Luminex cytokine assays were then done to measure the global cytokine profiles secreted by CD4+ T cells, as listed in Materials and Methods, stimulated with either the mutant or wild-type p53 peptide. Although IL-4 secretion was consistently higher in a subset of donor cells after mutant (p53220C) peptide stimulation, this difference was not statistically significant due to the sample size. Secretion of IL-4 and IL-5 (Fig. 4B) support a partial Th2 bias under these conditions. Interestingly other Th1 cytokines (IFN-{gamma}, tumor necrosis factor-{alpha}) were not detected in these cultures (data not shown), confirming a lack of Th1 bias often seen using DC-based IVS procedures (33). The lower limit of the assay may reflect insufficient sensitivity to detect cytokines produced by a low number of cells, such as the IL-5 secretion T cells portrayed in Fig. 4B. The quantity of IL-6 and IL-8 secreted by the CD4+ T cells derived from the three healthy donors were not consistent with either the antigen-presenting cell used, mature DC or PBMC, or the peptide stimulation, wild-type or mutant p53 peptides. Stimulations were done using either autologous DC or PBMC, with similar results, to rule out that alterations in cytokine secretion were associated with the particular induction conditions used.


   Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
References
 
Head and neck cancer patients are capable of generating anti-p53 antibodies because 16% (7 of 43) of the patients had reactive sera. The antibodies detected in the sera recognized both mutated and wild-type p53 proteins in accordance with previous studies (19, 34, 35). As was found with other tumor types, not all head and neck cancer patients with a p53 gene mutation had detectable antibody to the oncoprotein. Because immunoprecipitation detected one reactive serum whereas Western blot assays discovered seven sera with antibody present, the Western blot technique seemed to be much more sensitive. There was a relatively high incidence of anti-p53 antibodies detected in this patient population when compared with other studies, such as lung cancer patients (13%; ref. 19) or breast cancer patients (9%; ref. 36), although 23% of patients with colorectal carcinoma had serum p53 antibodies (37). In other studies of head and neck cancer patients from 18% to 44% of the patients had serum p53 antibodies (38, 39). Nearly half (7 of 15, 47%) of our patients with tumors harboring a p53 mutation had an immunologic response to both mutant and wild-type p53 protein. This finding implies that tolerance was abrogated in these patients.

Immunodominant epitopes derived from the p53 protein, the result of conformational changes caused by missense mutations, also do not seem to account for the antibody response against p53. Lubin et al. (40) and Schlichtholz et al. (41) found that immunodominant epitopes were localized predominantly in the amino terminus and, to a lesser extent, in the carboxy terminus. Using our Western blot technique with p53 fusion proteins (of evolutionarily conserved regions), no immunodominant epitopes were found to correlate with antibody production. Because almost all (six of seven) patients producing anti-p53 antibodies had either DQB1*0301 or DQB1*0501 alleles, we postulated that the peptide's amino acid structure and ability to bind to HLA class II binding grooves or motifs may be implicated in the anti-p53 response.

To test our hypothesis, the published class II allele-specific HLA-DQ7 (DQA1*0501-DQB1*0301) binding motif (32) was matched to the mutant p53 peptide sequences to determine if the peptides were ligands for the binding motif. Typically, peptides of varying lengths, often 12 to 24 residues, bind class II molecules and, therefore, binding motifs exhibit open ends, such that portions of entire proteins could bind if appropriate anchor residues were exposed (42). Therefore, the best possible fit of the mutated p53 amino acid sequence into the class II binding motifs was attempted. Four patients had the DQB1*0301 allele, and HLA-DR alleles associated with the HLA-DQ7 molecule. Three other patients possessed the DRB1*0101-DQB1*0501 haplotype of the HLA-DR1 molecule. The peptide sequences (14-mers and 11-mers) from patients that produced antibodies fit the motif extremely well; both the mandatory anchor residues and the preferred charged/polar and proline residues were present in the mutant p53 peptides. The patients that did not generate anti-p53 antibodies had mutant peptide sequences that did not fit the class II binding motifs. Only one of the p53 mutations resulted in the creation of a new anchor residue. In a vast majority of the cases (six of seven), the wild-type p53 peptide fit the motif, and mutation was nearby. In the case of HLA-DQ7 binding, the mutation was within two amino acid residues away from a potential anchor-binding residue. This suggested that the mutation was creating a neoantigen that could be presented by the class II complexes.

The competitive binding assay to the DQ7 molecule further suggests that p53 oncoprotein binding to certain class II molecules determines antibody production. The lack of binding of the DQ7 molecule by the mutated peptide of patients 226 shows that predicting which peptides will bind may depend upon other factors including the peptide's tertiary structure in vivo. In this particular case, the mutation caused a glutamic acid residue (charged) to be encoded instead of a glycine residue (small) which could result in steric hindrance of binding. Whereas this study is suggestive of binding of p53 oncoproteins to specific class II molecules as the determining factor in anti-p53 antibody production, the actual in vivo process involving conformation changes may be more complex.

Mutated and wild-type p53 molecules are also targets for class I–restricted peptide-induced CTL (2, 7). Less is known about the role of class II presentation of oncoproteins. The effect of stimulation of healthy, HLA-DQ7+ healthy donor PBMC was modeled in vitro using a mutant, HLA-DQ7–binding, mutant p53 peptide, and induced a Th2-biased cytokine profile. This peptide (p53220C) was identified in two of our head and neck squamous cell carcinoma patients with anti-p53 serum antibodies and was found to bind to HLA-DQ7 molecules with 10-fold higher affinity than the wild type peptide (p53210-223).

We observed that the enhanced binding of the mutant p53220C peptide was associated with lower IFN-{gamma} and higher IL-5 secretion by PBMC stimulated from healthy HLA-DQ7+ donors, whose T cells were stimulated with the mutant p53220C peptide. This is consistent with a skewing toward the Th2 population of CD4 T cells.` Further experiments are necessary outside the scope of the studies reported here to clarify more subtle effects on T-cell phenotypic subsets. We conclude that a form of immune escape may be exerted on antitumor effector T cells by tumors presenting mutated p53 peptides. This mechanism has been suggested in tumor bearing renal cell carcinoma or melanoma patients, which could be reversed after tumor removal (12). Because a missense mutation in p53 is among the most common alterations in human cancer, these finding have important implications for strategies to reverse immunosuppressive effects of tumor burden, such as cancer vaccines.


   Footnotes
 
Grant support: P50 CA96784 (D. Sidransky), P50 CA097190 (R.L. Ferris), and a FAMRI clinical investigator award (R.L. Ferris).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received 3/23/07; revised 8/17/07; accepted 8/23/07.


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 

  1. Greten TF, Jaffee EM. Cancer vaccines. J Clin Oncol 1999;17:1047–60.[Abstract/Free Full Text]
  2. Chada S, Mhashilkar A, Roth JA, Gabrilovich D. Development of vaccines against self-antigens: the p53 paradigm. Curr Opin Drug Discov Devel 2003;6:169–73.[Medline]
  3. Carbone DP, Ciernik IF, Kelley MJ, et al. Immunization with mutant p53- and K-ras-derived peptides in cancer patients: immune response and clinical outcome. J Clin Oncol 2005;23:5099–107.[Abstract/Free Full Text]
  4. Yanuck M, Carbone DP, Pendleton CD, et al. A mutant p53 tumor suppressor protein is a target for peptide-induced CD8+ cytotoxic T-cells. Cancer Res 1993;53:3257–61.[Abstract/Free Full Text]
  5. Houbiers JG, Nijman HW, van der Burg SH, et al. In vitro induction of human cytotoxic T lymphocyte responses against peptides of mutant and wild-type p53. Eur J Immunol 1993;23:2072–7.[Medline]
  6. Ropke M, Hald J, Guldberg P, et al. Spontaneous human squamous cell carcinomas are killed by a human cytotoxic T lymphocyte clone recognizing a wild-type p53-derived peptide. Proc Natl Acad Sci U S A 1996;93:14704–7.[Abstract/Free Full Text]
  7. Sirianni N, Ha PK, Oelke M, et al. Effect of human papillomavirus-16 infection on CD8+ T-cell recognition of a wild-type sequence p53264–272 peptide in patients with squamous cell carcinoma of the head and neck. Clin Cancer Res 2004;10:6929–37.[Abstract/Free Full Text]
  8. Pardoll DM. New strategies for enhancing the immunogenicity of tumors. Curr Opin Immunol 1993;5:719–25.[CrossRef][Medline]
  9. Knutson KL, Disis ML. Tumor antigen-specific T helper cells in cancer immunity and immunotherapy. Cancer Immunol Immunother 2005;54:721–8.[CrossRef][Medline]
  10. Hokey DA, Larregina AT, Erdos G, Watkins SC, Falo LD, Jr. Tumor cell loaded type-1 polarized dendritic cells induce Th1-mediated tumor immunity. Cancer Res 2005;65:10059–67.[Abstract/Free Full Text]
  11. McCarter M, Clarke J, Richter D, Wilson C. Melanoma skews dendritic cells to facilitate a T helper 2 profile. Surgery 2005;138:321–8.[CrossRef][Medline]
  12. Tatsumi T, Kierstead LS, Ranieri E, et al. Disease-associated bias in T helper type 1 (Th1)/Th2 CD4(+) T cell responses against MAGE-6 in HLA-DRB10401(+) patients with renal cell carcinoma or melanoma. J Exp Med 2002;196:619–28.[Abstract/Free Full Text]
  13. Bennouna J, Hildesheim A, Chikamatsu K, Gooding W, Storkus WJ, Whiteside TL. Application of IL-5 ELISPOT assays to quantification of antigen-specific T helper responses. J Immunol Methods 2002;261:145–56.[CrossRef][Medline]
  14. Koch WM, Brennan JA, Zahurak M, et al. p53 mutation and locoregional treatment failure in head and neck squamous cell carcinoma. J Natl Cancer Inst 1996;88:1580–6.[Abstract/Free Full Text]
  15. Zhang JY, Casiano CA, Peng XX, Koziol JA, Chan EK, Tan EM. Enhancement of antibody detection in cancer using panel of recombinant tumor-associated antigens. Cancer Epidemiol Biomarkers Prev 2003;12:136–43.[Abstract/Free Full Text]
  16. Shimada H, Ochiai T, Nomura F. Titration of serum p53 antibodies in 1,085 patients with various types of malignant tumors: a multiinstitutional analysis by the Japan p53 Antibody Research Group. Cancer 2003;97:682–9.[Medline]
  17. Sainger RN, Shah MH, Desai AA, et al. Clinical significance of serum p53 antibodies in oral cancer. Tumori 2006;92:134–9.[Medline]
  18. Saleh J, Brunner C, Golzer R, Nastainczyk W, Montenarh M. p53 autoantibodies from patients with head and neck cancer recognise common epitopes on the polypeptide chain of p53. Cancer Lett 2006;233:48–56.[CrossRef][Medline]
  19. Lubin R, Zalcman G, Bouchet L, et al. Serum p53 antibodies as early markers of lung cancer. Nat Med 1995;1:701–2.[CrossRef][Medline]
  20. Davidoff AM, Iglehart JD, Marks JR. Immune response to p53 is dependent upon p53/HSP70 complexes in breast cancers. Proc Natl Acad Sci U S A 1992;89:3439–42.[Abstract/Free Full Text]
  21. Elledge RM, Clark GM, Fuqua SA, Yu YY, Allred DC. p53 protein accumulation detected by five different antibodies: relationship to prognosis and heat shock protein 70 in breast cancer. Cancer Res 1994;54:3752–7.[Abstract/Free Full Text]
  22. Winter SF, Minna JD, Johnson BE, Takahashi T, Gazdar AF, Carbone DP. Development of antibodies against p53 in lung cancer patients appears to be dependent on the type of p53 mutation. Cancer Res 1992;52:4168–74.[Abstract/Free Full Text]
  23. Brennan JA, Boyle JO, Koch WM, et al. Association between cigarette smoking and mutation of the p53 gene in squamous-cell carcinoma of the head and neck. N Engl J Med 1995;332:712–7.[Abstract/Free Full Text]
  24. Ruppert JM, Stillman B. Analysis of a protein-binding domain of p53. Mol Cell Biol 1993;13:3811–20.[Abstract/Free Full Text]
  25. Apple RJ, Erlich HA, Klitz W, Manos MM, Becker TM, Wheeler CM. HLA DR-DQ associations with cervical carcinoma show papillomavirus-type specificity. Nat Genet 1994;6:157–62.[CrossRef][Medline]
  26. Kwok WW, Nepom GT, Raymond FC. HLA-DQ polymorphisms are highly selective for peptide binding interactions. J Immunol 1995;155:2468–76.[Abstract]
  27. Mailliard RB W-KA, Cai Q, Wesa A, et al. {alpha}-type-1 polarized dendritic cells: a novel immunization tool with optimized CTL-inducing activity. Cancer Res 2004;64:5934–7.[Abstract/Free Full Text]
  28. Hathaway B, Landsittel DP, Gooding W, et al. Multiplexed analysis of serum cytokines as biomarkers in squamous cell carcinoma of the head and neck patients. Laryngoscope 2005;115:522–7.[CrossRef][Medline]
  29. Linkov F, Lisovich A, Yurkovetsky Z, et al. Early detection of head and neck cancer: development of a novel screening tool using multiplexed immunobead-based biomarker profiling. Cancer Epidemiol Biomarkers Prev 2007;16:102–7.[Abstract/Free Full Text]
  30. Parham PBC, Bodmer WF. Use of a monoclonal antibody (W6/32) in structural studies of HLA-A,B,C, antigens. J Immunol 1979;123:342–9.[Abstract/Free Full Text]
  31. Temponi M, Kekish U, Hamby CV, Nielsen H, Marboe CC, Ferrone S. Characterization of anti-HLA class II monoclonal antibody LGII-612.14 reacting with formalin fixed tissues. J Immunol Methods 1993;161:239–56.[CrossRef][Medline]
  32. Falk K, Rotzschke O, Stevanovic S, Jung G, Rammensee HG. Pool sequencing of natural HLA-DR, DQ, and DP ligands reveals detailed peptide motifs, constraints of processing, and general rules. Immunogenetics 1994;39:230–42.[CrossRef][Medline]
  33. Chikamatsu K, Albers A, Stanson J, et al. P53(110–124)-specific human CD4+ T-helper cells enhance in vitro generation and antitumor function of tumor-reactive CD8+ T cells. Cancer Res 2003;63:3675–81.[Abstract/Free Full Text]
  34. Labrecque S, Naor N, Thomson D, Matlashewski G. Analysis of the anti-p53 antibody response in cancer patients. Cancer Res 1993;53:3468–71.[Abstract/Free Full Text]
  35. Marxsen J, Schmiegel W, Roder C, et al. Detection of the anti-p53 antibody response in malignant and benign pancreatic disease. Br J Cancer 1994;70:1031–4.[Medline]
  36. Angelopoulou K, Diamandis EP, Sutherland DJ, Kellen JA, Bunting PS. Prevalence of serum antibodies against the p53 tumor suppressor gene protein in various cancers. Int J Cancer 1994;58:480–7.[Medline]
  37. Angelopoulou K, Stratis M, Diamandis EP. Humoral immune response against p53 protein in patients with colorectal carcinoma. Int J Cancer 1997;70:46–51.[CrossRef][Medline]
  38. Kaur J, Srivastava A, Ralhan R. Serum p53 antibodies in patients with oral lesions: correlation with p53/HSP70 complexes. Int J Cancer 1997;74:609–13.[CrossRef][Medline]
  39. Bourhis J, Lubin R, Roche B, et al. Analysis of p53 serum antibodies in patients with head and neck squamous cell carcinoma. J Natl Cancer Inst 1996;88:1228–33.[Abstract/Free Full Text]
  40. Lubin R, Schlichtholz B, Bengoufa D, et al. Analysis of p53 antibodies in patients with various cancers define B-cell epitopes of human p53: distribution on primary structure and exposure on protein surface. Cancer Res 1993;53:5872–6.[Abstract/Free Full Text]
  41. Schlichtholz B, Legros Y, Gillet D, et al. The immune response to p53 in breast cancer patients is directed against immunodominant epitopes unrelated to the mutational hot spot. Cancer Res 1992;52:6380–4.[Abstract/Free Full Text]
  42. Germain RN. MHC-dependent antigen processing and peptide presentation: providing ligands for T lymphocyte activation. Cell 1994;76:287–99.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Couch, M. E.
Right arrow Articles by Sidransky, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Couch, M. E.
Right arrow Articles by Sidransky, D.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online