Radiol Oncol 1998; 32(1): 41-108
In vitro generation of cytotoxic T lymphocytes against mutated ras peptides
Antonio Juretic'l, Mirko Samija1, Zdenko Krajina1, Damir Eljuga1, Marko Turic1, Michael Heberer 2 and Giulio C Spagnoli 2
1 University Haspital far Tumars, Zagreb, Croatia and 2 Departments af Surgery and Research-ZLF, University af Basel, Basel, Switzerland
Mutatians af the ras prata-ancagenes represent frequent genetic alteratians in human cancers. These mutatians appear as single-paint mutatians causing single amina acid substitutians at residues 12, 13 ar 61 (activated p21 ras proteins). Therefare, peptides encompassing ras mutations appear to represent an appealing target far active immunotherapy procedures. By using a camputer program, selecting appropriate HLA-A2.1 binding motifs fram defined protein sequences, ras nonapeptides encompassing mutations in positions 12 and 61 with a putative binding capacity far HLA-A2.1 molecules were identified and synthesized. Generation of primary cytotoxic T cell lymphocyte (CTL) respanse was attempted by weekly restimulations of peripheral blood lymphocytes from healthy donors in the presence of irradiated autologous Epstein Barr virus transformed lymphablastoid cells (EBV cell lines) and IL-2 and IL-4. Periadical-ly, cultured cells were tested far their killing capacity using as target cells HLA-A2.1+ EBV cells previaus-ly pulsed with different combinations of the peptides under investigation. After eight rounds of restimulation reproducible cytotoxic activity against EBV target cells pulsed with two nonamers encompassing ras 61 Gin—Leu mutation was detectable in one donor. Thus, the results obtained indicate that it is possible to induce from PBMC of healthy donors CTL specific far peptides encompassing 61 GLN—LEU ras gene mutation following repeated weekly in vitro restimulations.
Key words: oncogene protein p21 (ras); piont mutatian; T-lymphacytes, cytotoxic
Introduction
Neoplastic transformation is caused by a stepwise accumulation of a series of genetic alterations affecting oncogenes and tumor supressor genes. The expression of mutated
Correspondence to: Prof. dr. sc. A. Juretic, Department of Radiation Oncology, University Hospital for Tumors, HR-10'000 Zagreb, Ilica 197, Croatia. Fax: +385-1-3775 536.
proteins encoded by these genes is restricted to abnormal cells thus raising the possibility to consider these molecules as tumor specific antigens. On the other hand, the longstanding goal of cancer immunotherapy is to stimulate the immune rejection of tumors. Based on the assumption that T lymphocytes might be able to eradicate cancer cells as effectively as they kill autologous virus-infected cells or allogeneic cells, tumor immunologists have
42
JureticA et al.
been trying to identify specific target antigens displayed by cancer cells that could make them recognizable to cytolytic T lymphocytes (CTL). The identification and selection of these potential human cancer antigens and specific epitopes as targets for CTL is now in a highly dynamic phase. Specific peptides that bind to human major histocompatibility complex (HLA) molecules have now been identified for melanoma-associated antigens. The identification of other human carcinoma-associated antigens and epitopes that can be recognized by human T cells is also under active investigation. Molecules, such as prostate specific antigen (PSA), c-erbB/2, MUC-1, point mutated ras, point mutated p53, and carcinoembryonic antigen (CEA) are possible such candidates.1,2,3
We investigated the potential antigenic epitopes encompassing mutations of ras proteins using as responders, peripheral blood lymphocytes from healthy donors. Mutations of the ras proto-oncogenes represent frequent genetic alterations in human cancers, detectable in 90 % of pancreatic carcinomas, 40-50 % of colorectal tumors, 20-40 % of lung carcinomas and approximately 30 % of acute myelogenous leukemias. These mutations appear as single-point mutations causing single amino acid substitutions at residues 12, 13 or 61 (activated p21 ras proteins).4'6
Therefore, we explored the possibility to define conditions permitting the use of synthetic peptides encompassing „mutant" residues of activated ras proteins for the induction of specific CTL responses. In the present study we applied a recently described protocol,7 which is based on stimulation of large numbers of naive lymphocytes with the antigen (peptides) and on concomitant application of Tetanus toxoid, IL-2 and of IL-4.
Materials and methods
Cells and media
The medium used throughout this study was RPMI 1640 supplemented with 2 mM L-glut-amine, 1 % non-essential amino acids, 1 % sodium piruvate, penicillin (100 U/ml), streptomycin (100 mg/ml) (all from Gibco Ltd, Paisley, UK) and with 10 % heat inactivated human AB-serum (Blutspendezentrum, SRK, Basel) (complete medium). Human peripheral blood mononuclear cells (PBMC) were obtained from the heparinized peripheral venous blood from a group of HLA-A2.1 healthy donors. Mononuclear cells were isolated by standard gradient centrifugation over Lymphoprep cushion (800 g for 20 minutes). After two washings cells were resus-pended in complete medium. PBMC were also used as a cellular source for the generation of Epstein-Barr virus (EBV) transformed lymphoblastoid cell lines.8
Synthetic peptides
Peptides were synthesized by solid-phase method using 9050 Millipore peptide synthesizer (Millipore, Volketswil, Switzerland). Synthesis was performed as suggested by the manufacturer. The peptides were purified to homogeneity and analyzed by HPLC. Their purity as analyzed by HPLC was routinely found to exceed 90 %. Synthesized peptides were designed by using a computer program (a gift of Dr. J. D'Amaro, Leiden, Holland;9) based on the amino acid motifs found in the known HLA-A2.1 binding peptides. This program represents a useful scanning tool for the identification of potential HLA-A2.l restricted peptides. We thus synthesized a panel of seven peptide nonamers encompassing ras mutations in positions 12 and 61. The sequences of these seven nonamers (singleletter code sequences) together with their arbitrary HLA-A2.1 binding scores are
T cell response against mutated ras peptides
43
Table l. Synthetic peptides used in vitro either for the generation of ras specific CTL or for the pulsing of EBV target cells in cytotoxic assays. An amino acid (aa.) sequence and HLA-A2.1 binding score of the seven synthesized nonapeptides encompassing ras oncogene mutations in positions 12 and 61 is indicated
Peptide	a.a.	Mutation position	a.a. change	Amino acid sequence	HLA-A2.1 binding scorex
poolx 1					
ras 1 ras 2 ras 3	61 61 61		Q toL QtoL QtoL	-G( nLEEYSAMR(8--L50D I L D TAG Ln - -i55ldtaglee63 -	96 48 48
pool 2					
ras 4 ras 5	61 61		QtoK QtoK	-t58 a g k e e y s a(( - -I55 L D T A K K E E63 -	96 48
pool 3					
ras 6 ras 7	12 12		G to V G to V	-V9 GAV G V G K S17 - -y4klvvvgav12-	128 96
x Initially, in vitro stimulation of PBMC was attempted by using pools of two or three synthesized peptides. Binding score, expressed as arbitrary units, was calculated by taking advantage of a specific computer program (9).
reported in Table l. A correlation can be expected between high score and binding to HLA-A2.1.9 The peptides were dissolved in DMSO (Sigma, Buchs, Switzerland) as a stock solution (at 20 mg/ml). For in vitro experiments they were further diluted in complete medium (10 mg/ml final concentration).
in the presence of irradiated, autologous EBV cells pulsed with combinations of appropriate peptides (pool 1, pool 2 or pool 3 peptides). Exogenous IL-2 and IL-4 were added twice per week at low doses (10 U/ml and 1 U/ml, respectively). Periodically, 51 Cr release tests (killing assays) were performed, by using, as target cells HLA-A2.1+ EBV cells,
Generation ofCTL.
Generation of primary CTL response was attempted by taking advantage of a recently described protocol.7 The initial in vitro stimulations were performed by using „bulk cultures" containing large numbers of PBMC in the presence of synthetic peptides. During the initial seven days the medium was also supplemented with tetanus toxoid (1 mg/ml) to provide „helper" function. Briefly, 25-75 x 106 PBMC were stimulated with mixtures of ras-derived peptides (pool 1, pool 2 or pool 3 peptides, Table 1) in 10 ml of complete medium in the presence of tetanus toxoid (1 mg/ml). Subsequently, Ficoll purified lym-phoblasts underwent weekly restimulations
Effector - to - target ratio
Fi^ue l. Specificity of bulk-cultured CTL undergoing in vitro re-stimulations with mutant ras peptides 1, 2 and 3 (pool 1 peptides). As target cells, HLA-A2.1+ EBV-trans-formed B cells, pulsed with indicated individual peptides, were used.
60
50
40
30
20
10
1.2: 1
44
]ureticA et al.
previously pulsed with the different combinations of peptides under investigation. Briefly, EBV cells were first labelled with a Na51Cr solution (Dupont, Regensdorf, Switzerland) as described.10 Afterwards, they were pulsed for 1 hr at 37 ° C with defined peptides at 10 mg/ml final concentration. Following extensive washings, target cells were resuspended at 105 cells/ml final concentration and added to different numbers of effector cells in U-bottom, 96-well trays. Cultures were incubated for 4 hours at 37 C in a 5 % CO2 humidified atmosphere. Supernatants were then collected and specific 51 Cr release was calculated according to the standard formula as described.10 Antibody blocking studies were performed by adding monoclonal antibodies in the form of hybridoma super-natants to killing assays, at 1:3 final dilution.
Results and discussion
Using a panel of mutated p21 ras peptides a significant specific killing could be detected after about 8 restimulation cycles in one PBMC donor out of five tested. We were able to demonstrate that the CTL recognized a p21 ras nonapeptide encompassing a Q—L (Gln—Leu) mutation in position 61. CTL recognized EBV target cells preincubated with either peptide 2 or peptide 3. The results
from one such representative experiment are presented in Figure l. Preincubation of EBV target cells with wild type peptides did not result in their recognition by CTL and subsequent killing (Table 2). Killing of EBV target cells preincubated with a mutant ras non-apeptide could be blocked by anti-HLA-A2.1 monoclonal antibodies, consisted with a specifically restricted recognition (Table 3). When further analyzed, our CTL appeared to be monoclonal in nature, since all CDS+ T cell clones generated express Vb14 gene product in combination with Jf32.7 and Cf32 (data not shown,11).
Our data are in agreement with similar results obtained by others, by taking advantage of different culture conditions.12 The characterization of mechanisms underlying HLA class I restricted antigen presentation and the identification of the peptide motifs allowing binding of antigenic epitopes to defined HLA determinants permit testing the immunogenicity of specific reagents in terms of capacity to induce CTL responses. Considering the role of mutated oncogenes or of tumor suppressor genes in the transformation processes and their specific expression in neoplastic cells, their products could represent the ultimate target for tumor specific active immunotherapies. Accordingly, we report here that we were able to generate in vitro a primary peptide specific CTL response
Table 2. Recognition of EBV target cells pulsed by wild type ras peptide by CTL generated with the pool 1 oi mutated ras peptides
Effector to target	% specific 51Cr release
ratio	EBV target cells preincubated with
exp. 1	mutant ras 2 peptide	wild type peptide
2:1	32	0
1:1	28	1
0.5:1	12	0
exp. 2	mutant ras 3 peptide	wild type peptide
5:1	56	13
2.5:1	35	2
1.2:1	22	0
T cell response against tmutated ras peptides Table 3. Effect of anti-HLA-A2.1 monoclonal antibody on the killing of EBV target cells
45
Effector to target	% specific 51Cr release
ratio	EBV target cells preincubated with ras 3 mutant peptide
monoclonal antibody	monoclonal antibody
anti-HLA-A2.1	anti-Mage3
10:1	7	69
5:1	5	72
2.5:1	1	45
1.2:1	3	32
against peptides encompassing mutation at position 61 in ras oncogene. The limitations of the experimental system described above are the number of restimulation cycles (8 rounds) and low frequency of responding donors (one out of five). Therefore, research for more practical immunization conditions is currently being pursued. In addition, the capacity to generate specific CTL should be comparatively analyzed in patients and healthy donors.
Acknowledgement
This work was supported by grants from Swiss Cancer League (grant 466 to A.J.), and by Swiss National Fond (grant 3100-039509 to A.J.).
References
1.	Cheever MA, Disis ML, Bernhard H, Gralow JR, Hand SL, Huseby ES, Qin HL, Takahashi M, Chen W. Immunity to oncogenic proteins. Immunol Rev 1995; 145: 33-59.
2.	Urban JL, Schreiber H. Tumor antigens. Annu Rev Immunol 1992; 10: 617-44.
3.	Van Pel A, van der Brüggen P, Coulie PG, Brichard VG, Lethe, B., Van den Eynde, Uyttenhove, C., Renauld, J-C., Boon, T. Genes coding for tumor antigens recognized by cytolytic T lymphocytes. Immunol Rev 1995; 145: 229-50.
4.	Barbacid M. ras genes. Ann Rev Biochem 1987; 56: 779-827.
5.	Barbacid M. ras oncogenes: their role in neoplasia. Eur J Clin Invest 1990; 20: 225-35.
6.	Bos JL. ras oncogenes in human cancer: a review. Cancer Res 1989; 49: 4682-9.
7.	Cerny A, Fowler P, Brothers MA, Houghton M, Schlicht H, Chisari F. Induction in vitro of a primary human antiviral cytotoxic T cell response. Eur J Immunol 1995; 25: 627-30.
8.	Spagnoli GC, SchaeferC, WillimannTE, KocherT, Amoroso A, Juretic A, Zuber M, Lüscher U, Harder F, Heberer M. Peptide-specific CTL in tumor-infiltrating lymphocytes from metastatic melanomas expressing MART-1/Melan-A, gp100 and Tyrosinase genes : a study in an unselected 12:oup of HLA-A2.1-positive patients. Int. J. Cancer (Fred. Oncol.) 1995; 64: 309-15.
9.	D'Amaro J, Houbiers JG, Drijfhout JW, Brandt RM, Schipper R, Bavinck JN, Melief CJM, Kast WM. A computer program for predicting possible cytotoxic T lymphocyte epitopes based on HLA class I peptide-binding motifs. Hum. Immunol 1995; 43:13-8.
10.	Juretic A, Spagnoli GC, von Bremen K, Hörig H, Filgueira L, Liischer U, Babst R, Harder F, Heberer M. Generation of lymphokine-activated killer activity in rodents but not in humans is nitric oxide dependent. Cell Immunol 1994; 157: 462-77.
11.	Juretic A, Juergens-Goebel J, Schaefer C, Noppen C, Williman T, Kocher T, Zuber M, Harder F, Heberer M, Spagnoli GC. Cytotoxic T lymphocyte responses against mutated p21 ras peptides: an analysis of specific T cell receptor gene usage. Int J Cancer 1996; 68: 471-8.
12.	Van Elsas A, Nijman HW, Van der Minne CE, Mourer JS, Kast WM, Melief CJM, Schrier Pl. Induction and characterization of cytotoxic T-lym-phocytes recognizing a mutated p21ras peptide presented by HLA-A*0201. Int J Cancer 1995; 61: 389-96.