T-Cell Immunity to Oncogenic Proteins Including Mutated ...€¦ · CHEEVER ef a/.:T-CELL IMMUNITY 103 that synthetic rus peptides elicited T cells specifically immune to the sensitizing

T-cell Immunity to Oncogenic Proteins Including Mutated RAS and Chimeric

BCR-ABL” MARTIN A. CHEEVER,b WE1 CHEN,b MARY L. DISIS,b

MASAZUMI TAKAHASHI,b AND DAVID J. PEACE“ bDivision of Medical Oncology

Mail Stop RM-I 7 University of Washington

Seattle, Washington 98195 ‘Department of Medicine

Loyola University Chicago May wood, Illinois 60 I53

Extensive studies in animal models have shown that immune T cells that are specifically immune to the malignant cells can cure advanced malignancy. T-cell therapy is proportionate to the number of immune T cells used, with larger doses inducing longer survival times and greater percentages of cures. Immunization in viuo (vaccination) with tumor or tumor antigen functions to increase antitumor immunity by inducing the growth of effector T cells in viuo, but in uiuo growth of T cells is severely limited by stringent normal control mechanisms. The use of T cells activated and grown in uitro can circumvent the normal in uiuo control mechanisms. By growing small numbers of immune T cells to large numbers in vitro and by transferring the cultured T cells in uivo, levels of immunity can be achieved that otherwise are impossible. Using this method it is now theoretically feasible to “flood” human cancer patients with immune T cells reactive against antigens expressed by cancer cells and capable of specifically killing cancer cells. The techniques of adoptive T-cell therapy with cultured T cells were recently applied to the treatment of human viral diseases3 The major remaining issue to be resolved before being able to test immune T-cell therapy against cancer in humans is to determine which antigens are appropriate targets for T-cell attack.

The existence of tumor-specific antigens first gained experimental credence in animal transplantation models approximately three decades a g ~ . ~ - ~ The advent of lymphocyte culture and cloning techniques has enabled investigators to isolate human T cells from tumor-bearing patients who exhibit apparent specific reactivity to autologous tumor cells. This indicates that many different types of human tumors express antigens recognizable by autologous T cells.’-13 However, molecu- lar identification of the antigens recognized, with few exception^,'^.^^ has been unsuccessful. Most studies to identify human tumor targets recognizable by au- tochthonous T cells have isolated cytotoxic T lymphocytes (CTL) from tumor- infiltrating cell populations or from peripheral blood and have attempted to identify the antigens recognized by the existent CTL. An alternative ploy presented in

This work was supported by NIH Grants R37 CA30558, ROI CA54561, ROI CA57851, and T32 CA09515.

101

102 ANNALS NEW YORK ACADEMY OF SCIENCES

this report is to assess whether known and already characterized proteins that are uniquely expressed by malignant cells can serve as T-cell targets.

IMMUNITY TO RAS PROTEINS

Ras protooncogenes encode a highly conserved family of 2 1-kD proteins collec- tively designated as p21 .I4.I5 ~ 2 1 " ~ proteins are intermediary signaling proteins that regulate cell growth and differentiation.I6 The p21 '(Is proteins bind to the inner aspect of the cell membrane, associate with guanosine nucleotides, and have intrinsic GTPase activity. In animals, carcinogens induce very specific and predict- able nucleotide substitutions in rus genomic DNA, usually at or near codons 12 or 61. The resultant single amino acid substitutions at residues 12 or 61 in the primary protein structure alter the tertiary structure and impart transforming activity.

In humans, a substantial proportion of malignancies of a variety of histologic types are associated with activating mutations in the same codons of ras. As examples, activating mutations of ras occur in approximately 90% of pancreatic adenocarcinoma, 60% of follicular carcinoma of the thyroid, 50% of colon adenocarcinoma, 40% of seminoma, 30% of lung adenocarcinoma, 30% of hepatocellular carcinoma, 20% of melanoma, 40% of myeloid leukemia, and 30% of multiple myeloma. 14.15*17-19 The human genome contains at least three rus proto-oncogene homologs, denoted as H-rus, K-rus, and N-ras, which are located on separate chromosomes.20 The gene structures and expressed proteins are highly conserved. In humans, each of the three rus genes can become activated through specific m u t a t i ~ n s . ~ ~ ~ I ~ . ~ I The mutations in human tumors presumably result from previous exposure to carcinogens.

Activating mutations of ras are prevalent in a wide variety of tumors. The same characteristic mutations of rus can be detected in patients with different types of tumors. Because activating amino acid substitutions occur in otherwise conserved regions of ~ 2 1 " ~ protein, only a restricted set of tumor-specific determinants is observed. Thus, activating mutations of rus engender a small, manageable set of shared tumor-specific determinants that should greatly facilitate the detection and analysis of immune responses to abnormal rus proteins among heterogeneous patient populations. Activating mutations of rus are intimately linked to the transforming process itself and may be indispensable for maintenance of the malignant phenotype. Activation of rus protooncogenes often occurs during the early stages of malignant transformation22; therefore, it may be possible to eliminate "premalig- nant" cells undergoing transformation with immunotherapy against rus determinants.

CD4+ Helperllnducer T Cells Immune to Synthetic ras Peptides Can Be Elicited by Immunization with Synthetic Peptides Identical to the

Mutated Segment

To determine whether the presence of a single transforming amino acid substitution can render normal ras protein immunogenic and thus a possible target for T- cell-mediated tumor therapy, B6 mice were immunized with synthetic peptides constructed to be identical to normal rus amino acid sequences but with amino acid substitutions identical to point-mutated rus. The result^^'^*^ demonstrated

CHEEVER ef a/.: T-CELL IMMUNITY 103

that synthetic rus peptides elicited T cells specifically immune to the sensitizing rus peptide but not to homologous rus peptides containing either the native amino acid sequence or alternative amino acid substitutions. T cells specific to rus peptides representing various activating amino acid substitutions of residue-12 or residue-61 could be elicited in various strains of mice. As an example, C57BL/6 (B6) mice were inoculated with adjuvant plus a synthetic peptide corresponding to residues 5-16 of p2lrUs protein containing the activating substitution of arginine for the native glycine at residue-12, designated as p5-16 [Arg-12]. One week later, lymphocytes obtained from the draining lymph nodes of mice immunized with peptide responded in uitro in proliferative assays to the p5-16 [Arg-12] peptide. A homologous p5-16 rus peptide bearing serine at residue-12 did not induce proliferation of p5-16 [Arg-l2]-primed lymph node cells, indicating that the immunogenic determinant was specifically defined by the Arg- 12 residue. Analogous results have been obtained with synthetic rus peptides corresponding to aberrant p2 1 ‘‘(IS proteins bearing alternative activating substitutions at either residue-12 or residue-6 1.

CD4+ T Cells Immune to the Majority of Potential ras Mutations Can Be Generated in Most Mice

Only a limited number of different transforming amino acid substitutions have been detected at residue position-12 or -61 in tumors that arise in uiuo. For example, only alanine, arginine, aspartic acid, cysteine, serine, or valine has been detected in lieu of the normal glycine at residue-12. The ability of several individual strains of mice to respond to each of the common activating substitutions at residue-12 has been systematically examined. BALB/c mice were immunized with rus peptides bearing normal glycine or various alternative amino acids at position-12. Mice did not respond to peptide corresponding to [Gly- 121, that is, the normal p21 ‘‘(Is sequence, and thus were presumably tolerant to native endogenous p21 sequences. Excluding the normal [Gly-12] rus peptide, BALB/c mice mounted strong proliferative responses to 3 of 5 rus peptides tested. The proliferative responses were strongest against the immunizing ~ e p t i d e , * ~ but cross-reactivity was observed between peptides bearing cysteine uersiis valine at residue-12. Other murine strains showed similar results but with responses to different sets of peptides. Thus, T cells from inbred strains of mice can specifically respond to many of the potential residue-12 substitutions of p21 “(Is that occur in transformed cells.

All of the synthetic peptides tested were immunogenic in some strains of mice, but not all of the peptides were immunogenic in all strains of mice examined. A lack of response to a particular synthetic peptide does not necessarily mean that the corresponding region of an activated p21 ‘‘(Is protein cannot be recognized by immune T cells. Subtle alterations in the size or the flanking regions of synthetic rus peptides can markedly affect the induction of T cells specifically immune to particular activating amino acid substitutions. For example, the rus peptide extending from residue 5 through 16 and corresponding to ~ 2 1 ‘ “ ~ proteins with an activating substitution of Arg-12 was not immunogenic in BALB/c mice. However, a slightly larger synthetic peptide which encompasses the same activating mutations, that is, Arg-12, was capable of eliciting specifically immune T cells in BALB/c mice (unpublished data). Similar results were obtained with peptides corresponding to p21 ‘(Is [Ser-121. The results indicate that the T-cell repertoire for aberrant p21 ‘[’’ proteins is potentially extensive in many murine haplotypes. There is no compelling reason to think that the T-cell repertoire would be less in humans.


In fact, inasmuch as individual outbred humans generally express two different alleles for each class I1 MHC molecule, the chance for response should be greater.

Ras Peptide-SpeciJic T Cells Respond to Intact p21 tps Protein

Peptide-specific T cells need not necessarily respond to the parental protein containing the peptide. Protein antigens generally must be internalized and de- graded to smaller peptide fragments by antigen-presenting cells (APC) for binding to class I1 MHC molecules, which is necessary for recognition by CD4+ T cells. Inadequate processing or inappropriate degradation of antigen may preclude class I1 MHC-restricted T-cell recognition. To determine if aberrant p21 rfls proteins are processed by syngeneic APC in a fashion appropriate for presentation to peptide- specific T cells, T cells immune to a ras peptide, p5-17 [Arg-12], were cultured with APC plus purified intact p21 )'(Is protein containing arginine at residue-12.24 A vigorous specific response was induced which was comparable to that induced by equivalent concentrations of p5-17 [Arg-l2] peptide. Fixation of the APC with gluteraldehyde prevented stimulation by the intact protein, but not by peptide, confirming that processing of the intact ~ 2 1 " ~ [Arg-12] protein was required for successful antigen presentation. Similar results were obtained with ras-specific T cells that were elicited with peptides corresponding to the aberrant region of p21 'o"

containing an activating substitution at residue-61, ~ 2 1 " ~ [Leudl]. Together, the results demonstrate that aberrant ras proteins bearing mutations at either residue- 61 or residue-12 can be appropriately processed and presented to specifically immune T cell^.^^.^^

Purified Aberrant p2ltSs Protein Elicits Specifically Immune T Cells in Vivo

The demonstration that intact p21 protein can be processed and presented to class I1 MHC-restricted T cells suggests that intact p2Ira" protein may be immunogenic in uivo. C3HIHen mice were immunized twice with p21'"" [Leu-611. Immunized mice developed CD4+ T cells that were specifically reactive to peptides corresponding to the altered portion of the p21 ras [Leu-611 protein. The results show that a T-cell epitope spanning the mutated region of a transforming p21 ras protein can be processed and presented in uivo in a manner appropriate for eliciting specific T-cell response^.^^ Thus, an abnormal p21 ras protein associated with malignant transformation can be antigenic in vivo. These results predict that in some circumstances transforming p2 1 ras proteins expressed by malignancies will be immunogenic and might result in priming of the host and the generation of detectable ras-specific T-cell responses. Although p21 )'('" is considered to be an intracellular protein, the protein has been detected in sera and bodily fluids of tumor-bearing patient^.^^.^' Thus, it is likely that p2lrflS derived from tumor in uivo is available in small amounts to APC for processing and presentation and is capable of stimulating class I1 restricted ras-specific T cells. The observation that p2lrUs protein can immunize in vivo increases the likelihood that primed T- and B-cell responses might be present in patients with ras-positive malignancies and begs the issue as to whether detection of immune responses to ~21"" can be used to indicate the presence of small foci of cancer in vivo.

CHEEVER er al.: T-CELL IMMUNITY 105

Tumors Expressing Aberrant ras Protein in vivo Elicit T Cells Specific for Corresponding ras Peptide

Studies examined whether aberrant p21 laS protein produced by tumor cells in vivo elicit T cells that proliferate specifically in response to the altered region of the ~ 2 1 ' " ~ protein. A Friend virus-induced leukemia cell line, designated as FBL, was transfected with the T24 Ha-rus-1 oncogene.28 The T24 oncogene encodes aberrant p2lrUs protein with valine substituted for the normal glycine at residue- 12.2' Syngeneic B6 mice were immunized twice at 2-week intervals with FBL cells transfected with the gene expressing p21 '"I [Vas-12] or with the normal Ha- rm-l gene. Spleen cells from mice immunized with FBL transfected with the T24 oncogene proliferated in response to the rus [Val-l2] peptide, whereas spleen cells from mice immunized with FBL transfected with normal Ha-rus-1 did not respond. The finding (unpublished data) that tumors which express aberrant p2lroS in vivo can elicit rus-specific T cells in vivo strongly implies that p2lrUs is available to APC for processing and presentation to class 11-restricted T cells in vivo.

Tumors Expressing Aberrant ras Protein Can Be Treated with T Cells Immune to the Mutated Segment

Presentation of the activated p21 rus in proximity to tumor cells may elicit class 11-restricted T cells that can mediate substantial therapeutic effects in vivo against tumors even if the latter do not express class I1 MHC molecules and cannot be directly recognized. Use of immune T cells in therapy is currently being evaluated in murine models. The Rus 4 tumor which expresses the transfected T24 H2-rus- 1 oncogene was used in therapy experiments in BALB/c x C57BL/6 F1 mice. Small doses of tumor cells injected intraperitoneally kill animals within 2 weeks. A T-cell line specific for the altered region of ~ 2 1 ' " ~ [Val-12] was tested for the ability to protect BALB/c x C57BL/6 F1 mice against challenge with RAS 4 tumor cells. In repeated experiments, mice injected with tumor alone all died by 3 weeks, whereas mice injected concurrently with tumor cells plus rus [Val-l2]- specific T cells all survived without evidence of tumor for longer than 8 weeks of observation (unpublished data). Whether the quantity of rus protein expressed by human tumors is adequate for stimulating therapeutic CD4+ T-cell responses is unknown.

Antibody to ras p21 Protein Can Be Detected in Patients with Colon Cancer

One step towards developing rus-specific T-cell therapy in humans is to determine if patients with tumors expressing mutant rus protein have existent T-cell responses to such proteins. Our studies have begun to examine patients for antibody responses to p21"" protein as an indicator of CD4+ T-cell responses. Nor- mally p2lroS is affixed to the inner aspect of the cell membrane; however, substantial amounts of ~21'"" are released into the sera and bodily fluids of tumor-bearing patient^.^^.^' Therefore, it is possible that individuals bearing tumors that express aberrant p21 can develop specific antibodies to the altered rus protein. In animal models, antibody specific for aberrant p21 protein can be elicited by immunization with synthetic rus peptides corresponding to the altered region of the protein.29 It is probable that intact p21 can similarly elicit antibodies specific for the altered domain of ~ 2 1 " ~ . In initial experiments, patients' sera were analyzed with an


enzyme-linked immunoassay (ELISA). A panel of purified p21 proteins with several known single amino acid substitutions was generated by prokaryotic expression of a mutated synthetic H-ras gene. Preliminary results demonstrated the existence of antibody to ras in three of six patients with colon cancer and none of six normal individuals (unpublished data). The antibody responses appear to be directed against both normal and mutated ras proteins. The sera responses to purified p2lrRs protein in ELISA were confirmed by Western analysis of two- dimensional gels. Whether some antibody is specific for only aberrant proteins is being evaluated.

The demonstration of ras-specific humoral responses in human cancer patients has several ramifications. First, antibody responses to small globular proteins such as p2 1 rus generally require cognate T-cell help. Therefore, detected humoral responses to p21 ras proteins containing a single amino acid substitution should signify the existence of CD4+ helper T cells which also can specifically recognize the aberrant p21 ras protein. Of note, the detection of primed CD4+ T-cell responses to ras-derived synthetic peptides in some patients with cancer30 provides additional inferential evidence that some patients with ras-positive malignancies have existent T-cell immunity to mutated p2 1 protein. Second, assays for ras-specific antibody responses may provide a highly sensitive method of identifying individuals who have been exposed to mutated p2 1 r"s protein. Accordingly, screening assays for ras-specific antibody may provide an early indication of incipient or occult malignancies undetectable by conventional methods. Third, analysis of clinical speci- mens for the presence of mutations of ras protooncogenes and corresponding ras- specific T cells is slow and inconvenient. It is possible that the assay for ras- specific antibody can predict the existence of indigenous ras-positive tumor cells. Current studies are evaluating whether antibody responses correlate with CD4+ T-cell responses. The detection of ras-specific antibody in patients may provide a rapid and efficient technique for identifying patients who have existent specific helper T-cell responses and ras-positive malignancies.

Synthetic Peptides Identical to the Mutated Segment of ras Protein Can Bind to Murine Class I MHC Molecules

Class I MHC-restricted T cells recognize peptides derived from the degradation of endogenously synthesized proteins. Determinant selection by class I MHC molecules is highly selective. Appropriate binding of antigenic peptides results in the formation of thermodynamically stable trimolecular complexes of peptide antigen, class I heavy chain and beta-2 microglobulin. The rules that determine the ability of antigen to be processed and complexed with class I MHC molecules are not completely understood. However, peptides derived from endogenous proteins that bind to particular class I MHC molecules share discernible sequence motif^.^' Thus far, only a limited number of class I MHC molecules have been analyzed for restriction motifs. However, it is now technically feasible to discern the critical anchoring residues of virtually any class I MHC molecules. We have observed that ras protein contains a peptide motif extending from residues 60 through 67 which includes putative anchoring residues appropriate for Kb-restricted binding.3' Utilizing a mutant cell line, denoted as RMA-S3*, which does not appropriately process endogenous antigen, we demonstrated that ras peptides can bind to murine class I MHC molecules.33 Whether or not the mutated segment of p21 ras is capable of binding to other class I MHC molecules including common human MHC antigens is an open experimental question.

CHEEVER el al.: T-CELL IMMUNITY 107

Cytotoxic T Cells Specific for the ras Peptides Can Be Generated by Primary in vitro Sensitization Using Appropriate ras Peptides

The demonstration that the region of p21 rus protein encompassing residue-61 includes a Kb restriction motif prompted studies to determine if class I-restricted CTL with specificity for the same region could be generated. C57BL/6(H-2b) spleen cells were cultured with ras [Leudl] peptide. After 7 days, primed T cells specific for Ras [Leudl ] peptide could be demonstrated. Ras [Leu-611 peptide- specific CTL could lyse EL4 target cells which were preincubated with ras [Leu-61] peptides to a greater extent than EL4 cells which were preincubated in medium alone or with irrelevant tryptic digests of ovalbumin.

The most important consideration was whether the ras peptide-specific CTL could lyse target cells expressing the corresponding p21 ras protein. Peptide- specific CTL need not necessarily lyse targets synthesizing this target protein. Under normal circumstances antigenic class I MHC-restricted peptide determinants are proteolytically excised from endogenously produced proteins. Potentially immunogenic determinants nested within an intact protein may be destroyed by inappropriate cleavage of proteins by proteosis during antigen processing. In addition, competing or suppressive determinants may mask CTL recognition of the potential target peptide segment. To establish whether endogenously expressed p21 rUE [Leudl] protein could be processed and presented to enable CTL recognition of the mutated segment of the protein, a fibroblast line of C57BL/6 origin was transfected with a mammalian expansion vector encoding p21 ‘(Is [Leudl ] protein and used as a target for testing CTL activity. The ras peptide-specific CTL could effectively and specifically lyse the ras transformed fibroblast line.33 Thus, activated p21 ras, a normal intracellular protein, is available for processing and presentation via the class I MHC molecule pathway, and recognition of oncogenic ras proteins by mammalian CTL is possible. The results predict that it will be possible to define circumstances for eliciting and therapeutically utilizing ras-specific CTL to treat established ras-positive malignancies.

IMMUNITY TO CHIMERIC BCR-ABL PROTEIN

The second system under investigation is immunity to bcr-abl protein. The hallmark of chronic myelogenous leukemia is the translocation of the human c- ab/ protooncogene from chromosome 9 to the specific breakpoint cluster (bcr) region on chromosome 22. The t(9;22) translocation results in the formation of a bcr-ablfusion gene that encodes at 210-kD chimeric protein with abnormal tyrosine kinase a ~ t i v i t y . ~ ~ . ~ ~ Bcr-abl proteins are appealing candidates for T-cell therapy for many reasons, including: ( I ) bcr-abl proteins are expressed only by malignant cells and shared by many individuals with chronic myelogenous leukemia (CML) and acute lymphocytic leukemia; (2) bcr-abl is associated with malignant transformation and maintenance of the malignant phenotype and thus antigen-negative varieties should not O C C U ~ ~ ~ . ~ ’ ; (3) the joining region segment is composed of a unique sequence of amino acids and thus is potentially immunogenic; (4) only three potential joining region segments are commonly observed; (5) detection of malignant cells and identification of particular joining region segments in individual patients is simple; (6) T-cell therapy of CML has already been shown to be effective in the form of allogeneic bone marrow transplantation with T cells directed against minor transplantation antigens3*; and (7) the potential problem of tolerance/anergy


which might prevent the generation of effective immune T cells from cancer patients can be circumvented by using donor bcr-abl-specific T cells as part of a sibling bone marrow transplantation procedure.

Immunization of Mice with Synthetic Peptides Corresponding to the Joining Region Segment of p210 bcr-abl Protein Can Elicit Peptide and Protein-Specijic

CD4+ Class 11 MHC-Restricted T Cells

BALB/c mice were immunized with peptides identical to the joining region segment composed of both bcr and c-abl amino acids.39 Peptide-specific T cells recognized only the chimeric joining region sequence. Peptide-specific T cells could also recognize and respond to p210 bcr-abl proteins. The response of peptide- specific CD4+ T cells to protein demonstrated that p210 bcr-abl protein can be processed by APC, so that the joining region segment is bound to class I1 MHC molecules in a configuration similar to that of the immunizing peptide and in a concentration high enough to stimulate peptide-specific T cells. In animals, cura- tive tumor therapy can be mediated by CD4+ T cells recognizing soluble tumor antigens. Whether bcr-abl protein is present in high enough concentrations in the extracellular environment in patients with CML to be an appropriate target for CD4' T-cell therapy has not yet been evaluated.

Class I MHC-Restricted bcr-abl Peptide-Specific CTL Can Be Elicited by Immunization with bcr-abl Peptides

In most normal effective immune responses, both class I and class I1 MHC- restricted T cells participate, and either subset alone can be effective in therapy. Which subset is most effective is most likely dependent on the availability of particular antigens to either the class I or class I1 antigen-processing pathways. Studies to elicit class I MHC-restricted CD8 + CTL have been greatly facilitated by the determination that CD8+ CTL recognize short peptides bound in the groove of class I MHC molecules and that peptides binding to particular class I MHC molecules have discernible amino acid sequence motifs. The most common bcr- abl proteins result from the transposition of the third bcr exon to the second c- abl exon termed the bcr3-abl2 chimeric protein. By comparing the amino acid sequence of the joining region segment of bcr3-a612 chimeric protein to the amino acid motifs of peptides capable of binding to MHC molecules, we determined that the peptide segments of the bcr3-abl2 joining region segment might bind to murine H-2Kd and to human HLA-A2.1. These peptides were synthesized.

The synthetic bcr3-abl2 peptides were shown, in fact, to be able to bind to murine H-2b class I MHC and to human HLA-A2.1. For binding assays, class I MHC stabilization of murine RMA-S and human T2 were used. Binding of peptides to MHC class I is the sine qua non of CTL generation. Methods for primary in uitro immunization were used to generate peptide-specific CTL. Priming of CTL has classically required in viuo immunization with stimulator cells synthesizing the nominated target protein and capable of processing and presenting that protein in the class I MHC pathway. Priming T cells in vitro has been difficult to achieve. However, the use of peptides with a precise amino acid sequence for binding in the groove of class I MHC molecules has allowed priming of CTL in uitro and in uiuo with exogenous peptides in some circumstance^.^^ Preliminary studies in mice showed that bcr3-ab12 peptide-specific CD8+ class I MHC-restricted CTL can be

CHEEVER e ta / . : T-CELL IMMUNITY 109

elicited by in vivo or in vitro immunization with synthetic peptides corresponding to the joining region segment of bcr-abl protein (unpublished data). The murine peptide-specific CTL recognized only the chimeric joining region sequence of bcr plus c-abl amino acids. Preliminary studies in humans showed that CTL reactive to bcr3-abI2 peptide can be generated by in vitro priming of normal donor peripheral blood lymphocytes with bcr3-abl2 peptide~.~'Thus, bcr-abl protein represents a potential tumor-specific antigen.

The observation that bcr-ablpeptides can bind to human class I MHC molecules and can elicit peptide-specific CTL shows that the generation of CML-specific CTL is possible. The remaining major step is to determine if bcr3-abl2 peptide- specific CTL can lyse leukemia cells transformed by bcr-abl chimeric protein or, alternatively, how best to generate bcr-abl-specific CTL capable of lysing the leukemia cells. Whether bcr-abl peptide-specific CTL can lyse CML cells is dependent upon whether the peptide-specific CTL have high enough affinity for the bcr- abl-peptidelMHC complex and whether the bcr-abl-peptide/MHC complex is present in high enough concentration on the surface of CML cells. Both issues are being evaluated.

SUMMARY

The process of malignant transformation can be ascribed to a series of characteristic and definable mutations of genes which encode proteins that control cell growth and differentiation. During the course of malignant transformation the cancer-related genes are altered by a variety of mechanisms including transloca- tions, deletions, and point mutations which commonly result in the expression of aberrant proteins. Our laboratory has focused on determining the extent to which cancer-specific proteins expressed by aberrant cancer-related genes can function as tumor-specific antigens. The current paper reviews our studies with two proto- type cancer-specific proteins, mutated p21 ros protein and chimeric p2106cr-"b' protein.

Ras protooncogenes are activated by point mutation in approximately 20% of human malignancies. The mutations occur primarily at codons 12 or 61 and result in the expression of p21r(Js proteins with single substituted amino acids. Only a limited number of amino acid substitutions occur. Murine studies demonstrate that immunization with synthetic peptides corresponding to the mutated segment can elicit both class I1 restricted CD4+ helpedinducer T-cell responses and class I restricted CD8+ cytotoxic T-cell responses specific for mutated ~21"" protein. In addition, the existence in vivo of tumors expressing mutated ras proteins can be detected by assaying for T-cell immunity to the mutated segment of ras protein. Preliminary human studies show that some patients with colon cancer have existent antibody responses to p21 protein, implying the possible existence of autochtho- nous T-cell immunity to mutated ras proteins in those patients.

In chronic myelogenous leukemia the human c-abl protooncogene from chromosome 9 is translocated to the specific breakpoint cluster (bcr) region on chromosome 22. The translocation results in the formation of a bcr-abl fusion gene that encodes at 210-kD chimeric protein. The joining region segment of chimeric bcr- abl protein is composed of a unique combination of c-abl and bcr amino acids and is expressed only by malignant cells. Studies demonstrate that immunization of mice with synthetic peptides corresponding to the joining region segment can elicit class I1 restricted CD4+ T-cell responses to p210hc"ub' proteins. Preliminary


studies show that bcr-abl peptides can bind in the groove of both murine and human class I MHC molecules and can elicit bcr-abl peptide-specific cytotoxic T lymphocytes (CTL). Whether bcr-abl peptide-specific CTL can lyse cells expressing bcr-abl protein is as yet unknown. In summary, the results of the studies reviewed confirm that cancer-specific oncogenic proteins can serve as tumor- specific antigens.

REFERENCES

I . CHEEVER, M. A., D. B. THOMPSON, J . P. KLARNET & P. D. GREENBERG. 1986. Antigen- driven long-term cultured T cells proliferate in uiuo, distribute widely, mediate specific tumor therapy and persist long-term as functional memory T cells. J. Exp. Med.

CHEN, W., V. A. REESE & M. A. CHEEVER. 1990. Adoptively transferred antigen- specific T cells can be grown and maintained in large numbers in uiuo for extended periods of time by intermittent restimulation with specific antigen plus IL-2. J. Immunol. 114: 3659-3666.

GREENBERG. 1992. Restoration of viral immunity in immunodeficient humans by the adoptive transfer of T cell clones. Science 257: 238-241.

BALDWIN, R. W. 1955. Immunity to methylcholanthrene-induced tumours in inbred rats following atrophy and regression of the implanted tumours. Brit. J . Cancer

PREHN, R. T. & J. M. MAIN. 1957. Immunity tomethycholanthrene-induced sarcomas. J. Natl. Cancer Inst. 18: 769-778.

KLEIN, G.. H. 0. S J ~ G R E N , E. KLEIN & K. E. HELLSTR~M. 1960. Demonstration of resistance against methycholanthrene-induced sarcomas in the primary autochtho- nous host. Cancer Res. 20: 1561-1572.

VOSE, B. M. & G. D. BONNARD. 1982. Human tumor antigens defined by cytotoxic and proliferating responses of cultured lymphoid cells. Nature 2 % 359-361.

MITSUYA, H., L. MATIS, M. MEGSON, P. BUNN. C. MURRAY, D. MANN, R. GALLO & S. BRODER. 1983. Generation of an HLA-restricted cytotoxic T cell line reactive against cultured tumor cells from a patient infected with human T cell leukemia/ lymphoma virus. J . Exp. Med. 158: 994-999.

cyte clones from peripheral blood and from tumor site detect intratumor heterogeneity of melanoma cells: Analysis of specificity and mechanisms of interaction. J. Immunol.

163: 1100-1112. 2.

3. RIDDELL, S. R., K. S. WATANABE, J. M. GOODRICH, C.-R. LI, M. E. AGHA & P. D.

4.

9: 652-657. 5 .

6.

7.

8.

9. ANICHINI, A., A. MAZZOCCHI, G. FOSSATI & G. PARMIANI. 1989. Cytotoxic T lympho-

142: 3692-3701. 10. COZZOLINO, F., M. TORCIA, A. M. CAROSSINO, R. GIORDANI, c. SELLI, c. T A L I N I ,

E. REALI, A. NOVELLI, V. PisToiA & M. FERRARINI. 1987. Characterization of cells from invaded lymph nodes in patients with solid tumors: Lymphokine requirement for tumor-specific lymphoproliferative response. J. Exp. Med. 166: 303-3 18.

SLOVIN, S. F., R. D. LACKMAN, S . FERRONE, P. E. KIELY & M. 1. MASTRANGELO. 1986. Cellular immune response to human sarcomas: Cytotoxic T cell clones reactive with autologous sarcomas. J. Immunol. 137: 3042-3048.

VAN DER BRUGGEN, P., C. TRAVERSARI, P. CHOMEZ, C . LURQUIN. E. DE PLAEN, B. VAN DEN EYNDE, A. KNUTH & T. BOON. 1991. A gene encoding an antigen recognized by cytolytic T lymphocytes on a human melanoma. Science 254: 1643-1647.

I. F. MCKENZIE, R. C. BAST, JR. & 0. J. F I N N . 1991. Cytotoxic T-lymphocytes derived from patients with breast adenocarcinoma recognize an epitope present on the protein core of a much molecule preferentially expressed by malignant cells. Cancer Res. 51: 2908-2916.

11.

12.

13. JEROME, K. R., D. L. BARND, K. M. B E N D T , ~ . M. BOYER,J.TAYLOR-PAPADlMlTRIOU,

14. BARBACID, M. 1987. Rus genes. Ann. Rev. Biochem. 56: 779-827.

CHEEVER eta/ . : T-CELL IMMUNITY 111

IS. 16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33

Bos, J. L. 1989. Ras oncogenes in human cancer. A review. Cancer Res. 49: 4682-4689. BOURNE, H. R., D. A. SANDERS & F. MCCORMICK. 1990. The GTPase superfamily:

A conserved switch for diverse cell functions. Nature 348: 125-132. ALMOGUERA, C., D. SHIBATA, K. FORRESTER, J. MARTIN, N. ARNHEIM & M.

PERUCHO. 1988. Most human carcinomas of the exocrine pancreas contain mutant c-K-ras genes. Cell 33: 549-554.

Bos, J. L., E. R. FEARON, S. R. HAMILTON, M. VERLAAN-DE VRIES, J. H. VAN BOOM, A. J. VAN DER EB & B. VOGELSTEIN. 1987. Prevalence of ras gene mutations in human colorectal cancers. Nature 327: 293-297.

NERI, A., J . P. MURPHY, L. CRO, D. FERRERO, C. TARELLA, L. BALDINI & R. DALLA-FAVERA. 1989. Ras oncogene mutations in multiple myeloma. J. Exp. Med. 7 0 1715-1725.

STAVNEZER, E., I. FOGH & M. H. WIGLER. 1983. Three human transforming genes are related to the viral ras oncogenes. Proc. Natl. Acad. Sci. USA 80: 21 12-21 16.

REDDY, F. P.. R. K. REYNOLDS, E. SANTOS & M. BARBACID. 1982. A point mutation is responsible for the acquisition of transforming properties by the T24 human bladder carcinoma oncogene. Nature 300: 149-152.

FORRESTER, K., C. ALMOGUERA, K. HAN, W. E. GRIZZLE & M. PERUCHO. 1987. Detection of high incidence of K-ras oncogenes during human colon tumorigenesis. Nature 32'1: 298-303.

CHEEVER, M. A., W. CHEN, H. NELSON, P. D. GREENBERG, V. K. LEE, K. D. CROSSLAND & D. J. PEACE. 1990. T cell immunity to the oncogenic form of ras protein can be induced by immunization with synthetic peptides. In Cellular Immunity and lmmunotherapy of Cancer. 295-302. M. T. Lotze & 0. J. Finn, eds. Wiley- Liss, Inc. New York.

PEACE, D. J., W. CHEN, H. NELSON & M. A. CHEEVER. 1991. T cell recognition of transforming proteins encoded by mutated ras proto-oncogenes. J. Immunol.

PEACE, D. J., J. W. SMITH, M. L . Disis, W. CHEN & M. A. CHEEVER. T cells specific for the mutated segment of activated p21"" protein can be elicited by immunization in uivo with ~21'"' protein. J. Immunother., in press.

NIMAN. H. L., A. M. H. THOMPSON, A. Yu, M. MARKMAN, J. J. WILLEMS, H. A. HERWIC, N. A. HABIB, C. B. WOOD, R. A. HOUGHTEN & R. A. LERNER. 1985. Anti-peptide antibodies detect oncogene-related proteins in urine. Proc. Natl. Acad. Sci. USA 82: 7924-7928.

NG, S. C., P. HAMER, D. PETIT, R. DELELLIS, H. WOLFE, D. GARLICK & W. CARNEY. 1989. Detection and quantitation of ras p21 expression in normal and tumor tissues by ELISA. FASEB J. 3: 151 1 (Abstr. 1657).

MIURA, K.,Y. INOUE,H. NAKAMORI, S. IWAI, E. OHTSUKA,M. I K E H A R A , ~ . NOGUCHI & S. NISHIMURA. 1986. Synthesis and expression o fa synthetic gene for the activated human c-Ha-ras protein. Jpn Cancer Res. n: 45-5 I .

PULLANO, T. G., E. SINN & W. P. CARNEY. 1989. Characterization of monoclonal antibody R256, specific for activated ras p21 with arginine at 12, and analysis of breast carcinoma for v-Harvey-ras transgenic mouse. Oncogene 4 1003- 1008.

E. THORSBY. 1993. Oncogene derived peptides. A new class of tumor rejection antigens. J. Cell Biochem. Suppl. 17D: 109 (Abstr. NZ 208).

FALK, K., 0. R~TZSCHKE, S. STEVANOVIC, G. JUNG & H A . RAMMENSEE. 1991. Allele-specific motifs revealed by sequencing of self-peptides eluted from MHC molecules. Nature 351: 290-296.

LJUNGGREN, H.-G., N. J. STAM, C. O H L ~ N , J. J . NEEFJES, P. H ~ G L U N D , M.-T. HEEMELS, J. BASTIN, T. N. M. SCHUMACHER, A. TOWNSEND, K. KARRE & H. L. PLOECH. 1990. Empty MHC class I molecules come out in the cold. Nature

PEACE, D. J., J. W. SMITH, W. CHEN, S . 4 . You, W. L. COSAND, J. BLAKE & M. A. CHEEVER. Lysis of ras oncogene-transformed cells by cytotoxic T lymphocytes specific for the oncogenic ~21'"' protein. Submitted.

146: 2059-2065.

GAUDERNACK, G . , T. GEDDE-DAHL, 3D, B. FOSSUM, B. H. OLSEN, J. A. ERIKSEN &

346: 476-480.

1 U ANNALS NEW YORK ACADEMY OF SCIENCES

34. KONOPKA, J. B. & 0. N. WITTE. 1984. An alteration of the human c-abl protein in K562 leukemia cells unmasks associated tyrosine kinase activity. Cell 37: 1035.

35. SHTIVELMAN, E., B. LIFSHITZ, R. P. GALE & E. CANAANI. 1985. Fused transcripts of abl and bcr genes in chronic myelogenous leukemia. Nature 315: 550.

36. YOUNG, J. C. & 0. N. WITTE. 1988. Selective transformation of primitive lymphoid cells by the bcrlabl oncogene expressed in long-term lymphoid or myeloid culture. Molec. Cell Biol. 8 4079-4087.

37. SZCZYLIK, C., T. SKORSKI, N. C. NICOLAIDES, L. MANZELLA, L. MALAGUARNERA, D. VENTURELLI, A. M. GEWIRTZ, & B. CALABRETTA. 1991. Selective inhibition of leukemia cell proliferation by BCR-ABL antisense oligodeoxynucleotides. Science

38. SULLIVAN, K. M., P. WEIDEN, R. STORB, R. P. WITHERSPOON, A. FEFER, L. FISHER, C. D. BUCKNER, C. ANASETTI, F. R. APPELBAUM, C. BADGER, P. BEATTY, W. BENSINGER, R. BERENSON, C. BIGELOW, M. A. CHEEVER, R. CLIFT, H. J. DEEG, K. DONEY, P. GREENBERG. J. A. HANSEN, R. HILL, T. LOUGHRAN, P. MARTIN, P. NEIMAN, F. PETERSON, J . SANDERS, J. SINGER, P. STEWART & E. D. THOMAS. 1989. Influence of acute and chronic graft-versus-host disease on relapse and survival after bone marrow transplantation from HLA-identical siblings as treatment of acute and chronic leukemia. Blood 73: 1720-1728.

39. CHEN, W., D. J . PEACE, D. K. ROVIRA, S.-G. You & M. A. CHEEVER. 1992. T cell immunity to the joining region of p210bcr-"b' protein. Proc. Natl. Acad. Sci. USA

40. CHEN, W., S. G. You, M. L. DISIS & M. A. CHEEVER. 1993. Evaluation of thejoining region segment of p2 I@cR"BL chimeric protein as a potential leukemia-specific antigen to elicit class I MHC-restricted CTL responses. J . Cell Biochem. (Abstr. Suppl. 17D): 107 (Abstr. NZ 203).

253: 562-565.

8 9 1468-1472.

Documents

T-Cell Immunity to Oncogenic Proteins Including Mutated ...€¦ · CHEEVER ef a/.:T-CELL IMMUNITY 103 that synthetic rus peptides elicited T cells specifically immune to the sensitizing