Heterogeneity in expression of human leukocyte antigens and melanoma-associated antigens in advanced melanoma

The study of tumor immunology has led to many innovative therapeutic strategies for the treatment of melanoma. The strategies are primarily dependent on melanoma-associated antigen peptide vaccination or T-cell-based therapy. These immunotherapies are totally reliant on proper copresentation of human leukocyte antigen class I molecules in sufficient quantity and the presence and availability of melanoma-associated antigenic peptides. Altered expression of either HLA class I molecules or melanoma antigens is known to occur. These defects lead to altered manufacture and copresentation of HLA class I molecules with melanoma-associated antigens to T-cells. Defects in any one combination can lead to loss of recognition of melanoma cells and their subsequent destruction by cytotoxic T-lymphocytes. Thus, these immunotherapy strategies can be thwarted by defects or heterogeneity of expression of human leukocyte antigen class I or of melanoma-associated antigens.     

Unbiased identification of target antigens of CD8+ T cells with combinatorial libraries coding for short peptides

Cytotoxic CD8(+) T cells recognize the antigenic peptides presented by class I major histocompatibility complex (MHC) molecules. These T cells have key roles in infectious diseases, autoimmunity and tumor immunology, but there is currently no unbiased method for the reliable identification of their target antigens. This is because of the low affinities of antigen-specific T cell receptors (TCR) to their target MHC-peptide complexes, the polyspecificity of these TCRs and the requirement that these TCRs recognize protein antigens that have been processed by antigen-presenting cells (APCs). Here we describe a technology for the unbiased identification of the antigenic peptides presented by MHC class I molecules. The technology uses plasmid-encoded combinatorial peptide libraries and a single-cell detection system. We validated this approach using a well-characterized influenza-virus–specific TCR, MHC and peptide combination. Single APCs carrying antigenic peptides can be detected among several million APCs that carry irrelevant peptides. The identified peptide sequences showed a converging pattern of mimotopes that revealed the parent influenza antigen. This technique should be generally applicable to the identification of disease-relevant T cell antigens.     

The specific T-cell response to antigenic peptides is influenced by bystander peptides

T lymphocytes recognize antigens in the form of peptides presented by major histocompatibility complex (MHC) molecules on the cell surface. Only a small proportion of MHC class I and class II molecules are loaded with foreign antigenic peptides; the vast majority are loaded with thousands of different self peptides. It was suggested that MHC molecules presenting self peptides may serve either to decrease (antagonistic effect) or increase (synergistic effect) the T cell response to a specific antigen. Here, we present our finding that transfected mouse fibroblasts presenting a single antigenic peptide covalently bound to a class II MHC molecule stimulated specific mouse T cell hybridoma cells to an interleukin-2 response less efficiently than fibroblasts presenting a similar amount of antigenic peptide in the presence of class II molecules loaded with heterogenous bystander peptides.     

Selective MHC expression in tumours modulates adaptive and innate antitumour responses

Progress towards developing vaccines that can stimulate an immune response against growing tumours has involved the identification of the protein antigens associated with a given tumour type. Epitope mapping of tumour antigens for HLA class I- and class II-restricted binding motifs followed by immunization with these peptides has induced protective immunity in murine models against cancers expressing the antigen. MHC class I molecules presenting the appropriate peptides are necessary to provide the specific signals for recognition and killing by cytotoxic T cells (CTL). The principle mechanism of tumour escape is the loss, downregulation or alteration of HLA profiles that may render the target cell resistant to CTL lysis, even if the cell expresses the appropriate tumour antigen. In human tumours HLA loss may be as high as 50%, inferring that a reduction in protein levels might offer a survival advantage to the tumour cells. Alternatively, MHC loss may render tumour cells susceptible to natural killer cell-mediated lysis because they are known to act as ligands for killer inhibitory receptors (KIRs). We review the molecular features of MHC class I and class II antigens and discuss how surface MHC expression may be regulated upon cellular transformation. In addition, selective loss of MHC molecules may alter target tumour cell susceptibility to lymphocyte killing. The development of clinical immunotherapy will need to consider not only the expression of relevant CTL target MHC proteins, but also HLA inhibitory to NK and T cells. Received: 20 March 1999 / Accepted: 3 May 1999     

Identification of five MAGE-A1 epitopes recognized by cytolytic T lymphocytes obtained by in vitro stimulation with dendritic cells transduced with MAGE-A1.

MAGE genes are expressed by many human tumors of different histological types but not by normal cells, except for male germline cells. The Ags encoded by MAGE genes and recognized by T cells are therefore strictly tumor-specific. Clinical trials involving therapeutic vaccination of cancer patients with MAGE antigenic peptides or proteins are in progress. To increase the range of patients eligible for therapy with peptides, it is important to identify additional MAGE epitopes recognized by CTL. Candidate peptides known to bind to a given HLA have been used to stimulate T lymphocytes in vitro. In some instances, CTL clones directed against these synthetic peptides have been obtained, but these clones often failed to recognize tumor cells expressing the relevant gene. Therefore, we designed a method to identify CTL epitopes that selects naturally processed peptides. Monocyte-derived dendritic cells infected with a recombinant canarypoxvirus (ALVAC) containing the entire MAGE-A1 gene were used to stimulate CD8+ T lymphocytes from the blood of individuals without cancer. Responder cell microcultures that specifically lysed autologous cells expressing MAGE-A1 were cloned using autologous stimulator cells either transduced with a retrovirus coding for MAGE-A1 or infected with recombinant Yersinia-MAGE-A1 bacteria. The CTL clones were tested for their ability to lyse autologous cells loaded with each of a set of overlapping MAGE-A1 peptides. This strategy led to the identification of five new MAGE-A1 epitopes recognized by CTL clones on HLA-A3, -A28, -B53, -Cw2, and -Cw3 molecules. All of these CTL clones recognized target cells expressing gene MAGE-A1.     

A Common Minimal Motif for the Ligands of HLA-B*27 Class I Molecules

CD8⁺ T cells identify and kill infected cells through the specific recognition of short viral antigens bound to human major histocompatibility complex (HLA) class I molecules. The colossal number of polymorphisms in HLA molecules makes it essential to characterize the antigen-presenting properties common to large HLA families or supertypes. In this context, the HLA-B*27 family comprising at least 100 different alleles, some of them widely distributed in the human population, is involved in the cellular immune response against pathogens and also associated to autoimmune spondyloarthritis being thus a relevant target of study. To this end, HLA binding assays performed using nine HLA-B*2705-restricted ligands endogenously processed and presented in virus-infected cells revealed a common minimal peptide motif for efficient binding to the HLA-B*27 family. The motif was independently confirmed using four unrelated peptides. This experimental approach, which could be easily transferred to other HLA class I families and supertypes, has implications for the validation of new bioinformatics tools in the functional clustering of HLA molecules, for the identification of antiviral cytotoxic T lymphocyte responses, and for future vaccine development.     

Clustering HLA Class I Superfamilies Using Structural Interaction Patterns

Human leukocyte antigen (HLA) class I molecules are critical components of the cell-mediated immune system that bind and present intracellular antigenic peptides to CD8⁺ T cell receptors. To understand the interaction mechanism underlying human leukocyte antigen (HLA) class I specificity in detail, we studied the structural interaction characteristics of 16,393 nonameric peptides binding to 58 HLA-A and -B molecules. Our analysis showed for the first time that HLA-peptide intermolecular bonding patterns vary among different alleles and may be grouped in a superfamily dependent manner. Through the use of these HLA class I ‘fingerprints’, a high resolution HLA class I superfamily classification schema was developed. This classification is capable of separating HLA alleles into well resolved, non-overlapping clusters, which is consistent with known HLA superfamily definitions. Such structural interaction approach serves as an excellent alternative to the traditional methods of HLA superfamily definitions that use peptide binding motifs or receptor information, and will help identify appropriate antigens suitable for broad-based subunit vaccine design.     

Identification of tumor antigens and T-cell epitopes,and its clinical application

The capability of antigen-specific CD8⁺ and CD4⁺ T lymphocytes to mediate antitumor immunity has generated remarkable interest in the identification of target antigens and their epitopes. The classical strategy to define tumor antigens is based on the employment of in vivo sensitized tumor-reactive T lymphocytes from cancer patients. These lymphocytes are used to screen an autologous tumor cDNA expression library for the target antigen. Alternatively, antibodies from the serum of cancer patients can be applied to screen a tumor-derived phage expression library for immunogenic cellular structures. In addition, potential target antigens have been selected by gene expression profiling searching for overexpressed gene products in neoplastic cells compared with normal tissues. B-cell target structures and overexpressed gene products have to be verified as T-cell antigens by the strategy of reverse immunology. Therefore, T cells are sensitized in vitro by autologous dendritic cells loaded with predicted antigenic peptide ligands for a given HLA allele or with the global antigen. These different approaches led to the identification of a still growing number of antigenic peptides providing the basis for the development of new active and passive immunotherapies and for the monitoring of spontaneous and vaccine-induced T-cell responses. Some of these antigens and/or their epitopes are now validated in different clinical regimens for their capability to mediate potent T-cell immunity in cancer patients.This work was presented at the first Cancer Immunology and Immunotherapy Summer School, 8–13 September 2003, Ionian Village, Bartholomeio, Peloponnese, Greece.     

A combined immuno-informatics and structure-based modeling approach for prediction of T cell epitopes of secretory proteins of Mycobacterium tuberculosis

The role of secretory proteins of Mycobacterium tuberculosis in pathogenesis and stimulation of specific host responses is well documented. They are also shown to activate different cell types, which subsequently present mycobacterial antigens to T cells. Therefore identification of T cell epitopes from this set of proteins may serve to define candidate antigens with vaccine potential. Fifty-two secretory proteins of M. tuberculosis H37Rv were analyzed computationally for the presence of HLA class I binding nonameric peptides. All possible overlapping nonameric peptide sequences from 52 secretory proteins were generated in silico and analyzed for their ability to bind to 33 alleles belonging to A, B and C loci of HLA class I. Fifteen percent of generated peptides are predicted to bind to HLA with halftime of dissociation T(1/2) >or=100 min and 73% of the peptides predicted to bind are mono-allelic in their binding. The structural basis for recognition of no-namers by different HLA molecules was studied employing structural modeling of HLA class I-peptide complexes and there exists a good correlation between structural analysis and binding prediction. Pathogen peptides that could behave as self- or partially self-peptides in the host were eliminated using a comparative study with the human proteome, thus reducing the number of peptides for analysis. The implications of the finding for vaccine development are discussed vis-à-vis the limitations of the use of subunit vaccine and DNA vaccine.     

Computational binding assays of antigenic peptides

Computer models can be combined with laboratory experiments for the efficient determination of (i) peptides that bind MHC molecules and (ii) T-cell epitopes. For maximum benefit, the use of computer models must be treated as experiments analogous to standard laboratory procedures. This requires the definition of standards and experimental protocols for model application. We describe the requirements for validation and assessment of computer models. The utility of combining accurate predictions with a limited number of laboratory experiments is illustrated by practical examples. These include the identification of T-cell epitopes from IDDM-, melanoma- and malaria-related antigens by combining computational and conventional laboratory assays. The success rate in determining antigenic peptides, each in the context of a specific HLA molecule, ranged from 27 to 71%, while the natural prevalence of MHC-binding peptides is 0.1–5%.     

Computational binding assays of antigenic peptides

Summary Computer models can be combined with laboratory experiments for the efficient determination of (i) peptides that bind MHC molecules and (ii) T-cell epitopes. For maximum benefit, the use of computer models must be treated as experiments analogous to standard laboratory procedures. This requires the definition of standards and experimental protocols for model application. We describe the requirements for validation and assessment of computer models. The utility of combining accurate predictions with a limited number of laboratory experiments is illustrated by practical examples. These include the identification of T-cell epitopes from IDDM-, melanoma-and malaria-related antigens by combining computational and conventional laboratory assays. The success rate in determining antigenic peptides, each in the context of a specific HLA molecule, ranged from 27 to 71%, while the natural prevalence of MHC-binding peptides is 0.1–5%.     

Recognition of HLA class I molecules by antisera directed to synthetic peptides corresponding to different regions of the HLA-B7 heavy chain

Antisera have been prepared in rabbits and in mice against different peptides corresponding to four hydrophilic and variable regions of HLA-B7 heavy chain (65-82, 99-118, 138-157, and 164-187). Specific antipeptide sera have been obtained with all synthetic peptides; for three of them which were more than 20 amino acids long, highly potent sera were elicited by injection of the free peptide. Three overlapping peptides included in region 138-157 have been used, and two different antigenic sites were detected in this region. HLA molecules solubilized in nonionic detergent were precipitated by antipeptide sera directed against regions 65-82, 138-157, and 164-187, but not by antipeptide serum directed against the less hydrophilic region 99-118. Analysis by two-dimensional electrophoresis of the isolated molecules confirmed the anti-HLA specificity of the antipeptide 65-82 and 138-157 sera. Variable numbers of HLA-related spots were found according to the antisera used. Antipeptide 138-157 serum precipitated numerous HLA molecules and therefore probably reacted with monomorphic determinants whereas antipeptide 65-82 appeared specific for a more limited number of HLA antigens. Such reagents directed against well-defined regions of the HLA class I heavy chain are of considerable interest, notably for the mapping of antigenic epitopes on the molecule and for the study of relationships between structure and function.     

Unusual viral ligand with alternative interactions is presented by HLA-Cw4 in human respiratory syncytial virus-infected cells

Short viral antigens bound to human major histocompatibility complex (HLA) class I molecules are presented on infected cells. Vaccine development frequently relies on synthetic peptides to identify optimal HLA class I ligands. However, when natural peptides are analyzed, more complex mixtures are found. By immunoproteomics analysis, we identify in this study a physiologically processed HLA ligand derived from the human respiratory syncytial virus matrix protein that is very different from what was expected from studies with synthetic peptides. This natural HLA-Cw4 class I ligand uses alternative interactions to the anchor motifs previously described for its presenting HLA-Cw4 class I molecule. Finally, this octameric peptide shares its C-terminal core with the H-2D(b) nonamer ligand previously identified in the mouse model. These data have implications for the identification of antiviral cytotoxic T lymphocyte responses and for vaccine development.     

Activation of human CD8-positive T cells via the CD8/HLA class I complex

Cross-linking of CD8 and HLA class I molecules with appropriate monoclonal antibodies (mAb) and goat anti-mouse Ig (GaMIg) antibody resulted in a marked proliferation of resting human CD8 cells in the presence of interleukin-2 (IL-2). These cells also expressed IL-2 receptor (IL-2R), transferrin receptor, HLA-DR and -DQ antigens. Activation of the cross-linked CD8 cells is apparently independent of accessory monocytes. Various anti-CD8 and anti-HLA class I mAb recognizing nonpolymorphic antigenic determinants were examined for the efficacy of activating CD8 cells. Among mAb specific for HLA class I molecules, PA2.6, MB40.5, BB7.7, A1.4, and W6/32 mAb markedly stimulated the proliferation of cross-linked CD8 cells, whereas BBM.1, Q1/28, and HC10 mAb were found inactive. Footprinting analysis of HLA class I molecules suggested that the activity of these anti-HLA class I mAb appeared to be related to the corresponding peptides they protect from enzymatic digestion. In contrast to the anti-HLA class I mAb, all anti-CD8 mAb examined (C8, OKT8A, and anti-Leu-2a) induced the proliferation of CD8-HLA class I cross-linked cells with similar efficacy. These results suggest that physical interaction between CD8 and at least one specific region of HLA class I molecules can trigger the activation of resting human CD8 cells.     

Identification of noncanonical melanoma-associated T cell epitopes for cancer immunotherapy

The identification of tumor-associated T cell epitopes has contributed significantly to the understanding of the interrelationship of tumor and immune system and is instrumental in the development of therapeutic vaccines for the treatment of cancer. Most of the known epitopes have been identified with prediction algorithms that compute the potential capacity of a peptide to bind to HLA class I molecules. However, naturally expressed T cell epitopes need not necessarily be strong HLA binders. To overcome this limitation of the available prediction algorithms we established a strategy for the identification of T cell epitopes that include suboptimal HLA binders. To this end, an artificial neural network was developed that predicts HLA-binding peptides in protein sequences by taking the entire sequence context into consideration rather than computing the sum of the contribution of the individual amino acids. Using this algorithm, we predicted seven HLA A*0201-restricted potential T cell epitopes from known melanoma-associated Ags that do not conform to the canonical anchor motif for this HLA molecule. All seven epitopes were validated as T cell epitopes and three as naturally processed by melanoma tumor cells. T cells for four of the new epitopes were found at elevated frequencies in the peripheral blood of melanoma patients. Modification of the peptides to the canonical sequence motifs led to improved HLA binding and to improved capacity to stimulate T cells.     

Identification of breast cancer peptide epitopes presented by HLA-A*0201

Cellular immune mechanisms detect and destroy cancerous and infected cells via the human leukocyte antigen (HLA) class I molecules that present peptides of intracellular origin on the surface of all nucleated cells. The identification of novel, tumor-specific epitopes is a critical step in the development of immunotherapeutics for breast cancer. To directly identify peptide epitopes unique to cancerous cells, secreted human class I HLA molecules (sHLA) were constructed by deletion of the transmembrane and cytoplasmic domain of HLA A*0201. The resulting sHLA-A*0201 was transferred and expressed in breast cancer cell lines MCF-7, MDA-MB-231, and BT-20 as well as in the immortal, nontumorigenic cell line MCF10A. Stable transfectants were seeded into bioreactors for production of > 25 mg of sHLA-A*0201. Peptides eluted from affinity purified sHLA were analyzed by mass spectroscopy. Comparative analysis of HLA-A*0201 peptides revealed 5 previously uncharacterized epitopes uniquely presented on breast cancer cells. These peptides were derived from intracellular proteins with either well-defined or putative roles in breast cancer development and progression: Cyclin Dependent Kinase 2 (Cdk2), Ornithine Decarboxylase (ODC1), Kinetochore Associated 2 (KNTC2 or HEC1), Macrophage Migration Inhibitory Factor (MIF), and Exosome Component 6 (EXOSC6). Cellular recognition of the MIF, KNTC2, EXOSC6, and Cdk2 peptides by circulating CD8+ cells was demonstrated by tetramer staining and IFN-gamma ELISPOT. The identification and characterization of peptides unique to the class I of breast cancer cells provide putative targets for the development of immune diagnostic tools and therapeutics.     

Lack of tumor recognition by cytolytic T lymphocyte clones recognizing peptide 195–203 encoded by gene <Emphasis Type="Italic">MAGE-A3</Emphasis> and presented by HLA-A24 molecules

Gene MAGE-A3 encodes tumor-specific antigenic peptides recognized by T cells on many tumors. MAGE-A3 peptides presented by HLA class I molecules have been identified using CD8 lymphocytes stimulated with cells that either expressed gene MAGE-A3 or were pulsed with candidate peptides. One antigen identified with the latter method is peptide MAGE-A3(195-203) IMPKAGLLI, presented by HLA-A24 molecules. It has been used to vaccinate advanced cancer patients. Here, we have used HLA/peptide tetramers to detect T cells recognizing this peptide. Their frequency was estimated to be 2 x 10(-8) of the blood CD8 cells in non-cancerous HLA-A24(+) individuals, which is tenfold lower than the reported frequencies of T cells against other MAGE peptides. In the blood of a patient vaccinated with MAGE-A3, the estimated frequency was 5 x 10(-7). Anti-MAGE-3.A24 cytolytic T cell clones were derived, that lysed peptide-pulsed cells with half-maximal effect at the low concentration of 500 pM. However, these CTL did not recognize a panel of HLA-A24(+) tumor cells that expressed MAGE-A3 at levels similar to those found in HLA-A1(+) tumor cells recognized by anti-MAGE-3.A1 CTLs. Furthermore, 293-EBNA cells transfected with MAGE-A3 and HLA-A24 constructs were hardly recognized by the anti-MAGE-3.A24 CTL clones. These results suggest that peptide MAGE-A3(195-203) is poorly processed and is not an appropriate target for cancer immunotherapy.     

Identification of potential HLA class I and class II epitope precursors associated with heat shock protein 70 (HSPA)

Heat shock protein 70 (HSPA) is a molecular chaperone which has been suggested to shuttle human leukocyte antigen (HLA) epitope precursors from the proteasome to the transporter associated with antigen processing. Despite the reported observations that peptides chaperoned by HSPA are an effective source of antigens for cross-priming, little is known about the peptides involved in the process. In this study, we investigated the possible involvement of HSPA in HLA class I or class II antigen presentation and analysed the antigenic potential of the associated peptides. HSPA was purified from CCRF-CEM and K562 cell lines, and using mass spectrometry techniques, we identified 44 different peptides which were co-purified with HSPA. The affinity of the identified peptides to two HSPA isoforms, HSPA1A and HSPA8, was confirmed using a peptide array. Four of the HSPA-associated peptides were matched with 13 previously reported HLA epitopes. Of these 13 peptides, nine were HLA class I and four were HLA class II epitopes. These results demonstrate the association of HSPA with HLA class I and class II epitopes, therefore providing further evidence for the involvement of HSPA in the antigen presentation process.     

Regressing and progressing metastatic lesions: resistance to immunotherapy is predetermined by irreversible HLA class I antigen alterations

Despite the significant efforts to enhance immune reactivity against malignancies the clinical effect of anti-tumor vaccines and cancer immunotherapy is still below expectations. Understanding of the possible causes of such poor clinical outcome has become very important for improvement of the existing cancer treatment modalities. In particular, the critical role of HLA class I antigens in the success of T cell based immunotherapy has led to a growing interest in investigating the expression and function of these molecules in metastatic cancer progression and, especially in response to immunotherapy. In this report, we illustrate that two types of metastatic lesions are commonly generated in response to immunotherapy according to the pattern of HLA class I expression. We found that metastatic lesions, that progress after immunotherapy have low level of HLA class I antigens, while the regressing lesions demonstrate significant upregulation of these molecules. Presumably, immunotherapy changes tumor microenvironment and creates an additional immune selection pressure on tumor cells. As a result, two subtypes of metastatic lesions arise from pre-existing malignant cells: (a) regressors, with upregulated HLA class I expression after therapy, and (b) progressors with resistance to immunotherapy and with low level of HLA class I. Tumor cells with reversible defects (soft lesions) respond to therapy by upregulation of HLA class I expression and regress, while tumor cells with structural irreversible defects (hard lesions) demonstrate resistance to immunostimulation, fail to upregulate HLA class I antigens and eventually progress. These two types of metastases appear independently of type of the immunotherapy used, either non-specific immunomodulators (cytokines or BCG) or autologous tumor vaccination. Similarly, we also detected two types of metastatic colonies in a mouse fibrosarcoma model after in vitro treatment with IFN-gamma. One type of metastases characterized by upregulation of all MHC class I antigens and another type with partial IFN-gamma resistance, namely with lack of expression of L(d)-MHC class I molecule. Our observations may shed new light on the understanding of the mechanisms of tumor escape and might have implications for improvement of the efficacy of cancer immunotherapy.