Purpose: MUC-1/DF-3 remains an attractive target for vaccine therapy. It is overexpressed in the majority of human carcinomas and multiple myeloma. Clinical trials using MUC-1-based vaccines have demonstrated safety, clinical responses, and the induction of T-cell responses directed against MUC-1. Previous studies in experimental models and in clinical trials have demonstrated that altering the amino acid sequence of a “self” epitope can lead to the generation of an enhancer agonist epitope capable of eliciting stronger T-cell responses than the native epitope can.

Experimental Design and Results: We describe here the identification of six novel class I HLA-A2 epitopes of MUC-1 that reside outside of the variable number of tandem repeat region. Each is shown to have the ability to activate human T cells as measured by IFN-γ production. One epitope (ATWGQDVTSV, at amino acid position 92–101 and designated P-92), which demonstrated the highest level of binding to HLA-A2 and which induced the highest level of IFN-γ in human T cells, was further studied for the generation of potential enhancer agonist epitopes. Of four potential agonists identified, one epitope (ALWGQDVTSV, designated P-93L) was identified as an enhancer agonist. Compared with the native P-92 peptide, the P-93L agonist (a) bound HLA-A2 at lower peptide concentrations, (b) demonstrated a higher avidity for HLA-A2 in dissociation assays, (c) when used with antigen-presenting cells, induced the production of more IFN-γ by T cells than with the use of the native peptide, and (d) was capable of more efficiently generating MUC-1-specific human T-cell lines from normal volunteers and pancreatic cancer patients. Most importantly, the T-cell lines generated using the agonist epitope were more efficient than those generated with the native epitope in the lysis of targets pulsed with the native epitope and in the lysis of HLA-A2 human tumor cells expressing MUC-1.

Conclusions: In addition to the identification of novel MUC-1 epitopes outside the variable number of tandem repeat region, the studies reported here describe the first agonist epitope of MUC-1. The employment of this agonist epitope in peptide-, protein-, and vector-based vaccines may well aid in the development of effective vaccines for a range of human cancers.

The tumor-associated antigen MUC-1, or MUC-1/DF-3, is overexpressed on the cell surface of many human adenocarcinomas, such as ovarian, breast, pancreatic, colorectal, and prostate carcinoma, and in hematological malignancies, including multiple myeloma and some B-cell non-Hodgkin lymphomas (1, 2, 3, 4, 5). Although MUC-1 is expressed on some normal epithelial tissue on lumenal surfaces, it has been demonstrated that the apical localization of MUC-1 is lost in tumor tissues. In addition, MUC-1 is underglycosylated in human adenocarcinomas as compared with normal tissues, and, thus, the antigenic epitopes of the protein core are more exposed (6). A high level of MUC-1 expression and secretion has also been shown to be associated with poor prognosis and high metastatic potential (7, 8, 9, 10). It was initially demonstrated that MHC-unrestricted cytotoxic T cells could be established from patients with pancreatic carcinoma (11), ovarian cancer (12), and multiple myeloma (13, 14); these T cells were shown to recognize the MUC-1 protein core in the 20-amino-acid (15, 16) variable number of tandem repeat (VNTR) region. More recently, it has been demonstrated that MHC-restricted T cells that recognize epitopes in the VNTR region could also be induced (17, 18). It has also been reported that CTLs could be generated to MUC-1 peptides in HLA-A2/Kb transgenic mice immunized with mannan-mucin (MUC-1) fusion protein containing five repeats of the 20-amino acid VNTR (19). Peptides within the VNTR have been studied by many investigators because VNTR is the most immunogenic region in MUC-1 when the whole cancer cells or mucin extracts are used to immunize hosts for the production of antibodies. Although the VNTR region is immunogenic for MHC unrestricted CTLs as well as for the production of MUC-1 specific antibodies (11, 17, 20), relatively limited information is available with respect to the immunogenicity of the region outside the VNTR.

Potential CTL epitope sequences outside the VNTR have now been identified (21, 22, 23, 24). Two 9-mer MUC-1 peptides (STAPPVHNV, amino acid 950–958, and STAPPAHGV, amino acid 130–138), which are located outside the VNTR region and which can induce HLA-A2-restricted human CTLs in vitro, have been identified (21). It has also been demonstrated that CTLs can be generated in BALB/c, C57BL/6, HLA-A0201/Kb, and double transgenic HLA-A0201/Kb × human MUC-1 (A2KbMUC-1) mice by immunization with human milk fat membrane antigen (22). A 9-mer peptide (amino acid 12–20, LLLLTVLTV) derived from the leader sequence of the MUC-1 protein has also been identified for its ability to bind to the HLA-A2 molecule and may be a potential CTL epitope for MUC-1-specific CTLs. The fact that MUC-1 is overexpressed and underglycosylated in tumor versus normal tissue, and that MHC-restricted and MHC-unrestricted human CTLs could be generated, which, in turn, was shown to lyse MUC-1-positive tumor cells, suggests that MUC-1 is a potential target for immunotherapy of many cancer types.

It has been demonstrated that one way to modify the immunogenicity of a peptide of a self-antigen is to alter the amino acid residues that interact with either the HLA molecule or the T-cell receptor. Some amino acid modification in the anchor residues of peptides may result in enhanced binding to MHC class I and enhanced T-cell activation, whereas most other amino acid modifications will either have no effect or act to antagonize T-cell activation (25, 26, 27, 28). Enhancement of T-cell activation by modifying HLA-anchor residues has been demonstrated for some human melanoma-associated antigens as well as for prostate-specific antigen (29, 30, 31), but has not been demonstrated for most antigens associated with solid tumors, leukemias, or lymphomas.

The studies described here report the identification and characterization of a MUC-1 CTL epitope, not previously identified, in the non-VNTR extracellular region of MUC-1. These studies also describe the generation of an enhancer agonist entity of this epitope. Although the native CTL epitope was shown capable of generating human CTLs in vitro, the studies reported here demonstrate enhanced binding of the agonist epitope to the HLA-A2 molecule and enhanced stability of the peptide-MHC complex as compared with the native peptide. T-cell lines, generated from the same normal individuals with either the native or the agonist peptide, showed higher levels of lysis of target cells and enhanced IFN-γ production when target cells were pulsed with the agonist versus the native peptide. Moreover, T-cell lines generated from two patients with pancreatic cancer using the agonist peptide showed higher levels of in vitro biological activity in terms of IFN-γ production and lysis of tumor cells, as compared with T-cell lines generated in parallel from the same patients using the native epitope. These studies thus provide the rationale for the potential utility of the novel agonist epitope in peptide- and/or vector-mediated immunotherapy regimens for the treatment of MUC-1-expressing tumors.

Cell Cultures

The human breast adenocarcinoma cell line MCF-7 (Ref. 32; HLA-A2 positive and MUC-1 positive), and SK-Mel-24 (HLA-A2 positive, MUC-1 negative) were purchased from American Type Culture Collection (Manassas, VA). The cultures were free of Mycoplasma and were maintained in complete medium [RPMI 1640 (Invitrogen Life Technologies, Inc., Carlsbad, CA) supplemented with 10% fetal bovine serum, 2 mm glutamine, 100 units/ml penicillin, and 100 μg/ml streptomycin (Invitrogen Life Technologies, Inc.)]. The C1R cell line is a human plasma leukemia cell line that does not express endogenous HLA-A or -B antigens (33). C1R-A2 cells are C1R cells that express a transfected genomic clone of HLA-A2.1 (34). These cells were obtained from Dr. William E. Biddison (National Institute of Neurological Disorders and Stroke, NIH, Bethesda, MD). The 174CEM-T2 cell line (T2) transport deletion mutant (35) was provided by Dr. Peter Cresswell (Yale University School of Medicine, New Haven, CT). C1R-A2 cells and T2 cells were Mycoplasma free and were maintained in RPMI 1640 complete medium and in Iscove’s modified Dulbecco’s complete medium (Invitrogen Life Technologies, Inc.), respectively.

Peptides

The amino acid sequence of MUC-1 was scanned for matches to consensus motifs for HLA-A2-binding peptides. We used the computer algorithm from the BioInformatics and Molecule Analysis Section of NIH (BIMAS) that was developed by Parker et al.(36), which ranks potential MHC-binding peptides according to the predictive one-half-time dissociation of peptide/MHC complexes. The HLA-A2 allele was chosen because it is the most commonly expressed class I allele. Ten-mer peptides from the nonvariable number of tandem repeat sequence were synthesized if they conformed to the respective consensus motif. A panel of 10-mer MUC-1 peptides (Table 1) and analogs with single amino acid substitution to positions P1-to-P10 of P-92 MUC-1 peptide (Fig. 1) were made by American Peptide Company (Sunnyvale, CA) with purity >90%. In addition, a carcinoembryonic antigen (CEA) peptide CAP1–6D (28) >96% pure was made by American Peptide Company (Sunnyvale, CA).

Flow Cytometric Analysis

Single-Color Flow Cytometric Analysis.

The method for single-color flow cytometric analysis has been described previously (37). Briefly, cells were washed three times with cold Ca2+- and Mg2+-free Dulbecco’s phosphate-buffered saline and then were stained for 1 h at 4°C using 1 μg of the monoclonal antibody against HLA-A2 (A2, 28, One Lambda, Inc., Canoga Park, CA), CD3, CD4, CD8, or CD56 (BD Biosciences, San Jose, CA). MOPC-104E (IgM; Cappel/Organon Teknika Corp., West Chester, PA) was used as a negative control. The cells were then washed three times and incubated with a 1:100 dilution of FITC-labeled goat antimouse immunoglobulin (IgG; Kirkegaard and Perry Laboratories, Gaithersburg, MD). The cells were immediately analyzed using a Becton Dickinson FACScan equipped with a blue laser with an excitation of 15 mW at 488 nm. Data were gathered from 10,000 live cells, were stored, and were used to generate results.

Dual-Color Flow Cytometric Analysis.

The procedure for dual-color flow cytometric analysis was similar to that for single-color analysis with the following exceptions. Anti-MHC-class II FITC/anti-CD11c phycoerythrin (PE), anti-MHC-class II FITC/anti-CD80 PE, anti-CD58 FITC/anti-CD54 PE, anti-MHC class I FITC/anti-MHC class II PE, and anti-IgG1 FITC/anti-IgG2a PE (isotype controls) were used for the analysis of dendritic cells (DCs); >96% of the DCs were CD11c and MHC class II positive.

The antibodies used for the analysis of T-cell lines were anti-CD56 FITC/anti-CD8 PE, anti-CD4 FITC/anti-CD8 PE, and anti-CD45R0-FITC/anti-CD49d PE; >98% of the T-1191-P92 and T-1191-P-93L cells were CD8 positive. Antibodies to CD4, CD8, CD28, CD45RO, CD56, CD49d, CD54, CD80, CD86, CD58, and CD11c were purchased from BD Biosciences. Antibodies to MHC-class I and MHC-class II were purchased from Serotec, Oxford, United Kingdom. Staining was done simultaneously for 1 h, after which cells were washed three times, resuspended as above, and immediately analyzed using a Becton Dickinson FACScan equipped with a blue laser with an excitation of 15 mW at 488 nm with the use of the CELLQuest program.

Results were expressed in percentage of positive cells and mean fluorescence intensity. Mean fluorescence intensity was used to express the levels of fluorescence determined by measuring the average for all of the cells in the gated fluorescence dot plot. The mean fluorescence intensity value was collected in log scale on the FACScan.

Peptide Binding to HLA-A2

Binding of P-92 and P-92 analogs to HLA-A2 molecules was evaluated by the up-regulation of HLA-A2 expression on T2 cells as demonstrated by flow cytometry (38). In this assay, increased stability (accumulation) of HLA-A2 molecules on the surface of T2 cells as a consequence of peptide binding is measured by increased binding of antibody directed against HLA-A2 molecule. Briefly, 1 × 106 cells in serum-free Iscove’s modified Dulbecco’s complete medium were incubated with peptides at a concentration of 50 μg/ml in 24-well culture plates at 37°C in 5% CO2. Flow cytometry for peptide binding was performed using T2 cells and single-color analysis. After cells were washed three times in Dulbecco’s phosphate-buffered saline, as described above, they were incubated for 1 h with a 1:100 dilution of anti-HLA-A2, 69-specific monoclonal antibody (One Lambda, Inc.) per 106 cells. UPC-10 (Cappel/Organon Teknika, West Chester, PA) was used as isotype control. The cells were then washed three times and incubated with 1:100 dilution of FITC-labeled antimouse IgG (BD Biosciences). Analysis was conducted with the FACScan, as described above. Cells were maintained on ice during all cell preparation and staining.

Culture of DCs from PBMCs

HLA-A2 normal donor peripheral blood mononuclear cells (PBMCs) were obtained from heparinized blood. PBMCs were separated using lymphocyte separation medium gradient (Organon Teknika, Durham, NC), as described previously (39). DCs were prepared using a modification of the procedure described by Sallusto and Lanzavecchia (40). PBMCs (1.5 × 108) were resuspended in AIM-V medium containing 2 mm glutamine, 50 μg/ml streptomycin, and 10 μg/ml gentamicin (Invitrogen Life Technologies, Inc.) and were allowed to adhere to a T-150 flask (Corning Costar Corp., Cambridge, MA). After 2 h at 37°C, the nonadherent cells were removed with a gentle rinse. The adherent cells were cultured for 6–7 days in AIM-V medium containing 100 ng/ml recombinant human granulocyte macrophage colony-stimulating factor (rhGM-CSF) and 20 ng/ml recombinant human interleukin (IL)-4. The culture medium was replenished every 3 days.

Recombinant Virus and Infection of DCs with Avipox Virus Containing MUC-1 (rF-MUC-1/TRICOM)

Recombinant fowlpox virus was constructed as described by Jenkins et al.(41). A plaque-purified isolate from the POXVAC-TC vaccine strain of fowlpox virus was used as the parental virus for this recombinant virus. The MUC-1, LFA-3, ICAM-1, and B7–1 sequences were inserted into the BamHIJ region of the fowlpox virus genome. In addition, the lacZ gene, under the control of the fowlpox C1 promoter, was included to identify and isolate recombinant viruses using a chromogenic assay for β-galactosidase. The MUC-1 gene that was inserted into fowlpox virus varies from the native MUC-1(42). It encodes a 30-amino-acid signal sequence, followed by the first 38 amino acids of the mature NH2-terminal sequence of the MUC-1 protein, 10 identical copies of the 20-amino acid repeated sequence, and the COOH-terminal portion of the protein. The 600-bp repeated sequence was produced using overlapping synthetic oligonucleotides containing codons with numerous third-base variations in each repeat sequence without changing the encoded amino acids. This was done to minimize duplicated nucleotide sequences, which are unstable in pox viruses, while maintaining the repeated amino acid sequences.

rF-MUC-1/triad of costimulatory molecules (TRICOM) is a recombinant fowlpox virus that contains the MUC-1 gene under the control of the 40K promoter (43), the human LFA-3 gene under the control of the vaccinia 30K (M2L) promoter (44), the human ICAM-1 gene under the control of the vaccinia I3 promoter (45), and the human B7–1 gene under the control of the synthetic early/late (sE/L) promoter (46). DCs (1 × 106) were incubated in 1 ml of Opti-MEM medium (Life Technologies, Inc.) at 37°C with rF-MUC-1/TRICOM or control avipox virus vector [fowlpox wild-type (FP-WT)]. Titration experiments demonstrated that 4 × 107 plaque-forming units/ml, equal to a multiplicity of infection of 40:1 for 2 h, were able to consistently induce transgene expression in approximately 35% of the infected DCs. DCs from different donors were used for the infection with rF-MUC-1/TRICOM with the range of efficiency of infection from 20 to 55%. The infected DCs were suspended in 10 ml of fresh, warm RPMI 1640 complete medium containing 100 ng/ml of recombinant human granulocyte macrophage colony-stimulating factor, 20 ng/ml recombinant human IL-4, and 20 ng/ml of TNF-α cultured for 24 h, and then subsequently used as antigen-presenting cells (APCs).

Generation of T-Cell Lines

Modification of the protocol described by Tsang et al.(47) was used to generate MUC-1-specific CTL. To generate T-cell lines T-1191-P-93L and T-1191-P-92, autologous DCs infected with rF-MUC-1/TRICOM were used as APCs. Autologous nonadherent cells were then added to APCs at an effector:APC ratio of 10:1. Cultures were then incubated for 3 days at 37°C in a humidified atmosphere containing 5% CO2. The cultures were then supplemented with recombinant human IL-2 at a concentration of 20 units/ml for 7 days; the IL-2 containing medium was replenished every 3 days. The 3-day incubation with peptide and 7-day IL-2 supplement constituted one in vitro stimulation (IVS) cycle. Primary cultures were restimulated with rF-MUC-1/TRICOM-infected autologous DCs, as described above, on day 11 to begin the next IVS cycle. rF-MUC-1/TRICOM-infected autologous DCs were used as APCs for three IVS cycles. Irradiated (23,000 rads) autologous EBV-transformed B cells were used as APCs after the third IVS cycle. For the restimulation with EBV-transformed B cells, peptides at a concentration of 50 μg/ml were used to pulse the autologous EBV-transformed B cells at a ratio of effector:APC of 1:3 for restimulation. Cultures were then incubated for 3 days at 37°C in a humidified atmosphere containing 5% CO2. After removal of the peptide-containing medium, the cultures were then supplemented with recombinant human IL-2 at a concentration of 20 units/ml for 7 days. T-cell lines from patients 18 and 23 (T-18-P-92, T-18-P93L, T-23-P-92, and T-23-P93L) were generated by stimulation of PBMCs with autologous DCs pulsed with the P-92 or P93L peptides using the same stimulation protocol as described above. The markers used for the analysis and identification of DCs were CD11c, MHC-class II, CD80, CD54, CD58, and CD83. CD3 was also used as a negative marker.

Cytotoxic Assay

Target cells (C1R-A2 or tumor cells) were labeled with 50 μCi of 111In-labeled oxyquinoline (Medi-Physics Inc., Arlington, IL) for 15 min at room temperature. Target cells (0.3 × 104) in 100 μl of RPMI 1640 complete medium were added to each of 96 wells in flat-bottomed assay plates (Corning Costar Corp.). Labeled C1R-A2 target cells were incubated with peptides at the concentration indicated for 60 min at 37°C in 5% CO2 before adding effector cells. No peptide was used when carcinoma cell lines were used as targets. Effector cells were suspended in 100 μl of RPMI 1640 complete medium supplemented with 10% pooled human AB serum and added to the target cells. The plates were then incubated at 37°C in 5% CO2 for 4 or 16 h. Supernatant was harvested for gamma counting with the use of harvester frames (Skatron, Inc., Sterling, VA). Determinations were carried out in triplicate, and SDs were calculated. Specific lysis was calculated with the use of the following formula (all values in cpm):

\[\mathrm{\%\ lysis}\ {=}\ \frac{\mathrm{Observed\ release\ {-}\ spontaneous\ release}}{\mathrm{Total\ release\ {-}\ spontaneous\ release}}\ {\times}\ 100\]

Spontaneous release was determined from wells to which 100 μl of RPMI 1640 complete medium were added. Total releasable radioactivity was obtained after treatment of targets with 2.5% Triton X-100.

Detection of Cytokines

Supernatants of T cells exposed for 24 h to peptide-pulsed autologous EBV-transformed B cells, in IL-2-free medium at various peptide concentrations, were screened for secretion of IFN-γ using an ELISA kit (R&D Systems, Minneapolis, MN). The results were expressed in pg/ml. A cytometric bead array (CBA) system (BD PharMingen, San Diego, CA) was also used to determine the secretion of multiple cytokines by specific T cells. The CBA system uses the fluorescence detection by flow cytometry to measure soluble analytes in a particles-based immunoassay. The BD PharMingen human Th1/Th2 cytokine CBA kit was used to measure IL-2, IL-4, IL-5, IL-10, and tumor necrosis factor α protein levels in a single sample. The cytokine capture beads were mixed with PE-conjugated detection antibodies and then were incubated with recombinant cytokine standards or test samples to form sandwich complexes. The sample results were generated in graphic and tabular format, using BD PharMingen CBA analysis software. The results were expressed in pg/ml.

Statistical Analysis

Statistical analysis of differences between means was done using a two-tailed paired t test (Stat View statistical software, Abacus Concepts, Berkeley, CA).

The primary amino acid sequence of human MUC-1 was analyzed for consensus motifs for novel HLA-A2 binding peptides. Twelve 10-mer peptides were identified, subsequently synthesized, and studied for binding to the HLA-A2 molecule in a T2 cell binding assay. The amino acid sequences and positions of these 10-mer peptides are shown in Table 1. The CEA CAP1–6D peptide and a nonspecific cross-reacting antigen peptide were used as a positive and negative control, respectively. The predicted binding of the 12 peptides is also given in Table 1. Three of these peptides (P-92, P-94, and P-1108) were shown to have the highest level of binding in the T2 assay.

Studies were then conducted to determine whether MUC-1-specific T-cell lines could be established from PBMCs of an apparently healthy donor. To accomplish this, autologous DCs infected with rF-MUC-1/TRICOM were used as APCs. rF-MUC-1/TRICOM is a replication-defective avipox vector containing the transgenes for MUC-1 and for a triad of human costimulatory molecules (B7–1, ICAM-1, and LFA-3, designated TRICOM). rF-MUC-1/TRICOM was shown to efficiently infect human DCs and hyperexpress each of the costimulatory molecules, as well as MUC-1, on the DC surface (Table 2). Approximately 96% of the cells were CD11c and MHC-class II positive. The specificity of the MUC-1-specific T cells generated (designated T-1191-MUC-1) was analyzed after IVS cycle 3 (see “Materials and Methods”) for their ability to release IFN-γ after stimulation with autologous B cells pulsed with each of the MUC-1 peptides listed in Table 1. The results shown in Table 3 demonstrate that when the T-1191-MUC-1 cells were stimulated with autologous B cells pulsed with peptides P-92, P-1135, P-94, P-1004, P-1069, and P-4, the T cells produced IFN-γ, whereas the use of autologous B cells pulsed with the other peptides did not result in IFN-γ production. The results shown in Tables 1 and 3 demonstrate that the P-92 peptide had the highest level of T2 binding, as well as the ability to activate T-1191-MUC-1 cells to produce the highest levels of IFN-γ; this peptide was thus chosen for further study.

Analysis of the primary and secondary HLA-A2 anchor amino acid residues at positions 2 and 10 of the P-92 peptide revealed that modification of amino acids at these positions could potentially enhance the binding ability of the peptide to the HLA-A2 molecule. Thus, six different analogs of P-92 were synthesized, as shown in Table 4, and were tested for their binding ability to T2 cells along with the native P-92 peptide. The CEA CAP-7 (HLA-A3 binding peptide), which has previously been shown not to bind to HLA-A2 (47), was used as a negative control. As shown in Table 4, four of the six analog peptides bound to HLA-A2 at higher levels than the P-92 peptide. Analogs P-93L and P-93I bound HLA-A2 with the greatest efficiency. Experiments were then conducted to compare the ability of the P-93L and P-93I peptides to bind HLA-A2 at various peptide concentrations. As seen in Fig. 1, the P-93L and P-93I peptides bound to HLA-A2 at higher levels than did P-92 at all peptide concentrations. The levels of binding were similar for P-93L and P-93I at the various peptide concentrations. These data thus indicated that both P-93L and P-93I with modification in the primary anchor position 2 (position 93 of the MUC-1 molecule) were potential agonists of peptide P-92.

Studies were then undertaken to examine the stability of the peptide-MHC complex for peptides P-92 (native), P-93L, and P-93I. Peptides were incubated with T2 cells overnight, the unbound peptides were washed off, and the cells were then incubated with Brefeldin A to block delivery of new class I molecules to the cell surface; at various time points cells were analyzed for the presence of peptide-HLA-A2 complexes. As shown in Fig. 2, P-93L-HLA-A2, P-93M-HLA-A2, and P-93I-HLA-A2 complexes were more stable than P-92-HLA-A2 complexes over the 8-h observation period, with P-93L-HLA-A2 complexes slightly more stable than the P-93I-HLA-A2 and P-93M-HLA-A2 complexes over the same period of time. Thus, both the binding of peptides to MHC and the stability of the peptide-MHC complex were shown to be greater for the P-93L, P-93I, and P-93M peptides than the native P-92 peptide.

Studies were then conducted to compare the ability of the P-93L and P-93I agonist peptides, at various peptide concentrations, to activate the T-1191-MUC-1 cells. As seen in Fig. 3, at each concentration of peptide, the pulsing of APCs with the P-93L peptide led to the greatest level of IFN-γ production by T-1191-MUC-1 cells as compared with the P-93I peptide or the native P-92 peptide. The P-93L agonist peptide was thus chosen for further study. The cytokine profile of T-1191-MUC-1 cells stimulated with APCs, pulsed with either the P-92 or the P-93L peptide, was then analyzed. A CBA assay was used for the analysis. Table 5 shows the levels of each of the six cytokines produced by the T-1191-MUC-1 cell line stimulated with APCs pulsed with no peptide, P-92 peptide, and P-93L peptide. These results demonstrated greater production of type 1 cytokines IL-2 and IFN-γ by T cells stimulated with the P-93L peptide than with the P-92 peptide. Low or undetectable levels of type 2 cytokines IL-4 and IL-10 were seen with either peptide. Tumor necrosis factor α could not be detected in the supernatants at the 24-h time point.

To further compare the biological activity of the native P-92 peptide and the agonist P-93L peptide, two additional T-cell lines were established. This was accomplished, using as APCs, autologous DCs infected with rF-MUC-1/TRICOM and autologous PBMCs as effectors from an apparently healthy donor. After three IVSs, the T-cell lines were stimulated with autologous B cells pulsed with either the P-92 or the P-93L peptide. These two cell lines were designated T-1191-P-92 and T-1191-P-93L, respectively. The two cell lines were shown to be >98% CD8 positive, 99% CD49d positive, <2% CD56 positive, and >75% CD45RO positive cells. The two T-cell lines were then analyzed for their ability to lyse peptide-pulsed targets. T-1191-P-93L was shown to lyse C1R-A2 cells pulsed with P-93L peptide to a greater extent than cells pulsed with the P-92 peptide (Fig. 4, squares). T-1191-P-92 also lysed target cells pulsed with the P-93L peptide to a greater extent than those pulsed with the P-92 peptide (Fig. 4, triangles). The data in Fig. 4 also show that when target cells are pulsed with the native peptide, the T-cell line established with the agonist P-93L peptide lyses target cells at greater levels than the T-cell line established with the native peptide. This was seen at two different E:T ratios. T-1191-P-93L cells and T-1191-P-92 cells did not lyse C1R-A2 cells when pulsed with control CEA CAP1–6D peptide (Fig. 4, circles). Studies were then conducted to determine whether these two T-cell lines could lyse the MUC-1-positive and HLA-A2-positive breast carcinoma cell line MCF-7. The MUC-1-negative HLA-A2-positive SK-Mel-24 melanoma cell line was used as a negative control. As shown in Table 6, MCF-7 cells were lysed by both the T-1191-P-92 and the T-1191-P-93L cells. No lysis was observed against the SK-Mel-24 cells. T-1191-P-93L cells were shown to lyse MCF-7 cells to a greater degree as compared with the T-1191-P-92 cell line. The addition of unlabeled C1R-A2 cells pulsed with the P-92 peptide and P-93L peptide, but not with the CEA CAP1–6D control peptide, decreased the cytotoxic activity of both T-cell lines, demonstrating the MUC-1 specificity of the lysis (Table 6). The cytotoxic activity of these T-cell lines against MCF-7 cells was also shown to be HLA-A2 restricted, as demonstrated by the inhibition of lysis with the addition of anti-HLA-A2, 69 antibody, but not with the control antibody UPC-10 (Table 7).

The T-1991-P-92 and T-1191-P-93L T-cell lines were derived from an apparently healthy individual. Studies were then conducted to determine whether additional T-cell lines could be established from two patients with pancreatic cancer (patients 23 and 18). Four MUC-1-specific T-cell lines were generated and were designated T-23-P-92, T-23-P-93L, T-18-P-92, and T-18-P-93L. The T-cell lines T-18-P-92 and T-18-P-93L were generated from patient 18 by stimulation of PBMCs with autologous DCs pulsed with the P-92 and P-93L peptides, respectively. T-cell lines T-23-P-92 and T-23-P-93L were generated from patient 23 by stimulation of PBMCs by autologous DCs pulsed with the P-92 and P-93L peptides, respectively. As seen in Table 8, all four T-cell lines from both pancreatic cancer patients could be stimulated to produce IFN-γ when stimulated with DCs pulsed with either the P-92 or the P-93L peptide. No IFN-γ production was observed, however, when these T cells were stimulated in a similar manner with the CEA peptide CAP1–6D. For all four T-cell lines, greater levels of IFN-γ production were seen when the P-93L agonist peptide was used as compared with the P-92 native peptide. It also should be noted that the T-cell lines derived using the agonist peptide always showed higher levels of stimulation than did the T-cell lines derived from the native peptide, when a given peptide was used for stimulation. Studies were then conducted to determine whether the T-cell lines derived from cancer patient 23 could lyse MUC-1-positive HLA-A2-positive cancer cells. The SK-Mel-24 cell line (MUC-1 negative and HLA-A2 positive) was used as a negative control for specificity. As can be seen in Tables 9 and 10, the T-23-P-93L line, T-23-P-92 line, T-18-P-93L line, and T-18-P-92 line showed lysis of the MCF-7 carcinoma cells at two different E:T ratios, but showed no lysis of the melanoma cell line. In concordance with the results shown above, the T-cell line (T-23-P-93L, T-18-P-93L), derived using the agonist peptide, demonstrated greater lysis of the tumor cells than the T-cell line (T-23-P-92, T-18-P-92) derived using the native peptide. This was seen at two different E:T ratios.

This study identifies a MUC-1-derived HLA-A2-restricted CTL epitope (P-92) that maps outside the VNTR region. Human T-cell lines established with the P-92 peptide were shown to lyse peptide-pulsed targets and MCF-7 carcinoma cells in a MHC-restricted manner. The addition of the unlabeled P-92-pulsed C1R-A2 cells was shown to eliminate the CTL activity against MCF-7 cells. These results collectively indicate that the P-92 peptide represents a naturally processed CTL epitope. In the studies reported here, we have also modified the primary anchor residues of the P-92 peptide to enhance the binding affinity of the peptide to the HLA-A2 molecule. Six analogs were analyzed, and one of them, designated P-93L, was shown to be superior to the native P-92 epitope in terms of the affinity of binding to the MHC molecule, the avidity of the peptide-MHC complex, and the ability to activate CTL in vitro.

It has been demonstrated that residues strongly associated with good binding of 10-mer peptides to HLA-A2.1 molecules at the primary binding position 2 are Leu and Met, and Val, Leu, and Ile at position 10 (48). Replacement of Thr at position 2 (amino acid position 93) of the P-92 peptide with Leu improved the binding of the peptide to HLA-A2.1. This observation may partly be attributable to the fact that Thr at position 2 can decrease the affinity of the peptide binding to HLA-A2.1 as well as decrease the stability of the peptide-MHC complex (48). Substitution of Thr with Ile or Met at position 2 also resulted in the improvement of the binding of the P-92 peptide to HLA-A2.1. Analog peptides made by amino acid substitution with Leu at position 2 (amino acid position 93) and at position 10 (amino acid position 101) resulted in reduced binding to HLA-A2 as compared with a single substitution with Leu at position 2. In addition, substitution of Thr with Ile at position 2 and substitution of Val with Leu at position 10 resulted in less efficient binding of the P-92 peptide to the HLA-A2.1 molecule. Likewise, substitution of Thr with Met at position 2 and substitution of Val with Leu at position 10 resulted in less efficient binding of the P-92 peptide to the HLA-A2 molecule. This may have resulted from the negative effects such as repulsive electrostatic interaction, steric hindrances, and conformation changes. It has been reported that, although a human melanoma gp-100 peptide YLEPGPVTA (amino acid position 280–288) contains the predicted motif residues Tyr at position 1, Leu at position 2, and Ala at position 9 (which support good binding to the HLA-A2.1 molecule), the actual binding affinity of this peptide was not high. The lower binding may be attributable to residues at other positions of the peptide, such as a negatively charged Glu at position 3 (30, 49). Amino acids strongly associated with poor binding to the HLA-A2.1 molecule at the secondary anchor position are Asp, Glu, and Pro at position 1; Asp and Glu at position 3; Arg, Lys, His and Ala at position 4; Pro at position 5; Arg, Lys and His at position 7; Asp, Glu, Arg, Lys, and His at position 8; and Arg, Lys, and His at position 9 (50). No residue associated with poor binding to HLA-A2.1 at the secondary anchor position is present in P-92 peptide.

Other MUC-1-derived HLA-A0201-restricted CTL 9-mer epitopes from the non-VNTR region have been reported. Three 9-mer peptides, MUC79–87 (TLAPATEPA), MUC167–175 (ALGSTAPPV), and MUC264–272 (FLSFHISNL; Ref. 24), were analyzed in HLA-A2/Kb-transgenic mice and were shown to elicit peptide-specific CTLs and to protect mice against challenge with a MUC-1-positive A2Kb-expressing murine tumor. Another peptide, MUC13–21 (LLLTVLTVV), was found to bind well to the HLA-A2 molecule but did not display immunogenicity in vivo(23). Peptide MUC353–361 (NLTISDVSV), which was found to be immunogenic in HLA-A2.1/Db-β2 microglobulin single-chain transgenic mice, was a poor binder to HLA-A2 molecule (23, 24). Two MUC-1-specific epitopes, designated M1.1 and M1.2, were previously shown to lyse both peptide-pulsed targets and human carcinoma cells expressing MUC-1 (21). Autologous DCs pulsed with these peptides were used to vaccinate patients with metastatic breast cancer and ovarian cancer. CTL responses from vaccinated patients were demonstrated principally to the M1.2 peptide (51); it is interesting to note that in addition to the generation of T cells reactive with the MUC-1-derived peptide, T cells specific for CEA and MAGE-3 were also detected after multiple vaccinations. We have demonstrated that production of lymphotactin by a MUC-1-specific T-cell line occurs when the line is stimulated with autologous B cells pulsed with the P-93L peptide but not with the native P-92 peptide (52). Enhanced expression of lymphotactin by CD8+ CEA-specific T cells has also been reported when a CEA agonist peptide (CAP1–6D) was used for the stimulation (53). A clinical trial using DC pulsed with the agonist CAP1–6D peptide has demonstrated clinical responses in patients with CEA-expressing carcinomas (54), whereas in another Phase 1 clinical trial using DCs pulsed with native CAP-1 peptide, only modest CAP-1-specific responses were observed (55). On the basis of these clinical responses, together with the results from the CEA agonist peptide study and the ability of lymphotactin to mainly recruit activated T cells, it is conceivable that the enhanced specific antitumor immune responses observed with the agonist CAP1–6D peptide may be due in part to the increased lymphotactin production.

Several other clinical trials involving MUC-1-based vaccines have been reported. These vaccines consisted of a synthetic 105-mer MUC-1 peptide that has five repeated immunodominant epitopes, MUC-1-cDNA-transfected DCs, recombinant vaccinia virus encoding MUC-1 and IL-2, STn-KLH [Theratope(R)], and a recombinant vaccinia virus containing MUC-1 (56, 57, 58, 59, 60). These vaccine trials were shown to be safe, with some evidence of antitumor activity.

The P-93L epitope, moreover, is the first agonist epitope to be described for MUC-1. The MUC-1 agonist epitope can potentially be used as a peptide vaccine in adjuvant, in autologous peptide-pulsed DC therapy, or as part of a recombinant viral or bacterial vector. This investigation also provides evidence for the concept that the alteration of a single residue in the CTL peptide epitope of a relatively weakly immunogenic “self-antigen” can be used to activate specific T cells more efficiently than the native antigen.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Requests for reprints: Jeffrey Schlom, Laboratory of Tumor Immunology and Biology, Center for Cancer Research, National Cancer Institute, NIH, 10 Center Drive, Room 8B09, MSC 1750, Bethesda, MD 20892. Phone: (301) 496-4343; Fax: (301) 496-2756; E-mail: js141c@nih.gov

Fig. 1.

Binding of MUC-1 peptide P-92 and the agonists P-93L and P-93I to HLA-A2 molecule. Peptides were analyzed for binding to the T2 cell line as described in “Materials and Methods.” A, peptides were used at concentrations of 0–50 μg/ml. B, peptides were used at concentrations of 0–12.5 μg/ml. P-92 MUC-1 peptide (□), P-93L (▪), P-93I (▴). Results are expressed in mean fluorescence intensity (MFI).

Fig. 1.

Binding of MUC-1 peptide P-92 and the agonists P-93L and P-93I to HLA-A2 molecule. Peptides were analyzed for binding to the T2 cell line as described in “Materials and Methods.” A, peptides were used at concentrations of 0–50 μg/ml. B, peptides were used at concentrations of 0–12.5 μg/ml. P-92 MUC-1 peptide (□), P-93L (▪), P-93I (▴). Results are expressed in mean fluorescence intensity (MFI).

Close modal
Fig. 2.

Comparison of the stability of the complex of the P-92, P-93L (L-93), P-93I (I-93), or P-93M (M-93) peptide with HLA-A2. T2 cells were incubated overnight with P-92 (□), P-93L (▪), P-93I (▴), and P-93M (•) peptide at a concentration of 50 μg/ml and then were washed free of unbound peptide and incubated with Brefeldin A to block delivery of new class I molecules to the cell surface. At the indicated times, cells were stained for the presence of surface peptide-HLA-A2 complexes. Results are expressed in relative percentage of binding compared with 100% at time 0.

Fig. 2.

Comparison of the stability of the complex of the P-92, P-93L (L-93), P-93I (I-93), or P-93M (M-93) peptide with HLA-A2. T2 cells were incubated overnight with P-92 (□), P-93L (▪), P-93I (▴), and P-93M (•) peptide at a concentration of 50 μg/ml and then were washed free of unbound peptide and incubated with Brefeldin A to block delivery of new class I molecules to the cell surface. At the indicated times, cells were stained for the presence of surface peptide-HLA-A2 complexes. Results are expressed in relative percentage of binding compared with 100% at time 0.

Close modal
Fig. 3.

Ability of autologous B cells pulsed with native MUC-1 peptide P-92 (□), P-93L peptide (▪), and P-93I peptide (▴) to induce IFN-γ production by MUC-1-specific T cells. Peptides were used at concentrations of 0–6.25 μg/ml. Results are expressed in pg/ml.

Fig. 3.

Ability of autologous B cells pulsed with native MUC-1 peptide P-92 (□), P-93L peptide (▪), and P-93I peptide (▴) to induce IFN-γ production by MUC-1-specific T cells. Peptides were used at concentrations of 0–6.25 μg/ml. Results are expressed in pg/ml.

Close modal
Fig. 4.

Cytotoxicity of the MUC-1-specific T-cell lines against C1R-A2 cells pulsed with P-92 and P-93L peptide. T-1191-P-93L against C1R-A2 pulsed with P-93L peptide (▪), T-1191-P-93L against C1R-A2 pulsed with P-92 peptide (□), T-1191-P-93L against C1R-A2 pulsed with CAP1–6D peptide (•), T-1191-P-92 against C1R-A2 pulsed with P-93L peptide (▴), T-1191-P-92 against C1R-A2 pulsed with P-92 peptide (▵), T-1191-P-92 against C1R-A2 pulsed with CAP1–6D peptide (○). E:T ratio = 25:1 and 12.5:1 in a 16-h 111In release assay. Bars, SD.

Fig. 4.

Cytotoxicity of the MUC-1-specific T-cell lines against C1R-A2 cells pulsed with P-92 and P-93L peptide. T-1191-P-93L against C1R-A2 pulsed with P-93L peptide (▪), T-1191-P-93L against C1R-A2 pulsed with P-92 peptide (□), T-1191-P-93L against C1R-A2 pulsed with CAP1–6D peptide (•), T-1191-P-92 against C1R-A2 pulsed with P-93L peptide (▴), T-1191-P-92 against C1R-A2 pulsed with P-92 peptide (▵), T-1191-P-92 against C1R-A2 pulsed with CAP1–6D peptide (○). E:T ratio = 25:1 and 12.5:1 in a 16-h 111In release assay. Bars, SD.

Close modal
Table 1

Binding of human MUC-1 peptides to HLA-A2 molecules

PeptideAmino acid position in MUC-1SequencePredicted binding to HLA-A2aT2 bindingb
P-92 92–101 ATWGQDVTSV POS 740 
P-94 94–103 WGQDVTSVPV NEG 591 
P-1108 1108–1117 REGTINVHDV NEG 482 
P-4 4–13 GTQSPFFLLL NEG 467 
P-1105 1105–1114 LAFREGTINV NEG 461 
P-1004 1004–1013 TLASHSTKTD NEG 442 
P-1069 1069–1078 LQRDISEMFL NEG 433 
P-1162 1162–1171 ALLVLVCVLV POS 431 
P-1135 1135–1144 TISDVSVSDV POS 422 
P-1172 1172–1181 ALAIVYLIAL POS 372 
P-1169 1169–1178 VLVALAIVYL POS 369 
P-1177 1177–1186 YLIALAVCQC POS 338 
CAP1-6D NAc YLSGADLNL POS 975 
NCA NA YRPGENLNL NEG 365 
PeptideAmino acid position in MUC-1SequencePredicted binding to HLA-A2aT2 bindingb
P-92 92–101 ATWGQDVTSV POS 740 
P-94 94–103 WGQDVTSVPV NEG 591 
P-1108 1108–1117 REGTINVHDV NEG 482 
P-4 4–13 GTQSPFFLLL NEG 467 
P-1105 1105–1114 LAFREGTINV NEG 461 
P-1004 1004–1013 TLASHSTKTD NEG 442 
P-1069 1069–1078 LQRDISEMFL NEG 433 
P-1162 1162–1171 ALLVLVCVLV POS 431 
P-1135 1135–1144 TISDVSVSDV POS 422 
P-1172 1172–1181 ALAIVYLIAL POS 372 
P-1169 1169–1178 VLVALAIVYL POS 369 
P-1177 1177–1186 YLIALAVCQC POS 338 
CAP1-6D NAc YLSGADLNL POS 975 
NCA NA YRPGENLNL NEG 365 
a

Predicted binding on the basis of reported motif (37); POS, positive; NEG, negative.

b

Results are expressed in mean fluorescence intensity. CAP1-6D is an HLA-A2 binding carcinoembryonic antigen peptide that was used as a positive control. NCA peptide was used as a negative control.

c

NA, not applicable; NCA, nonspecific cross-reacting antigen.

Table 2

Phenotypic analysis of dendritic cells (DCs) infected with rF-MUC-1/triad of costimulatory molecules (TRICOM)

Flow cytometric analysis of surface marker expression on DCs. DCs used were cultured in AIM-V medium containing 100 ng/ml recombinant human granulocyte macrophage colony-stimulating factor and 20 ng/ml recombinant human interleukin 4 for 7 days. DCs used for infection with fowlpox wild-type (FP-WT) or rF-MUC-1/TRICOM were cultured as described in “Materials and Methods.” Results indicate the percentage of positive cells; numbers in parentheses represent mean fluorescence intensity.

DCs infected withCD80CD54CD58Class IMUC-1
Uninfected 4.8 (13.6) 59.5 (62.3) 68.1 (18.2) 99.7 (271.8) 5.0 ( 95.7) 
FP-WT 7.4 (13.3) 79.3 (83.4) 74.0 (19.5) 99.5 (176.3) 3.1 ( 50.7) 
rF-MUC-1/TRICOM 30.9 (35.7) 84.5 (133.9) 79.8 (27.4) 99.9 (189.3) 31.6 ( 213.1) 
DCs infected withCD80CD54CD58Class IMUC-1
Uninfected 4.8 (13.6) 59.5 (62.3) 68.1 (18.2) 99.7 (271.8) 5.0 ( 95.7) 
FP-WT 7.4 (13.3) 79.3 (83.4) 74.0 (19.5) 99.5 (176.3) 3.1 ( 50.7) 
rF-MUC-1/TRICOM 30.9 (35.7) 84.5 (133.9) 79.8 (27.4) 99.9 (189.3) 31.6 ( 213.1) 
Table 3

Production of IFN-γ by T-1191-MUC-1 cells stimulated with autologous B cells pulsed with MUC-1 peptides

T-1191-MUC-1 cells were used as effectors at in vitro stimulation (IVS)-3. T cells were stimulated with irradiated autologous EBV-transformed B cells pulsed with different MUC-1 peptides at a concentration of 25 μg/ml, and an effector:antigen-presenting cells ratio of 1:3. Twenty-four-h culture supernatants were collected and screened for the secretion of IFN-γ.

PeptideProduction of IFN-γ (pg/ml)
With peptideNo peptide
P-92 380.8 <52.3 
P-1135 347.2 <52.3 
P-94 323.6 <52.3 
P-1004 305.6 <52.3 
P-1069 288.0 <52.3 
P-4 260.0 <52.3 
P-1105 <52.3 <52.3 
P-1108 <52.3 <52.3 
P-1162 <52.3 <52.3 
P-1169 <52.3 <52.3 
P-1172 <52.3 <52.3 
P-1177 <52.3 <52.3 
PeptideProduction of IFN-γ (pg/ml)
With peptideNo peptide
P-92 380.8 <52.3 
P-1135 347.2 <52.3 
P-94 323.6 <52.3 
P-1004 305.6 <52.3 
P-1069 288.0 <52.3 
P-4 260.0 <52.3 
P-1105 <52.3 <52.3 
P-1108 <52.3 <52.3 
P-1162 <52.3 <52.3 
P-1169 <52.3 <52.3 
P-1172 <52.3 <52.3 
P-1177 <52.3 <52.3 
Table 4

MUC-1 peptide analogs

Amino acid sequences of the parental P-92 peptide (amino acid positions 92–101 of MUC-1) and analog peptides. Amino acids are shown by the single-letter code. Substitution amino acids are indicated in bold and italic.

Amino acid sequenceInitial designationT2 bindinga
ATWGQDVTSV P-92 (native) 510 
AIWGQDVTSV I-93 823 
ALWGQDVTSV L-93 821 
ALWGQDVTSL L-93/L-101 736 
AMWGQDVTSV M-93 723 
AMWGQDVTSL M-93/L-101 325 
AIWGQDVTSL I-93/L-101 280 
Amino acid sequenceInitial designationT2 bindinga
ATWGQDVTSV P-92 (native) 510 
AIWGQDVTSV I-93 823 
ALWGQDVTSV L-93 821 
ALWGQDVTSL L-93/L-101 736 
AMWGQDVTSV M-93 723 
AMWGQDVTSL M-93/L-101 325 
AIWGQDVTSL I-93/L-101 280 
a

Results are expressed in mean fluorescence intensity. HLA-A3 peptide (T2 binding = 200) was used as a negative control, and CAP1-6D (T2 binding = 875) was used as a positive control.

Table 5

Cytometric bead array assay for the production of cytokines by MUC-1 peptide-stimulated T cells

The production of interleukin (IL)-2, IL-4, IL-5, IL-10, tumor necrosis factor (TNF)-α, and IFN-γ was analyzed. Standards at concentrations of each cytokine at 0 pg/ml, 312 pg/ml, and 5000 pg/ml were used to determine the concentrations of these six cytokines in the samples. T-1191 MUC-1 cells at in vitro stimulation (IVS)-3 were used as effectors. T cells were stimulated with irradiated autologous EBV-transformed B cells pulsed without peptide or with P-92 or P-93L peptides at a concentration of 25 μg/ml, and an effector:antigen-presenting cell ratio of 1:3. Twenty-four-h culture supernatants were collected and screened for the secretion of cytokines. Results are expressed in pg/ml/106 cell/ml.

PeptideCytokines
IL-2IL-4IL-5IL-10TNF-αIFN-γ
None <20 <20 <20 <20 <20 <20 
       
P-92 58.8 <20 <20 <20 <20 266 
P-93L 366.9 <20 140.9 <20 <20 650 
PeptideCytokines
IL-2IL-4IL-5IL-10TNF-αIFN-γ
None <20 <20 <20 <20 <20 <20 
       
P-92 58.8 <20 <20 <20 <20 266 
P-93L 366.9 <20 140.9 <20 <20 650 
Table 6

Ability of MUC-1-specific T-cell lines T-1191-P-92 and T-1191-P-93L to lyse a MUC-1-expressing tumor cell line (MCF-7)

Results are expressed in percentages of specific lysis at E:T = 25:1. The numbers in parentheses are the SD. MCF-7 (human breast carcinoma cell line) cells are MUC-1 positive and HLA-A2 positive. SK-Mel-24 (melanoma) cells are MUC-1 negative and HLA-A2 positive. T-1191-P-92 cells and T-1191-P-93L cells were used at in vitro stimulation (IVS)-5. The T-1191-P-92 cell line was passaged on the native P-92 peptide, and the T-1191-P-93L cell line was passaged on the agonist P-93L peptide, from IVS-3 to IVS-5. MCF-7 cells were labeled with 111In-labeled MCF-7 cells, and unlabeled C1R-A2 cells were used at a ratio of 1:10. C1R-A2 cells were incubated with or without P-92 peptide (25 μg/ml), P-93L peptide (25 μg/ml), or CAP1-6D control peptide (25 μg/ml).

TargetT-1191-P-92T-1191-P-93L
MCF-7 16.5 (2.7)a 24.5 (4.5)a 
MCF-7 + C1R-A2 15.6 (3.2)a 21.6 (2.9)a 
MCF-7 + C1R-A2 + P-92 4.2 (2.2) 6.1 (1.9) 
MCF-7 + C1R-A2 + P-93L 3.0 (1.5) 3.5 (1.2) 
MCF-7 + C1R-A2 + CAP1-6D 17.1 (3.8)a 20.8 (3.9)a 
SK-Mel-24 0.5 (1.1) 1.4 (0.8) 
TargetT-1191-P-92T-1191-P-93L
MCF-7 16.5 (2.7)a 24.5 (4.5)a 
MCF-7 + C1R-A2 15.6 (3.2)a 21.6 (2.9)a 
MCF-7 + C1R-A2 + P-92 4.2 (2.2) 6.1 (1.9) 
MCF-7 + C1R-A2 + P-93L 3.0 (1.5) 3.5 (1.2) 
MCF-7 + C1R-A2 + CAP1-6D 17.1 (3.8)a 20.8 (3.9)a 
SK-Mel-24 0.5 (1.1) 1.4 (0.8) 
a

Statistically significant lysis (P < 0.01, two-tailed t test) when comparing lysis of MCF-7 cells versus SK-Mel-24 cells. There is also a statistical significance (P < 0.01, two-tailed t test) when comparing lysis of MCF-7 + C1R-A2 versus MCF-7 + C1R-A2 + P-92 peptide or MCF-7 + C1R-A2 versus MCF-7 + C1R-A2 + P-93L.

Table 7

Anti-HLA-A2 antibody inhibition of MCF-7 cell lysis by T-cell lines

MCF-7 cells (1 × 107) were labeled with 111In and were incubated for 1 h in the presence of medium containing no antibody, anti-HLA-A2, 69 (1:100, 1:1000 dilution), or UPC-10 (10 μg/ml). Cells were then used as targets in an 18-h cytotoxic assay. The results are expressed in percentage of specific lysis at E:T ratios of 12.5:1. The numbers in parentheses are the SD.

AntibodyT-1191-P-93LT-1191-P-92
None 18.2 (1.2) 13.0 (2.5) 
Anti-HLA-A2, 69 (1:100) 4.4 (0.14)a 4.2 (1.06)a 
Anti-HLA-A2, 69 (1:1000) 16.3 (2.7) 12.2 (1.3) 
UPC-10 15.7 (1.8) 11.9 (1.0) 
AntibodyT-1191-P-93LT-1191-P-92
None 18.2 (1.2) 13.0 (2.5) 
Anti-HLA-A2, 69 (1:100) 4.4 (0.14)a 4.2 (1.06)a 
Anti-HLA-A2, 69 (1:1000) 16.3 (2.7) 12.2 (1.3) 
UPC-10 15.7 (1.8) 11.9 (1.0) 
a

Statistically significant lysis (P < 0.01, two-tailed t test) when comparing percentage of lysis of none versus the percentage of lysis with anti-HLA-A2, 69. UPC-10 is a control antibody.

Table 8

Production of IFN-γ by T-cell lines generated from two pancreatic cancer patients stimulated with P-92 and agonist P-93L peptide

Cells from four MUC-1-specific T-cell lines established from two pancreatic cancer patients (patients 23 and 18) were used as effector cells at in vitro stimulation (IVS)-4. These T-cell lines were established by stimulation with P-92-pulsed autologous dendritic cells (DCs; T-23-P-92 and P-18-P-92) or P-93L pulsed autologous DCs (T-23-P-93L and T-18-P-93L). For IFN-γ production, T-cell lines were stimulated with irradiated HLA-A-positive allogeneic DCs pulsed with either P-92 or P-93L peptide at a concentration of 25 μg/ml and an effector-to-antigen-presenting cell ratio of 10:1. Twenty-four-h culture supernatants were collected and screened for the secretion of IFN-γ. Results are expressed as pg IFN-γ produced.

T-cell linePeptide
P-92P-93LCAP1-6D
T-23-P-92 299.8 644.5 <26 
T-23-P-93L 400.5 973.0 <26 
T-18-P-92 168.0 366.6 <26 
T-18-P-93L 378.2 524.1 <26 
T-cell linePeptide
P-92P-93LCAP1-6D
T-23-P-92 299.8 644.5 <26 
T-23-P-93L 400.5 973.0 <26 
T-18-P-92 168.0 366.6 <26 
T-18-P-93L 378.2 524.1 <26 
Table 9

Ability of T-cell lines from a pancreatic cancer patient, generated with agonist peptide P-93L, to lyse cancer cells expressing native MUC-1a

T-cell lineTargetE:T ratios
25:112.5:1
T-23-P-93L MCF-7 24.6 (1.0)b,c 17.9 (0.2)b,c 
T-23-P-93L SK-Mel-24 1.3 (1.1) 0.8 (0.6) 
    
T-23-P-92 MCF-7 14.4 (0.6)b 10.1 (0.8)b 
T-23-P-92 SK-Mel-24 0.5 (1.6) 0.4 (1.9) 
T-cell lineTargetE:T ratios
25:112.5:1
T-23-P-93L MCF-7 24.6 (1.0)b,c 17.9 (0.2)b,c 
T-23-P-93L SK-Mel-24 1.3 (1.1) 0.8 (0.6) 
    
T-23-P-92 MCF-7 14.4 (0.6)b 10.1 (0.8)b 
T-23-P-92 SK-Mel-24 0.5 (1.6) 0.4 (1.9) 
a

Results are expressed as percentage lysis.

b

Statistical significance (P < 0.01, two-tailed t test) when comparing lysis of MCF-7 cells versus SK-Mel-24 cells by T-23-P-93L and T-23P-92 cells.

c

Statistical significance (P < 0.01, two-tailed t test) when comparing lysis of MCF-7 cells by T-23-P-93L and T-23-P-92 cells.

Table 10

Ability of T cells from a pancreatic cancer patient generated with agonist peptide P-93L to lyse cancer cells expressing native MUC-1

T-cell lineTargetE:T ratios
25:112.5:1
T-18-P-93L MCF-7 2.5 (0.8)a,b 14.6 (0.5)a,b 
T-18-P-93L SK-Mel-24 1.1 (0.9) 1.0 (0.8) 
    
T-18-P-92 MCF-7 12.1 (0.3)a 8.4 (0.4)a 
T-18-P-92 SK-Mel-24 0.9 (1.4) 1.0 (1.2) 
T-cell lineTargetE:T ratios
25:112.5:1
T-18-P-93L MCF-7 2.5 (0.8)a,b 14.6 (0.5)a,b 
T-18-P-93L SK-Mel-24 1.1 (0.9) 1.0 (0.8) 
    
T-18-P-92 MCF-7 12.1 (0.3)a 8.4 (0.4)a 
T-18-P-92 SK-Mel-24 0.9 (1.4) 1.0 (1.2) 
a

Statistical significance (P < 0.01, two-tailed t test) when comparing lysis of MCF-7 cells versus SK-Mel-24 cells by T-18-P-93L and T-18-P-92 cells.

b

Statistical significance (P < 0.01, two-tailed t test) when comparing lysis of MCF-7 cells by T-18-P-93L and T-18-P-92 cells.

1
Ho JJ, Bi N, Yan PS, Yuan M, Norton KA, Kim YS Characterization of new pancreatic cancer-reactive monoclonal antibodies directed against purified mucin.
Cancer Res
,
51
:
372
-80,  
1991
.
2
Zotter S, Hageman PC, Lossnitzer A, Mooi WJ, Hilgers J Tissue and tumor distribution of human polymorphic epithelial mucin.
Cancer Rev
,
11–12
:
55
-101,  
1988
.
3
Duperray C, Klein B, Durie GB, et al Phenotype analysis of human myeloma cell lines.
Blood
,
73
:
566
-72,  
1989
.
4
Mark AS, Mangkomkanok M B-cell lymphoma marking only with anti-epithelial membrane antigen.
Cancer (Phila.)
,
63
:
2152
-5,  
1989
.
5
Kufe D, Inghirami G, Abe M, Hayes D, Justi-Wheeler H, Schlom J Differential reactivity of a novel monoclonal antibody (DF3) with human malignant versus benign breast tumors.
Hybridoma
,
3
:
223
-32,  
1984
.
6
Burchell J, Gendler S, Taylor-Papadimitriou J, et al Development and characterization of breast cancer reactive monoclonal antibodies directed to the core protein of the human milk mucin.
Cancer Res
,
47
:
5476
-82,  
1987
.
7
Bresalier RS, Niv Y, Byrd JC, et al Mucin production by human colonic carcinoma cells correlates with their metastatic potential in animal models of colon cancer metastasis.
J Clin Investig
,
87
:
1037
-45,  
1991
.
8
Kobayashi H, Terao T, Kawashima Y Serum sialyl Tn as an independent predictor of poor prognosis in patients with epithelial ovarian cancer.
J Clin Oncol
,
10
:
95
-101,  
1992
.
9
Nakamori S, Ota DM, Cleary KR, Shirotani K, Irimura Y MUC-1 mucin expression as a marker of progression and metastasis of human colorectal carcinoma.
Gastroenterology
,
106
:
353
-61,  
1994
.
10
McGuckin MA, Walsh MD, Hohn BG, Ward BG, Wright RG Prognostic significance of MUC-1 epithelial mucin expression in breast cancer.
Hum Pathol
,
26
:
432
-9,  
1995
.
11
Barnd DL, Lan MS, Metzgar RS, Finn OJ Specific, major histocompatibility complex-unrestricted recognition of tumor-associated mucins by human cytotoxic T cells.
Proc Natl Acad Sci USA
,
86
:
7159
-63,  
1989
.
12
Ioannides CG, Fisk B, Jerome KR, Irimura T, Wharton JT, Finn OJ Cytotoxic T cells from ovarian malignant tumors can recognize polymorphic epithelial mucin core peptides.
J Immunol
,
151
:
3693
-703,  
1993
.
13
Takahashi T, Makiguchi Y, Hinoda Y, et al Expression of MUC1 on myeloma cells and induction of HLA-unrestricted CTL against MUC1 from a multiple myeloma patient.
J Immunol
,
153
:
2102
-9,  
1994
.
14
Noto H, Takahashi T, Makiguchi Y, Hayashi T, Hinoda Y, Imai K Cytotoxic T lymphocytes derived from bone marrow mononuclear cells of multiple myeloma patients recognize an underglycosylated form of MUC1 mucin.
Int Immunol
,
9
:
791
-8,  
1997
.
15
Gendler S, Taylor-Papadimitriou J, Duhig T, Rothbard J, Burchell J A highly immunogenic region of human polymorphic epithelial mucin expressed by carcinomas is made up of tandem repeats.
J Biol Chem
,
263
:
12820
-3,  
1988
.
16
Siddiqui J, Abe M, Hayes D, Shani E, Yunis E, Kufe D Isolation and sequencing of a cDNA coding for the human DF3 breast carcinoma-associated antigen.
Proc Natl Acad Sci USA
,
85
:
2320
-3,  
1988
.
17
Apostolopoulos V, Loveland BE, Pietersz GA, McKenzie IF CTL in mice immunized with human mucin 1 are MHC-restricted.
J Immunol
,
155
:
5089
-94,  
1995
.
18
Apostolopoulos V, Pietersz GA, McKenzie IF Cell-mediated immune responses to MUC1 fusion protein coupled to mannan.
Vaccine
,
14
:
930
-8,  
1996
.
19
Apostolopoulos V, Karanikas V, Haurum JS, McKenzie IF Induction of HLA-A2-restricted CTLs to the mucin 1 human breast cancer antigen.
J Immunol
,
159
:
5211
-8,  
1997
.
20
McKenzie IF, Xing PX Mucins in breast cancer: recent advances.
Cancer Cells
,
2
:
75
-8,  
1990
.
21
Brossart P, Heinrich KS, Stuhler G, et al Identification of HLA-A2 restricted T-cell epitopes derived from the MUC1 tumor antigen for broadly applicable vaccine therapies.
Blood
,
93
:
4309
-17,  
1999
.
22
Pietersz GA, Li W, Osinski C, Apostolopoulos V, McKenzie IF Definition of MHC-restricted CTL epitopes from non-variable number of tandem repeat sequence of MUC1.
Vaccine
,
18
:
2059
-71,  
2000
.
23
Carmon L, El-Shami KM, Paz A, et al Novel breast-tumor-associated MUC-1-derived peptides: characterization in Db−/− × β 2 microglobulin (β2m) null mice transgenic for a chimeric HLA-A2.1/Db-β2 microglobulin single chain.
Int J Cancer
,
85
:
391
-7,  
2000
.
24
Heukamp LC, van der Burg SH, Drijfhout JW, Melief CJM, Taylor-Papadimitriou J, Offringa R Identification of three non-VNTR MUC-1 derived HLA-A*0201-restricted T-cell epitopes that induce protective anti-tumor immunity in HLA-A2/K(b)-transgenic mice.
Int J Cancer
,
91
:
385
-90,  
2001
.
25
Grey HM, Ruppert J, Vitiello A, et al Class I MHC-peptide interactions: structural requirements and functional implications.
Cancer Surv
,
22
:
37
-49,  
1995
.
26
De Magistris MT, Alexander J, Coggeshall M, et al Antigen analog-major histocompatibility complexes act as antagonists of the T cell receptor.
Cell
,
68
:
625
-34,  
1992
.
27
Jameson SC, Bevan MJ T cell receptor antagonists and partial agonists.
Immunity
,
2
:
1
-11,  
1995
.
28
Zaremba S, Barzaga E, Zhu M, Soares N, Tsang KY, Schlom J Identification of an enhancer agonist cytotoxic T lymphocyte peptide from human carcinoembryonic antigen.
Cancer Res
,
57
:
4570
-7,  
1997
.
29
Valmori D, Fonteneau JF, Lizana CM, et al Enhanced generation of specific tumor-reactive CTL in vitro by selected Melan-A/MART-1 immunodominant peptide analogues.
J Immunol
,
160
:
1750
-8,  
1998
.
30
Kawakami Y, Eliyahu S, Jennings C, et al Recognition of multiple epitopes in the human melanoma antigen gp-100 by tumor-infiltrating T lymphocytes associated with in vivo tumor regression J.
Immunol
,
154
:
3961
-8,  
1995
.
31
Terasawa H, Tsang KY, Gulley J, Arlen P, Schlom J Identification and characterization of a human agonist cytotoxic T-lymphocyte epitope of human prostate-specific antigen.
Clin Cancer Res
,
8
:
41
-53,  
2002
.
32
Soule HD, Vazguez J, Long A, Albert S, Brennan M A human cell line from a pleural effusion derived from a breast carcinoma.
J Natl Cancer Inst (Bethesda)
,
51
:
1409
-16,  
1973
.
33
Anderson KS, Alexander J, Wei M, Cresswell P Intracellular transport of class I MHC molecules in antigen processing mutant cell lines.
J Immunol
,
151
:
3407
-19,  
1993
.
34
Storkus WJ, Howell DN, Salter RD, Dawson JR, Cresswell P NK susceptibility varies inversely with target cell class I HLA antigen expression.
J Immunol
,
138
:
1657
-9,  
1987
.
35
Hogan KT, Shimojo N, Walk SF, et al Mutations in the α2 helix of HLA-A2 affect presentation but do not inhibit binding of influenza virus matrix peptide.
J Exp Med
,
168
:
725
-36,  
1988
.
36
Parker KC, Bednarek MA, Coligan JE Scheme for ranking potential HLA-A2 binding peptides based on independent binding of individual peptide side-chains.
J Immunol
,
152
:
163
-75,  
1994
.
37
Guadagni F, Witt PL, Robbins PF, Schlom J, Greiner JW Regulation of carcinoembryonic antigen expression in different human colorectal tumor cells by interferon-γ.
Cancer Res
,
50
:
6248
-55,  
1990
.
38
Nijman HW, Houbiers JG, Vierboom MP, et al Identification of peptide sequences that potentially trigger HLA-A2.1 restricted cytotoxic T lymphocytes.
Eur J Immunol
,
23
:
1215
-9,  
1993
.
39
Boyum A A one-stage procedure for isolation of granulocytes and lymphocytes from human blood. General sedimentation properties of white blood cells in 1 g gravity field.
Scand J Clin Lab Investig
,
97(Suppl)
:
51
-76,  
1968
.
40
Sallusto F, Lanzavecchia A Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colony stimulating factor plus interleukin-4 and down-regulated by tumor necrosis factor α.
J Exp Med
,
179
:
1109
-18,  
1994
.
41
Jenkins S, Gritz L, Fedor CH, O’Neil EM, Cohen LK, Panicali DL Formation of lentivirus particles in mammalian cells infected with recombinant fowlpox virus.
AIDS Res Hum Retroviruses
,
7
:
991
-8,  
1991
.
42
Gendler SJ, Lancaster CA, Taylor-Papadimitriou J, et al Molecular cloning and expression of human tumor-associated polymorphic epithelial mucin.
J Biol Chem
,
265
:
15286
-93,  
1990
.
43
Gritz L, Destree A, Cormier N, et al Generation of hybrid genes and proteins by vaccinia virus-mediated recombination: application to human immunodeficiency virus type 1 env.
J Virol
,
64
:
5948
-57,  
1990
.
44
Morgan JR, Roberts BE Organization of RNA transcripts from a vaccinia virus early gene cluster.
J Virol
,
51
:
283
-97,  
1984
.
45
Schmitt JFC, Stunnenberg HG Sequence and transcriptional analysis of the vaccinia virus HindIII I fragment.
J Virol
,
62
:
1889
-97,  
1988
.
46
Chakrabarti S, Sisler JR, Moss B Compact, synthetic, vaccinia virus early/late promoter for protein expression.
Biotechniques
,
23
:
1094
-7,  
1997
.
47
Tsang KY, Zaremba S, Nieroda CA, Zhu MZ, Hamilton JM, Schlom J Generation of human cytotoxic T cells specific for human carcinoembryonic antigen epitopes from patients immunized with recombinant vaccinia-CEA vaccine.
J Natl Cancer Inst (Bethesda)
,
87
:
982
-90,  
1995
.
48
Ruppert J, Sidney J, Celis E, Kubo RT, Grey HM, Sette A Prominent role of secondary anchor residues in peptide binding to HLA-A2.1 molecules.
Cell
,
74
:
929
-37,  
1993
.
49
Cox AL, Skipper J, Chen Y, et al Identification of a peptide recognized by five melanoma-specific human cytotoxic T cell lines.
Science (Wash. DC)
,
264
:
716
-9,  
1994
.
50
Rammensee HG, Friede T, Stevanoviic S MHC ligands and peptide motifs: first listing.
Immunogenetics
,
41
:
178
-228,  
1995
.
51
Brossart P, Wirths S, Stuhler G, Reichardt VL, Kanz L, Brugger W Induction of cytotoxic T-lymphocyte responses in vivo after vaccinations with peptide-pulsed dendritic cells.
Blood
,
96
:
3102
-8,  
2000
.
52
Palena C, Arlen P, Zeytin H, Greiner JW, Schlom J, Tsang KY Enhanced expression of lymphotactin by CD8+ T cells is selectively induced by enhancer agonist peptides of tumor-associated antigens.
Cytokine
,
24
:
128
-42,  
2003
.
53
Palena C, Schlom J, Tsang KY Differential gene expression profiles in a human T-cell line stimulated with a tumor-associated self-peptide versus an enhanced agonist peptide.
Clin Cancer Res
,
9
:
1616
-27,  
2003
.
54
Fong L, Hou Y, Rivas A, et al Altered peptide ligand vaccination with Flt3 ligand expanded dendritic cells for tumor immunotherapy.
Proc Natl Acad Sci USA
,
98
:
8809
-14,  
2001
.
55
Morse MA, Deng Y, Coleman D, et al A phase I study of active immunotherapy with carcinoembryonic antigen peptide (CAP-1)-pulsed, autologous human cultured dendritic cells in patients with metastatic malignancies expressing carcinoembryonic antigen.
Clin Cancer Res
,
5
:
1331
-8,  
1999
.
56
Goydos JS, Elder E, Whiteside TL, Finn OJ, Lotze MT A Phase I trial of a synthetic mucin peptide vaccine. Induction of specific immune reactivity in patients with adenocarcinoma.
J Surg Res
,
63
:
298
-304,  
1996
.
57
Pecher G, Haring A, Kaiser L, Thiel E Mucin gene (MUC1) transfected dendritic cells as vaccine: results of a Phase I/II clinical trial.
Cancer Immunol. Immunother
,
51
:
669
-73,  
2002
.
58
Scholl SM, Balloul JM, Le Goc G, et al Recombinant vaccinia virus encoding human MUC-1 and IL-2 as immunotherapy in patients with breast cancer.
J Immunother
,
23
:
570
-80,  
2000
.
59
Miles D, Papazisis K Rationale for the clinical development of STn-KLH (Theratope(R)) and anti-MUC-1 vaccines in breast cancer.
Clin Breast Cancer
,
3(Suppl)
:
S134
-8,  
2003
.
60
Akagi J, Hodge JW, McLaughlin JP, et al Therapeutic antitumor response after immunization with an admixture of recombinant vaccinia viruses expressing a modified MUC1 gene and the murine T-cell costimulatory molecule B7.
J Immunother
,
20
:
38
-47,  
1997
.