Identification and translational validation of novel mammaglobin-A CD8 T cell epitopes.

Mammaglobin-A (MAM-A) is a secretory protein that is overexpressed in 80 % of human breast cancers. Its near-universal expression in breast cancer as well as its exquisite tissue specificity makes it an attractive target for a breast cancer prevention vaccine, and we recently initiated a phase 1 clinical trial of a MAM-A DNA vaccine. Previously, we have identified multiple MAM-A CD8 T cell epitopes using a reverse immunology candidate epitope approach based on predicted binding, but to date no attempt has been made to identify epitopes using an unbiased approach. In this study, we used human T cells primed in vitro with autologous dendritic cells expressing MAM-A to systematically identify MAM-A CD8 T cell epitopes. Using this unbiased approach, we identified three novel HLA-A2-restricted MAM-A epitopes. CD8 T cells specific for these epitopes are able to recognize and lyse human breast cancer cells in a MAM-A-specific, HLA-A2-dependent fashion. HLA-A2(+)/MAM-A(+) breast cancer patients have an increased prevalence of CD8 T cells specific for these novel MAM-A epitopes, and vaccination with a MAM-A DNA vaccine significantly increases the number of these CD8 T cells. The identification and translational validation of novel MAM-A epitopes has important implications for the ongoing clinical development of vaccine strategies targeting MAM-A. The novel MAM-A epitopes represent attractive targets for epitope-based vaccination strategies, and can also be used to monitor immune responses. Taken together these studies provide additional support for MAM-A as an important therapeutic target for the prevention and treatment of breast cancer.

A novel melanoma gene (MG50) encoding the interleukin 1 receptor antagonist and six epitopes recognized by human cytolytic T lymphocytes.

We identified a novel 8.1-kb human melanoma gene, MG50, derived from subtractive hybridization with a squamous lung carcinoma cell line, LU-1. 6.8 kb containing an open reading frame were sequenced, and the structure of the encoded 1496 amino acid protein was deduced. With HLA-A2.1-transduced Drosophila cells as antigen-presenting cells, we identified six epitopes restricted by HLA-A2.1 that elicited CTLs in vitro. Reactivity of the CTLs to melanoma cells containing MG50 indicated that the epitopes were displayed naturally. Significant cross-reactivity of CTLs immunized against a melanoma cell line that lacked HLA-A2.1 indicated that at least four of the epitopes were also recognized in a different HLA class I context, most likely HLA-A*6802. By quantitative reverse transcription, MG50 message was found in one of two skin melanoma cell lines, an ocular melanoma cell line, two of four metastatic skin melanomas, two of three mammary carcinomas, one of two colon carcinomas, and an ovarian carcinoma. Of six normal tissues, MG50 was found only in a specimen of normal skin and was absent from a congenital nevus. It is likely that MG50 is the gene for the interleukin 1 receptor antagonist because a reported sequence of cDNA from the latter had a sequence of 528 bases in the 3' region, a long contiguous base sequence, and 176 encoded amino acids identical with those of MG50. MG50 is one of the few melanoma-associated antigens that is not a differentiation antigen or a mutated protein. Because of its nature, it may prove to be important in the pathogenesis of the tumors in which it is found, as well as an immunogen and target for immunotherapy.

Dendritic cells transduced by multiply deleted HIV-1 vectors exhibit normal phenotypes and functions and elicit an HIV-specific cytotoxic T-lymphocyte response in vitro.

Dendritic cells (DCs) genetically modified to continually express and present antigens may be potent physiologic adjuvants for induction of prophylactic or therapeutic immunity. We have previously shown that an env and nef deleted HIV-1 vector (HIV-1 delta EN) pseudotyped with VSV-G transduced monocyte-derived macrophages as well as CD34(+) precursors of DCs. Here we extended these findings with HIV-1 delta EN to highly differentiated human DCs derived in culture from circulating monocytes (DCs). In addition, a new vector derived from HIV-1 delta EN but further deleted in its remaining accessory genes vif, vpr, and vpu (HIV-1 delta EN V(3)) was also tested. Both vectors efficiently transduced DCs. Transduction of DCs did not significantly alter their viability or their immunophenotype when compared with untransduced DCs. Furthermore, the phagocytic potential of immature DCs, as well as their ability to differentiate into mature DCs capable of stimulating T-cell proliferation, was not affected. Finally, DCs transduced by the HIV-1 delta EN vector were able to elicit a primary antiviral cytotoxic T-cell response in autologous CD8 T cells. These results suggest that HIV-1-based vectors expressing viral antigens may be useful for in vivo active immunization as well as ex vivo priming of cytotoxic T cells for adoptive T-cell therapy. (Blood. 2000;96:1327-1333)

Evaluation of the human choroidal melanoma rabbit model for studying microcirculation patterns with confocal ICG and histology.

The aim of this study was to develop consistently focal elevated choroidal masses of human choroidal melanoma in immunosuppressed rabbits and to correlate the visualization of prognostically significant microcirculation patterns from confocal indocyanine green angiography with histologic microcirculation patterns. A human choroidal melanoma cell line (OCM1) was implanted in the choroid of 40 rabbit eyes using three different techniques: transscleral choroidal injection of a cell suspension, injection of a cell suspension in a surgically induced cyclodialysis cleft, and implantation of solid tumor fragments in a surgically induced cyclodialysis cleft. The rabbits were immunosuppressed with daily injections of Cyclosporin A to prevent host versus graft reaction. The eyes were studied weekly with indirect ophthalmoscopy and fundus photography to monitor tumor growth and indocyanine green angiography using a confocal scanning laser ophthalmoscope to identify microcirculation patterns in vivo and correlate these findings with the histologic demonstration of tumor microcirculation patterns. A tumor mass was identified by indirect ophthalmoscopy in 16 of the 40 implanted rabbit eyes (40%). Each of these tumors was confirmed histologically to represent a focal elevated choroidal mass. All 16 elevated choroidal masses grow in eyes in which solid tumor fragments were implanted. In total, a melanoma was identified histologically in 28 of the implanted 40 eyes (70%). In addition to the 16 eyes where the melanoma appeared as a focal elevated choroidal mass, 4 eyes contained a focal elevated mass in the sclera and 8 eyes contained a flat choroidal tumor. Histologically, microcirculation patterns were identified only in the 16 eyes with focal elevated choroidal masses. Confocal indocyanine green angiography imaged microcirculation patterns in 13 of these 16 eyes (81%). The surgical implantation of small solid fragments of human choroidal melanoma in immunosuppressed rabbit eyes provides the best method to consistently obtain focal elevated choroidal masses. These focal elevated choroidal masses resemble booth the localization and the growth pattern of choroidal melanomas in humans. In addition, they also contain microcirculation patterns similar to those seen in humans that are detectable with confocal indocyanine green angiography. The use of indocyanine green angiography with this animal model may be especially useful in designing and evaluating anti-microcirculation treatments directed at uveal melanoma.

Anti-MUC1 class I restricted CTLs in metastatic breast cancer patients immunized with a synthetic MUC1 peptide.

Sixteen metastatic breast cancer patients were immunized with a low dose (5 micrograms) of a 16 amino acid MUC1 peptide (GVTSAPDTRPAPGSTA) conjugated to KLH (BP16-KLH) plus DETOX adjuvant and evaluated for antibody titers against MUC1 peptide and KLH and for cytotoxic lymphocyte (CTL) activity using class 1 HLA-matched MUC1-positive tumor targets. All patients generated strong anti-KLH IgG responses. Only 3 patients developed an anti-MUC1 IgG response, which was weak in magnitude. As it is controversial whether human cancer patients generate class-1-restricted CTL against MUC1, we examined anti-MUC1 CTL activity of PBLs following 4 immunizations with BP16-KLH. The generation of MUC1-specific CTLs required only a 6-day in vitro stimulation of patients' T-cells with synthetic MUC1-peptide-pulsed autologous APCs. The assay for CTL activity was a 4 hour 51Cr release from labeled adenocarcinoma target cells. Eleven of the 16 immunized patients were tested for CTL activity using class-1-matched adenocarcinoma target cell lines. Evidence for class-1-restricted killing of MUC1-expressing tumor cell lines was obtained in 7 of these 11 patients.

System A amino acid transport in cultured human tumor cells: implications for tumor imaging with PET.

The A system of amino acid transport is concentrative and thought to be a regulator of cell growth. The [11C]methyl alpha-aminoisobutyric acid (MeAIB) is prospectively an ideal tracer for transport measurements with PET, as it is not metabolized and concentrates in cells only via System A transport. We examined the factors governing [14C]MeAIB accumulation by cultured human erythroleukemic (K562) cells. Experiments were performed in growth medium and phosphate-buffered saline (PBS) +/- cycloheximide (an inhibitor of protein synthesis) on logarithmically growing cells, as well as cells that had reached a growth plateau. Both inward transport rate and net uptake of MeAIB were positively correlated with cell growth rate and showed a strong inverse relationship to amino acid supply. The observations are consistent with a body of evidence from animal tumor cells, and they suggest that the correlation between System A transport and tumor cell proliferation may be obscured in vivo by variations in amino acid supply. Thus, while [11C]MeAIB might be useful as a PET radiotracer of System A transport per se, this compound may be limited in its ability to provide measurements of tumor growth rate.

Establishment and characterization of primary and metastatic uveal melanoma cell lines.

We report on the establishment and characterization of 2 primary (EOM-3, EOM-29) and 3 metastatic uveal melanoma cell lines (OMM-1, OMM-2, OMM-3) and further cytogenetic characterization of a previously described primary uveal melanoma cell line (OCM-1). Only a few long-term growing primary uveal melanoma cell lines have as yet been established, while of metastatic uveal melanoma cell lines we have found no descriptions. The morphology of the in vitro cultured cells varied from spindle to epithelioid. The cell lines were characterized by immunocytochemistry, electron microscopy and cytogenetical analysis. The relative growth rate was determined by bromodeoxyuridine (BUdR) incorporation. The melanocytic origin of the cell lines was determined by positive staining with antibodies identifying melanoma-associated antigens. Melanosomes and pre-melanosomes were indeed observed by electron microscopy in all cell lines. The stem-cell karyotype was found to be normal in 3 cell lines (EOM-29, OMM-2, OMM-3) and abnormal in 3 others (EOM-3, OCM-1, OMM-1) showing a net loss of chromosome 6. The OCM-1 and the OMM-1 cell lines even demonstrated a large amount of structural chromosomal aberrations, the former being near-tetraploid and the latter triploid. The EOM-29 cell line, cultured from a ciliary body melanoma, did not show the previously described chromosome 3 and 8 abnormalities.

Evidence of sodium-dependent glucose transport in human erythroleukemia cells.

Sodium-dependent transport of D-glucose has been reported only in epithelial cells of small intestine and kidney, and well-differentiated tumors thereof. We observed a two-fold decrease (p < 0.05) in the intracellular distribution volume (Vic, defined as steady-state intracellular uptake divided by extracellular concentration) of the non-metabolized D-glucose analog 3-O-methylglucose (3-O-MG) when logarithmically growing K562 cells (an anaplastic human erythroleukemia) were incubated 3 h in choline-substituted, phosphate buffered saline (PBS) rather than Na+ PBS, each containing a glucose concentration ([Glu]) of 5.6 mM. Electromechanically measured cellular volume Vc differed < 10% between the different media. In Na+ PBS, Vic (3-O-MG) was approximately twice Vc and declined progressively when [Glu] was reduced to 2.8 and 0.1 mM. We conclude that, in a balanced salt medium containing glucose as the only energy source, K562 cells express a concentrative mechanism having characteristics consistent with Na(+)-dependent transport of glucose.

Non-fastidious, melanoma-specific CD8+ cytotoxic T lymphocytes from choroidal melanoma patients.

To characterize the anti-melanoma reactivity of CD8+ cytotoxic T lymphocytes (CTL) from choroidal melanoma patients, CTL clones were isolated from the peripheral blood of three patients after mixed lymphocyte/tumor cell culture (MLTC). Clones were derived from lymphocytes stimulated by allogeneic (OCM-1, A24, A28) or autologous (OCM-3, A1, A30) melanoma cells. Their reactivity against a panel of HLA-typed melanoma and nonmelanoma cells was assessed, to determine whether a single CTL clone could recognize and lyse a variety of allogeneic melanoma cell lines. While proportionately more clones derived from autologous MLTC were melanoma-specific than allogeneic MLTC (42% versus 14%), melanoma-specific CTL were recovered from both. Notably, a novel melanoma specificity was identified. These CTL clones were termed non-fastidious because they were capable of lysing melanoma cells with which they had no HLA class I alleles in common. Nonetheless, lysis was mediated by the HLA class I molecule. Since lysis was specific for melanoma cells, these CTL appeared to recognize a shared melanoma peptide(s). Because of their prevalence, we propose that non-fastidious CTL are integral to human anti-melanoma T cell immunity. This reinforces clinical findings that allogeneic melanomas can substitute for autologous tumors in active specific immunotherapy. By circumventing the need for autologous melanoma, it is possible to treat patients after removal of the primary choroidal melanoma in an attempt to prevent metastasis.

Sustained regression of a primary choroidal melanoma under the influence of a therapeutic melanoma vaccine.

PURPOSE: To determine whether active specific immunotherapy with lysates of cutaneous melanoma cells, administered with immunologic adjuvant DETOX (Ribi ImmunoChem Research, Inc, Hamilton, MT), is effective in shrinking a primary choroidal melanoma, in an elderly patient already blind in the nontumorous eye. An 81-year-old man was referred with a primary choroidal melanoma of the left eye, with virtual blindness of the right eye due to macular degeneration. He was begun on active specific immunotherapy with an experimental melanoma vaccine (melanoma theraccine) and DETOX on weeks 1, 2, 3, 4, and 6, respected after a hiatus of 2 weeks. After a response was noted, monthly injections were given. RESULTS: The patient had a significant shrinkage of his choroidal melanoma from a height of 4.2 mm to 2.4 mm within 2 months. This was sustained by continual treatment for 21 months until September 1991. After the patient failed to return for 9 months while recuperating from a stroke, the lesion regrew to a height of 3.7 mm and developed an additional lobe. On resumption of monthly treatments, the lesion shrank to 3.4 mm within 3 months, lost the additional lobe, and has since remained stable. No metastases have been found over a period of nearly 4 years on quarterly computed tomographic (CT) scanning of the chest and abdomen, and magnetic resonance imaging of the head. CONCLUSION: Active specific immunotherapy with cutaneous melanoma lysates has caused a clinically useful protracted regression of a primary choroidal melanoma in an elderly patient in whom surgery and radiation therapy were contraindicated. This may represent the first case of a primary choroidal melanoma, and perhaps the only primary tumor, successfully treated with systemic immunotherapy alone. A formal trial of active specific immunotherapy for primary choroidal melanoma in selected patients may be warranted.

Differential S100 beta expression in choroidal and skin melanomas: quantitation by the polymerase chain reaction.

PURPOSE: S100 beta, a member of a calcium-binding protein family (S100s), is an important clinical marker for skin melanoma. In contrast, uveal melanomas appeared to express S100 beta protein less frequently and to a lesser degree. This study was performed to verify and extend this finding to the mRNA level. METHODS: A quantitative polymerase chain reaction (PCR)-based method was used. A ratio, comparing the S100 beta PCR fragment to that of beta-actin (an internal reference gene), was generated to compare S100 beta mRNA expression among samples. RESULTS: The ratios for skin melanomas (1.2 to 3.9; three tissues and two cell lines) were significantly higher than that for choroidal melanomas (0.1 to 0.63; seven of eight primary tumors and four of four cell lines). Only one choroidal melanoma biopsy had a ratio greater than 1. The PCR products from choroidal melanoma were identical in size and sequence to the S100 beta, as determined by gel electrophoresis and RNA conformational polymorphism. Because the ratios were also low in choroidal melanoma cell lines, the S100 beta phenotype appears to be genetically stable. CONCLUSION: S100 beta is differentially expressed at the RNA and protein levels by skin and choroidal melanomas, which are derived from distinct populations of melanocytes. However, choroidal melanomas expressing little or no S100 beta were significantly stained by antiserum specific for the S100 protein family. Taken together, these data suggest that choroidal melanocytes express another, perhaps even novel, S100 protein(s).

Detection of melanoma-reactive CD4+ HLA-class I-restricted cytotoxic T cell clones with long-term assay and pretreatment of targets with interferon-gamma.

Twenty-five CD4+ cytotoxic T lymphocyte (CTL) clones were obtained from the peripheral blood or tumor tissues of melanoma patients undergoing active specific immunotherapy. Melanoma-reactive T cells were cloned by limiting dilution using either autologous or allogeneic melanoma cells to stimulate their proliferation. Sixteen of the clones reacted against autologous melanoma cells but not against the autologous lymphoblastoid cell line, which we defined as "melanoma-specific." Optimal demonstration of the lytic activity of CD4+ CTL required a 16-h incubation period and an effector:target cell ratio of 40:1. In addition, a 24-h pre-incubation of the target melanoma cells with 100 U interferon (IFN) gamma consistently augmented lysis by these CD4+ CTL, increasing it from a mean level of 20% to one of 52%. Lysis by 8 of the 11 melanoma-reactive CD4+ T cell clones was exclusively HLA-class-I-restricted, as judged by blocking with monoclonal antibodies (mAb). Five of these HLA class-I-restricted clones were reactive only with the autologous melanoma cells, while the other 3 clones were also reactive with allogeneic melanoma cells. In all cases, the T cells and melanoma targets shared at least one HLA class I allele, usually HLA-A2, HLA-C3 or HLA-B62. Interestingly, lysis by 2 of the 11 clones was inhibited by both anti-HLA-class-I or -HLA-class-II mAb, while lysis by 1 other clone was inhibited by neither. HLA class I molecules and several accessory molecules were maximally expressed by the melanoma target cells, both in terms of distribution and copy number before IFN gamma treatment. Thus, IFN gamma may have acted by increasing the expression of melanoma-associated epitopes as presented by HLA class I (or HLA class II) molecules. A proportion of human CD4+ CTL appeared to recognize melanoma-associated epitopes presented by the HLA class I molecule, although their lytic potency may be less than that of their CD8+ counterparts.

Active specific immunotherapy of melanoma with allogeneic cell lysates. Rationale, results, and possible mechanisms of action.

Since 1985, we have conducted clinical trials with a therapeutic melanoma vaccine (melanoma theraccine). Mechanical lysates of two melanoma cell lines chosen for their complementary characteristics were combined with the adjuvant DETOX and injected subcutaneously on weeks 1, 2, 3, 4, and 6 for one or two courses and then monthly in patients with objective clinical responses. Of 106 patients, 20 had objective clinical regression of tumor masses, 5 with complete responses. The median duration of response was 21 months. Twelve patients lived at least 2 years, with a median survival of nearly 3 years. Two of them are free of disease for > 2 and > 6 years, respectively. However, it was not necessary to achieve complete remissions to cause an increase in survival, and most of the long-surviving patients have one or more (stable) residual nodules. The pace of the disease process was clearly slowed in those individuals. A rise in the level of cytotoxic T-lymphocyte precursors in the blood (pTC) correlated with clinical response. Only those patients who had a rise in pTC had a remission. In addition to "classical" CD8+ Tc, CD4+ Tc were cloned from the blood of immunized patients. Melanoma-specific Tc of both types that killed autologous melanoma but not matched lymphoblastoid cells were detected. Allogeneic melanoma cell lines were also killed, with mainly HLA-A2/28 and HLA-B12/44/45 degenerate restriction. CD4+ Tc were restricted by HLA Class I antigens, as judged by their killing of HLA Class II-negative melanomas and blocking by anti-class I antibodies. Other CD4+ clones were blocked by both anti-HLA Class I or anti-Class II MHC monoclonal antibodies, and only two were blocked only by anti-HLA Class II. Immunohistory revealed CD4+ and CD8+ T cells in lesions under rejection, but the predominant cells were macrophages, suggesting delayed-type hypersensitivity as a possible mechanism. Clinical responses were found most often in patients with HLA-A2/28, -B12/44/45, and -C3, particularly when two or more of those alleles were present. This may have been due either to (1) similarity of MHC antigens between one of the immunizing melanomas and the patient's melanoma or (2) the intrinsic importance of these MHC molecules in presenting melanoma-associated antigens to Tc in vivo. IFN-alpha 2 b salvaged 8 of 18 patients who failed with the theraccine, regardless of MHC phenotype, perhaps through upregulation of MHC and tumor epitopes on the autochthonous tumor.

Clonal analysis of in vivo activated CD8+ cytotoxic T lymphocytes from a melanoma patient responsive to active specific immunotherapy.

To study in vivo activated cytolytic T cells, CD8+ T cells clones were isolated from a melanoma patient (HLA A2, A11) treated with active specific immunotherapy for 5 years. CD8+ T lymphocytes, purified by fluorescence-activated cell sorting, were cloned directly from the peripheral blood without antigen-presenting cells in the presence of irradiated autologous melanoma cells and recombinant interleukin-2 (IL-2) and IL-4. These conditions were inhibitory to de novo in vitro immunization. Of the 28 cytolytic CD8+ T cell clones, 21 lysed the autologous melanoma cell line (M7) but not the autologous lymphoblastoid cell line (LCL-7) nor the two melanoma cell line, M1 (HLA A28) and M2 (HLA A28, A31), used to immunize the patient. The remaining 7 clones were also melanoma-specific, although their reactivities were broader, lysing several melanoma cell lines but not HLA-matched lymphoblastoid cells. Eight clones from the first group, ostensibly self-MHC-restricted, were expanded for further analysis. All expressed cluster determinants characteristic of mature, activated T cells, but not those of thymocytes, naive T cells, B cells or natural killer (NK) cells. They also expressed CD13, a myeloid marker. Of the 8 clones, 3 expressed both CD4 and CD8, but dual expression was not correlated with specificity of lysis. Two CD8+ and 2 CD4+ CD8+ clones were specific for the autologous melanoma cells, the other 4 were also reactive against other HLA-A2-positive melanomas. Cytotoxicity for both singly and doubly positive clones was restricted by HLA class I but not class II antigens. Analysis of the RNA expression of the T cell receptor (TCR) V alpha and V beta gene segments revealed heterogeneous usage by the A2-restricted clones and, perhaps, also by the broadly melanoma-specific clones. Apparent TCR-restricted usage was noted for the self-MHC-restricted clones; 2 of the 4 expressed the V alpha 17/V beta 7 dimer. Since the T cell clones were derived from separate precursors of circulating cytotoxic T lymphocytes (CTL), the V alpha 17/V beta 7 TCR was well represented in the peripheral blood lymphocytes of this patient. In summary, we show that melanoma cells presented their own antigens to stimulate the proliferation of melanoma-reactive CD8+ CTL. CTL with a range of melanoma specificities and different TCR alpha beta dimers were encountered in this patient, perhaps as a result of hyperimmunization.(ABSTRACT TRUNCATED AT 400 WORDS)

Prevention of metastasis of intraocular melanoma in mice treated with difluoromethylornithine.

The antimetastatic potential of a novel chemotherapeutic agent, alpha-difluoromethylornithine (DFMO), was evaluated in a murine model of intraocular melanoma. In vivo studies demonstrated that DFMO retarded the growth and spontaneous metastasis of murine intraocular melanomas. Further studies indicated that oral DFMO also exercised antimetastatic effects against the blood-borne stage of melanoma metastases. In vitro studies revealed that DFMO exerted impressive antiproliferative effects on three murine melanoma cell lines, four human cutaneous melanoma cell lines, one human uveal melanoma cell line, and one conjunctival melanoma cell line. DFMO inhibited in vitro DNA synthesis in human cutaneous melanoma cell lines by 84%-98% and that in two human ocular melanoma cell cultures by 62% and 86%, respectively. DFMO possesses several characteristics that render it an attractive chemotherapeutic agent for potential use in the management of uveal melanoma. These include its antiproliferative effect against a wide range of murine and human melanomas, its extremely low toxicity, and its ease of administration.

Evaluation of human monoclonal antibody (2-139-1) in cutaneous melanocytic neoplasms in fixed tissue sections.

Despite the growing list of xenogeneic monoclonal antibodies (MAb) that recognize malignant melanoma-associated antigens (MAA) in formalin-fixed, paraffin-embedded tissue, none has been able to detect epitopes found in malignant melanomas and not in melanocytic nevi. A human MAb, 2-139-1, that showed promise in this regard was evaluated against 85 melanocytic neoplasms, including malignant melanoma and histological simulators, particularly Spitz's nevus. MAb 2-139-1 stained 18 (53%) of 34 melanomas, eight (57%) of 14 dysplastic nevi, six (38%) of 16 Spitz's nevi, and three (14%) of 21 banal nevi, which included three small congenital nevi. We observed a significant increasing trend in reactivity (% positive cells x intensity) associated with the potential for malignancy (p for linear trend = 0.002). We conclude that human MAb 2-139-1 is applicable to the study of melanocytic neoplasms in routinely processed tissue. Although the ability of this MAb to separate benign from malignant cells is not absolute, our results suggest that the expression of the 2-139-1 epitope may be an early event in melanocytic tumor progression.

Novel gene sequences expressed by human melanoma cells identified by molecular subtraction.

Despite the variety of approaches used, only a limited number of tumor-associated antigens have been described for each histological type of tumor. In this report, we present a new strategy involving molecular subtraction to identify novel melanoma-associated gene sequences. Toward this end, 156 complementary DNA clones were isolated with a subtracted melanoma complementary DNA probe (melanoma minus lung carcinoma) after screening 2 x 10(4) independent recombinants of a melanoma expression library by in situ plaque hybridization. These clones were then polymerase chain reaction amplified, screened for duplication, and categorized into 53 discrete genes. By applying poissonian distribution to the numbers of duplicate isolates, we found most of the genes to be rare messages, present at less than 1 copy/200 molecules of mRNA in a typical somatic cell. Messages specific for a type of tissue are usually expressed in this range. The expression of the 53 genes was further studied in human tumor cell lines and normal tissues. Partial sequence data obtained for 20 complementary DNA clones revealed 8 novel human genes. The mRNA transcripts for 5 of the novel genes were identified by Northern blot analysis. Thus, molecular subtraction appears to be applicable for the identification of novel tumor-associated sequences. Some of the potential advantages and limitations of this technology are discussed, including its application to the molecular characterization of immunogenic melanoma-associated antigens.

Lymphocytes cytotoxic to uveal and skin melanoma cells from peripheral blood of ocular melanoma patients.

To study antitumor immunity in patients with choroidal melanoma, T cells were generated from the peripheral blood of choroidal melanoma patients by mixed lymphocyte/tumor cell culture (MLTC). Because autologous tumors are generally unavailable, an allogeneic choroidal melanoma cell line, OCM-1, was used as the specific stimulus. Lymphocyte cultures from 27 patients were characterized by cell-surface phenotypes, patterns of reactivity towards cells of the melanocytic origin and T-cell-receptor gene usage. Antimelanoma reactivity was found in cell-sorter-purified CD4+ and CD8+ T cell subsets. To analyze this reactivity, sorter-purified CD4+ and CD8+ cells from a MLTC were cloned by limiting dilution in the presence of exogenous interleukin-2 and interleukin-4 as well as irradiated OCM-1. Under these conditions, CD4+ T cells did not proliferate, perhaps because of the absence of antigen-presenting cells. However, CD8+ grew vigorously and 29 cytolytic CD8+ T cell clones were isolated. On the basis of their pattern of lysis of OCM-1, a skin melanoma cell line M-7 and its autologous lymphoblastoid cell line LCL-7, the clones were categorized into three groups. Group 1, representing 52% of the clones, lysed all three target cells, and are alloreactive. However, since OCM-1 and M-7 did not share class I antigens, these clones recognized cross-reactive epitope(s) of the histocompatibility locus antigen (HLA) molecule. Group 2, constituting 28% of the clones, lysed both the ocular and skin melanoma cell lines but not LCL-7, and were apparently melanoma-specific. Unlike classical HLA-restricted cytolytic T lymphocytes, these T cells might mediate the lysis of melanoma cells via other ligands or a more degenerate type of HLA restriction. For the latter, the HLA-A2 and -A28 alleles would have to act interchangeably as the restriction element for shared melanoma-associated antigen(s). Group 3, representing only 10% of the T cell clones, was cytotoxic only to OCM-1, but not to M-7 or LCL-7. These clones may recognize antigens unique to ocular melanoma cells. Our data suggest that choroidal melanoma patients can recognize melanoma-associated antigens common to both ocular and cutaneous melanoma cells, and presumbly their autologous tumor. Thus, choroidal melanoma, like its skin counterpart, may be responsive to immunotherapeutic regimens such as active specific or adoptive cellular immunotherapy.

S100 immunophenotypes of uveal melanomas.

To determine whether ocular melanomas are immunophenotypically identical to cutaneous melanomas, 34 primary and metastatic choroidal melanomas representing all major histotypes defined by the Callender's classification, plus one melanoma of the iris and one conjunctival melanoma, were subjected to a panel of immunostains designed to distinguish anaplastic biopsies of cutaneous melanomas from carcinomas and lymphomas. All ocular melanomas were found to express the intermediate filament vimentin but not keratin, and all but 2 were melanotic by immunostaining. Thirty-three of 34 (97%) choroidal melanomas were strongly stained with a rabbit polyclonal antibody (P-S100) developed against the S100 protein family. In contrast, none of 14 spindle cell type primary lesions was stained with a monoclonal antibody (MAB-079) specific for both S100 alpha and S100 beta, the best-characterized S100 polypeptides. Furthermore, only 2 of 5 epithelioid and 3 of 10 mixed-cell-type melanomas were weakly reactive. Overall, 14.7% (5 of 29) were stained. In comparison, MAB079 stained 85% of all cutaneous melanomas. Five metastases of choroidal melanomas (spindle B, epithelioid, and mixed cell types) from different organ sites also were stained by P-S100 but not by MAB079. These findings were corroborated by immunostaining with another monoclonal antibody (MAB4D4) specific for S100 beta. Differential staining by the polyclonal but not the monoclonal antibodies suggests the possible presence of a variant S100 polypeptide(s) in choroidal melanomas. Since S100 alpha, S100 beta, and related proteins appear to be physiologically important, additional studies of these S100 proteins may shed light on the etiology or pathology of choroidal melanomas.