The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Keogh, E.
Right arrow Articles by Sette, A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Keogh, E.
Right arrow Articles by Sette, A.
The Journal of Immunology, 2001, 167: 787-796.
Copyright © 2001 by The American Association of Immunologists

Identification of New Epitopes from Four Different Tumor-Associated Antigens: Recognition of Naturally Processed Epitopes Correlates with HLA-A*0201-Binding Affinity1

Elissa Keogh*, John Fikes*, Scott Southwood*, Esteban Celis{dagger}, Robert Chesnut* and Alessandro Sette2,*

* Epimmune, San Diego, CA 92121; and {dagger} Mayo Clinic, Rochester, MN 55905


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Forty-two wild-type and analogue peptides derived from p53, carcinoembryonic Ag, Her2/neu, and MAGE2/3 were screened for their capacity to induce CTLs, in vitro, capable of recognizing tumor target lines. All the peptides bound HLA-A*0201 and two or more additional A2 supertype alleles with an IC50 of 500 nM or less. A total of 20 of 22 wild-type and 9 of 12 single amino acid substitution analogues were found to be immunogenic in primary in vitro CTL induction assays, using normal PBMCs and GM-CSF/IL-4-induced dendritic cells. These results suggest that peripheral T cell tolerance does not prevent, in this system, induction of CTL responses against tumor-associated Ag peptides, and confirm that an HLA class I affinity of 500 nM or less is associated with CTL epitope immunogenicity. CTLs generated by 13 of 20 of the wild-type epitopes, 6 of 9 of the single, and 2 of 5 of the double substitution analogues tested recognized epitopes generated by endogenous processing of tumor-associated Ags and expressed by HLA-matched cancer cell lines. Further analysis revealed that recognition of naturally processed Ag was correlated with high HLA-A2.1-binding affinity (IC50 = 200 nM or less; p = 0.008), suggesting that high binding affinity epitopes are frequently generated and can be recognized as a result of natural Ag processing. These results have implications for the development of cancer vaccines, in particular, and for the process of epitope selection in general.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Cytotoxic T cells (CTLs) are a major component of antitumor immune responses. CTLs can directly lyse tumor cells and also secrete cytokines such as IL-2, TNF, GM-CSF, and IFN-{gamma}, which can have indirect antitumor effects. CTLs constitute a major component of tumor-infiltrating lymphocytes (TIL).3 These cells have been associated with spontaneous tumor regression in humans (1, 2). In vivo CTL induction in transgenic mice gives rise to responses recognizing tumor cells associated with tumor regression or protection from tumor challenge (3, 4). Adoptive transfer experiments in humans have also demonstrated the efficacy of antitumor CTLs (5). Finally, human trials have demonstrated that epitope-specific CTLs can be induced in cancer patients, and their induction correlated, in multiple instances, with partial or complete tumor responses (6, 7, 8, 9).

Tumor-specific CTLs recognize complexes between MHC class I receptors and peptides derived from tumor-associated Ags (TAA) (10, 11). By simultaneously targeting multiple TAA, the likelihood of the outgrowth of tumor cells (tumor escape) that do not express any of the tumor Ags is decreased. Accordingly, we have proposed (12) to use a mixture of epitopes derived from carcinoembryonic Ag (CEA), p53, Her2/neu, and melanoma Ag (MAGE). The combination of these four Ags also addresses different tumor types, such as breast, colon, and lung cancers (13, 14, 15).

The efficacy of epitope-based vaccines may also be increased when specialized APC such as dendritic cells (DC) are used (16, 17, 18). Several phase 1 safety studies have indeed provided evidence that DC-based vaccines are generally safe and well tolerated (7, 19). Finally, the inclusion of multiple epitopes from each TAA, including dominant and subdominant epitopes, offers an opportunity to elicit a response of breadth superior to that observed naturally (20).

TAA-derived epitopes have been identified by various means. Numerous epitopes have been identified by sequencing peptides eluted from purified MHC from the tumor cells of cancer patients and recognized by TIL cells (21, 22). Other epitopes, such as the ones derived from the MAGE family of Ags, have been identified by expression cloning (23). However, both processes are relatively laborious and provide identification of only a relatively small subset of epitopes. Another strategy identifies candidate epitopes by their MHC-binding motif and class I affinity (24, 25, 26). High affinity peptides are then tested for in vitro immunogenicity with PBMCs from normal donors and their ability to induce tumor-reactive CTLs (12, 20, 27, 28, 29).

A potential problem in the development of CTL epitope-based vaccines is the large degree of MHC polymorphism (30). A possible solution to this problem was pointed out when it was found that HLA class I molecules can be divided into several families or supertypes based on similar peptide-binding repertoires (31, 32, 33). For example, the A2 supertype consists of at least eight related molecules. Of these, the most frequently observed are HLA-A*0201, A*0202, A*0203, A*0206, and A*6802. The A2 supertype is expressed in all major ethnicities in the 39–46% range. Because many peptides that bind A*0201 also exhibit degenerate binding (binding to multiple alleles) (34, 35, 36), an A2 supertype multiepitope vaccine could be designed to provide broad, nonethically biased population coverage.

Previous studies with infectious disease Ags demonstrated that immunogenicity could be predicted on the basis of class I binding (37, 38) and supertype cross-reactivity (30, 39, 40, 41, 42). The issue of the relation between class I affinity and TAA epitope immunogenicity is of relevance because tissue-specific and developmental TAA are expressed on normal tissues at some point in time or location within the body. It is possible that T cells specific to these TAA might be functionally inactivated by T cell tolerance. For example, CEA is expressed in fetal tissue (43), MAGE2/3 are expressed by the testis and the placenta (44), Her2/neu is an oncogene that is homologous to the epidermal growth factor receptor (45), and p53 is ubiquitously expressed albeit at low levels in normal cells (46). However, a number of investigators have demonstrated CTL responses to tumor epitopes in both normal donors (12, 47) and cancer patients (14, 48), which would indicate that T cell tolerance to these TAA, if it exists at all, is incomplete. For instance, several investigators have shown that the use of professional APC such as DC (20, 49, 50) or adjuvants (8) is able to overcome tolerance to specific TAA. However, the concern exists that T cells recognizing high affinity epitopes have been selectively eliminated, leaving a repertoire capable of recognizing only low affinity epitopes. The present study was designed to specifically address this question.

An approach that we and others have taken to increase the likelihood of overcoming tolerance is the development of fixed anchor analogues (51) that demonstrate improved HLA-A*0201 affinity and supertype binding. For example, Sarobe (52), Vierboom (53), and Irvine (54) demonstrated that fixed anchor epitopes derived from TAA and infectious disease Ags showed improved immunogenicity in mice. Other investigators have demonstrated that when analogues with binding higher than the corresponding wild-type peptide were used to stimulate cells from cancer patients, in vitro, a peptide-specific CTL response was detected after far fewer restimulations than are required with the wild-type peptide (8, 55). Most importantly, killing of tumor cell lines was also observed. Similar results were also obtained with fixed anchor analogues and PBMCs from normal donors (12). Based on these results, a much stronger CTL response to these engineered epitopes would be anticipated in vivo. Indeed, in clinical trials, Rosenberg et al. (8) have observed tumor regression using an analogue of the gp100.209 epitope in conjunction with IL-2 therapy, demonstrating the value of analogue peptides as immunotherapeutics.

Besides the potential advantages in terms of overcoming T cell tolerance, the use of analogues can also expand the number of potential epitopes, which is relevant in the case of small tumor Ags, such as p53. In addition, analogues can be engineered to increase population coverage of a given epitope and further enhance the immunogenicity of known epitopes. Another potential advantage is to increase peptide manufacturability and stability by modifications, such as substituting aminobutyric acid for cysteine or methionine residues (56).

In this study, we report the results of testing 22 TAA wild-type and 20 analogue HLA-A2 supertype binders for their capacity to induce CTL responses in vitro, using normal PBMC donors and GM-CSF- and IL-4-induced DC (57). This has also allowed us to identify a large number of new epitopes. Furthermore, it allowed for the testing of 1) whether immunogenicity in this system correlates with binding affinity; 2) how frequently CTLs generated by analogues with improved MHC-binding capacity are also associated with recognition of wild-type sequences; and 3) whether high binding affinity is predictive of a peptide being generated by natural processing.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Motif analysis and HLA-binding assays

Briefly, protein sequences from the four tumor Ags (p53, CEA, Her2/neu, and MAGE2/3) were scanned, using a customized program, to identify 8-, 9-, 10-, and 11-mer sequences containing the HLA-A2 supertype main anchor motif. That motif is leucine (L), isoleucine (I), valine (V), methionine (M), alanine (A), or threonine (T) at both position 2 and the C terminus. Nonamer and decamer sequences were further characterized by evaluating the presence of A2-preferred secondary anchor residues (51) by use of an A2-specific algorithm.

All peptides were tested for their capacity to bind purified HLA-A*0201 molecules in vitro. Peptides exhibiting an IC50 of 500 nM or less were then tested for their capacity to bind other predominant A2 supertype molecules (A*0202, A*0203, A*0206, and A*6802) (36). A F6->Y analogue of the HBVcore peptide (sequence FLYSDYFPSV; 51) was used as the radiolabeled probe for the A*0201, A*0202, A*0203, and A*0206 assays. A C4->A analogue of the HBVpol 646–654 peptide (sequence FTQAGYPAL) was used as the radiolabeled probe for the A*6802 assay. The average IC50 of the F6->Y analogue of the HBVcore peptide were 5, 4.3, 10, and 3.7 nM for the A*0201, A*0202, A*0203, and A*0206 assays, respectively. The average IC50 of the C4->A analogue of the HBVpol 646–654 peptide in the A*6802 assay was 8 nM. Peptides that bound at least two of these allelic molecules (A2 supertype binders) were selected for in vitro immunogenicity testing.

Target cell lines

The .221A2.1 cell line, produced by transferring the HLA-A2.1 gene into the HLA-A, -B, -C null mutant human B-lymphoblastoid cell line .221 (58), was pulsed with appropriate peptides and used as the target line to measure the activity of HLA-A2.1-restricted CTLs. The tumor cell lines used as target cells for each Ag were: HT29 (A2-, CEA+) and SW403 or KATO III (both A2+, CEA+) for CEA, HT29 (A2-, Her2/neu+) and SW403 (A2+, Her2/neu+) for Her2/neu, 888 mel (A2-, MAGE3+) and 624 mel (A2+, MAGE2/3+) for MAGE2/3, and Saos-2 (A2+, p53-) and Saos-2/175 or BT549 (both A2+, p53+) for p53. The HLA-typed melanoma cell lines (624 mel and 888 mel) were a generous gift from Y. Kawakami and S. Rosenberg, National Cancer Institute (Bethesda, MD). The colon adenocarcinoma cell lines SW403 and HT-29, the osteosarcoma line Saos-2, and the breast tumor line BT549 were obtained from the American Type Culture Collection (ATCC, Manassas, VA). The gastric cancer line, KATO III, was obtained from the Japanese Cancer Research Resources Bank. The Saos-2/175 (Saos2 transfected with the p53 gene containing a mutation at position 175) was obtained from A. Levine, Princeton University (Princeton, NJ). All cell lines, except those from the ATCC, were grown in RPMI 1640 medium supplemented with antibiotics, sodium pyruvate, nonessential amino acids, and 10% (v/v) heat-inactivated FCS and in situ IFN-{gamma} assays. The melanoma, colon, and gastric cancer cells were treated with 100 U/ml IFN-{gamma} (Genzyme, Cambridge, MA) for 48 h at 37°C before use as targets in the 51Cr release and in situ IFN-{gamma} assays. The p53 tumor targets were treated with 20 ng/ml IFN-{gamma} and 3 ng/ml TNF-{alpha} for 24 h before assay (26).

Purification of PBMCs

PBMCs were collected by leukapheresis from healthy male and female donors. The donors were screened for common infectious diseases, and were class I typed serologically (One Lambda, Canoga Park, CA). Healthy HLA-A2-positive individuals were A2.1 subtyped by standard PCR methods. The PBMCs were purified using standard Ficoll-Paque (Amersham Pharmacia Biotech AB, Uppsala, Sweden) density-gradient centrifugation and frozen at 50 x 106 cells/ml. Each donor was used only once in testing a given peptide.

Primary CTL induction cultures

Generation of DC. Monocytes were purified from previously frozen PBMCs by plating 10 x 106 cells in 3 ml complete medium in each well of a six-well plate. After 2 h at 37°C, the nonadherent cells were removed, and 3 ml complete medium containing 50 ng/ml GM-CSF and 1000 U/ml IL-4 was added. On day 7, the DC were collected, washed, and pulsed with 40 µg/ml peptide at a cell concentration of 1–2 x 106/ml in the presence of 3 µg/ml {beta}2-microglobulin for 4 h at 20°C, and then irradiated (4200 rad).

Induction of CTLs with DC and peptide. CD8+ T cells were isolated by positive selection with Dynal (Great Neck, NY) immunomagnetic beads according to the manufacturer’s instructions. Typically, 200~250 x 106 PBMCs were processed to obtain 24 x 106 CD8+ T cells (enough for a 48-well plate). A total of 0.25 ml CD8+ T cells (2 x 106 cells/ml) was cocultured with 0.25 ml cytokine-generated DC (1 x 105 cells/ml) in each well of a 48-well plate in the presence of 10 ng/ml IL-7. Human rIL-10 was added the next day at a final concentration of 10 ng/ml, and human rIL-2 was added on day 2 at 10 IU/ml.

Restimulation of the induction cultures with peptide-pulsed adherent cells. Adherent cells were generated as described (57). Seven and 14 days after the primary induction, the cells were restimulated with irradiated, adherent cells, pulsed with 10 µg/ml peptide in the presence of 3 µg/ml {beta}2-microglobulin in RPMI 1640 with 5% human Ab serum for 2 h at 37°C. Induction cultures (CD8+ cells) were brought to 0.5 ml with fresh medium, and the cells were transferred to the wells containing the peptide-pulsed adherent cells. Human rIL-10 was added at a final concentration of 10 ng/ml 24 h later, and human rIL-2 was added after 48 h and again 2–3 days later at 50 IU/ml (57). Seven days after the second restimulation, the cultures were assayed for peptide-specific CTL activity in a 51Cr release assay or in situ IFN-{gamma} ELISA. Positive cultures were expanded as described below and tested again for peptide specificity as well as endogenous recognition of tumor targets (26, 59).

Measurement of CTL lytic activity by 51Cr release

Cytotoxicity was determined in a standard 51Cr release assay by assaying individual wells at a single E:T. Cell counts were not performed at this step because of the large sample number, but were estimated to be ~1 x 106 per ml. Peptide-pulsed targets were prepared by incubating the cells with 10 µg/ml peptide overnight at 37°C. Adherent target cells were removed from culture flasks with trypsin-EDTA. Target cells were labeled with 200 µCi 51Cr sodium chromate (DuPont, Wilmington, DE) for 1 h at 37°C, washed twice, resuspended at 106 cells/ml, and diluted 1/10 with K562 cells (an NK-sensitive erythroblastoma cell line used to reduce nonspecific lysis) at a concentration of 3.3 x 106/ml. Target cells (100 µl) and 100 µl effectors were plated in 96-well round-bottom plates and incubated for 5 h at 37°C. A total of 100 µl supernatant was collected from each well, and the percentage of lysis was determined according to the formula: ((cpm of the test sample - cpm of the spontaneous 51Cr release sample)/(cpm of the maximal 51Cr release sample - cpm of the spontaneous 51Cr release sample)) x 100. Maximum and spontaneous release were determined by incubating the labeled targets with 1% Triton X-100 and medium alone, respectively. A positive culture was defined as one in which the specific lysis (sample - background) was 10% or higher in the case of individual wells and was 15% or more at the two highest E:T ratios when expanded cultures were assayed.

Cold target inhibition

Ag specificity was confirmed by cold target inhibition experiments, which used unlabeled .221A2.1 cells pulsed for 16 h at 37°C with 10 µg/ml peptide or irrelevant peptide (HBVcore 18-27) to compete for recognition of 51Cr-labeled, peptide-pulsed .221A2.1 cells. The ratio of cold (inhibitor) targets to radiolabeled targets ranged from 90:1 down to 1:1.

In situ measurement of human IFN-{gamma} production

In brief, Immulon 2 plates were coated with mouse anti-human IFN-{gamma} mAb (BD PharMingen, San Diego, CA) overnight at 4°C. The plates were washed and blocked for 2 h, after which the CTLs (100 µl/well) and targets (100 µl/well) were added to each well, leaving empty wells for the standards and blanks (which received medium only). Again, cell counts were not performed at this step, but were estimated to be ~1 x 106/ml. For expanded cultures, 1 x 105 CTLs/well were mixed with 1 x 105 targets (negative control) or peptide-pulsed or endogenous targets. All wells were brought to 200 µl with medium and incubated for 48 h at 37°C with 5% CO2.

Human rIFN-{gamma} (BD PharMingen) was added to the standard wells starting at 400 pg/100 µl/well, and the plate incubated for 2 h at 37°C. The plates were washed, 100 µl biotinylated mouse anti-human IFN-{gamma} mAb (BD PharMingen) was added to each well, and the plates were incubated for 2 h at room temperature. After washing again, 100 µl/well HRP-streptavidin (Zymed, San Francisco, CA) were added and incubated for 1 h at room temperature. The plates were then washed six times with wash buffer, 100 µl/well developing solution (tetramethylbenzidine 1:1) was added, and the plates were allowed to develop for 5–15 min. The reaction was stopped with 50 µl/well 1 M H3PO4 and read at OD 450. A culture was considered positive if it measured at least 50 pg IFN-{gamma}/well above background and was twice the background level of expression.

CTL expansion

Those cultures that demonstrated activity against peptide-pulsed targets and/or tumor targets were expanded over a 2-wk period with anti-CD3 Abs (60, 61). Briefly, 5 x 104 CD8+ cells were added to a T25 flask containing the following: 1 x 106 irradiated (4200 rad) PBMCs (autologous or allogeneic) per ml, 2 x 105 irradiated (8000 rad) EBV-transformed cells per ml, and OKT3 at 30 ng/ml in RPMI 1640 containing 10% (v/v) human Ab serum, nonessential amino acids, sodium pyruvate, 25 µM 2-ME, L-glutamine, and penicillin/streptomycin. Human rIL-2 was added 24 h later at a final concentration of 200 IU/ml and every 3–4 days thereafter with fresh medium at 50 IU/ml. The cells were split if the concentration exceeded 1 x 106/ml, and the cultures were assayed between days 13 and 15.

Statistical analysis

Probability was determined by {chi}2 analysis.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Immunogenicity of A2 supertype cross-reactive peptides derived from four different TAA

A total of 22 peptides (5 CEA, 7 Her2/neu, 9 MAGE2/3, and 1 p53) (Table IGo) was selected for in vitro immunogenicity testing on the basis of an HLA-A*0201-binding affinity of 500 nM or less, because this affinity threshold had previously been shown to correlate with immunogenicity and antigenicity (37, 38). In addition, the peptides were selected on the basis of their cross-reactive binding (below the 500 nM threshold) to at least two other A2 supertype alleles (A*0202, A*0203, A*0206, or A*6802) because cross-reactive binding peptides have been shown to frequently represent naturally processed epitopes (30, 40, 41).


View this table:
[in this window]
[in a new window]
  		Table I. Immunogenicity of supertype cross-reactive TAA-derived peptides1

 
These peptides were tested using peptide-loaded DC for their capacity to induce peptide-specific CTLs using PBMCs from normal donors (27), as described in Materials and Methods. Seven days after the second restimulation, CTL recognition of both peptide-pulsed and tumor target cell lines was measured by 51Cr release assay or IFN-{gamma} in situ ELISA. Each peptide was tested in two to eight different donors (except CEA.687, which was tested in a single donor) and was considered positive if specific lysis was >=10% above background in the 51Cr release assay or twice background and at least 50 pg/well above background in the in situ IFN-{gamma} ELISA in at least one of the 48 replicate wells tested. A number of the Her2/neu peptides were also independently tested (14) with PBMCs from ovarian cancer patients, and those data are also included in Table IGo for reference purposes only.

Representative data are shown in Fig. 1Go. Individual cultures induced and restimulated with the wild-type peptide, CEA.687, were tested for recognition of peptide-pulsed target cells (Fig. 1GoA). Induction of specific CTLs was noted in 15 of the 48 cultures tested. The CTLs from positive wells were expanded and further tested against both peptide-pulsed targets and the tumor cell target SW403 at E:T ratios ranging from 1:1 to 10:1. Representative results are shown in Fig. 1GoB. Both targets were recognized, with 70 and 67% specific lysis detected at the 3:1 E:T ratio. Further evidence of the specificity of this epitope was demonstrated by cold target inhibition (Fig. 1GoC).



View larger version (15K):
[in this window]
[in a new window]
  		FIGURE 1. CEA.687-specific CTLs recognize peptide-pulsed and tumor target cells. A, Individual cultures were tested against .221A2.1 without peptide ({square}) or .221A2.1 + CEA.687 ({blacksquare}). Well number 2 is negative and is included only for purposes of comparison. B, CTLs from a positive well were expanded and tested for recognition of .221A2.1 without peptide ({square}), .221A2.1 + CEA.687 ({blacksquare}), HT29 (A2-, CEA+; {triangleup}), or SW 403 (A2+, CEA+; {blacktriangleup}). C, The Ag specificity of the CTLs was tested by cold target inhibition with .221A2.1 pulsed with irrelevant peptide, HBc18–27 ({blacksquare}) and .221A2.1 pulsed with CEA.687 ({square}).

 
Table IGo shows a summary of the results obtained using these peptides in terms of numbers of positive CTL induction experiments over the total number performed. The overall frequency of experiments yielding positive results was 63 of 86 (73%), and recognition of naturally processed Ag was noted in 27 of 61 (44%) of them. Thus, the induction of CTLs capable of recognizing the TAA peptides studied does not appear to be a rare event, but rather can be demonstrated in most of the donors studied. The number of wells yielding peptide-specific positive results for each peptide of the total wells tested is also shown in Table IGo. The range of responses from 1 to 15 of 48 is possibly reflective of donor to donor variation in the frequency of epitope-specific CTL precursors.

As can be seen, also in Table IGo, 20 of the 22 peptides tested (91%) were immunogenic in at least one donor, demonstrating that in this assay system, peripheral T cell tolerance did not prevent induction of responses against the A*0201 cross-reactive, tumor-derived epitopes tested. Furthermore, of the 20 peptides for which a response was detected, 13 (65%) induced CTLs that were capable of recognizing tumor cell lines, which endogenously express the corresponding tumor Ag. It should be noted that these are minimal estimates, as it is likely that if additional in vitro immunogenicity experiments were to be performed, additional positive results might be obtained, thus raising the overall fraction of peptides positively demonstrated to be generated by natural TAA processing. Finally, it should be noted that similar frequencies of immunogenicity and naturally processed Ag recognition were observed irrespective of the Ag considered. This suggests that the results obtained are not limited to a particular TAA, but rather could apply to TAA in general, at least with respect to differentiation and overexpressed TAA.

Fixed anchor analogues are immunogenic and induce CTLs that recognize wild-type peptide and tumor cell lines

Thirteen analogues derived from the same TAA (specifically, one Her2/neu, nine p53, and three CEA-derived peptides) were also tested for in vitro immunogenicity following the protocol described above. It should be noted that, in general, corresponding wild-type epitopes were not tested side by side in our series of experiments. This strategy is consistent with the original goal of the study, which was to identify A2 supertype TAA-derived epitopes. Wild-type peptides, which had lower binding affinity or A2 supertype cross-reactivity, were not tested for in vitro immunogenicity to concentrate on the peptides most likely to yield positive results. In the few cases in which a direct comparison can be made, a trend toward higher immunogenicity was observed. Analogues with increased HLA-binding capacity were obtained by modifying one of the two main anchor positions; specifically, L, V, or M was introduced at position 2, or V was introduced at the C terminus. These substitutions were introduced based on the results of Ruppert et al. (51), which showed that these particular residues are associated with optimal A*0201-binding capacity. Several analogues were also generated by introducing an aminobutyric acid (B) for cysteine at nonanchor positions (56). The sequences of these 13 analogues tested are listed in Table IIGo, which also details their A*0201 and A2 supertype cross-reactive binding capacity, as compared with the wild-type peptides. It can be noted that the 13 analogues tested in this study were selected on the basis of improved binding to HLA-A*0201 and other frequent alleles of the A2 supertype. More specifically, these analogues fall into one of the following three categories: 1) improved A*0201-binding capacity, above the 500 nM threshold, in the case of weakly binding wild-type peptides; 2) improved cross-reactive binding capacity, allowing binding to at least three common allelic forms of the A2 supertype, including A*0201; or 3) at least 3-fold improvement of A*0201 binding of an epitope already binding <=500 nM.


View this table:
[in this window]
[in a new window]
  		Table II. Immunogenicity of supertype cross-reactive TAA-derived analogs

 
Representative data from a CTL culture stimulated with one of these fixed anchor analogues, p53.149 M2, are shown in Fig. 2GoA. The CTLs induced and restimulated with the analogue peptide were tested for their ability to lyse both analogue and wild-type peptide-pulsed target cells. Eight of the 48 cultures tested recognized analogue-pulsed targets, and of these, five were also able to recognize the wild-type peptide. A single culture was expanded and assayed for recognition of analogue-pulsed target cells, wild-type pulsed cells, and tumor target cells. A positive response was measured for all three targets (Fig. 2GoB).



View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 2. p53.149 M2-specific CTLs recognize wild-type peptide-pulsed and tumor target cells. A, Individual cultures were tested against .221A2.1 without peptide ({square}), .221A2.1 + p53.149 M2 (), or .221A2.1 + p53.149 ({blacksquare}). A positive response was 10 percentage points above background. Well number 2 is negative and is included only for purposes of comparison. B, CTLs from a positive well were expanded and tested against .221A2.1 without peptide ({square}), .221A2.1 + p53.149 M2 ({blacksquare}), .221A2.1 + p53.149 ({blacktriangleup}), Saos-2 (A2+, p53-; {circ}), or Saos-2/175 (A2+, p53+; •).

 
The results of the in vitro primary immunogenicity testing of the 13 A*0201 analogues are also shown in Table IIGo. As previously done in the case of wild-type peptides, these analogue epitopes were evaluated for their capacity to induce CTLs recognizing analogue-pulsed target cells. In this case, their capacity to recognize wild-type peptide-pulsed target cells was also tested. Finally, the same CTLs were expanded and subsequently tested for their ability to recognize the tumor cell lines endogenously expressing the TAA with which they were induced. This strategy is commonly used, and many previous studies have used expanded T cell populations, lines, or clones to demonstrate that a given epitope is generated in the course of natural processing (see, for example, Refs. 26, 59, 60).

The overall success rate of these experiments was similar to that seen with the wild-type peptides, in that 68% (32 of 47) of these CTL inductions were positive with respect to peptide recognition, and 39% (7 of 18) of these CTLs recognized the naturally processed epitopes generated by tumor cell lines (Table IIGo). The fraction of analogue-specific positive cultures observed in each donor was also similar to that observed with wild-type peptides (data not shown). The number of wells yielding wild-type peptide-specific CTLs of the total wells tested is also shown in Table IIGo and, in general, represents a portion of the analogue-specific cultures. Positive wells ranged between 1 and 22 of the 48 wells tested and are similar to that observed for wild-type peptides.

Most importantly, the data in Table IIGo also show that all of the 13 analogues tested (100%) were immunogenic. This is slightly better, but not significantly different from the 91% that was seen with the wild-type TAA peptides. Of the 13 immunogenic analogues, 12 were tested for recognition of the wild-type peptide and 9 (75%) were positive in at least one experiment (Table IIGo). When tested against HLA-matched tumor targets known to express the appropriate endogenous Ag, lysis of tumor cells or IFN-{gamma} release was detected for six of them (overall rate of recognition, 6 of 13 or 46% of the analogues tested). This is somewhat lower than the figure of 13 of 20 (65%) obtained in the case of wild-type peptides.

Effect of double anchor substitutions on immunogenicity and endogenous recognition

The results presented in the previous section were obtained with analogue peptides that contained a preferred amino acid substitution at one primary anchor position. In an additional series of experiments, preferred substitutions were introduced at both primary anchor positions. Similarly to the case of single substituted analogues, peptides that demonstrated improved A*0201 binding and/or supertype cross-reactive binding over the corresponding wild-type peptides were tested in primary CTL induction experiments (Table IIIGo). The overall number of experiments in which a peptide-positive response was induced was 20 of 25 (80%), and 4 of 9 (44%) of these CTLs recognized the endogenously expressed TAA epitope.


View this table:
[in this window]
[in a new window]
  		Table III. Immunogenicity of supertype cross-reactive TAA-derived double anchor substitution analogs

 
As in the case of singly substituted analogues, seven of seven (100%) of the peptides tested were immunogenic. These double substituted analogues were able to induce analogue-specific CTLs that recognized wild-type peptide in four of seven cases or 57%. These results are similar to those obtained in the case of single amino acid substitution analogues presented in Table IIGo (75%). The fraction of wells yielding wild-type peptide-reactive CTLs is also shown in Table IIIGo. The range of responses is similar to that seen with wild-type peptides (Table IGo). However, when tumor recognition is considered, the double analogues were positive in only two of four (50%) of the cases, while in the case of single substitution analogues, recognition of the tumor targets was observed in 67% (six of nine) of the cases. Even this 50% value may be artificially high because the two peptides that yielded positive CTL inductions are both analogues of Her2/neu.369, a previously known epitope.

Binding affinity correlates with immunogenicity and predicts tumor recognition

As mentioned above, peptides, both wild type and analogue, were chosen for in vitro screening based on HLA-A*0201-binding affinity and cross-reactive binding capacity (30). In previous studies targeting various infectious pathogens, a cutoff of 500 nM binding affinity and cross-reactive binding capacity were determined to accurately predict epitopes generated during the natural course of disease. We analyzed the present data, generated with peptides derived from TAA to examine whether A2 binding, supertype cross-reactive tumor peptides would also frequently represent naturally processed epitopes. A summary of the results of the in vitro screening of both the wild-type and analogue peptides is shown in Table IVGo. Consistent with previous classifications (37), peptides with an affinity <=50 nM are considered high binders, whereas intermediate binders are associated with affinities in the 51–500 nM range. For purposes of analysis, the intermediate binders have been further subdivided into two groups (51–200 and 201–500 nM).


View this table:
[in this window]
[in a new window]
  		Table IV. Summary of results from screening of TAA wild-type peptides and fixed anchor analogs

 
In the case of high binders, 100% of the wild-type peptides were immunogenic, and 80% of them were shown to elicit CTLs that specifically recognized tumor cells (Table IVGo). In the 51–200 nM range, very similar figures were obtained. CTL inductions positive for peptide and tumor cells were noted for 86 and 71% of the peptides, respectively. By contrast, in the case of high affinity, singly substituted analogues, four of five (80%) of the peptides were shown to induce CTLs capable of cross-reacting with wild-type sequence, and tumor recognition was noted in three of six (50%) of the peptides considered. Similar frequencies (67 and 50% for wild-type and tumor recognition, respectively) were observed in the case of peptides in the 51–200 nM HLA-A*0201 affinity range.

In the 201–500 nM range, although most peptides (four of five wild type and one of one analogue) were positive for induction of CTLs recognizing wild-type peptide, tumor recognition was not detected. When the results are compared with all peptides in the <200 nM affinity range, a significant difference is demonstrated (0 of 5 vs 19 of 28 for peptides with binding affinities <=200 nM, p = 0.008).

Identification of novel epitopes from CEA, Her2/neu, MAGE2/3, and p53

The present studies have resulted in the identification of 11 new epitopes: one wild-type and two analogue epitopes for CEA, one wild-type and two analogue Her2/neu epitopes, one MAGE3 wild-type epitope, and four p53 analogue epitopes (Table VGo). Of these, eight represent epitopes from nonredundant regions of their respective proteins. All are characterized by a binding affinity <=200 nM and an ability to induce CTLs in normal donors that also recognize tumor cells demonstrated to express the Ag in question.


View this table:
[in this window]
[in a new window]
  		Table V. Identification of novel tumor-associated Ag-derived epitopes

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
A total of 22 high affinity and cross-reactive wild-type and 20 analogue peptides derived from the Her2/neu, p53, MAGE2/3, and CEA TAA was screened for in vitro immunogenicity with PBMCs from normal donors. CTL responses were induced in a total of 19 of the cases. It would thus appear that peripheral T cell tolerance does not prevent, at least in this experimental system, induction of responses directed against epitopes derived from TAA. Although we and others have previously identified a number of these peptides as epitopes (e.g., CEA (12, 13); p53 (26, 62); Her2/neu (12, 63); and MAGE2/3 (12, 64)), this work describes new epitopes for each TAA. It should be noted that our main goal was to identify new epitopes. Only a few previously or independently identified epitopes were included in the study. Certain epitopes, such as the ones described previously (12, 50), were not included because we did not seek to exhaustively test all known epitopes from the four TAA considered. We have identified 11 new epitopes: three epitopes for CEA, three Her2/neu epitopes, one MAGE3 epitope, and four p53 epitopes, nearly doubling the available number of vaccine candidates. All of these epitopes are characterized by cross-reactive binding to multiple molecules of the A2 supertype, thus enabling broad and nonethically biased population coverage.

This analysis has also addressed three additional issues. The first of these was whether immunogenicity in this system correlates with binding affinity. A binding affinity cutoff of <=500 nm was previously found to be highly predictive of immunogenicity in the case of infectious disease epitopes (37). This same affinity threshold was found to be indicative of immunogenicity for 91% of the TAA-derived wild-type peptides and either 100 or 75% of the analogue peptides, depending on whether recognition of the parental analogue or the wild-type peptide version is considered. It should be noted that a stringent side-by-side comparison with nonself-derived epitopes was not performed. However, in previous experiments using the same protocol, the HBc18–27 peptide (57) induced CTLs in the 12 different CTL induction experiments performed. In all cases, the resulting CTLs recognized both peptide-pulsed cells and HBV core-transfected target cells. The overall number of peptide-positive cultures was between 1 of 48 and 10 of 48, depending on the donor, which is similar to what was reported above in the case of TAA-derived peptides. The lower frequency of positive CTL inductions observed in the case of most of the TAA-derived epitopes, as compared with this isolated example of a nonself, dominant CTL epitope, may be reflective of T cell tolerance and/or T cell anergy phenomena. Taken together, our data support the previously proposed MHC 500 nM affinity threshold for human HLA class I-restricted responses.

The second issue addressed by the present study was whether analogues with improved MHC-binding capacity are also associated with improved immunogenicity. It can be noted that corresponding wild-type epitopes were not in general tested side by side in our series of experiments. This is because the original goal of this study was to identify A2 supertype TAA-derived epitopes. Wild-type peptides, which had lower binding affinity or A2 supertype cross-reactivity, were not tested for in vitro immunogenicity to concentrate on the peptides most likely to yield positive results. In the few cases in which a direct comparison can be made, a trend toward higher immunogenicity was observed. These results are in agreement with what was reported by Parkhurst et al. (55) for a class I-restricted gp100 epitope, and Topalian et al. (65) for a class II-restricted tyrosinase epitope. Similarly, Kawashima et al. (12) demonstrated that an analogue with improved A2.1 binding, CEA.24 M2V9, induced CTLs from PBMCs of a normal donor. These CTLs lysed analogue-pulsed targets and tumor targets. Ag specificity was further demonstrated by cold target inhibition. However, when peptides with similar binding affinities are considered, it can be noted that 6 of 12 (50%) analogues in the 200 nM or less affinity range induced CTLs capable of recognition of tumor cell lines. By comparison, 13 of 17 (76%) of wild-type peptides in the same affinity range induced CTLs for which tumor recognition was demonstrated (Table IVGo, p = 0.11).

Another observation with implications for the use of fixed anchor analogues was that CTLs induced with analogues with substitutions at both position 2 and the C terminus were less frequently able to recognize the endogenously expressed epitope than single substitution analogues. It is possible that the double substitution alters the conformation of the peptide/MHC complex enough to elicit a different set of CTLs, largely noncross-reactive with the wild-type epitope. However, in certain instances, this analoging strategy may be useful, as demonstrated by the study of Kawashima et al. (12), which identified a CEA-derived double analogue associated with remarkable tumor-specific immunogenicity in vitro. Further studies in vivo will establish the relative merit of the use of wild-type or analogue peptide to involve anti-TAA responses in humans.

Another issue addressed in this study was whether high binding affinity is predictive of the peptide being generated by natural processing (Table IVGo). A cutoff of <=200 nM was found to be predictive of tumor recognition. Considering all peptides (wild type or analogues) that induced CTLs, and bound A*0201 <=200 nM, recognition of tumor cell lines known to express the naturally processed Ag was noted for 19 of 28 (68%). This is in contrast with CTL-inducing peptides binding with IC50 > 200 nM, for which recognition of naturally processed epitopes was demonstrated in 0 of 5 (p = 0.008). To more fully assess the immunological relevance of binding to an increased number of A2 supertype molecules, it will be interesting to demonstrate successful induction of CTL response using PBMCs expressing different A2 supertype molecules. These experiments are currently planned in our laboratory.

These results may be related to a faster dissociation rate of lower binding epitopes, which might hinder detection of naturally processed Ag (24, 66). Alternatively, higher affinity peptides might be more likely to be generated by natural Ag processing. Paz and coworkers (67) used engineered constructs of an OVA epitope to demonstrate that processing intermediates are transported via TAP to the endoplasmic reticulum, where they bind to the appropriate class I molecule. After binding to MHC, further trimming generates the optimal peptide, and further degradation is prevented because the MHC molecule renders the peptide inaccessible to proteases. However, in the present study, we show that recognition of naturally processed Ag can be demonstrated for lower binding epitopes when higher affinity analogues are used as in vitro immunogens. This suggests that this effect is linked to events in the T cell induction, and not to an inherent defect in the processing of lower affinity epitopes, being generated in the course of natural processing of TAA. These results are opposite to previous reports (68), which suggested that recognition of naturally processed epitopes might be skewed toward lower affinity epitopes. The reasons for this discrepancy are not clear, but might be related to the different assay systems used. The majority of previous studies on low affinity epitopes examined dominant, spontaneously occurring recall responses from TIL of cancer patients, whereas the present study has analyzed the T cell repertoire capable of responding to deliberate in vitro immunization with the epitope.

Although the present study demonstrates an important role of HLA-binding affinity in determining both immunogenicity and recognition of naturally processed Ag production, we emphasize that several other processes are likely to contribute to immunogenicity and processing efficiency of tumor epitopes. Among them, differences in T cell activity, expression of tumor Ags, proteosomal cleavage, and modulation by various lymphokines are all well recognized (67, 69, 70).

In conclusion, we report identification of 11 new epitopes derived from the TAA Her2/neu, CEA, p53, and MAGE2/3. Our results confirm previous results that suggested 500 nM as an affinity threshold associated with immunogenicity for CTL responses. High binding affinity epitopes appear to be more frequently generated in the course of natural processing. Overall, these results emphasize the importance of HLA binding as a selection criterion for epitopes destined for immunotherapy protocols.


   Acknowledgments
 
The expert secretarial assistance of Robin Delp and Denise Porter is gratefully acknowledged.


   Footnotes
 
1 This publication was supported by National Institutes of Health Contract NOI-AI-95362. Back

2 Address correspondence and reprint requests to Dr. Alessandro Sette, Epimmune, 5820 Nancy Ridge Drive, San Diego, CA 92121. E-mail address: asette{at}epimmune.com Back

3 Abbreviations used in this paper: TIL, tumor-infiltrating lymphocyte; CEA, carcinoembryonic Ag; DC, dendritic cells; MAGE, melanoma Ag; TAA, tumor-associated Ags. Back

Received for publication December 26, 2000. Accepted for publication May 15, 2001.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Zorn, E., T. Hercend. 1999. A MAGE-6-encoded peptide is recognized by expanded lymphocytes infiltrating a spontaneously regressing human primary melanoma lesion. Eur. J. Immunol. 29:602.[Medline]
  2. Kawakami, Y., S. Eliyahu, C. H. Delgado, P. F. Robbins, K. Sakaguchi, E. Appella, J. R. Yannelli, G. J. Adema, T. Miki, S. A. Rosenberg. 1994. Identification of a human melanoma antigen recognized by tumor-infiltrating lymphocytes associated with in vivo tumor rejection. Proc. Natl. Acad. Sci. USA 91:6458.[Abstract/Free Full Text]
  3. Toes, R. E., W. M. Kast, R. J. Blom, S. C. Bakker, R. Offringa, C. J. Melief. 1996. Efficient tumor eradication by adoptively transferred cytotoxic T-cell clones in allogeneic hosts. Int. J. Cancer 66:686.[Medline]
  4. Vierboom, M. P. M., H. W. Nijman, R. Offringa, E. I. H. Van der Voort, T. Van Hall, L. Van den Broek, G. J. Fleuren, P. Kenemans, W. M. Kast, C. J. M. Melief. 1997. Tumor eradication by wild-type p53-specific cytotoxic T lymphocytes. J. Exp. Med. 186:695.[Abstract/Free Full Text]
  5. Kawakami, Y., S. Eliyahu, C. Jennings, K. Sakaguchi, X. Kang, S. Southwood, P. F. Robbins, A. Sette, E. Appella, S. A. Rosenberg. 1995. Recognition of multiple epitopes in the human melanoma antigen gp100 by tumor-infiltrating T lymphocytes associated with in vivo tumor regression. J. Immunol. 154:3961.[Abstract]
  6. Murphy, G., B. Tjoa, H. Ragde, G. Kenny, A. Boynton. 1996. Phase I clinical trial: T-cell therapy for prostate cancer using autologous dendritic cells pulsed with HLA-A0201-specific peptides from prostate-specific membrane antigen. Prostate 29:371.[Medline]
  7. Nestle, F. O., S. Alijagic, M. Gilliet, Y. S. Sun, S. Grabbe, R. Dummer, G. Burg, D. Schadendorf. 1998. Vaccination of melanoma patients with peptide- or tumor lysate-pulsed dendritic cells. Nat. Med. 4:328.[Medline]
  8. Rosenberg, S. A., J. C. Yang, D. J. Schwartzentruber, P. Hwu, F. M. Marincola, S. L. Topalian, N. P. Restifo, M. E. Dudley, S. L. Schwarz, P. J. Spiess, et al 1998. Immunologic and therapeutic evaluation of a synthetic peptide vaccine for the treatment of patients with metastatic melanoma. Nat. Med. 4:321.[Medline]
  9. Bellone, M., G. Iezzi, M. A. Imro, M. P. Protti. 1999. Cancer immunotherapy: synthetic and natural peptides in the balance. Immunol. Today 20:457.[Medline]
  10. Gilboa, E.. 1999. The makings of a tumor rejection antigen. Immunity 11:263.[Medline]
  11. Boon, T., J.-C. Cerottini, B. Van den Eynde, P. van der Bruggen, A. Van Pel. 1994. Tumor antigens recognized by T lymphocytes. Annu. Rev. Immunol. 12:337.[Medline]
  12. Kawashima, I., S. J. Hudson, V. Tsai, S. Southwood, K. Takesako, E. Appella, A. Sette, E. Celis. 1998. The multi-epitope approach for immunotherapy for cancer: identification of several CTL epitopes from various tumor-associated antigens expressed on solid epithelial tumors. Hum. Immunol. 59:1.[Medline]
  13. Tsang, K. Y., S. Zaremba, C. A. Nieroda, M. Z. Zhu, J. M. Hamilton, J. Schlom. 1995. Generation of human cytotoxic T cells specific for human carcinoembryonic antigen epitopes from patients immunized with recombinant vaccinia-CEA vaccine. J. Natl. Cancer Inst. 87:982.[Abstract/Free Full Text]
  14. Rongcun, Y., F. Salazar-Onfray, J. Charo, K. J. Malmberg, K. Evrin, H. Maes, K. Kono, C. Hising, M. Petersson, O. Larsson, et al 1999. Identification of new HER2/neu-derived peptide epitopes that can elicit specific CTL against autologous and allogeneic carcinomas and melanomas. J. Immunol. 163:1037.[Abstract/Free Full Text]
  15. Soussi, T.. 1996. The humoral response to the tumor-suppressor gene-product p53 in human cancer: implications for diagnosis and therapy. Immunol. Today 17:354.[Medline]
  16. Steinman, R. M.. 1996. Dendritic cells and immune-based therapies. Exp. Hematol. 24:859.[Medline]
  17. Brossart, P., M. J. Bevan. 1997. Presentation of exogenous protein antigens on major histocompatibility complex class I molecules by dendritic cells: pathway of presentation and regulation by cytokines. Blood 90:1594.[Abstract/Free Full Text]
  18. Murphy, G. P., B. A. Tjoa, S. J. Simmons, J. Jarisch, V. A. Bowes, H. Ragde, M. Rogers, A. Elgamal, G. M. Kenny, O. E. Cobb, et al 1999. Infusion of dendritic cells pulsed with HLA-A2-specific prostate-specific membrane antigen peptides: a phase II prostate cancer vaccine trial involving patients with hormone-refractory metastatic disease. Prostate 38:73.[Medline]
  19. Holtl, L., C. Rieser, C. Papesh, R. Ramoner, M. Herold, H. Klocker, C. Radmayr, A. Stenzl, G. Bartsch, M. Thurnher. 1999. Cellular and humoral immune responses in patients with metastatic renal cell carcinoma after vaccination with antigen pulsed dendritic cells. J. Urol. 161:777.[Medline]
  20. Tsai, V., S. Southwood, J. Sidney, K. Sakaguchi, Y. Kawakami, E. Appella, A. Sette, E. Celis. 1997. Identification of subdominant CTL epitopes of the GP100 melanoma-associated tumor antigen by primary in vitro immunization with peptide-pulsed dendritic cells. J. Immunol. 158:1796.[Abstract]
  21. Cox, A. L., J. Skipper, Y. Chen, R. A. Henderson, T. L. Darrow, J. Shabanowitz, V. H. Engelhard, D. F. Hunt, Jr C. L. Slingluff. 1994. Identification of a peptide recognized by five melanoma-specific human cytotoxic T cell lines. Science 264:716.[Abstract/Free Full Text]
  22. Wolfel, T., J. Schneider, K. H. Meyer Zum Buschenfelde, H. G. Rammensee, O. Rotzschke, K. Falk. 1994. Isolation of naturally processed peptides recognized by cytolytic T lymphocytes (CTL) on human melanoma cells in association with HLA-A2.1. Int. J. Cancer 57:413.[Medline]
  23. Van der Bruggen, P., C. Traversari, P. Chomez, C. Lurquin, E. De Plaen, B. Van den Eynde, A. Knuth, T. Boon. 1991. A gene encoding an antigen recognized by cytolytic T lymphocytes on a human melanoma. Science 254:1643.[Abstract/Free Full Text]
  24. Sette, A., J. Sidney, M. F. del Guercio, S. Southwood, J. Ruppert, C. Dahlberg, H. M. Grey, R. T. Kubo. 1994. Peptide binding to the most frequent HLA-A class I alleles measured by quantitative molecular binding assays. Mol. Immunol. 31:813.[Medline]
  25. Kast, W. M., R. M. P. Brandt, J. Sidney, J.-W. Drijfhout, R. T. Kubo, H. M. Grey, C. J. M. Melief, A. Sette. 1994. Role of HLA-A motifs in identification of potential CTL epitopes in human papillomavirus type 16 E6 and E7 proteins. J. Immunol. 152:3904.[Abstract]
  26. Theobald, M., J. Biggs, D. Dittmer, A. J. Levine, L. A. Sherman. 1995. Targeting p53 as a general tumor antigen. Proc. Natl. Acad. Sci. USA 92:11993.[Abstract/Free Full Text]
  27. Celis, E., V. Tsai, C. Crimi, R. DeMars, P. A. Wentworth, R. W. Chesnut, H. M. Grey, A. Sette, H. M. Serra. 1994. Induction of anti-tumor cytotoxic T lymphocytes in normal humans using primary cultures and synthetic peptide epitopes. Proc. Natl. Acad. Sci. USA 91:2105.[Abstract/Free Full Text]
  28. Feltkamp, M. C. W., M. P. M. Vierboom, W. M. Kast, C. J. M. Melief. 1994. Efficient MHC class I-peptide binding is required but does not ensure MHC class I-restricted immunogenicity. Mol. Immunol. 31:1391.[Medline]
  29. Oiso, M., M. Eura, F. Katsura, M. Takiguchi, Y. Sobao, K. Masuyama, M. Nakashima, K. Itoh, T. Ishikawa. 1999. A newly identified MAGE-3-derived epitope recognized by HLA-A24-restricted cytotoxic T lymphocytes. Int. J. Cancer 81:387.[Medline]
  30. Sette, A., J. Sidney. 1998. HLA supertypes and supermotifs: a functional perspective on HLA polymorphism. Curr. Opin. Immunol. 10:478.[Medline]
  31. Barber, L. D., B. Gillece-Castro, L. Percival, X. Li, C. Clayberger, P. Parham. 1995. Overlap in the repertoires of peptides bound in vivo by a group of related class I HLA-B allotypes. Curr. Biol. 5:179.[Medline]
  32. Del Guercio, M. F., J. Sidney, G. Hermanson, C. Perez, H. M. Grey, R. T. Kubo, A. Sette. 1995. Binding of a peptide antigen to multiple HLA alleles allows definition of an A2-like supertype. J. Immunol. 154:685.[Abstract]
  33. Sidney, J., S. Southwood, M. F. del Guercio, H. M. Grey, R. W. Chesnut, R. T. Kubo, A. Sette. 1996. Specificity and degeneracy in peptide binding to HLA-B7-like class I molecules. J. Immunol. 157:3480.[Abstract]
  34. Tanigaki, N., D. Fruci, A. Chersi, G. Falasca, R. Tosi, R. H. Butler. 1994. HLA-A2-binding peptides cross-react not only within the A2 subgroup but also with other HLA-A-locus allelic products. Hum. Immunol. 39:155.[Medline]
  35. Sidney, J., M. F. del Guercio, S. Southwood, V. H. Engelhard, E. Appella, H. G. Rammensee, K. Falk, O. Rotzschke, M. Takiguchi, R. T. Kubo. 1995. Several HLA alleles share overlapping peptide specificities. J. Immunol. 154:247.[Abstract]
  36. Sidney, J., H. M. Grey, R. T. Kubo, A. Sette. 1996. Practical, biochemical and evolutionary implications of the discovery of HLA class I supermotifs. Immunol. Today 17:261.[Medline]
  37. Sette, A., A. Vitiello, B. Reherman, P. Fowler, R. Nayersina, W. M. Kast, C. J. Melief, C. Oseroff, L. Yuan, J. Ruppert. 1994. The relationship between class I binding affinity and immunogenicity of potential cytotoxic T cell epitopes. J. Immunol. 153:5586.[Abstract]
  38. Wentworth, P. A., A. Sette, E. Celis, J. Sidney, S. Southwood, C. Crimi, S. Stitely, E. Keogh, N. C. Wong, B. Livingston, et al 1996. Identification of A2-restricted hepatitis C virus-specific cytotoxic T lymphocyte epitopes from conserved regions of the viral genome. Int. Immunol. 8:651.[Abstract/Free Full Text]
  39. Threlkeld, S. C., P. A. Wentworth, S. A. Kalams, B. M. Wilkes, D. J. Ruhl, E. Keogh, J. Sidney, S. Southwood, B. D. Walker, A. Sette. 1997. Degenerate and promiscuous recognition by CTL of peptides presented by the MHC class I A3-like superfamily: implications for vaccine development. J. Immunol. 159:1648.[Abstract]
  40. Bertoni, R., J. Sidney, P. Fowler, R. W. Chesnut, F. V. Chisari, A. Sette. 1997. Human histocompatibility leukocyte antigen-binding supermotifs predict broadly cross-reactive cytotoxic T lymphocyte responses in patients with acute hepatitis. J. Clin. Invest. 100:503.[Medline]
  41. Doolan, D., S. L. Hoffman, S. Southwood, P. A. Wentworth, J. Sidney, R. W. Chesnut, E. Keogh, E. Appella, T. B. Nutman, A. A. Lal, et al 1997. Degenerate cytotoxic T cell epitopes from P. falciparum restricted by multiple HLA-A and HLA-B supertype alleles. Immunity 7:97.[Medline]
  42. Scognamiglio, P., D. Accapezzato, M. A. Casciaro, A. Cacciani, M. Artini, G. Bruno, M. L. Chircu, J. Sidney, S. Southwood, S. Abrignani, et al 1999. Presence of effector CD8+ T cells in hepatitis C virus-exposed healthy seronegative donors. J. Immunol. 162:6681.[Abstract/Free Full Text]
  43. Thompson, J. A., F. Grunert, W. Zimmermann. 1991. Carcinoembryonic antigen gene family: molecular biology and clinical perspectives. J. Clin. Lab. Anal. 5:344.[Medline]
  44. De Plaen, E., K. Arden, C. Traversari, J. J. Gaforio, J. P. Szikora, C. De Smet, F. Brasseur, P. van der Bruggen, B. Lethe, C. Lurquin. 1994. Structure, chromosomal localization, and expression of 12 genes of the MAGE family. Immunogenetics 40:360.[Medline]
  45. Yamamoto, T., S. Ikawa, T. Akiyama, K. Semba, N. Nomura, N. Miyajima, T. Saito, K. Toyoshima. 1986. Similarity of protein encoded by the human c-erb-B-2 gene to epidermal growth factor receptor. Nature 319:230.[Medline]
  46. Levine, A. J., J. Momand, C. A. Finlay. 1991. The p53 tumor suppressor gene. Nature 351:453.[Medline]
  47. Gnjatic, S., Z. Z. Cai, M. Viguier, S. Chouaib, J. G. Guillet, J. Choppin. 1998. Accumulation of the p53 protein allows recognition by human CTL of a wild-type p53 epitope presented by breast carcinomas and melanomas. J. Immunol. 160:328.[Abstract/Free Full Text]
  48. Jaeger, E., H. Bernhard, P. Romero, M. Ringhoffer, M. Arand, J. Karbach, C. Ilsemann, M. Hagedorn, A. Knuth. 1996. Generation of cytotoxic T-cell responses with synthetic melanoma-associated peptides in vivo: implications for tumor vaccines with melanoma-associated antigens. Int. J. Cancer 66:162.[Medline]
  49. Tjandrawan, T., D. M. Martin, M. J. Maeurer, C. Castelli, M. T. Lotze, W. J. Storkus. 1998. Autologous human dendriphages pulsed with synthetic or natural tumor peptides elicit tumor-specific CTLs in vitro. J. Immunother. 21:149.
  50. Brossart, P., G. Stuhler, T. Flad, S. Stevanovic, H. G. Rammensee, L. Kanz, W. Brugger. 1998. Her-2/neu-derived peptides are tumor-associated antigens expressed by human renal cell and colon carcinoma lines and are recognized by in vitro induced specific cytotoxic T lymphocytes. Cancer Res. 58:732.[Abstract/Free Full Text]
  51. Ruppert, J., J. Sidney, E. Celis, R. T. Kubo, H. M. Grey, A. Sette. 1993. Prominent role of secondary anchor residues in peptide binding to HLA-A2.1 molecules. Cell 74:929.[Medline]
  52. Sarobe, P., C. D. Pendleton, T. Akatsuka, D. Lau, V. H. Engelhard, S. M. Feinstone, J. A. Berzofsky. 1998. Enhanced in vitro potency and in vivo immunogenicity of a CTL epitope from hepatitis C virus core protein following amino acid replacement at secondary HLA-A2.1 binding positions. J. Clin. Invest. 102:1239.[Medline]
  53. Vierboom, M. P., M. C. Feltkamp, A. Neisig, J. W. Drijfhout, J. Ter Schegget, J. J. Neefjes, C. J. Melief, W. M. Kast. 1998. Peptide vaccination with an anchor-replaced CTL epitope protects against human papillomavirus type 16-induced tumors expressing the wild-type epitope. J. Immunother. 21:399.
  54. Irvine, K. R., M. R. Parkhurst, E. P. Shulman, J. P. Tupesis, M. Custer, C. E. Touloukian, P. F. Robbins, A. G. Yafal, P. Greenhalgh, R. P. Sutmuller, et al 1999. Recombinant virus vaccination against "self" antigens using anchor-fixed immunogens. Cancer Res. 59:2536.[Abstract/Free Full Text]
  55. Parkhurst, M. R., M. L. Salgaller, S. Southwood, P. F. Robbins, A. Sette, S. A. Rosenberg, Y. Kawakami. 1996. Improved induction of melanoma-reactive CTL with peptides from the melanoma antigen gp100 modified at HLA-A*0201-binding residues. J. Immunol. 157:2539.[Abstract]
  56. Sette, A., J. Alexander, E. Appella, E. Celis, F. V. Chisari, R. T. Kubo, B. Livingston, K. Sakaguchi, J. Sidney, A. Vitiello, et al 1999. Review: the use of synthetic peptides to characterize CTL responses in acute and chronic viral diseases. R. Ahmed, and I. Chen, eds. Persistent Viral Infections 677.-689. Wiley, New York.
  57. Tsai, V., I. Kawashima, E. Keogh, K. Daly, A. Sette, E. Celis. 1998. In vitro immunization and expansion of antigen-specific cytotoxic T lymphocytes for adoptive immunotherapy using peptide-pulsed dendritic cells. Crit. Rev. Immunol. 18:65.[Medline]
  58. Shimizu, Y., R. DeMars. 1989. Production of human cells expressing individual transferred HLA-A,-B,-C genes using an HLA-A,-B,-C null human cell line. J. Immunol. 142:3320.[Abstract]
  59. Van den Eynde, B., P. Hainaut, M. Herin, A. Knuth, C. Lemoine, P. Weynants, P. van der Bruggen, R. Fauchet, T. Boon. 1989. Presence on a human melanoma of multiple antigens recognized by autologous CTL. Int. J. Cancer 44:634.[Medline]
  60. Wang, R. F., S. L. Johnston, G. Zeng, S. L. Topalian, D. J. Schwartzentruber, S. A. Rosenberg. 1998. A breast and melanoma-shared tumor antigen: T cell responses to antigenic peptides translated from different open reading frames. J. Immunol. 161:3598.
  61. Crossland, K. D., V. K. Lee, W. Chen, S. R. Riddell, P. D. Greenberg, M. A. Cheever. 1991. T cells from tumor-immune mice nonspecifically expanded in vitro with anti-CD3 plus IL-2 retain specific function in vitro and can eradicate disseminated leukemia in vivo. J. Immunol. 146:4414.[Abstract]
  62. Nijman, H. W., S. H. Van der Burg, M. P. Vierboom, J. G. Houbiers, W. M. Kast, C. J. Melief. 1994. p53, a potential target for tumor-directed T cells. Immunol. Lett. 40:171.[Medline]
  63. Fisk, B., T. L. Blevins, J. T. Wharton, C. G. Ioannides. 1995. Identification of an immunodominant peptide of HER-2/neu protooncogene recognized by ovarian tumor-specific cytotoxic T lymphocyte lines. J. Exp. Med. 181:2109.[Abstract/Free Full Text]
  64. Van der Bruggen, P., J. Bastin, T. Gajewski, P. G. Coulie, P. Boel, C. De Smet, C. Traversari, A. Townsend, T. Boon. 1994. A peptide encoded by human gene MAGE-3 and presented by HLA-A2 induces cytolytic T lymphocytes that recognize tumor cells expressing MAGE-3. Eur. J. Immunol. 24:3038.[Medline]
  65. Topalian, S. L., M. I. Gonzales, M. Parkhurst, Y. F. Li, S. Southwood, A. Sette, S. A. Rosenberg, P. F. Robbins. 1996. Melanoma-specific CD4+ T cells recognize nonmutated HLA-DR-restricted tyrosinase epitopes. J. Exp. Med. 183:1965.[Abstract/Free Full Text]
  66. Van der Burg, S. H., M. J. Visseren, R. M. Brandt, W. M. Kast, C. J. Melief. 1996. Immunogenicity of peptides bound to MHC class I molecules depends on the MHC-peptide complex stability. J. Immunol. 156:3308.[Abstract]
  67. Paz, P., N. Brouwenstijn, R. Perry, N. Shastri. 1999. Discrete proteolytic intermediates in the MHC class I antigen processing pathway and MHC I-dependent peptide trimming in the ER. Immunity 11:241.[Medline]
  68. Engelhard, V. H.. 1994. Structure of peptides associated with class I and class II MHC molecules. Annu. Rev. Immunol. 12:181.[Medline]
  69. Alexander-Miller, M. A., G. R. Leggatt, J. A. Berzofsky. 1996. Selective expansion of high- or low-avidity cytotoxic T lymphocytes and efficacy for adoptive immunotherapy. Proc. Natl. Acad. Sci. USA 93:4102.[Abstract/Free Full Text]
  70. Yewdell, J. W., J. R. Bennink. 1999. Immunodominance in major histocompatibility complex class I-restricted T lymphocyte responses. Annu. Rev. Immunol. 17:51.[Medline]



This article has been cited by other articles:


Home page
 JCOHome page
M. Barve, J. Bender, N. Senzer, C. Cunningham, F. A. Greco, D. McCune, R. Steis, H. Khong, D. Richards, J. Stephenson, et al.
Induction of Immune Responses and Clinical Efficacy in a Phase II Trial of IDM-2101, a 10-Epitope Cytotoxic T-Lymphocyte Vaccine, in Metastatic Non-Small-Cell Lung Cancer
J. Clin. Oncol., September 20, 2008; 26(27): 4418 - 4425.
[Abstract] [Full Text] [PDF]