Cancer Research Cancer Research Funding Available Protein Translation and Cancer
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
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
Services
Right arrow Email this article to a friend
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
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Harada, M.
Right arrow Articles by Robbins, P. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Harada, M.
Right arrow Articles by Robbins, P. F.
[Cancer Research 61, 1089-1094, February 1, 2001]
© 2001 American Association for Cancer Research

Immunology

Use of an in Vitro Immunoselected Tumor Line to Identify Shared Melanoma Antigens Recognized by HLA-A*0201-restricted T Cells

Mamoru Harada, Yong F. Li, Mona El-Gamil, Steven A. Rosenberg and Paul F. Robbins1

Department of Virology, Medical Institute of Bioregulation, Kyushu University, Fukuoka 812-8582, Japan [M. H.]; and Surgery Branch, National Cancer Institute, NIH, Bethesda, Maryland 20892 [Y. F. L., M. E-G., S. A. R., P. F. R.]


   ABSTRACT
Top
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
An immunoselected melanoma cell line that had lost expression of the dominant melanoma antigens MART-1 and gp100 was generated in an attempt to identify previously unknown tumor antigens. After repeated stimulation with the autologous immunoselected tumor line, a number of HLA-A*0201-restricted T-cell clones were established from the peripheral blood of a single melanoma patient. One T-cell clone (C-22) recognized 14 of 16 HLA-A2+ melanoma cell lines, as well as HLA-A2+ melanocytes but recognized neither HLA-A2+ fibroblasts nor autologous B cells. Screening of an autologous cDNA library resulted in the isolation of a transcript identical to an entry in the expressed sequence tag database. Northern blot analysis revealed that this gene was expressed in most melanoma cell lines and melanocytes but not in normal tissues. The peptide epitope (AMFGREFCYA) recognized by clone C-22 was identified based on studies of the recognition of truncated cDNAs and the use of the consensus HLA-A*0201 binding motif. A second T-cell clone (C-29) was found to recognize a new tyrosinase-related protein 2 epitope (455-463; YAIDLPVSV) in an HLA-A*0201-restricted manner. Together, these results provide additional targets that can be used for the development of immunotherapeutic protocols in HLA-A2+ melanoma patients and demonstrate the utility of immunoselected tumor lines for the identification of new melanoma antigens.


   INTRODUCTION
Top
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Shared antigens recognized by human melanoma reactive T cells have been identified. A number of products, including the MAGE (1) , and ESO-1 (2) gene products, are nonmutated proteins and are not expressed in normal tissues, with the exception of testis. These antigens are expressed in a wide variety of tumor types including breast, lung, and ovarian cancer. In addition, melanoma-reactive T cells have been shown to recognize antigens such as MART-1/MelanA (3) and gp100 (4) , nonmutated self proteins the expression of which in normal tissues is limited to melanocytes and retina. Expression of these molecules is found in melanomas but not other types of cancer.

Several peptides from the MART-1 and gp100 antigens appear to represent immunodominant epitopes in HLA-A2 melanoma patients (5 , 6) . T-cell clones established from cultures of HLA-A2-restricted melanoma reactive T cells are thus predominantly reactive with these antigens. In this study, an in vitro immunoselected tumor line that had lost the ability to stimulate HLA-A2-restricted anti-MART-1 and anti-gp100 T cells was used for the in vitro stimulation of PBMC2 from an HLA-A*0201 melanoma patient. Use of this tumor cell line for the stimulation of in vitro mixed lymphocyte tumor cultures resulted in the generation of HLA-A2-restricted T-cell clones that appeared to recognize additional tumor antigens. Screening of an autologous tumor cell library with a T-cell clone resulted in the isolation of a cDNA clone that encoded a previously undescribed antigen. Use of a second clone resulted in the identification of a TRP-2 epitope that is distinct from a previously described epitope recognized by HLA-A2-restricted, TRP-2 reactive T cells (7) .


   MATERIALS AND METHODS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Cell Lines.
All of the melanoma (mel) and fibroblast cell lines were established in the Surgery Branch of the National Cancer Institute. Most melanoma cell lines were maintained in RPMI 1640 containing 5% fetal bovine serum. F002-S is a melanoma cell line that is recognized by MART-1- and gp100-reactive T cells. The MART-1 and gp100 antigen-loss variant melanoma cell line, termed F002-R mel, was established from F002-S mel using a modification of a previously described method used for in vitro immunoselection (8) . The parental F002-S mel line (2 x 105 cells/well) was cultured in a 24-well plate with both anti-MART-1 CTL clone (2 x 105 cells/well) and anti-gp100 CTL clone (2 x 105 cells/well) in a total volume of 2 ml of RPMI 1640 containing 10% human serum. Surviving tumor cells were expanded and then subjected to a repeat cycle of immunoselection. This process was continued for three cycles, with each cycle lasting approximately 2 weeks. The F002-R mel were cultured with RPMI 1640 containing 10% human serum before being used for in vitro stimulations. 293-A2 is a stable transfectant of the 293 human kidney cell line expressing the HLA-A*0201 gene. The 293 and 293-A2 cells were maintained in DMEM containing 7.5% FCS. Melanocytes were obtained from Clonetics (San Diego, CA).

Mixed Lymphocyte Tumor Culture and Limiting Dilution Cloning.
PBMCs (2 x 106) from patient F002 were cultured in a 24-well plate with the autologous melanoma cell line F002-R mel that had been selected for resistance to anti-MART-1 and anti-gp100 CTLs. The F002-R mel (1 x 105) were irradiated with 12,000 rad and cultured with the PBMCs in a final volume of 2 ml. On day 3, human rIL-2 was added at a concentration of 150 IU/ml, and cultures were restimulated on a weekly basis with 1 x 105 F002-R mel, with IL-2 added the following day at 150 IU/ml. One week after the fifth stimulation with the F002-R mel, the lymphocytes were harvested and tested for their ability to respond to the F002-R mel. The lymphocytes from two wells that responded to the F002-R mel were then cloned by plating the cells at 3, 1, and 0.3 cells/well in 96-well microtiter plates in 0.2 ml with 3,000 rad-irradiated allogeneic PBMCs (5 x 104 cells/well), 12,000 rad-irradiated autologous EBV B cells (1 x 104 cells/well), 30 ng/ml anti-CD3 monoclonal antibody (OKT3; OrthoBiotech, Rartan, NJ), and 150 IU/ml human rIL-2. Half of the medium was changed, and fresh medium containing 300 IU/ml of IL-2 was added twice a week. Fifty-two clones and clones that showed a response to F002-R mel were further characterized. The T-cell clones C-22 and C-29 used in this study were established from wells plated at 1 cell/well. Growth was observed in 34% of the wells plated at 1 cell/well, and thus C-22 and C29 would likely represent T-cell clones.

Flow Cytometric Analysis.
To examine the expression of HLA-A2 molecules, melanoma cells were cultured with the supernatant of the anti-HLA-A2 monoclonal antibody-producing murine hybridoma BB7.2 (IgG1) and followed by FITC-conjugated goat antimouse Fc.

cDNA Library Construction and Screening.
A cDNA library was constructed using 5 mg of poly(A)+ RNA from the F002-S mel, which had been doubly selected using the FastTrack poly(A) isolation kit (Invitrogen, San Diego, CA). Using oligo(dT) primers and the Orient Express cDNA synthesis kit (Novagen, Madison, WI), the cDNA was cloned into the modified eukaryotic expression vector VR1012 (Vical Inc., San Diego, CA), which had been modified to contain a single HindIII site in the multiple cloning site of the vector. Expression of genes cloned into this site is driven by the cytomegalovirus I-E promoter. After the transformation of DH10B cells, the volume needed to obtain approximately 100 bacterial transformants was calculated and used to inoculate deep well culture blocks (Edge Biosystems, Gaithersburg, MD) for 48 h. Plasmid DNA was then prepared using the QIAprep 96 Miniprep Kit (Qiagen Inc., Valencia, CA). Transfection was carried out by mixing 105 293 or 293-A2 cells in 0.1 ml of DMEM without serum with 200 ng of plasmid DNA in 0.1 ml DMEM containing 20 µl/ml of lipofectamine (Life Technologies, Gaithersburg, MD). The following day, the transfection mixture was removed and 1–2 x 104 T cells were added in 0.2 ml of AIM-V medium containing 2% human serum and 30 IU/ml of IL-2. The supernatant was harvested 18–24 h later, and IFN-{gamma} release was measured using an ELISA that was carried out with an antibody pair obtained from Endogen (Cambridge, MA).

Sequencing.
DNA sequencing reactions carried out using the BigDye Terminator Sequencing kit (PE-Applied Biosystems, Foster City, CA) were analyzed using an ABI Prism 310 automated capillary electrophoresis instrument (Perkin-Elmer).

Northern Blot Analysis.
Total RNA was isolated using the RNeasy kit (Qiagen). Total RNA from human normal tissues was purchased from Clontech (Palo Alto, CA). Total RNA (10 mg) was subjected to electrophoresis in a 1.2% agarose formaldehyde gel and transferred to a nylon membrane. A 1650-bp DNA fragment of the gene, isolated in this study, was labeled with [32P]dCTP by the random priming method. Prehybridization and hybridization were performed using the QuickHyb hybridization buffer (Stratagene). Membranes were washed twice with 1 x SSC/0.1% SDS at room temperature for 15 min and twice with 0.1 x SSC/0.1% SDS at 60°C for 30 min, and autoradiography was performed at -70°C.

Peptide Synthesis and Purification.
Peptides were synthesized on the Gilson AMS222 multiple peptide synthesizer using standard N-(9-fluorenyl)methoxycarbonyl chemistry and were purified using either a R2 reverse-phase high-performance liquid chromatography column (PerSeptive Biosystems, Framingham, MA) or a C8 column (VYDAC, Hesperia, CA). Separations were carried out using an acetonitrile gradient in water with 0.05% trifluoroacetic acid. The purity of the peptides was confirmed by mass spectrometry.


   RESULTS
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
In Vitro Generation of Specific Tumor Antigen Loss Variants.
Initially, an attempt was made to derive T-cell clones that recognized novel melanoma antigens by stimulating PBMC from patient F002 with an autologous F002-S mel line. The analysis of nine HLA-A2-restricted clones derived from this patient indicated, however, that all of them recognized MART-1 (data not shown). This finding was in accordance with previous studies demonstrating the immunodominance of responses in HLA-A2 patients to MART-1 and gp100 (5 , 9) and with the recent report demonstrating that HLA-A2 melanoma patients contain a relatively high frequency of MART-1-specific CD8+ T cells (10) .

An attempt was then made to derive a variant from the F002-S mel cell line that lacked expression of MART-1 and gp100, which could be used to generate responses against additional melanoma antigens. Coculture of the F002-S mel cell line with anti-MART-1 and anti-gp100 CTL clones resulted in the production of a cell line, designated F002-R mel, that failed to be recognized by T cells of either specificity (Table 1)Citation . The expression of HLA-A2 on the cell surface was comparable on both the F002-S mel and in vitro immunoselected F002-R mel cell lines (data not shown).


View this table:
[in this window]
[in a new window]

  		Table 1 Inability of F002-R mel to stimulate either anti-MART-1 CTLs or anti- gp100 CTLs

 
Isolation of an HLA-A2-restricted Clone That Recognizes a Novel Tumor Antigen.
Repeated in vitro stimulations of autologous PBMC were then carried out using the F002-R mel line, and 52 T-cell clones or clones that recognized the autologous tumor cell lines were established. One of the clones that was characterized further, C-22, recognized 14 of 16 HLA-A2+ melanoma cell lines (Table 2)Citation . The clone C-22 failed to respond to HLA-A2 negative melanoma cell lines, as well as an autologous EBV B cell line. Clone C-22 also failed to respond to an HLA-A2-expressing fibroblast cell line derived from patient 1383 but recognized the 1383 melanoma cell line that had been derived from the same patient. In addition, clone C-22 responded to an HLA-A2+ melanocyte cell line but failed to respond to an HLA-A2 negative melanocyte cell line, which suggests that this clone recognized a melanocyte differentiation antigen in the context of HLA-A2.


View this table:
[in this window]
[in a new window]

  		Table 2 Response of the C-22 cells against a panel of target cell lines

 
Cloning of the cDNA That Encodes the Antigen Recognized by Clone C-22.
The apparent lack of expression of MART-1 and gp100 in the F002-R cell line suggested that clone C-22, which demonstrated a comparable response to this as well as the parental cell line, did not recognize these antigens. To further exclude the possibility that this clone recognized MART-1 and gp100, as well as additional melanoma antigens that have previously been identified, the 293-A2 cell line was transiently transfected with constructs that encoded either MART-1, gp100, TRP-1, TRP-2, ESO-1, or tyrosinase (Table 3)Citation . The failure of cells transfected with these constructs to stimulate clone C-22 indicated that this clone might recognize a previously undescribed melanoma antigen.


View this table:
[in this window]
[in a new window]

  		Table 3 Recognition of the 1A1 cDNA by the C-22 cells

1 x 104 T cells were incubated with 5 x 104 293 cells, 5 x 104 293-A2 cells or 2 x 104 melanoma cells for 18 h, and the release of cytokine was measured using IFN-{gamma} ELISA.

 
A cDNA library was then constructed using poly(A)+ RNA isolated from the parental F002-S cell line, and cDNA pools containing approximately 100 cDNA clones were screened as described previously (11) . After the screening of 768 pools, 4 positive pools were isolated. Isolation of individual cDNA clones was then carried out from one of the positive pools, and a 1650-bp cDNA clone (1A1) that was isolated from this positive pool was recognized by clone C-22 after transfection into 293-A2 cells (Table 3)Citation . Several entries in the expressed sequence tag database possessed a nucleotide sequence that was nearly identical to that of the 1A1 clone. These included several sequences derived from melanocyte and melanoma cDNA libraries, as well as a single sequence derived from an endometrial carcinoma. The 1A1 cDNA clone appears to encode a previously undescribed 530 amino acid protein (Fig. 1)Citation and was designated AIM-1.



View larger version (56K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 1. Nucleotide and predicted amino acid sequence of the 1A1 cDNA gene. The antigenic epitope is underlined. The bold letters indicate the start and stop codons. The sequence data are available from EMBC/GenBank/DDBJ under accession no. AF172849.

 
Expression of the AIM-1 Gene.
Northern blot analysis was then carried out to examine the expression of the AIM-1 gene (Fig. 2)Citation . Total RNA (10 mg) from melanomas as well as normal melanocytes, fibroblasts, EBV B cells, and a variety of normal tissue types was subjected to electrophoresis. Equivalent loading of the RNA samples was confirmed by ethidium bromide staining (data not shown). The AIM-1 gene was found to be expressed in three melanoma cell lines but not in a fibroblast cell line (Fig. 2A)Citation . The AIM-1 gene was also expressed in one of the two melanocyte cell lines that were tested. As shown in Fig. 2BCitation , no expression of the AIM-1 gene was detected in the 1102 mel line, and expression of this gene product in the 1363 mel line appeared to be relatively low compared with the other three mel lines. These results of Northern blot analysis correlated with the functional assay (Table 2)Citation , because the HLA-A2+1102 mel cell line failed to stimulate C-22 T cells, and the HLA-A2+1363 mel cell line only weakly stimulated C-22 T cells.



View larger version (40K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 2. Northern blot analysis of melanoma cell lines, melanocytes and fibroblasts. Total RNA (10 mg) from each sample was subjected to electrophoresis in a 1.2% agarose formaldehyde gel. The equality of each RNA loaded was confirmed by the ethidium bromide-staining.

 
Northern blot analysis revealed that two bands hybridized to the AIM-1 probe under conditions of high stringency. The lower molecular weight band of approximately 1.7 kb may correspond to the 1A1 cDNA clone. The higher molecular weight band of approximately 2.8 kb could either represent an alternatively spliced product of the AIM-1 gene or result from cross-hybridization to a closely related gene. A summary of the results of Northern blot analysis as well as assays carried out is shown in Table 4Citation . The AIM-1 gene was expressed in 8F0366 melanocyte but not in either the autologous F002 EBV B cells or an allogeneic fibroblast cell line, and the AIM-1 gene did not appear to be expressed at significant levels in any of the 15 normal tissues that were examined. On the basis of previous results with the MART-1 and gp100 genes, expression of the AIM-1 gene would be expected in normal retinal tissues; however, this could not be examined because of the limited availability of this tissue. As noted above, T-cell recognition was only observed in cell lines that expressed the AIM-1 gene product and HLA-A2. Overall, these results indicated that the AIM-1 gene encoded a previously undescribed melanocyte differentiation antigen that was expressed in a high percentage of melanoma cell lines.


View this table:
[in this window]
[in a new window]

  		Table 4 Expression of the AIM-1 gene in different cell lines and normal tissues

 
Identification of AIM-1 T-cell Epitope.
To identify the epitope of AIM-1 recognized by clone C-22, truncated AIM-1 variants were generated. As shown in Table 3Citation , the C-22 T-cell clone recognized 293-A2 cells transfected with a construct encoding the first 84 amino acids of AIM-1. This sequence was scanned using the HLA-A*0201 binding motif (12) , and 14 peptides that were predicted to bind with high or intermediate affinity to HLA-A*0201 were synthesized. One of these peptides, AIM-1 41–50 (AMFGREFCYA), was found to induce IFN-{gamma} production from the C-22 cells after incubation with autologous EBV B cells, and dose titration experiments demonstrated that both of these peptides could stimulate the C-22 cells when pulsed on autologous EBV B cells at a minimum dose of approximately 10 ng/ml (Fig. 3)Citation . A number of HLA-A2 binding tumor antigen epitopes have been identified that contain amino acids such as alanine or threonine at one of the two primary anchor residue positions, indicating that amino acids with short side chains can be readily tolerated at these positions.



View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 3. Dose-dependent recognition of AIM-1 peptide by C-22 T cells. T cells (20,000) were incubated for 18 h with 4 x 104 F002 EBV B cells, which were cultured with the indicated peptides at the indicated doses for 90 min, and the release of cytokine was measured using IFN-{gamma} ELISA. As a control peptide, FluM1 peptide (influenza matrix protein 58-66; GILGFVFTL) was used.

 
Recognition of a Product of TRP-2 by Clone C-29.
C-29 was another T-cell clone established from autologous PBMC by repeated stimulation with in vitro immunoselected F002-R mel. The clone C-29 recognized seven of the seven HLA-A2+ melanoma cell lines that were tested (Table 5)Citation . The clone failed to respond to HLA-A2 negative melanoma cell lines, as well as an autologous EBV B cell line. Clone C-29 also failed to respond to an HLA-A2 expressing fibroblast cell line derived from patient 1300 but recognized the 1300 melanoma cell line that had been derived from the same patient.


View this table:
[in this window]
[in a new window]

  		Table 5 Response of the C-29 cells against a panel of targets

2 x 104 (experiment 1) or 1 x 104 T cells (experiment 2) were incubated with 2 x 104 melanoma cells, 5 x 104 293 cells or 5 x 104 293-A2 cells for 18 h, and the release of cytokine was measuring using IFN-{gamma} ELISA.

 
To determine whether this clone could recognize previously described melanoma antigens, the 293-A2 cell line was transiently transfected with constructs encoding either MART-1, gp100, tyrosinase, TRP-1, or TRP-2. Clone C-29 recognized 293-A2 cells, but not 293 cells, transfected with the TRP-2 construct. These results indicated that the clone C-29 recognized a TRP-2-derived epitope in the context of HLA-A2.

Identification of a New TRP-2 T-cell Epitope.
To identify the epitope of TRP-2 recognized by clone C-29, 22 peptides that were predicted to bind with high or intermediate affinity to HLA-A*0201 were synthesized. One of these peptides, TRP-2 455-463 (YAIDLPVSV), was found to induce IFN-{gamma} production from the clone C-29 cells after incubation with autologous EBV B cells. Dose titration experiments demonstrated that the half maximal dose of the TRP-2 455-463 peptide required to stimulate the C-29 cells was approximately 10 ng/ml when pulsed on autologous EBV B cells, although doses as low as 1 pg/ml appeared to be weakly stimulatory (Fig. 4)Citation . Substitution of a leucine or methionine residue for the alanine residue at position 2 resulted in the generation of a peptide that was significantly less stimulatory than the parental peptide (data not shown).



View larger version (14K):
[in this window]
[in a new window]
[Download PPT slide]
  		Fig. 4. Dose-dependent recognition of TRP-2 peptide by C-29 T cells. T cells (20,000) were incubated for 18 h with 4 x 104 F002 EBV B cells, which were cultured with the indicated peptides at the indicated doses for 90 min, and the release of cytokine was measured using IFN-{gamma} ELISA. IFN-{gamma} production by the C-29 cells in the presence of F002 EBV B that had not been incubated with any specific peptide was less than 8 pg/ml. As a control peptide, FluM1 peptide (influenza matrix protein 58-66; GILGFVFTL) was used.

 
T Cells Appear To Recognize Unknown Antigens Expressed on the F002-R Melanoma.
Additional T-cell clones generated by stimulating with the immunoselected F002-R variant were also tested for their ability to recognize known melanoma antigens. One of the six additional HLA-A2-restricted T-cell clones that were isolated recognized the MART-1 27-35 peptide as well as the F002-R melanoma, indicating that loss of expression of MART-1 was not complete in the F002-R cell line. Preliminary results indicate that additional T-cell clones isolated from the F002-R mixed lymphocyte tumor culture may recognize previously undescribed antigens in the context of HLA-A*0201, HLA-A*0101, or B*1501 (data not shown). These results suggest that the identification of subdominant antigens may be facilitated through the use of immunoselected tumor cell lines.


   DISCUSSION
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
It has been reported that responses of patient TIL to gp100 but not to MART-1 correlate with clinical responses after adoptive immunotherapy with autologous TILs (9) . This finding suggests that the gp100 antigen represents a good target for antitumor vaccine therapies. Vaccination with a modified gp100 peptide in combination with high dose IL-2 appeared to significantly enhance clinical response rates in comparison with those seen after treatment with IL-2 alone (13) . However, several reports suggest that metastatic melanoma lesions are heterogeneous in the expression of melanoma antigens (14 , 15) , and vaccination with a melanoma antigen-derived peptide may have resulted in the selection of tumor antigen loss variants in vivo (16) . Although this observation may provide immunological evidence for the specificity of melanoma antigen vaccination, it suggests that the growth of tumor antigen loss variants may, in at least some patients, be responsible for the lack of response to these therapies. Vaccination with combinations of peptides may prevent the emergence of tumor antigen loss variants.

Preliminary observations revealed that nine of nine HLA-A2-restricted T-cell clones that were made from the PBMC of patient F002 after stimulation with autologous F002-S mel recognized the MART-1 27-35 peptide. To avoid the predominant expansion of HLA-A2-restricted MART-1 27-35 reactive, as well as gp100 reactive CTLs, a cell line that appears to have down-regulated expression of these antigens, termed F002-R, was produced. The F002-R cell line was generated by culturing the original parental cell line with CTLs that were reactive with MART-1 as well as gp100. Previous observations demonstrated that the parental F002 cell line naturally expressed relatively low levels of gp100 (16) . The F002 cell line was derived from a patient after vaccination with an immunodominant epitope of gp100, and in vivo selection may have played a role in the generation of a cell line that expressed reduced levels of gp100. Heterogeneous expression of gp100 was also seen in this tumor, and the clonal variation in antigen expression already seen in vivo could have facilitated the generation of a resistant tumor cell line.

The F002-R mel line was then used for the in vitro stimulation of autologous PBMC. Of the 52 T-cell clones that recognized the F002-R mel line, 7 clones recognized melanomas in an HLA-A2-restricted manner. Three of the 7 HLA-A2-restricted clones recognized a single epitope within AIM-1. The recognition of the same epitope, as well as HLA-A2+ melanomas by three T-cell clones, indicates that AIM-1 represents an antigen recognized by melanoma reactive T cells. The strict correlation between recognition of the HLA-A2+ melanomas and melanocytes and AIM-1 gene expression further demonstrates that AIM-1 represents the antigen recognized by these T cells.

The AIM-1 gene was expressed in melanocytes but not in other normal tissues, and the C-22 T-cell clone recognized an HLA-A2+ melanocyte cell line. A search of the nucleotide databases demonstrated that the AIM-1 gene was nearly identical to several cDNA clones of unknown function derived from melanomas and melanocytes, indicating that it may represent a melanosomal differentiation antigen. In addition, a single cDNA clone with a nearly identical sequence was isolated from an endometrial carcinoma. The significance of this observation is unclear, because previous reports have failed to describe melanocyte differentiation antigens that are expressed either in tumors derived from this tissue or in normal endometrium. Searches of protein databases indicated that the AIM-1 gene product shares some similarity with a number of plant sucrose transporter proteins; however, proteins that function as sucrose transporters have not been described in animal cells. The AIM-1 gene product contains a single consensus N-linked glycosylation site but does not appear to contain either a conventional signal sequence or transmembrane region, making it unlikely that this represents a transporter protein. The absence of readily identifiable protein motifs in the AIM-1 gene product may make it difficult to determine the function of this molecule in normal melanocytes.

The HLA-A*0201-binding motif was used to identify candidate epitopes from AIM-1. Target cells pulsed with 1 of the 14 peptides that were tested, AIM-1 41–50 (AMFGREFCYA), strongly stimulated C-22 cells. A number of recently described tumor antigen epitopes have been found to contain an optimal amino acid at one of the two major anchor residue positions along with amino acids with short-side chains such as alanine or threonine at the additional anchor position, and thus the AIM-1 41–50 peptide appears to conform to an extended binding motif for this HLA class I allele.

The melanosomal enzyme TRP-2 has been shown previously (17) to represent a melanoma antigen recognized by tumor-reactive CTLs in both mouse and human. In C57BL/6 mice, CTL lines reactive with the murine B16 melanoma were found to predominantly recognize TRP-2 180-188 in the context of H-2Kb (17) . In the human, the TRP-2 197-205 peptide was shown to be recognized by two independent TILs in the context of either HLA-A31 or HLA-A33 (18) . In addition, the repeated in vitro stimulation of PBMC from HLA-A2+ melanoma patients with the TRP-2 180-188 peptide, identical to the murine TRP-2 epitope, was found to induce HLA-A*0201-restricted melanoma-reactive CTLs (7) . In the present study, the previously unidentified TRP-2 455-463 peptide was found to be recognized by HLA-A*0201-restricted melanoma-reactive T cells.

In conclusion, a previously undescribed shared melanoma antigen and a new TRP-2 epitope that are recognized by HLA-A*0201-restricted T cells have been identified using mixed tumor lymphocyte cultures with an in vitro immunoselected autologous tumor line. The HLA-A*0201 molecule represents the most commonly expressed MHC class I allele in Caucasian melanoma patients, with an estimated frequency of approximately 50%. Therefore, these findings could provide useful additional targets that can be used for immunotherapeutic protocols in HLA-A2 positive melanoma patients. In addition, these studies suggest that the identification of previously unknown tumor antigens may be facilitated by the use of immunoselected tumor cell lines.


   ACKNOWLEDGMENTS
 
We would like to thank Drs. F. Marincola and A. Riker for providing cell lines and Dr. C. Maccalli for assistance in some experiments. We also thank Dr. M. Parkhurst and J. Riley for preparation of synthetic peptides and A. Mixon for performing FACS analysis.


   FOOTNOTES
 
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.

1 To whom requests for reprints should be addressed, at Surgery Branch, National Cancer Institute, NIH, Room 2B42, Bethesda, MD 20892-1502. Phone: (301) 496-9383; Fax: (301) 496-0011; E-mail: Paul_Robbins{at}nih.gov Back

2 The abbreviations used are: PBMC, peripheral blood mononuclear cell; EBV B, Epstein Barr Virus transformed B; TIL, tumor-infiltrating lymphocyte; IL-2, interleukin-2; AIM, antigen isolated from immunoselected melanoma; TRP, tyrosinase-related protein. Back

Received 4/14/00. Accepted 11/20/00.


   REFERENCES
Top
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. van der Bruggen P., Traversari C., Chomez P., Lurquin C., De Plaen E., Van den Eynde B., Knuth A., Boon T. A gene encoding an antigen recognized by cytolytic T lymphocytes on a human melanoma.. Science (Washington DC), 254: 1643-1647, 1991.[Abstract/Free Full Text]
  2. Chen Y. T., Scanlan M. J., Sahin U., Tureci O., Gure A. O., Tsang S., Williamson B., Stockert E., Pfreundschuh M., Old L. J. A testicular antigen aberrantly expressed in human cancers detected by autologous antibody screening.. Proc. Natl. Acad. Sci. USA, 94: 1914-1918, 1997.[Abstract/Free Full Text]
  3. Kawakami Y., Eliyahu S., Delgado C. H., Robbins P. F., Rivoltini L., Topalian S. L., Miki T., Rosenberg S. A. Cloning of the gene coding for a shared human melanoma antigen recognized by autologous T cells infiltrating into tumor.. Proc. Natl. Acad. Sci. USA, 91: 3515-3519, 1994.[Abstract/Free Full Text]
  4. Kawakami Y., Eliyahu S., Delgado C. H., Robbins P. F., Sakaguchi K., Appella E., Yannelli J. R., Adema G. J., Miki T., Rosenberg S. A. Identification of a human melanoma antigen recognized by tumor infiltrating lymphocytes associated with in vivo tumor rejection.. Proc. Natl. Acad. Sci. USA, 91: 6458-6462, 1994.[Abstract/Free Full Text]
  5. Kawakami Y., Eliyahu S., Sakaguchi K., Robbins P. F., Rivoltini L., Yannelli J. B., Appella E., Rosenberg S. A. Identification of the immunodominant peptides of the MART-1 human melanoma antigen recognized by the majority of HLA-A2 restricted tumor infiltrating lymphocytes.. J. Exp. Med., 180: 347-352, 1994.[Abstract/Free Full Text]
  6. Kawakami Y., Eliyahu S., Jennings C., Sakaguchi K., Kang X-Q., Southwood S., Robbins P. F., Sette A., Appella E., Rosenberg S. A. Recognition of multiple epitopes in the human melanoma antigen gp100 associated with in vivo tumor regression.. J. Immunol., 154: 3961-3968, 1995.[Abstract]
  7. Parkhurst M. R., Fitzgerald E. B., Southwood S., Sette A., Rosenberg S. A., Kawakami Y. Identification of a shared HLA-A*0201-restricted T-cell epitope from the melanoma antigen tyrosinase-related protein 2 (TRP2).. Cancer Res., 58: 4895-4901, 1998.[Abstract/Free Full Text]
  8. Topalian S. L., Kasid A., Rosenberg S. A. Immunoselection of a human melanoma resistant to specific lysis by autologous tumor-infiltrating lymphocytes: possible mechanisms for immunotherapeutic failures.. J. Immunol., 144: 4487-4495, 1990.[Abstract]
  9. Kawakami Y., Dang N., Wang X., Tupesis J., Robbins P. F., Wang R. F., Wunderlich J. R., Yannelli J. R., Rosenberg S. A. Recognition of shared melanoma antigens in association with major HLA-A alleles by tumor infiltrating T lymphocytes from 123 patients with melanoma.. J. Immunother., 23: 17-27, 2000.
  10. Pittet M. J., Valmori D., Dunbar P. R., Speiser D. E., Lienard D., Lejeune F., Fleischhauer K., Cerundolo V., Cerottini J. C., Romero P. High frequencies of naive melan-A/MART-1-specific CD8(+) T cells in a large proportion of human histocompatibility leukocyte antigen (HLA)-A2 individuals.. J. Exp. Med., 190: 705-716, 1999.[Abstract/Free Full Text]
  11. Robbins P. F., El-Gamil M., Li Y. F., Kawakami Y., Loftus D., Appella E., Rosenberg S. A. A mutated ß-catenin gene encodes a melanoma-specific antigen recognized by tumor infiltrating lymphocytes.. J. Exp. Med., 183: 1185-1192, 1996.[Abstract/Free Full Text]
  12. Parker K. C., Bednarek M. A., Coligan J. E. Scheme for ranking potential HLA-A2 binding peptides based on independent binding of individual peptide side-chains.. J. Immunol., 152: 163-175, 1994.[Abstract]
  13. Rosenberg S. A., Yang J. C., Schwartzentruber D. J., Hwu P., Marincola F. M., Topalian S. L., Restifo N. P., Dudley M. E., Schwarz S. L., Spiess P. J., Wunderlich J. R., Parkhurst M. R., Kawakami Y., Seipp C. A., Einhorn J. H., White D. E. Immunologic and therapeutic evaluation of a synthetic peptide vaccine for the treatment of patients with metastatic melanoma [see comments].. Nat. Med., 4: 321-327, 1998.[Medline]
  14. Cormier J. N., Hijazi Y. M., Abati A., Fetsch P., Bettinotti M., Steinberg S. M., Rosenberg S. A., Marincola F. M. Heterogeneous expression of melanoma-associated antigens and HLA-A2 in metastatic melanoma in vivo.. Int. J. Cancer, 75: 517-524, 1998.[Medline]
  15. Jager E., Ringhoffer M., Karbach J., Arand M., Oesch F., Knuth A. Inverse relationship of melanocyte differentiation antigen expression in melanoma tissues and CD8+ cytotoxic-T-cell responses: evidence for immunoselection of antigen-loss variants in vivo.. Int. J. Cancer, 66: 470-476, 1996.[Medline]
  16. Riker A., Cormier J., Panelli M., Kammula U., Wang E., Abati A., Fetsch P., Lee K. H., Steinberg S., Rosenberg S., Marincola F. Immune selection after antigen-specific immunotherapy of melanoma.. Surgery (St. Louis), 126: 112-120, 1999.[Medline]
  17. Zeh H. J., III, Perry-Lalley D., Dudley M. E., Rosenberg S. A., Yang J. C. High avidity CTLs for two self-antigens demonstrate superior in vitro and in vivo antitumor efficacy. J. Immunol., 162: 989-994, 1999.[Abstract/Free Full Text]
  18. Wang R. F., Johnston S. L., Southwood S., Sette A., Rosenberg S. A. Recognition of an antigenic peptide derived from tyrosinase-related protein-2 by CTL in the context of HLA-A31 and -A33.. J. Immunol., 160: 890-897, 1998.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 GeneticsHome page
U. Gunnarsson, A. R. Hellstrom, M. Tixier-Boichard, F. Minvielle, B. Bed'hom, S. Ito, P. Jensen, A. Rattink, A. Vereijken, and L. Andersson
Mutations in SLC45A2 Cause Plumage Color Variation in Chicken and Japanese Quail
Genetics, February 1, 2007; 175(2): 867 - 877.
[Abstract] [Full Text] [PDF]

Home page
 Mol Biol EvolHome page
M. Soejima, H. Tachida, T. Ishida, A. Sano, and Y. Koda
Evidence for Recent Positive Selection at the Human AIM1 Locus in a European Population
Mol. Biol. Evol., January 1, 2006; 23(1): 179 - 188.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
A. Paschen, M. Song, W. Osen, X. D. Nguyen, J. Mueller-Berghaus, D. Fink, N. Daniel, M. Donzeau, W. Nagel, H. Kropshofer, et al.
Detection of Spontaneous CD4+ T-Cell Responses in Melanoma Patients against a Tyrosinase-Related Protein-2-Derived Epitope Identified in HLA-DRB1*0301 Transgenic Mice
Clin. Cancer Res., July 15, 2005; 11(14): 5241 - 5247.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
G. Denkberg, E. Klechevsky, and Y. Reiter
Modification of a Tumor-Derived Peptide at an HLA-A2 Anchor Residue Can Alter the Conformation of the MHC-Peptide Complex: Probing with TCR-Like Recombinant Antibodies
J. Immunol., October 15, 2002; 169(8): 4399 - 4407.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
H. T. Khong and S. A. Rosenberg
Pre-Existing Immunity to Tyrosinase-Related Protein (TRP)-2, a New TRP-2 Isoform, and the NY-ESO-1 Melanoma Antigen in a Patient with a Dramatic Response to Immunotherapy
J. Immunol., January 15, 2002; 168(2): 951 - 956.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
J. Du and D. E. Fisher
Identification of Aim-1 as the underwhite Mouse Mutant and Its Transcriptional Regulation by MITF
J. Biol. Chem., January 4, 2002; 277(1): 402 - 406.
[Abstract] [Full Text]

Home page
 J. Immunol.Home page
S. Akilesh, M. E. Dudley, P. A. Eden, and D. C. Roopenian
Efficient Chromosomal Mapping of a Methylcholanthrene- Induced Tumor Antigen by CTL Immunoselection
J. Immunol., November 1, 2001; 167(9): 5143 - 5149.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
A. K. Sharma, J. J. Kuhns, S. Yan, R. H. Friedline, B. Long, R. Tisch, and E. J. Collins
Class I Major Histocompatibility Complex Anchor Substitutions Alter the Conformation of T Cell Receptor Contacts
J. Biol. Chem., June 8, 2001; 276(24): 21443 - 21449.
[Abstract] [Full Text] [PDF]

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
Services
Right arrow Email this article to a friend
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
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Harada, M.
Right arrow Articles by Robbins, P. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Harada, M.
Right arrow Articles by Robbins, P. F.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online