Specificity of peptide binding by the HLA-A2.1 molecule

The HLA-A2 molecule contains a putative peptide binding site that is bounded by two alpha-helices and a beta-pleated sheet floor. Previous studies have demonstrated that the influenza virus matrix peptide M1 55-73 can sensitize target cells for lysis by HLA-A2.1-restricted virus-immune CTL and can induce CTL that can lyse virus-infected target cells. To assess the specificity of peptide binding by the HLA-A2.1 molecule, we examined the ability of seven variant M1 peptides to be recognized by a panel of M1 55-73 peptide-specific HLA-A2.1-restricted CTL lines. The results demonstrate that five out of the seven variant M1 55-73 peptides could be recognized by A2.1-restricted M1 55-73 peptide-specific CTL lines. The two variant peptides that were not recognized by any CTL could bind to HLA-A2.1 as indicated by their ability to compete for presentation of the M1 55-73 peptide. In addition, 5 of a panel of 24 unrelated peptides tested could also compete for M1 55-73 presentation by HLA-A2.1. One peptide derived from the sequence of a rotavirus protein could sensitize HLA-A2.1+ targets for lysis by M1 55-73 peptide-specific CTL. We conclude from these studies that: 1) the HLA-A2.1 molecule can bind a broad spectrum of peptides; 2) T cells selected for the ability to recognize one peptide plus a class I molecule can actually recognize an unrelated peptide presented by that same class I molecule; and 3) a stretch of three adjacent hydrophobic amino acids may be an important common feature of peptides that can bind to HLA-A2.1.     

In vitro induction of HLA-A2402-restricted and carcinoembryonic-antigen-specific cytotoxic T lymphocytes on fixed autologous peripheral blood cells

HLA-A2402-restricted and carcinoembryonic-antigen(CEA)-specific cytotoxic T lymphocytes (CTL) were induced by culturing human peripheral blood mononuclear cells (PBMC) on formalin-fixed autologous adhesive PBMC that had been loaded with CEA-bound latex beads. The CTL killed the CEA-producing HLA-type matched cancer cells, but not the non-producers of CEA, at an effector/target ratio of 10 within 24 h. On the basis of available HLA-A24-binding peptides, we have also attempted to identify the epitope peptide recognized by the CTL. The peptide CEA652(9), TYACFVSNL, stimulated the CTL most strongly when pulsed on HLA-A2402-expressing target cells. The other nine peptides so far tested were also active, but less efficient in their effect on CTL. The CTL failed to kill target cells pulsed with the HLA-A2-binding CEA peptide, CAP-1. The CTL were also generated on the fixed adherent cells previously pulsed with the peptide CEA652(9). Cytotoxic activity of the CTL was inhibited by monoclonal antibodies against CD3, CD8, and MHC class I molecules. These results suggest that human autologous CTL will be inducible on the autologous fixed PBMC without use of the cultured target cancer cells if tumor antigenic protein is available. Received: 31 December 1997 / Accepted: 4 May 1998     

CTL recognition of a novel HLA-A*0201-binding peptide derived from glioblastoma multiforme tumor cells

Genetic instability of tumor cells can result in translation of proteins that are out of frame, resulting in expression of neopeptides. These neopeptides are not self-proteins and therefore should be immunogenic. By eluting peptides from human glioblastoma multiforme (GBM) tumor cell surfaces and subjecting them to tandem mass spectrometry, we identified a novel peptide (KLWGLTPKVTPS) corresponding to a frameshift in the 3′ beta-hydroxysteroid dehydrogenase type 7 (HSD3B7) gene. HLA-binding algorithms predicted that a 9-amino acid sequence embedded in this peptide would bind to HLA-A*0201. We confirmed this prediction using an HLA-A*0201 refolding assay followed by live cell relative affinity assays, but also showed that the 12-mer binds to HLA-A*0201. Based on the 9-mer sequence, optimized peptide ligands (OPL) were designed and tested for their affinities to HLA-A*0201 and their abilities to elicit anti-peptide and CTL capable of killing GBM in vitro. Wild-type peptides as well as OPL induced anti-peptide CTL as measured by IFN-γ ELISPOTS. These CTL also killed GBM tumor cells in chromium-51 release assays. This study reports a new CTL target in GBM and further substantiates the concept that rational design and testing of multiple peptides for the same T-cell epitope elicits a broader response among different individuals than single peptide immunization.     

Cross-reactive cytotoxic T lymphocytes against human immunodeficiency virus type 1 protease and gamma interferon-inducible protein 30

The gamma interferon (IFN-gamma)-inducible protein 30 (IP-30) signal peptide -11 to -3 (LLDVPTAAV) is a prominent self peptide expressed with the class I human histocompatibility leukocyte antigen A2 (HLA-A2). Stimulation of peripheral blood mononuclear cells (PBMC) from HLA-A2 human immunodeficiency virus type 1 (HIV-1)-infected individuals with an HLA-A2-restricted HIV protease (PR) peptide 76-84 (LVGPTPVNI) activated cytotoxic T lymphocytes (CTL) against the IP-30 signal peptide. Since HIV-1 PR 76-84 stimulated CD8+ T cells from these individuals to secrete IFN-gamma, we tested whether the activation of IP-30-specific CTL in vitro resulted from T-cell cross-reactivity or from up-regulation of IP-30 by IFN-gamma. Neither high levels of exogenous IFN-gamma nor incubation of PBMC with other HIV peptides triggering substantial IFN-gamma release activated IP-30-specific CTL. Although the IP-30 signal peptide did not stimulate IFN-gamma release from freshly isolated PBMC, it activated CTL in vitro against itself and HIV PR 76-84. Peptide-stimulated IFN-gamma release, cold target inhibition, and HLA-A2/immunoglobulin dimer-mediated binding and depletion of effector cells all indicated that in vitro stimulation with HIV PR 76-84 or the IP-30 signal peptide activated a comparable population of cross-reactive effector cells. Neither IP-30 nor HIV PR 76-84 activated CTL against themselves following in vitro stimulation of PBMC from non-HIV-infected HLA-A2 individuals. Peptide titrations indicated higher-avidity T-cell interactions with HIV PR 76-84 than with the IP-30 signal peptide. These data indicate that HIV PR 76-84 is a heteroclitic variant of the IP-30 signal peptide -11 to -3, which has implications for immune memory and autoimmunity.     

Identification of a cyclin B1-derived CTL epitope eliciting spontaneous responses in both cancer patients and healthy donors

With the aim to identify cyclin B1-derived peptides with high affinity for HLA-A2, we used three in silico prediction algorithms to screen the protein sequence for possible HLA-A2 binders. One peptide scored highest in all three algorithms, and the high HLA-A2-binding affinity of this peptide was verified in an HLA stabilization assay. By stimulation with peptide-loaded dendritic cells a CTL clone was established, which was able to kill two breast cancer cell lines in an HLA-A2-dependent and peptide-specific manner, demonstrating presentation of the peptide on the surface of cancer cells. Furthermore, blood from cancer patients and healthy donors was screened for spontaneous T-cell reactivity against the peptide in IFN-γ ELISPOT assays. Patients with breast cancer, malignant melanoma, or renal cell carcinoma hosted powerful and high-frequency T-cell responses against the peptide. In addition, when blood from healthy donors was tested, similar responses were observed. Ultimately, serum from cancer patients and healthy donors was analyzed for anti-cyclin B1 antibodies. Humoral responses against cyclin B1 were frequently detected in both cancer patients and healthy donors. In conclusion, a high-affinity cyclin B1-derived HLA-A2-restricted CTL epitope was identified, which was presented on the cell surface of cancer cells, and elicited spontaneous T-cell responses in cancer patients and healthy donors.     

Substitution analog peptide derived from HER-2 can efficiently induce HER-2-specific, HLA-A24 restricted CTLs

In order to broaden the possibility for anti-HER-2/neu (HER-2) immune targeting, it is important to identify HLA-A24 restricted peptide epitopes derived from HER-2, since HLA-A24 is one of the most common alleles in Japanese and Asian people. In the present study, we have screened HER-2-derived, HLA-A24 binding peptides for cytotoxic T lymphocyte (CTL) epitopes. A panel of HER-2-derived peptides with HLA-A24 binding motifs and the corresponding analogs designed to enhance HLA-A24 binding affinity were selected. Identification of HER-2-reactive and HLA-A24 restricted CTL epitopes were performed by a reverse immunology approach. To induce HER-2-reactive and HLA-A24 restricted CTLs, PBMCs from healthy donors were repeatedly stimulated with monocytes-derived, mature DCs pulsed with HER-2 peptide. Subsequent peptide-induced T cells were tested for the specificity by enzyme linked immunospot, cytotoxicity and tetramer assays. CTL clones were then obtained from the CTL lines by limiting dilution. Of the peptides containing HLA-A24 binding motifs, 16 peptides (nine mers) including wild type peptides (IC₅₀<1,000 nM) and substituted analog peptides (IC₅₀<50 nM) were selected for the present study. Our studies show that an analog peptide, HER-2(905AA), derived from HER-2(905) could efficiently induce HER-2-reactive and HLA-A24 restricted CTLs. The reactivity of the HER-2(905AA)-induced CTL (CTL905AA) was confirmed by different CTL assays. The CTL905AA clones also were able to lyse HER-2(+), HLA-A24(+) tumor cells and cytotoxicity could be significantly reduced in cold target inhibition assays using cold targets pulsed with the HER-2(905) wild type peptide as well as the inducing HER-2(905AA) analog peptide. A newly identified HER-2(905) peptide epitope is naturally processed and presented as a CTL epitope on HER-2 overexpressing tumor cells, and an MHC anchor-substituted analog, HER-2(905AA), can efficiently induce HER-2-specific, HLA-A24 restricted CTLs.     

Cytotoxicity of amyloidogenic immunoglobulin light chains in cell culture

Light-chain amyloidosis (AL) is a devastating protein-misfolding disease characterized by abnormal proliferation of plasma cells in the bone marrow that secrete monoclonal immunoglobulin light chains that misfold and form amyloid fibrils, thus causing organ failure and death. Numerous reports on different protein-misfolding diseases show that soluble oligomeric species populated by amyloidogenic proteins can be quite toxic to cells. However, it is not well established whether the soluble immunoglobulin light-chain species found in circulation in patients with AL are toxic to cells in target organs. We determined the cellular toxicity of two well-characterized light-chain variable domain proteins from cardiac AL patients and their corresponding germline protein, devoid of somatic mutations. Our results show that the soluble form of the AL proteins we characterized are toxic to cardiomyocytes, and that the species found in cell culture correspond, for the most part, to the species present in circulation in these patients.     

Identification of NY-ESO-1 peptide analogues capable of improved stimulation of tumor-reactive CTL

Expression of NY-ESO-1 in a high proportion of different human tumors makes this protein a very attractive vaccine target. NY-ESO-1 peptides, recognized by HLA-A2-restricted CTL, have recently been described. However, it remains unclear how efficiently tumors generate these epitopes, and whether peptide analogues can be used for optimal expansion and activation of NY-ESO-1-specific HLA-A2-restricted CTL. By generating unique CTL clones, we demonstrate that NY-ESO-1-positive tumor cells are efficiently killed by HLA-A2-restricted CTL specific for the peptide epitope NY-ESO-1 157-165. Presentation of this epitope is not affected by the presence or absence of the proteasome subunits low molecular proteins 2 and 7 and is not blocked by proteasome inhibitors, while it is impaired in the TAP-deficient cell line LBL 721.174. NY-ESO-1 157-165 peptide analogues were compared for their antigenicity and immunogenicity using PBL from melanoma patients. Three peptides, containing the carboxyl-terminal cysteine substituted for either valine, isoleucine, or leucine, were recognized at least 100 times more efficiently than the wild-type peptide by specific CTL. Peptide analogues were capable of stimulating the expansion of NY-ESO-1-specific CTL from PBL of melanoma patients much more efficiently than wild-type peptide. These findings define the processing requirements for the generation of the NY-ESO-1 157-165 epitope. Identification of highly antigenic NY-ESO-1 peptide analogues may be important for the development of vaccines capable of expanding NY-ESO-1-specific CTL in cancer patients.     

Two preferentially expressed proteins protect vascular endothelial cells from an attack by peptide-specific CTL

Vascular endothelial cells (EC) are an exposed tissue with intimate contact with circulating Ag-specific CTL. Experimental in vitro and clinical data suggested that endothelial cells present a different repertoire of MHC class I-restricted peptides compared with syngeneic leukocytes or epithelial cells. This endothelial-specific peptide repertoire might protect EC from CTL-mediated cell death. The HLA-A*02-restricted peptide profile of human EC and syngeneic B lymphoblastoid cells was biochemically analyzed and compared. For EC selective peptides, source protein expression, peptide binding affinity, and peptide-HLA-A*02 turnover were measured. The significance of abundant peptide presentation for target cell recognition by immunodominant CTL was tested by small interfering RNA treatment of EC to knock down the source proteins. High amounts of two peptides, PTRF(56-64) and CD59(106-114), were consistently detected in EC. This predominance of two endothelial peptides was explained by cell type-specific source protein expression that compensated for poor HLA-A*02 binding affinity and short half-live of peptide/HLA-A*02 complexes. Knocking down the source proteins containing the abundant endothelial peptide motifs led to a nearly 100-fold increase of surface expression of SMCY(311-319), an immunodominant minor histocompatibility Ag, as detected by cytotoxicity assays using SMCY(311-319)-specific CTL. We conclude that EC express and present preferentially two distinct HLA-A*02-restricted peptides at extraordinary high levels. These abundant self-peptides may protect EC from CTL-mediated lysis by competing for HLA-A*02 binding sites with immunodominant scarcely expressed antigenic peptides.     

Target epitope in the Tax protein of human T-cell leukemia virus type I recognized by class I major histocompatibility complex-restricted cytotoxic T cells.

A trans-acting regulatory gene product p40tax (Tax) of human T-cell leukemia virus type I (HTLV-I) is one of the main target antigens recognized by cytotoxic T lymphocytes (CTL) specific for HTLV-I. A CTL epitope within the Tax protein was identified in this report. HTLV-I-specific CD8+ CTL lines established from two HTLV-I carriers with HTLV-I-associated myelopathy or Sj?gren syndrome were previously demonstrated to kill predominantly the target cells expressing HTLV-I Tax. The CTL from two patients showed significant levels of cytotoxicity to autologous target cells pulsed with a synthetic peptide of 24 amino acids corresponding to the amino-terminal sequences of the Tax protein. Allogeneic target cells were also sensitized for CTL by this peptide when the target cells have HLA-A2. Tax-specific cytotoxicity, detected as cytolysis of the target cells infected with vaccinia virus-HTLV-I recombinant expressing Tax protein, was almost completely inhibited by competitor cells pulsed with the synthetic peptide. This indicates that a major CTL epitope is present in this peptide. Further analysis using shorter peptides revealed that the core sequence of the CTL epitope was LLFGYPVYV at positions 11 through 19. This sequence can be aligned with the HLA-A2-specific motifs reported recently.     

Dendritic cells infected with a vaccinia vector carrying the human gp100 gene simultaneously present multiple specificities and elicit high-affinity T cells reactive to multiple epitopes and restricted by HLA-A2 and -A3

To investigate the ability of human dendritic cells (DC) to process and present multiple epitopes from the gp100 melanoma tumor-associated Ags (TAA), DC from melanoma patients expressing HLA-A2 and HLA-A3 were pulsed with gp100-derived peptides G9154, G9209, or G9280 or were infected with a vaccinia vector (Vac-Pmel/gp100) containing the gene for gp100 and used to elicit CTL from autologous PBL. CTL were also generated after stimulation of PBL with autologous tumor. CTL induced with autologous tumor stimulation demonstrated HLA-A2-restricted, gp100-specific lysis of autologous and allogeneic tumors and no lysis of HLA-A3-expressing, gp100+ target cells. CTL generated by G9154, G9209, or G9280 peptide-pulsed, DC-lysed, HLA-A2-matched EBV transformed B cells pulsed with the corresponding peptide. CTL generated by Vac-Pmel/gp100-infected DC (DC/Pmel) lysed HLA-A2- or HLA-A3-matched B cell lines pulsed with the HLA-A2-restricted G9154, G9209, or G9280 or with the HLA-A3-restricted G917 peptide derived from gp100. Furthermore, these DC/Pmel-induced CTL demonstrated potent cytotoxicity against allogeneic HLA-A2- or HLA-A3-matched gp100+ melanoma cells and autologous tumor. We conclude that DC-expressing TAA present multiple gp100 epitopes in the context of multiple HLA class I-restricting alleles and elicit CTL that recognize multiple gp100-derived peptides in the context of multiple HLA class I alleles. The data suggest that for tumor immunotherapy, genetically modified DC that express an entire TAA may present the full array of possible CTL epitopes in the context of all possible HLA alleles and may be superior to DC pulsed with limited numbers of defined peptides.     

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.     

Efficient induction of antitumor cytoxic T lymphocytes from a healthy donor using HLA-A2-restricted MAGE-3 peptide in vitro

 The antigenic peptides encoded by tumor-rejection antigen genes, MAGE-1 and -3, have been identified, and various methods have been utilized for the in vitro induction of MAGE-specific, cytotoxic T lymphocytes (CTL) from peripheral blood mononuclear cells (PBMC) using synthetic peptides. However, all of these methods are technically demanding and thus have a relatively limited usefulness. We herein report a simple and efficient method for the in vitro induction of specific CTL by using the HLA-A2-restricted MAGE-3 peptide from the PBMC of a healthy donor. CTL responses could thus be efficiently induced from unseparated PBMC by stimulation with freshly isolated, peptide-pulsed PBMC as antigen-presenting cells and by using interleukin-7 and keyhole limpet hemocyanin for the primary culture. The induced CTL could thus recognize and lyse not only HLA-A2 target cells pulsed with the peptide but also HLA-A2 tumor cells expressing MAGE-3, in an HLA-class-I-restricted manner. This simple method may, therefore, become a useful tool for investigating the potential peptides for tumor antigens as well as for developing various immunotherapeutic approaches for human malignant tumors. Received: 15 October 1996 / Accepted: 6 December 1996     

An HLA-A3-binding prostate acid phosphatase-derived peptide can induce CTLs restricted to HLA-A2 and -A24 alleles

We previously reported peptide vaccine candidates for HLA-A3 supertype (-A3, -A11, -A31, -A33)-positive cancer patients. In the present study, we examined whether those peptides can also induce cytotoxic T lymphocyte (CTL) activity restricted to HLA-A2, HLA-A24, and HLA-A26 alleles. Fourteen peptides were screened for their binding activity to HLA-A*0201, -A*0206, -A*0207, -A*2402, and -A*2601 molecules and then tested for their ability to induce CTL activity in peripheral blood mononuclear cells (PBMCs) from prostate cancer patients. Among these peptides, one from the prostate acid phosphatase protein exhibited binding activity to HLA-A*0201, -A*0206, and -A*2402 molecules. In addition, PBMCs stimulated with this peptide showed that HLA-A2 or HLA-A24 restricted CTL activity. Their cytotoxicity toward cancer cells was ascribed to peptide-specific and CD8⁺ T cells. These results suggest that this peptide could be widely applicable as a peptide vaccine for HLA-A3 supertype-, HLA-A2-, and -A24-positive cancer patients.     

Review: immunoglobulin light chain amyloidosis--the archetype of structural and pathogenic variability

AL amyloidosis is caused by deposition in target tissue of amyloid fibrils constituted by monoclonal immunoglobulin light chains. The amyloidogenic plasma cells derive from a transformed memory B cell that can be identified by anti-idiotype monoclonal antibodies. Comparison of the primary structures of amyloidogenic and nonamyloidogenic light chains does not show any common structural motif in the amyloidogenic variants but reveals peculiar replacements which can destabilize the folding state. Reduced folding stability now appears to be a unifying property of amyloidogenic light chains. The tendency of these proteins to populate a partially unfolded intermediate state is a key event in the self-association that progresses to the formation of oligomers and fibrils. The mechanism of organ damage caused by AL amyloid deposition is not known, but clinical findings suggest that the process of amyloid fibril formation itself exerts tissue toxic effects independently of the amount of amyloid deposited. Since the disease is caused by the neoplastic expansion of the plasma cell population synthesizing the amyloidogenic light chains, the clone represents the prime therapeutic target of conventional chemotherapy and experimental immunotherapy. In common with other types of amyloidosis the therapeutic strategy can take advantage of drugs able to improve the reabsorption of the amyloid deposits or able to bind and stabilize the light chain in the native-like folded state.     

Structural identity between HLA-A2 antigens differentially recognized by alloreactive cytotoxic T lymphocytes

Alloreactive CTL raised against HLA-A2 Ag often display heterogeneous recognition of HLA-A2+ target cells. This heterogeneity has been found to reflect structural polymorphism among the corresponding target Ag, thus defining HLA-A2 subtypes. A previous study (van der Poel et al. 1986. Human Immunol. 16:247) established the existence of a new HLA-A2.4 variant, A2-SCHU, that was distinguished from A*0206 (A2.4a) by HLA-A2-specific alloreactive CTL. The same CTL subdivided HLA-A2.1 Ag into two subgroups. In the present study, the molecular basis of this heterogeneity has been examined by double-label comparative peptide mapping analysis of differentially recognized A2.1 and A2.4 Ag. In addition, we have determined the complete sequence of polymerase chain reaction-amplified full length cDNA from A2-SCHU. The results show that: 1) A2-SCHU is indistinguishable from A*0206 by peptide mapping; 2) the cDNA sequence of A2-SCHU is identical to that of A*0206; and 3) two differentially recognized A2.1 Ag are both indistinguishable from A*0201 by comparative peptide mapping. These results indicate that differential recognition by alloreactive CTL can occur among structurally identical class I HLA Ag and suggest that allorecognition by such CTL may involve corecognition of endogenous peptides, presumably derived from polymorphic proteins.     

Epitope recognition of conserved HIV envelope sequences by human cytotoxic T lymphocytes.

CTL constitute an essential part of the immune response against the HIV. CTL recognize peptides derived from viral proteins together with the MHC class I molecules on the surface of infected cells. The CTL response could be important in prevention or control of infection with HIV by destroying virus-producing cells. In this study we have attempted to identify peptide epitopes recognized by HIV-specific CTL. Using synthetic peptides, we have identified six conserved peptidic epitopes on the gp120 envelope glycoprotein recognized by polyclonal human CTL in association with HLA-A2 class I transplantation Ag. These results were confirmed by two approaches: i) blocking of CTL activities with antibodies specific for three of these conserved peptides; and ii) construction of doubly transfected P815-A2 target cells, using deletions of the HIV env gene. Vaccination or immunotherapy in HLA-A2 individuals can thus be considered using highly conserved HIV env peptidic sequences.     

The minimum peptide epitope from the influenza virus matrix protein. Extra and intracellular loading of HLA-A2.

Influenza virus matrix protein-derived peptides were synthesized based on the amino acid motifs for HLA-A2 bound self peptides. Among these peptides a nonamer (amino acids 58 through 66: G I L G F V F T L) was found to be 100 to 1000 times more effective than the commonly used peptide 57-68 (K G I L G F V F T L T V) in sensitizing HLA-A2+ target cells to lysis by influenza virus specific cytotoxic T lymphocytes. The sensitizing activity of the 12-mer 57-68 was not due to contamination with shorter and more active peptides. Intracellular expression of peptide 58-66 (mediated by a stable expression plasmid with DNA coding for this peptide) also sensitized HLA-A2+ cells to lysis. Peptide 58-66 stimulated human PBMC to generate CTL that recognized peptides 58-66 and 57-68 in association with HLA-A2.