Synthetic Peptides Induce a Cytotoxic Response against Human Papillomavirus Type-18

Gynecologic Oncology 82, 77– 83 (2001) doi:10.1006/gyno.2001.6205, available online at http://www.idealibrary.com on

Synthetic Peptides Induce a Cytotoxic Response against Human Papillomavirus Type-18 Mario R. Castellanos, M.D.,* Roberta L. Hayes, Ph.D.,* ,† and Mitchell A. Maiman, M.D.* ,‡ *Department of Medicine, †Cancer Immunotherapy Program, Nalitt Institute for Cancer and Blood Related Diseases, and ‡Department of Obstetrics and Gynecology, Staten Island University Hospital, Staten Island, New York 10305 Received August 4, 1999; published online May 22, 2001

[4]. These oncogenic effects are mediated by the binding of E6 and E7 proteins to the products of the tumor suppressor genes p53 and retinoblastoma (rb), respectively [5, 6]. Once formed, the E6 –p53 and the E7–rb protein complexes are degraded and the peptides are translocated into the endoplasmic reticulum. There they may bind to major histocompatibility complex (MHC) class I molecules, prior to being transported to the cell surface [7–10]. Cytotoxic T-lymphocytes (CTL) recognize these viral peptides when they are bound to the surface MHC class I molecules. As a result of the association of HPV with cervical carcinoma and the intracellular processing and presentation of viral peptides to the cell surface, E6 and E7 related peptides represent unique tumor antigens. Being unique tumor antigens, they are attractive antigenic targets for specific cervical cancer immunotherapy [11, 12]. Initial studies demonstrated that peptides from HPV type-16 were able to stimulate murine CTL in vitro and in vivo models. More importantly, these peptides were also able to stimulate human CTL against HPV-16-positive cervical cancer cell lines [10 –14]. These immunogenicity studies have resulted in active clinical trials that use HPV type-16 E6 and E7 peptides to develop or augment cell-mediated immunity in patients with cervical intraepithelial neoplasia and cervical cancer. The development of cervical cancer immunotherapy has focused on HPV type-16 peptide antigens, since HPV type-16 is the most common HPV type found in squamous cell carcinoma of the cervix. However, HPV type-18 is the second most common HPV type associated with squamous cell carcinoma of the cervix and is the predominant type in cervical adenocarcinoma [15]. Therefore, novel immunotherapies for cervical cancer must consider HPV type-18 positive tumors. To date, there are limited data on HPV type-18 peptides for the purpose of developing cervical cancer immunotherapy [16]. In this study we investigate synthetic peptides, derived from HPV type-18 E6 and E7 proteins, that have a high binding activity for the human leukocyte antigen (HLA) A2 class I molecule. We show that two of these synthetic peptides are able to induce specific CTL activation from human peripheral

Objectives. Over 90% of cervical carcinomas express human papillomavirus (HPV) E6 and E7 proteins. These unique antigens are ideal targets for the development of cytotoxic T-lymphocytes (CTL) for antitumor immunotherapy. In this study we identify peptides from HPV-18 E6 and E7 proteins that bind to HLA class I molecules. We further show that these peptides are able to induce peptide-specific CTL from an HLA-A2-positive (ⴙ) peripheral blood donor in vitro. Methods. A computer-assisted algorithm was devised to identify peptides from HPV-18 E6 and E7 proteins that bind to HLA-A2 molecules. Peptides that were predicted to bind were synthesized and their binding activity was determined. HLA-A2 ⴙ irradiated stimulator cells pulsed with HPV-18 peptides were incubated with HLA-A2 ⴙ peripheral blood mononuclear cells. Cytotoxicity assays were performed to assess specific cell lysis. Results. Of 295 possible sequences, the computer-assisted algorithm predicted 10 peptides that would have a high probability of binding to HLA-A2. The 4 strongest binding peptides were analyzed for their ability to induce cytotoxic cells against HPV-18 peptide-pulsed targets. Two of the peptides induced significant lysis. Conclusions. There are limited data on peptide-based immunotherapy for HPV-18 ⴙ tumors. The combination of our computerassisted algorithm and binding assay permits rapid selection of potential CTL epitopes. We identified two peptides that were able to induce peptide-specific lysis. These two epitopes are candidates for a peptide-based vaccine against HPV-18 ⴙ tumors. The model described has broad applications and can be used in the development of immunotherapy for other types of cancers. © 2001 Academic Press

Key Words: papillomavirus, human; cervical cancer; immunotherapy; vaccine; HLA-A2 antigens; cytotoxic T-lymphocytes.

INTRODUCTION Worldwide, cervical carcinoma is one of the leading causes of cancer-related deaths in women [1–3]. A major risk factor for the development of cervical carcinoma is human papillomavirus (HPV), in particular the viral oncogenes E6 and E7 Presented in part at the 30th Annual Meeting of the Society of Gynecologic Oncologists, San Francisco, CA, March 20 –24, 1999. 77

0090-8258/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.

78

CASTELLANOS, HAYES, AND MAIMAN

blood mononuclear cells (PBMC) and that these cytotoxic cells are able to lyse target cells pulsed with HPV type-18 peptides. MATERIALS AND METHODS Peptide Selection Using a computer-assisted algorithm we previously designed the HPV-18 E6 and E7 proteins for analysis [17]. Briefly, consecutive overlapping 9- and 10-amino-acid peptides that possess the specific primary and secondary anchoring amino acids for HLA-A2 were identified [18, 19]. Based on the presence of favorable and unfavorable amino acids, the peptides were scored as having an indeterminate, low, medium, or high probability of binding to HLA-A2 molecules. The sequences predicted to have a high probability of binding were synthesized and the binding activity was determined in a semiquantitative T2 assay. Peptide Synthesis The HPV-18 E6 and E7 related peptides were synthesized at a commercial laboratory (Genemed Synthesis, San Francisco, CA) according to standard f-MOC synthesis methods and purified using high-pressure liquid chromatography. The peptides obtained were greater than 90% pure; results were verified by mass spectrometry. An immunogenic HPV type-16 E7(11) peptide epitope (YMLDLQPETT) was tested as a positive control and was used to assess the ability of the donor PBMC to respond to in vitro stimulation. Cells and Reagents Peptides were dissolved in ammonium hydroxide plus water, neutralized with acetic acid, and diluted to a concentration of 25–50 mg/ml in phosphate-buffered saline (PBS). The T2 cell line was purchased from the American Type Culture Collection (ATCC; Rockville, MD). The cells were grown in Dulbecco’s modified Eagle’s medium with 4.5 g/L glucose, 15 mM Hepes, 1.5 mM L-glutamine, and 5% fetal bovine serum (Sigma Chemical Co.) at 37°C, 5% CO 2, and humidified air. Prior to the stabilization assay the cells were switched to a 27°C incubator overnight to decrease the turnover of unoccupied HLA-A2 molecules. The murine hybridoma cell line HB82, which produces an anti-HLA-A2 antibody (BB7.2), was purchased from the ATCC and the monoclonal antibody was obtained from culture supernatants of these cells. The FITCconjugated rabbit anti-mouse immunoglobulin was obtained from Zymed Laboratories, Inc. (San Francisco, CA), and was used at a final concentration of 1:75 dilution. T2 Cell Assay The binding activity of the computer-selected peptides was assayed semi-quantitatively by measuring peptide-induced stabilization of HLA-A2 molecules on the transporter-deficient

T2 [(174 ⫻ CEM. T2) human hybrid B and T lymphoblast] cell line, as determined by flow cytometry. The T2 assay was performed as described previously [20]. Briefly, T2 cells [2 ⫻ 10 5/well] were cultured in 100 ␮l of X-vivo 10 serum-free medium (Bio Whittaker, Walkersville, MD) in sterile 96-well U-bottom plates at 27°C for either 18 or 30 h in the presence of 2– 4 ⫻ 10 ⫺5 mmol of synthetic peptide and 1.5 ␮g of human ␤ 2 microglobulin (Sigma Chemical Co.) After the incubation period, the T2 cells were incubated with either 150 ␮l of anti-HLA-A2 murine antibody or PBS for 30 min at 8°C. The plates were then centrifuged at 5000 rpm for 7 min and the supernatants were removed. The cell pellets were washed and then resuspended in fluorescein-labeled rabbit anti-murine Ig (1:75) in PBS containing 0.02% sodium azide and then incubated for an additional 30 min at 8°C. The cells were washed with PBS plus azide to remove excess antibody, fixed with 1% paraformaldehyde for 15 min at RT, and transferred to test tubes for analysis of the presence of green fluorescence using a Coulter-XL flow cytometer. Generation of Cytotoxic Cells The donor responder cells used for in vitro stimulation and induction of HPV-18-specific cytotoxic cells were obtained from HLA-A2-positive, male, primary brain tumor patients undergoing leukapheresis prior to immune therapy. All clinical investigations were performed with the approval of the SIUH Institutional Review Committee and each blood donor gave informed consent for the use of their cells for research purposes. The mononuclear cells obtained by leukapheresis were further isolated by standard density gradient separation using Accu-Prep lymphocyte separation medium (Nycomed, Oslo, Norway) and cryopreserved in medium containing 7.5% DMSO. Prior to stimulation with the HPV-18-specific peptides, the responder cells were thawed and washed with 1⫻ Hanks’ balanced salt solution (HBSS) to remove residual preservative and then immediately cultured with gamma-irradiated stimulators (cobalt radiation source, 2000 rad). Responders were initially cultured at a concentration of 1.5 ⫻ 10 6/ml in six-well flat-bottom plates in X-vivo 10 serum-free medium with irradiated peptide-pulsed T2 stimulators, at a stimulator to responder ratio of 10:1. The T2 stimulator cells were pulsed 4 –18 h with peptide and ␤2-microglobulin prior to irradiation. Twenty-four hours after the addition of stimulators, interleukin-2 (IL-2) was added to each culture at a concentration of 25 IU/ml and then refreshed every 4 days thereafter. Cultures were split and given fresh X-vivo 10 medium at 3– 4 days if needed, determined by monitoring the metabolic pH culture indicator, phenol red, and cell density. Each week similar responder cultures were pooled, centrifuged, washed, recounted, and restimulated with fresh peptidepulsed, gamma-irradiated T2 stimulator cells. After 4 weeks of restimulation the responder cells were harvested and tested on various target cells for lytic activity using a 51Cr-release assay.

79

HPV-18 CYTOTOXIC CELLS

Each group was assayed using homologous peptide-pulsed T2 cells as well as T2 cells without peptide. In addition, some of the responder groups were tested for their ability to lyse both natural killer (NK) sensitive and NK cell-resistant target cells (K562 and Raji cells, respectively) to determine whether IL-2 had caused any nonspecific or MHC-unrestricted cytolytic activity in these cultures. 51

Chromium-Release Cytotoxicity Assay

Immune effector cells have characteristic patterns of lysing target cells that can be used to distinguish the killer cell of origin. K562 was used as an indicator of classic natural killer function; NK-resistant Raji cell targets were used to identify IL-2 activated nonrestricted lymphokine activated killer (LAK) cells. To determine HLA-A2 restricted peptide specific lysis, the TAP-deficient T2 cell line was pulsed with 5 ␮g of peptide in the presence of ␤2-microglobulin for 4 –18 h at 27°C in a CO 2 incubator. After peptides were loaded, each appropriate target cell group was labeled with 100 ␮Ci of sodium– 51chromate for 1–2 h. After being radiolabeled, the target cells were washed twice with 1⫻ HBSS to remove excess free 51chromium. The effector cells and target cells were then incubated for 4 h in 96-well round-bottom microtiter plates in serum-free medium. The amount of isotope released into the culture medium during this incubation is proportional to the amount of effectorcell-mediated lysis. Spontaneous isotope release was determined by culturing 51Cr-labeled cells in medium alone. Maximum isotope release was determined by the addition of 1% sodium dodecyl sulfate. The percentage of specific target cells that have been lysed is calculated according to the formula below. Multiple target well determinations, in twofold serial dilutions, were performed in each experiment to generate the cytotoxicity curves. The spontaneous and maximum release values are the mean of eight wells.

%Specific lysis ⫽

experimental cpm ⫺ spontaneous release in cpm ⫻ 100 maximum cpm ⫺ spontaneous cpm RESULTS

Peptide Selection From a total of 295 possible overlying nanomer or decamer peptides from the HPV-18 E6 and E7 proteins, 10 (3.4%) peptides were predicted as having a high probability of binding to HLA-A2 molecules based on favorable amino acid combinations (Table 1). The relative binding activity of each of these peptides was confirmed with the T2 HLA stabilization assay. Peptides Ex1, Ex2, Ex9, and Ex10 demonstrated the highest

TABLE 1 HPV Type-18 E6 and E7 Peptides Predicted to Have High Binding Activity for HLA-A2 Molecules a Peptide

Location

Length

Sequence

Ex1 Ex2 Ex3 Ex4 Ex5 Ex6 Ex7 Ex8 Ex9 Ex10

E6 (13) E6 (47) E6 (24) E6 (92) E6 (97) E6 (149) E7 (7) E7 (12) E7 (88) E7 (5) b

9 9 10 10 10 10 9 10 10 11

KLPDLCTEL FAFKDLFVV SLQDIEITCV KLTNTGLYNL GLYNLLIRCL RLQRRRETQV TLQDIVLHL VLHLEPQNEI QLFLNTLSFV KATLQDIVLHL

a The list does not represent any preference within the predicted high binding group. b An 11-amino-acid peptide was also designed which contained three smaller sequences, ATLQDIVLHL, TLQDIVLHL, and KATLQDIVLHL, each predicted to have moderate probability of binding.

binding activity and were selected for the immunogenicity studies. Generation of Cytotoxic Cells The four binding peptides were evaluated for their ability to induce a specific in vitro cytotoxic response using peripheral blood mononuclear cells from two HLA-A2-positive donors that were not likely to have been exposed to HPV-18 infection. These donors’ reactive cells were stimulated weekly for 4 weeks with the four synthetic HPV peptides, and then chromium-release cytotoxicity assays were performed. The antigen processing defective T2 cells, which express unoccupied surface HLA-A2 molecules, were loaded separately with the four HPV-18 peptides and served as targets cells to assess peptidespecific lysis. From donor A, CTLs stimulated by peptides Ex9 and Ex10 were recognized and lysed T2 cells were loaded with the corresponding HPV-18 peptide. When the donor PBMC were incubated with T2 cells without peptide, minimal or no lysis was observed (Figs. 1a and 1b). In order to assess each donor’s capacity to generate CTL, an immunogenic control peptide (CP) was used. Cytotoxic T-lymphocytes stimulated with CP were able to similarly lyse T2 cells loaded with CP but not T2 cells without peptide (data not shown). From donor B, CTL stimulated by Ex1, Ex2, and Ex9 were able to lyse T2 cells loaded with the corresponding homologous peptide but not T2 cells without peptide (Figs. 2a–2c, respectively). These results demonstrate that the HPV-18 peptides were able to generate peptide-specific cytotoxic cells and that the peptides induced a heterogeneous immune response in each donor tested. Assessment of Nonspecific Cell Lysis The specificity of lysis was further assessed for nonspecific lysis. Low levels of IL-2 were used throughout the 4 weeks of

80

CASTELLANOS, HAYES, AND MAIMAN

FIG. 1. Donor A. PBMC stimulated with irradiated T2 stimulator cells pulsed with peptides Ex9 (a) or Ex10 (b) generated specific cytotoxicity against T2 target cells pulsed with the corresponding peptide. Minimal lysis was observed on T2 cells without peptide.

peptide stimulation in order to replace the helper cell function. Since IL-2 might activate NK cells, and the T2 cells express other unmatched HLA molecules which might induce an allogeneic response, donor cytotoxic cells were assessed on both the NK cell-sensitive K562 cell line and the LAK cell indicator, the Raji cell line. No significant lysis was observed (data not shown).

DISCUSSION The association of HPV with 90 –95% of malignant and premalignant cervical lesions [2, 5, 6] provides a unique opportunity for developing peptide-based vaccine immunotherapy, targeting all stages of disease. This strategy recognizes the role of HPV-encoded proteins in the development and main-

HPV-18 CYTOTOXIC CELLS

81

FIG. 2. Donor B. PBMC stimulated with irradiated T2 cells pulsed with peptides Ex1, Ex2, or Ex9 for 4 weeks in vitro lysed T2 target cells pulsed with the corresponding peptide (a, b, and c, respectively). There was minimal lysis of T2 target cells without peptide by these responder cytotoxic cells.

tenance of the transformed state [4]. The HPV oncoproteins E6 and E7 provide ideal targets for vaccine development. Several lines of evidence demonstrate the importance of immune surveillance in patients infected with HPV, including the observation that patients infected with the human immunodeficiency virus have a disproportionate increase in incidence and a faster rate of progression of cervical cancer [21]. Studies have focused on HPV type-16, but HPV type-18 is also an important

etiologic agent [3, 6], especially since it has been implicated in rapidly developing and potentially more aggressive cervical carcinomas [22, 23]. Studies examining peptide epitopes from HPV type-18 for the purpose of developing cervical cancer immunotherapy are very limited. There is only one published study identifying peptides from HPV type-18 E6 that were able to induce CTL [16]. In this study we examined both HPV type-18 E6 and E7

82

CASTELLANOS, HAYES, AND MAIMAN

FIG. 2—Continued

peptides for the purpose of identifying cytotoxic T lymphocyte epitopes. In the first part of this study we were able to identify and select 10 peptides that have relatively high binding activity for HLA-A2 molecules and thus are more likely to be immunogenic [12]. From a pool of 295 possible decamer or nanomer sequences, the determination of HLA allele-specific primary and secondary motifs permitted analysis of peptides to be made without having to rely on random experimental testing. The selected peptides were assayed using a simple semi-quantitative method, which confirmed that 10 peptides had a high binding affinity for HLA-A2 molecules. Peptides Ex1, Ex2, Ex9, and Ex10 demonstrated the highest binding activity and were selected for further analysis of their cytotoxicity inducing ability. Using two different sources of unprimed “normal” donor mononuclear cells, we successfully induced peptide-specific cytotoxicity following 4 weekly stimulations with gammairradiated peptide-pulsed T2 cells in the presence of low-dose IL-2, which was added to circumvent loss of helper cell function. Functional effector cell specificity was then determined by comparing the lysis of homologous peptide-pulsed and -unpulsed T2 cells, as well as the responder cells’ ability to lyse NK-sensitive and -resistant target cells. Both donors A and B generated reasonable levels of cytotoxicity above background against peptide Ex9; Ex1 induced a minor response in both patients which was higher in Patient B, and the lytic response observed against Ex10 and Ex2 seemed to be more donor specific, in that Patient A responded better to Ex10 while Patient B responded moderately well to Ex2. In our laboratory, the amount of the HLA-A2 stabilization on the T2 cell line

generally plateaued at approximately 35% positivity. This heterogeneity seen in the T2 cell line maybe responsible for the relatively low amounts of cytotoxicity seen using peptidepulsed T2 target cells in these pilot studies. Although the absolute levels of cytotoxicity were easily lost by effector to target cell ratio titration, this is to be expected in a bulk culture of responder cells, which contains both B and T lymphocytes as well as NK cells, and monocytes. In order to confirm and characterize the nature of the actual responding cytotoxic cells, further analysis must be performed. One method might be to clone the responder cells so that a more homogenous group of cells can be studied. Another alternative is to use cell surface marker analysis in conjunction with intracellular cytokine determinations in order to identify what percentage of the “bulk cell” population activated by specific peptide was CTLs, B cells, or NK cells. Cytotoxic T lymphocytes should be CD3 and CD8␣␤ positive, B cells should be CD19⫹, and activated NK cells should be CD57, -56, or -11 positive. In addition, the production of interferon-␥ or tumor necrosis factor ␣ has been shown to correlate with cytotoxicity and cell lysis. Therefore, using flow cytometry one should be able to quantify the relative percentages of CTL activated in a bulk stimulated population, as well as correlate this information with immune function. Although the two unprimed donors that we tested both responded to Ex9, and both sets of donors were able to induce a cytotoxic response in vitro to this peptide, we need to confirm whether these are protective epitopes in vivo. In addition, further studies need to be done to evaluate the CTL response to these peptides in patients with persistent HPV infection, pre-

HPV-18 CYTOTOXIC CELLS

malignant cervical lesions, and cervical cancer. The results of these studies will facilitate the translation of this type of immunotherapy into the patient care setting. In summary, we have identified four peptides that can induce the development of cytotoxic cells in vitro in unprimed donors. Individually, or in combination, these peptides may represent a potential vaccine against cervical cancers caused by HPV type-18.

12.

13.

ACKNOWLEDGMENTS Funding for this project was made possible by the Staten Island University Hospital Research Fund, the Department of Medicine, and the Nalitt Institute for Cancer and Blood-Related Diseases.

REFERENCES 1. Cannistra SA, Niloff JM. Cancer of the uterine cervix. N Engl J Med 1996;334:1030 – 8. 2. Parkin DM, Pisani P, Ferlay J. Estimates of the worldwide incidence of eighteen major cancers in 1985. Int J Cancer 1993;54:594 – 606. 3. Birley HD. Human papillomaviruses, cervical cancer and the developing world. Ann Trop Med Parasitol 1995;89:453– 63. 4. Turek PL, Smith EM. The genetic program of genital human papillomaviruses in infection and cancer. Obstet Gynecol Clin North Am 1996;23: 735–58. 5. Stoler MH, Rhodes CR, Whitbeck A, Wolinsky SM, Chow LT, Broker TR. Human papillomavirus type-16 and 18 gene expression in cervical neoplasias. Hum Pathol 1992;23:117–28.

14.

15.

16.

17.

18.

6. Pfister H. The role of human papillomavirus in anogenital cancer. Obstet Gynecol Clin North Am 1996;23:579 –95. 7. Nijenhuis M, Schmitt S, Armandola E, Obst R, Brunner J, Hammerling GJ. Identification of a contact region for peptide on the TAP1 chain of the transporter associated with antigen processing. J Immunol 1996;156: 2186 –95. 8. Tanaka K, Tanahashi N, Tsurumi C, Yokota K, Shimbara N. Proteasome and antigen processing. Adv Immunol 1997;64:1–38. 9. Yang Y, Sempe P, Peterson PA. Molecular mechanisms of class I major histocompatibility complex antigen processing and presentation. Immunol Res 1996;15:208 –33. 10. Ressing ME, van Driel WJ, Celis E, Sette A, Brandt R, Hartman M, Anholts J, Schreuder G, Harmsel W, Jan Fleuren G, Trimbos BJ, Kast M, Melief C. Occasional memory cytotoxic T-cell response of patients with human papillomavirus type-16-positive cervical lesions against a human leukocyte antigen-A*0201-restricted E7-encoded epitope. Cancer Res 1996;56:582– 8. 11. Alexander M, Salgaller ML, Celis E, Sette A, Barnes W, Rosenberg SA, Steller MA. Generation of tumor-specific cytotoxic T lymphocytes from

19.

20.

21.

22.

23.

83

peripheral blood of cervical cancer patients by in vitro stimulation with a synthetic human papillomavirus type-16 E7 epitope. Am J Obstet Gynecol 1996;175:1586 –93. Ressing ME, Sette A, Brandt R, Ruppert J, Wentworth PA, Hartman M, Osefoff C, Grey HM, Melief C, Kast W. Human CTL epitopes encoded by human papillomavirus type-16 E6 and E7 identified through in vivo and in vitro immunogenicity studies of the HLA-A*201-binding peptides. J Immunol 1995;154:5934 – 43. Feltkamp MC, Vreugdenhil GR, Vierboom MPM, Ras E, van der Burg SH, ter Schegget J, Melief CJM, Kast W. Cytotoxic T lymphocytes raised against a subdominant epitope offered as a synthetic peptide eradicate human papillomavirus type-16-induced tumors. Eur J Immunol 1995;25: 2638 – 42. Feltkamp MC, Smits HL, Vierboom MPM, Minnar RP, de Jongh BM, Drijfhout JW. Vaccination with cytotoxic T lymphocyte epitope-containing peptide protects against a tumor induced by human papillomavirus type-16-transformed cells. Eur J Immunol 1994;23:2242–9. Bosch FX, Manos MM, Munoz N, Sherman M, Jansen AM, Peto J, Schiffman MH, Moreno V, Kurman R, Shah KV. Prevalence of human papillomavirus in cervical cancer: a worldwide perspective. International biology study on cervical cancer (IBSCC) study group. J Natl Cancer Inst 1995;87:796 – 802. Yoon H, Chung M, Min S, Lee HG, Yoo WD, Chung KT, Jung N, Park SN. Synthetic peptides of human papillomavirus type-18 E6 harboring HLA-A2.1 motif can induce peptide-specific cytotoxic T-cells from peripheral blood mononuclear cells of healthy donors. Virus Res 1998;54: 23–9. Castellanos MR, Hayes RL. A simple algorithm to identify potential cytotoxic T-lymphocytes epitopes from HLA-A2(⫹) donors. Submitted for publication. Kast W, Brandt R, Sidney J, Drijfhout J, Kubo RT, Grey HM, Melief C, Sette A. Role of HLA-A motifs. Identification of potential CTL epitopes in human papillomavirus type-16 E6 and E7 proteins. J Immunol 1994; 152:3904 –12. Ruppert J, Sidney J, Esteban C, Kubo RT, Grey HM, Sette A. Prominent role of secondary anchor residues in peptide binding to HLA-A2.1. Mol Cell 1993;74:929 –38. Regner M, Claesson MH, Bregenholt S, Ropke M. An improved method for the detection of peptide-induced upregulation of HLA-A2 molecules on TAP-deficient T2 cells. Exp Clin Immunogenet 1996;13:30 –5. Wright TC, Sun XW. Anogenital papillomavirus infection and neoplasia in immunodeficient women. Obstet Gynecol Clin North Am 1996;23:861– 93. Kurman RJ, Schiffman MH, Lancaster WD, Reid R, Jenson AB, Temple GF, Lorincz AT. Analysis of individual human papillomavirus types in cervical, neoplasia: a possible role for type-18 in rapid progression. Am J Obstet Gynecol 1988;159:293– 6. Hildesheim A, Hadjimichael O, Schwartz PE, Wheeler CM, Barnes W, Lowell DM, Willett J, Schiffman M. Risk factors for rapid-onset cervical cancer. Am J Obstet Gynecol 1999;180:571–7.