Introduction

Many viral infections are controlled by virus-specific CD8 + cytotoxic T lymphocytes (CTL).^{1, 2, 3} It is well known that CTL recognize only a few short sequences (8–11 amino acids long) in a protein⁴ in the context of MHC Class I molecules, named 'CTL epitopes'. The identification of such epitopes is important in the study of the role of CTL in anticancer and antiviral vaccines. Although some of the CTL epitopes in proteins can be predicted using published MHC Class I binding algorithms if the MHC Class I haplotype is known, not all epitopes are correctly identified by these methods. Ideally, exhaustive CTL epitope mapping should be carried out with all possible peptides which can arise from a protein sequence. Taking into account the known range of common CTL epitope lengths of 8, 9 or 10 amino acids, and assuming no post-translational modification of the sequence, this implies that about 3 (n - 7) peptides are required, where n is the number of amino acid residues in the protein of interest. In practice, the synthesis of large numbers of peptides can be accomplished but at a very substantial cost, ruling out this approach for many investigators. Significant, or even the dominant, epitope(s) could be missed entirely by an approach based on prediction methods^{5, 6} or methods using only 8 or 9 mer overlapping peptides.⁷ Here we present a quick and easy method based on IFN- ELISPOT assay and a unique set of overlapping peptides, which reduces the number of separate peptide syntheses by about threefold, while still covering all the peptides required. This brings the exhaustive scanning of proteins for CD8 T cell epitopes within the reach of many more investigators. Once the CD8 T cell epitope peptides have been identified by this method, the CTL activity of the peptides could be confirmed using standard chromium release assays or tumour protection assays.⁸ We have used this method to identify CTL epitopes of cervical cancer associated human papillomavirus type 16 E7 protein in five different types of mice to determine the frequency of occurrence of CTL epitopes in this oncoprotein.

Materials and Methods

Synthesis of peptides

A set of overlapping peptides (Pepset) spanning the whole length of the E7 protein of human papillomavirus type 16 (HPV16) was made by Mimotopes Pty Ltd, Melbourne, Australia (Table 1). The synthesis of peptides was controlled so that in each peptide sample, there are roughly an equimolar mixture of 8, 9 and 10 mer peptides. In addition, some of the peptides containing serines were synthesized to produce equimolar mixtures of serines and phosphoserines to mimic post-translational modifications of serines.

Table 1 - Sequences of the Pepset peptides used in the study of functional CTL epitopes in the E7 oncoprotein of cervical cancer associated human papillomavirus type 16.

Full table

The H-2^b CTL epitope peptides of E7 and ovalbumin (RAHYNIVTF and SIINFEKL, respectively) were bought from Mimotopes Pty Ltd, Melbourne, Australia, and from Auspep Pty Ltd, Melbourne, Australia.

E7 protein and its mutant W7 protein in which the D^b restricted CTL epitope RAHYNIVTF⁶ was mutated to RAHWTSATK to completely abolish CTL activity.⁹ These were prepared and purified, and endotoxin extracted, as previously described.¹⁰

Immunization of mice and preparation of splenocytes

Female BALB/c (H-2^d), CBA/CaH (H-2 ^k), FVB/N (H-2^q) and A2K^bH2^b human HLA2.1 transgenic mice were purchased from ARC, Perth, Australia. Major histocompatibility Class I¹¹ and MHC Class II knockout mice¹² were purchased from WEHI, Melbourne, and ANU, Canberra, Australia. Mice were housed in the germ-free animal facility at Princess Alexandra Hospital and used between the ages of 6 and 10 weeks. Mice were immunized once with 20 g E7 protein together with the adjuvant Quil-A (10 g) subcutaneously at the tail base as described previously.^{8, 13} Spleens were removed at different time points after immunization, and single cell suspensions of splenocytes were prepared and assayed for IFN- secretion using an ELISPOT assay. Interferon- secretion activity was observed to peak at day six after immunization and therefore this time point was used in all subsequent experiments.

IFN- ELISPOT assay

This assay is based on a modification of a published ELISPOT assay method.¹⁴ Ninety-six-well (Millipore MultiScreen-HA Cat MHABS4510) ELISPOT plates were coated under sterile conditions with the capture antibody (BD Biosciences Cat 551216, Clone R4-6A2) at 8 g/mL in freshly prepared filter sterilized 0.1 mol/L NaHCO₃ pH = 8.4 (binding buffer). Then 75 L of this capture antibody solution was added per well and incubated overnight 16 h at 4°C. The plates were then blocked by the addition of 200 L of filtered sterile complete media (RPMI supplemented with 10% FCS, 100 U/mL penicillin, 100 g/mL streptomycin, 10^-5 mol/L 2-mercaptoethanol and 2 mmol/L glutamine) and incubated for 2 h at room temperature (23°C). The plates were washed once with 200 L of medium, and 100 L of splenocyte preparation from immunized mice (10⁷ cells per mL in complete medium) was added to each well. The peptides and rIL-2 (GibcoBRL Cat 13238-050) were also added in 100 L of complete medium to each well to give a final peptide concentration of 0.1 mol/L and an IL-2 concentration of 50 U/mL (5 ng/mL). The plates were incubated at 37°C in a 5% CO₂ oven for 16–20 h. Then the plates were washed six times with PBST (0.2 mL Tween-20/L of phosphate buffered saline). After adding 75 L of biotinylated detection mAb (BD Biosciences Cat 554410, Clone XMG1.2) at a concentration of 1 g/mL in PBST containing 0.1 mg/mL BSA to each well, the plates were incubated for 3 h at room temperature in the dark. The plates were washed six times with PBST, and 75 L of Avidin-horse radish peroxidase solution (Sigma Cat A-3151, 1 mg/mL stock solution in PBS was diluted 1 : 400 in PBST containing 0.1 mg/mL BSA), was added to each well and the plates were incubated for 1 h at room temperature. The plates were first washed three times with PBST, and then washed three times with PBS. Diaminobenzidine (DBA) developing reagent (Sigma Cat D0426) was prepared as recommended, and 75 L of this substrate solution was added to each well and kept at room temperature until dark spots appeared (1–3 min). Colour development was stopped by thoroughly washing the plates in tap water. The plates were allowed to air-dry overnight. Spots were counted using a stereomicroscope (Nikon SMZ 10 A) with reflected light, or alternatively the wells were photographed using a camera (Panasonic BP330) mounted on the microscope and was printed on paper and counted using a colony counter pen (BelArt Products, Pequannock, NJ, USA). In all experiments, data presented are representative of at least two separate experiments.

To confirm that CD8 epitope peptides detected by this method were CD8 restricted epitopes, ELISPOT assay was repeated using splenocytes from H-2^b syngeneic MHC Class I and Class II knockout E7 immunized mice. In addition, the contribution of CD4⁺ and CD8⁺ cells to the spot numbers were tested using splenocytes treated in vitro with antibodies against CD4 and CD8 molecules. The splenocytes were treated for 1 h at room temperature (23°C) with either anti CD8 antibody (2.43, ATCC) or anti CD4 antibody (GK1.5, ATCC) at a concentration of 300 g/mL prior to addition to wells in the ELISPOT assay.

Results

Optimization of the ELISPOT assay

Optimal peptide dose to be used in the ELISPOT assay

To determine the optimal in vitro peptide concentration for the ELISPOT assay, mice were immunized with E7/Quil-A and 6 days after immunization, splenocytes were removed and ELISPOT assays performed using different concentrations of the minimal CTL epitope peptide RAHYNIVTF. A peptide concentration of 10^-7 mol/L (0.1 mol/L) to 10^-6 mol/L (1.0 mol/L) gave the maximal number of IFN--secreting cells (Figure 1a). This CTL peptide concentration is also optimal for the ovalbumin (H-2^b) CTL epitope SIINFEKL, and for the HIV gp120 (H-2^d) CTL epitope RGPGRAFVTI (data not shown). Therefore, it is possibly the optimal dose for most minimal CTL epitope peptides.

Figure 1.

(a) Groups of five C57BL/6 J (H-2^b) mice were immunized with human papillomavirus E7 protein and Quil-A adjuvant (E7/Quil-A). Splenocytes were taken 6 days after the immunization, pooled and were assayed for IFN- production in an ELISPOT assay. The splenocytes were challenged in vitro with varying concentrations of the D^b restricted (CTL) epitope peptide RAHYNIVTF of E7 protein. The ELISPOT plates were developed 16–20 h after the incubation. Interferon--producing cell numbers are presented. (b) Groups of five C57BL/6 J mice were immunized with E7/Quil-A at different time intervals, and spleens were removed on the same day and splenocytes from each group were pooled, and tested in an IFN- ELISPOT assay. The spot numbers corresponding to mice from each group are presented. (c) Groups of five C57BL/6 J mice were immunized with E7/Quil-A. Splenocytes were taken six days after the immunization, pooled, and placed into a 96-well ELISPOT plate containing different individual Pepset peptides in each well, and assayed for IFN- producing cells. The spot number appearing in each consecutive well is presented. (d) Groups of five C57BL/6 J mice were immunized with E7/IFA, W7/IFA, GF001/IFA, or as a control Ova/IFA. W7 is a CTL epitope crippled analogue of E7 protein. Six days after the immunization, the mice were injected subcutaneously in the scruff of the neck with E7 producing cell tumour cell line TC-1 (10⁶ cells per mouse). The mice were killed 10–12 days after the tumour challenge and the tumour was dissected out and weighed. Tumour weights observed for each group are shown.

Full figure and legend (13K)

Optimal splenocyte harvest time

To determine the time interval between immunization and the time of harvest of the splenocytes that gives the highest number of spots in an IFN- ELISPOT assay (peak CTL activity), mice immunized with E7/QA on different days were assayed. Peak CTL activity was observed at day six after immunization (Figure 1b), and subsequent assays performed at days five, six, and seven confirmed this observation (data not shown). Mice immunized with ovalbumin also demonstrated maximum number of IFN--secreting cells 6 days after the immunization (data not shown). Day six after immunization was therefore selected as the splenocyte harvest time.

Pepset analysis detects the unique H-2^b CTL epitope in E7 protein

We used the pepset method on the well-studied E7 oncoprotein of cervical cancer associated human papillomavirus type 16^{6, 8, 15, 16, 17} to confirm that we could detect the previously published sole H-2^b restricted CTL epitope RAHYNIVTF. An ELISPOT assay using splenocytes from E7 protein immunized C57BL/6 J mice showed IFN- secreting T cells reactive with peptide mixtures 50 and 51 containing the RAHYNIVTF peptide (Figure 1c) (average 118 spots and 60 spots, respectively). These data agree with the published report that E7 contains only one functional CTL epitope (RAHYNIVTF) restricted by D^b.⁶

The ELISPOT spots are due to the presence of the CD8 cytotoxic T cell epitope in the whole protein

To confirm that IFN- secretion could be attributable to RAHYNIVTF restricted CD8⁺ T cells and did not reflect CD4⁺ T cell reactivity, C57BL/6 J mice were immunized with W7 protein (which is an E7 mutant protein whose CTL epitope has been crippled by mutations). Interferon--secreting T cells were observed in E7 immunized mice but not in W7 immunized mice (183 11 vs. 2 1 spots SEM). Lack of a H-2^b restricted CD8 T cell epitope in W7 was confirmed in a tumour protection assay (Figure 1d) in which immunization with E7 or with CTL epitope peptide (001 = RAHYNIVTF) protects mice from a tumour challenge with a tumour expressing E7. W7 immunization did not induce a host protective immune response as the mean of the tumour weights observed between the W7 immunized mice and the control ovalbumin immunized groups were not statistically different (P = 0.3166 by t-test).

CD8⁺ T cells mediate RAHYNIVTF directed CTL response

To show that IFN- secretion was from CD8⁺ T cells, splenocytes were preincubated with antimouse CD8 mAb (2.43) or with antimouse CD4 mAb (GK 1.5) to block the CD8 or CD4 molecules, respectively. The number of IFN- secreting T cells was reduced to background levels when anti CD8 mAb was used, but was not altered when anti CD4 mAb was used (Figure 2). Furthermore MHC Class I and II KO mice were immunized with E7 protein and subjected to ELISPOT assay. No IFN- secreting T cells were seen in MHC Class I KO mouse splenocytes. In contrast, in class II KO mouse splenocytes, IFN- secreting cells were observed (Figure 2).

Figure 2.

Groups of three either C57BL/6 J mice or syngeneic MHC Class I and MHC Class II knock out mice were immunized with E7/Quil-A. Six days after immunization splenocytes were prepared from these mice. Splenocytes from the C57BL/6 J mice were preincubated with anti CD8 mAb (2.43), or with nothing, or as a control with anti CD4 mAb (GK1.5). Splenocytes from all groups were subjected to an IFN- ELISPOT assay. The spot number for each group is shown.

Full figure and legend (7K)

Determination of E7 protein CTL epitopes in mice of different MHC Class I haplotypes

We wished to determine the presence of CTL epitopes in E7 protein corresponding to other mouse MHC Class I types and to human HLA A2.1. BALB/c (H-2^d), CBA/CaH (H-2 ^k), FVB/N (H-2^q) and A2K^bH2^b human HLA2.1 transgenic mice¹⁸ were tested by this method for the presence of CTL epitopes. When A2K^bH2^b mouse splenocytes were assayed, IFN- secreting cells were observed in wells containing the H-2^b restricted CTL epitope peptide RAHYNIVTF (50 and 51) (Figure 3a). No putative A2.1 epitopes were detected in this group of mice, confirming the results of Bauer et al.¹⁷ that no CTL activity corresponding to A2.1 haplotype is induced when E7 protein was used in immunizations. Furthermore no significant IFN--secreting cells were observed in splenocytes from either E7 protein immunized BALB/c, CBA/CaH or FVB/N mice showing absence of CTL epitopes for E7 in these mice (Figure 3b–d).

Figure 3.

Groups of five (a) A2K^bH2^b mice, (b) BALB/c mice, (c) CBA/CaH mice, and (d) FVB/N mice were immunized with E7/Quil-A. Splenocytes were taken 6 days after the immunization, pooled, and placed into a 96-well ELISPOT plate containing different individual Pepset peptides in each well, and assayed for IFN--producing cells. The spot number appearing in each consecutive well is presented. As a positive assay control, splenocytes from E7 immunized C57BL/6 J mice were added to the last well of all assays with the known D^b restricted CTL epitope peptide RAHYNIVTF.

Full figure and legend (15K)

Discussion

We have studied the E7 oncoprotein of the cervical cancer associated human papillomavirus type 16 (HPV 16) in mice with different MHC Class I haplotypes to determine the frequency of the presence of functional CD8 T cell epitopes in this protein. Furthermore, in this study we developed a method that could be used to determine CTL epitopes in a protein, which is rapid and easy to perform.

We used an IFN- ELISPOT based method. The ELISPOT assay detects cytokine release in response to specific T cell epitope peptides at a single cell level.^{14, 19, 20, 21} Used in conjunction with uniquely designed Mimotope Pepset peptides, the ELISPOT assay has the advantage that it can easily cover in a single assay all possible 8, 9 and 10 mer CTL epitopes, as well as post translational modifications of peptides containing phosphorylated serines of the protein. In this study a Pepset was synthesized to obtain a tandem set of peptides offset by one amino acid, that contain all the possible 8, 9 and 10 mer peptides covering the whole sequence of the HPV16 E7 protein. In addition, the serine molecules in these peptide sets were synthesized to obtain an equimolar mixture of phosphorylated and non-phosphorylated serines. Using comprehensive coverage of different overlapping peptides we have studied the functional CTL epitopes of E7 protein in five types of mice having different MHC Class I molecules. Only one functional H-2^b CTL epitope was identified in all mice haplotypes tested. Although E7 is a multifunctional protein^{22, 23, 24, 25} it is a relatively small protein of 98 amino acids long. Our data agree with the prediction that the odds of a given 8–11 amino acid stretch of a protein binding any given allomorph with an affinity better than 500 nmol/L is about 1/200, and the odds of achieving immunodominance is about 1/2000.^{26, 27, 28} This indicates that it may be advantageous for the virus to pack all the necessary functional domains into a minimal size protein so that the probability of a CTL epitope occurring in the protein by chance will be minimized, and therefore escape the host immune attack.

Here we also have described a simple method to determine CTL epitopes in proteins using tandem overlapping multiple synthetic peptides and an IFN- ELISPOT assay. Interferon- ELISPOT assays have been extensively used to determine CTL epitopes in peptides.^{21, 29} This method has the advantage that once the peptide set is synthesized for a protein of interest, it could potentially be used to determine CD8 T cell epitopes in many different species of animals, even when the identity of the MHC Class I molecule of the animal is not known. The ease and the cost effectiveness of this method is a major advantage over the other published methods.

References

1. Harty JT, Tvinnereim AR, White DW. CD8+ T cell effector mechanisms in resistance to infection. Annu. Rev. Immunol. 2000; 18: 275–308. | Article | PubMed | ISI | ChemPort |
2. Zinkernagel RM, Bachmann MF, Kundig TM et al. On immunological memory. Annu. Rev. Immunol. 1996; 14: 333–67. | Article | PubMed | ISI | ChemPort |
3. Sedlik C, Dadaglio G, Saron MF et al. In vivo induction of a high-avidity, high-frequency cytotoxic T-lymphocyte response is associated with antiviral protective immunity. J. Virol. 2000; 74: 5769–75. | Article | PubMed | ChemPort |
4. Falk K, Rotzschke O, Stevanovic S, Jung G, Rammensee HG. Allele-specific motifs revealed by sequencing of self-peptides eluted from MHC molecules. Nature 1991; 351: 290–6. | Article | PubMed | ISI | ChemPort |
5. Amaro D, Houbiers JG, Drijfhout JW et al. A computer program for predicting possible cytotoxic T lymphocyte epitopes based on HLA class I peptide-binding motifs. Hum. Immunol. 1995; 43: 13–18. | Article | PubMed |
6. Feltkamp MC, Smits HL, Vierboom MP et al. Vaccination with cytotoxic T lymphocyte epitope-containing peptide protects against a tumor induced by human papillomavirus type 16-transformed cells. Eur. J. Immunol. 1993; 23: 2242–9. | PubMed | ISI | ChemPort |
7. Li H, Natarajan K, Malchiodi EL, Margulies DH, Mariuzza RA. Three-dimensional structure of H-2Dd complexed with an immunodominant peptide from human immunodeficiency virus envelope glycoprotein 120. J. Mol. Biol. 1998; 283: 179–91. | Article | PubMed | ISI | ChemPort |
8. Fernando GJ, Stewart TJ, Tindle RW, Frazer IH. Th2-type CD4+ cells neither enhance nor suppress antitumor CTL activity in a mouse tumor model. J. Immunol. 1998; 161: 2421–7. | PubMed | ISI | ChemPort |
9. Fernando GJ, Khammanivong V, Leggatt GR, Liu WJ, Frazer IH. The number of long-lasting functional memory CD8+ T cells generated depends on the nature of the initial nonspecific stimulation. Eur. J. Immunol. 2002; 32: 1541–9. | Article | PubMed | ISI | ChemPort |
10. Fernando GJ, Murray B, Zhou J, Frazer IH. Expression, purification and immunological characterization of the transforming protein E7, from cervical cancer-associated human papillomavirus type 16. Clin. Exp. Immunol. 1999; 115: 397–403. | Article | PubMed | ChemPort |
11. Koller BH, Marrack P, Kappler JW, Smithies O. Normal development of mice deficient in beta 2M, MHC class I proteins, and CD8+ T cells. Science 1990; 248: 1227–30. | Article | PubMed | ISI | ChemPort |
12. Kontgen F, Suss G, Stewart C, Steinmetz M, Bluethmann H. Targeted disruption of the MHC class II Aa gene in C57BL/6 mice. Int. Immunol. 1993; 5: 957–64. | PubMed | ISI | ChemPort |
13. Fernando GJ, Stewart TJ, Tindle RW, Frazer IH. Vaccine-induced Th1-type responses are dominant over Th2-type responses in the short term whereas pre-existing Th2 responses are dominant in the longer term. Scand. J. Immunol. 1998; 47: 459–65. | Article | PubMed | ChemPort |
14. Miyahira Y, Murata K, Rodriguez D et al. Quantification of antigen specific CD8+ T cells using an ELISPOT assay. J. Immunol. Meth. 1995; 181: 45–54. | Article | ISI | ChemPort |
15. Chen LP, Thomas EK, Hu SL, Hellstrom I, Hellstrom KE. Human papillomavirus type 16 nucleoprotein E7 is a tumor rejection antigen. Proc. Natl. Acad. Sci. USA 1991; 88: 110–14. | Article | PubMed | ChemPort |
16. Ressing ME, Sette A, Brandt RM et al. Human CTL epitopes encoded by human papillomavirus type 16, E6 and E7 identified through in vivo and in vitro immunogenicity studies of HLA-A*0201-binding peptides. J. Immunol. 1995; 154: 5934–43. | PubMed | ISI | ChemPort |
17. Bauer M, Wagner H, Lipford GB. HPV type 16 protein E7 HLA-A2 binding peptides are immunogenic but not processed and presented. Immunol. Lett. 2000; 71: 55–9. | Article | PubMed | ISI | ChemPort |
18. Vitiello A, Marchesini D, Furze J, Sherman LA, Chesnut RW. Analysis of the HLA-restricted influenza-specific cytotoxic T lymphocyte response in transgenic mice carrying a chimeric human-mouse class I major histocompatibility complex. J. Exp. Med. 1991; 173: 1007–15. | Article | PubMed | ISI | ChemPort |
19. Czerkinsky C, Andersson G, Ekre HP et al. Reverse ELISPOT assay for clonal analysis of cytokine production. I. Enumeration of gamma-interferon-secreting cells. J. Immunol. Meth. 1988; 110: 29–36. | Article | ISI | ChemPort |
20. Murali-Krishna K, Altman JD, Suresh M et al. Counting antigen-specific CD8 T cells: a re-evaluation of bystander activation during viral infection. Immunity 1998; 8: 177–87. | Article | PubMed | ISI | ChemPort |
21. Scheibenbogen C, Romero P, Rivoltini L et al. Quantitation of antigen-reactive T cells in peripheral blood by IFN-gamma-ELISPOT assay and chromium-release assay: a four-centre comparative trial. J. Immunol. Meth 2000; 244: 81–9. | Article | ChemPort |
22. Akerman GS, Tolleson WH, Brown KL et al. Human papillomavirus type 16, E6 and E7 cooperate to increase epidermal growth factor receptor (EGFR) mRNA levels, overcoming mechanisms by which excessive EGFR signaling shortens the life span of normal human keratinocytes. Cancer Res. 2001; 61: 3837–43. | PubMed | ChemPort |
23. Gonzalez SL, Stremlau M, He X, Basile JR, Munger K. Degradation of the retinoblastoma tumor suppressor by the human papillomavirus type 16, E7 oncoprotein is important for functional inactivation and is separable from proteasomal degradation of E7. J. Virol. 2001; 75: 7583–91. | Article | PubMed | ISI | ChemPort |
24. Katich SC, Zerfass-Thome K, Hoffmann I. Regulation of the Cdc25A gene by the human papillomavirus Type 16, E7 oncogene. Oncogene 2001; 20: 543–50. | Article | PubMed | ChemPort |
25. Nees M, Geoghegan JM, Hyman T et al. Papillomavirus type 16 oncogenes downregulate expression of interferon- responsive genes and upregulate proliferation-associated and NF-kappaB- responsive genes in cervical keratinocytes. J. Virol. 2001; 75: 4283–96. | Article | PubMed | ISI | ChemPort |
26. Sette A, Sidney J, del Guercio MF et al. Peptide binding to the most frequent HLA-A class I alleles measured by quantitative molecular binding assays. Mol. Immunol. 1994; 31: 813–22. | Article | PubMed | ISI | ChemPort |
27. Sette A, Vitiello A, Reherman B et al. The relationship between class I binding affinity and immunogenicity of potential cytotoxic T cell epitopes. J. Immunol. 1994; 153: 5586–92. | PubMed | ISI | ChemPort |
28. Yewdell JW, Bennink JR. Immunodominance in major histocompatibility complex class I-restricted T lymphocyte responses. Annu. Rev. Immunol. 1999; 17: 51–88. | Article | PubMed | ISI | ChemPort |
29. Tobery TW, Wang S, Wang XM et al. A simple and efficient method for the monitoring of antigen-specific T cell responses using peptide pool arrays in a modified ELISPOT assay. J. Immunol. Meth. 2001; 254: 59–66. | Article | ChemPort |

Acknowledgements

The authors wish to thank the National Health and Medical Research Council (NHMRC) of Australia for the grant 102553 which funded this research project. We also thank Dr T. C. Wu for TC-1 tumour cell line, Dr William Heath and Dr Guna Karupiah for MHC Class I and II knockout mice, Dr Linda Sherman for A2K^bH2^b mice, Mr David Wiseman and Ms Caron Maxim for the excellent animal care.