First Comparative Delineation of the T Cell Receptor Repertoire in Primary and Multiple Subsequent/Coexisting Metastatic Melanoma Sites
1998; Elsevier BV; Volume: 111; Issue: 6 Linguagem: Inglês
10.1046/j.1523-1747.1998.00450.x
ISSN1523-1747
AutoresRobert Strohal, Christine Brna, Georg Stingl, Ulrike Mossbacher, Gottfried Fischer, Hubert Pehamberger,
Tópico(s)Cancer Immunotherapy and Biomarkers
ResumoAt present, very little is known about the types and heterogeneity of T cell responses and immunodominant epitopes of melanoma-associated antigens at coexisting sites of primary melanoma and metastatic lesions. To address this issue, we compared the T cell receptor (TCR) gene usage, complemetary-determining region 3 diversity, and melanoma-associated antigens expression patterns of primary and metastatic melanoma specimens from three patients with partially homologous HLA class-1 types. Results obtained showed an overall predominance of a very limited number of TCRV regions with AV13 and BV14 being most frequently overexpressed. Sequencing of the dominating TCR transcripts confirmed the restricted usage of certain TCR specificities and, in two of the three patients, identified several identical TCR clonotypes at more than one metastatic site. Nevertheless, we failed to detect TCR transcripts that were common to all tumor deposits in a given patient and, within the majority of coexisting metastases, tumor-infiltrating lymphocytes preferentially used individual site-specifically expanded TCR β-chain VJ segment combinations. This occurrence of individual responses simultaneously executed at and influenced in their specificity by the different sites of tumor growth, has important implications for the type of strategies chosen in the development of efficacious vaccines for patients with metastatic melanoma. At present, very little is known about the types and heterogeneity of T cell responses and immunodominant epitopes of melanoma-associated antigens at coexisting sites of primary melanoma and metastatic lesions. To address this issue, we compared the T cell receptor (TCR) gene usage, complemetary-determining region 3 diversity, and melanoma-associated antigens expression patterns of primary and metastatic melanoma specimens from three patients with partially homologous HLA class-1 types. Results obtained showed an overall predominance of a very limited number of TCRV regions with AV13 and BV14 being most frequently overexpressed. Sequencing of the dominating TCR transcripts confirmed the restricted usage of certain TCR specificities and, in two of the three patients, identified several identical TCR clonotypes at more than one metastatic site. Nevertheless, we failed to detect TCR transcripts that were common to all tumor deposits in a given patient and, within the majority of coexisting metastases, tumor-infiltrating lymphocytes preferentially used individual site-specifically expanded TCR β-chain VJ segment combinations. This occurrence of individual responses simultaneously executed at and influenced in their specificity by the different sites of tumor growth, has important implications for the type of strategies chosen in the development of efficacious vaccines for patients with metastatic melanoma. complemetary-determining region cytotoxic T-lymphocytes melanoma-associated antigens nodular melanoma tumor-infiltrating lymphocytes Several authors have shown that CD3+ T cell receptor (TCR) α/β-bearing T cells that represent the vast majority of tumor-infiltrating lymphocytes (TIL) within primary and metastatic melanoma, can be propagatedin vitro by growth factors such as IL-2 and then exhibit specific HLA class-1-restricted lytic activity for autologous tumor targets (Rosenberg et al., 1986Rosenberg S.A. Spiess P. Lafreniere R.A. new approach to the adoptive immunotherapy of cancer with tumor-infiltrating lymphocytes.Science. 1986; 233: 1318-1321Crossref PubMed Scopus (1432) Google Scholar;Muul et al., 1987Muul L.M. Spiess P.J. Director E.P. Rosenberg S.A. Identification of specific cytolytic immune responses against autologous tumor in humans bearing malignant melanoma.J Immunol. 1987; 138: 989-995PubMed Google Scholar;Itoh et al., 1988Itoh K. Platsoucas C.D. Balch C.M. Autologous tumor-specific T lymphocytes in the infiltrate of human metastatic melanomas. Activation by interleukin 2 and autologous tumor cells, and involvement of the T cell receptor.J Exp Med. 1988; 168: 1419-1441Crossref PubMed Scopus (330) Google Scholar). By using tumor-specific cytotoxic T lymphocytes (CTL) derived from either autologous peripheral blood lymphocytes (PBL) or TIL, a variety of melanoma-associated antigens (MAA), together with their HLA-binding motifs, could be identified (Van den Eynde and Brichard, 1995Van den Eynde B. Brichard V.G. New tumor antigens recognized by T cells.Curr Opin Immunol. 1995; 7: 674-681Crossref PubMed Scopus (66) Google Scholar). These include tumor-specific antigens (Mage, Bage, Gage) and melanocyte-specific differentiation antigens (MART-1/Melan-A, tyrosinase, Pmel17/gp100, gp75). Sequencing of TCR moieties in CTL clones recognizing a given MAA in the context of the appropriate HLA class-I molecules, demonstrated that differences in the fine specificities of these CTL clones correlate with the selective usage of particular TCRVJ genes (Sensi et al., 1993Sensi M. Salvi S. Castelli C. et al.T cell receptor (TCR) structure of autologous melanoma-reactive cytotoxic T lymphocyte (CTL) clones: tumor- infiltrating lymphocytes overexpress in vivo the TCR β;-chain sequence used by an HLA-A2-restricted and melanocyte-lineage-specific CTL clone.J Exp Med. 1993; 178: 1231-1246Crossref PubMed Scopus (106) Google Scholar,Sensi et al., 1995Sensi M. Traversari C. Radrizzani M. et al.Cytotoxic T-lymphocyte clones from different patients display limited T-cell-receptor variable-region gene usage in HLA-A2-restricted recognition of the melanoma antigen Melan-A/MART-1.Proc Natl Acad Sci USA. 1995; 92: 5674-5678Crossref PubMed Scopus (111) Google Scholar;Van der Bruggen et al., 1994Van der Bruggen P. Szikora J.P. Boel P. Wildmann C. Somville M. Sensi M. Boon T. Autologous cytolytic T lymphocytes recognize a MAGE-1 nonapeptide on melanomas expressing HLA-Cw*1601.Eur J Immunol. 1994; 24: 2134-2140Crossref PubMed Scopus (212) Google Scholar;Boel et al., 1995Boel P. Wildmann C. Renauld J.C. Coulie P. Boon T. Bage van der Bruggen P. a new gene encoding an antigen recognized on human melanomas by cytolytic T lymphocytes.Immunity. 1995; 2: 167-175Abstract Full Text PDF PubMed Scopus (545) Google Scholar;Maeurer et al., 1995Maeurer M.J. Martin D.M. Storkus W.J. Lotze MTTCR. usage in CTLs recognizing melanoma/melanocyte antigens.Immunol Today. 1995; 16: 603-604Abstract Full Text PDF PubMed Scopus (15) Google Scholar;Zarour et al., 1996Zarour H. de Smet C. Lehmann F. et al.The majority of autologous cytolytic T lymphocyte clones derived from peripheral blood lymphocytes of a melanoma patient recognize an antigenic peptide derived from Pmel17/gp100.J Invest Dermatol. 1996; 107: 63-67Crossref PubMed Scopus (55) Google Scholar). Semiquantitative polymerase chain reaction (PCR) analyses of TCR transcripts in TIL of human melanoma specimens, showed that their repertoire is often skewed towards 1–3 predominantly expressed TCRAV or TCRBV regions indicative of clonal/oligoclonal expansion and/or recruitment of tumor-reactive T cell populations at lesional sites (Sensi et al., 1995Sensi M. Traversari C. Radrizzani M. et al.Cytotoxic T-lymphocyte clones from different patients display limited T-cell-receptor variable-region gene usage in HLA-A2-restricted recognition of the melanoma antigen Melan-A/MART-1.Proc Natl Acad Sci USA. 1995; 92: 5674-5678Crossref PubMed Scopus (111) Google Scholar;Strohal et al., 1994aStrohal R. Paucz L. Pehamberger H. Stingl G. T-cell receptor repertoire of lymphocytes infiltrating cutaneous melanoma is predominated by Vα specificities present in T-cells of normal human skin.Cancer Res. 1994 a; 54: 4734-4739PubMed Google Scholar). Because the vast majority of published data were generated from solitary tumor lesions only, the question still remains whether the same TCR transcripts dominating the TCR repertoire in single melanoma specimens (primary melanoma, metastases) are also preferentially used within subsequent and/or coexisting melanoma metastases at different organ sites. In order to address this issue, we comparatively delineated TCRAV/BV gene usage, complemetary-determining region 3 (CDR3) diversity, and MAA expression patterns (MAGE-1, MAGE-2, MAGE-3, MAGE-4, MAGE-6, BAGE, GAGE, MART-1/Melan-A, gp100) of one primary melanoma and multiple coexisting metastases from three different HLA-typed patients. By doing so, we attempted to gather information not only about the diversity and nature of the anti-melanoma immune response but also about the type, structure, and heterogeneity of the immunodominant epitopesin situ. We obtained 12 different autopsy specimens of tumor tissue from three female patients (1622, 1464, 1214) with metastatic melanoma. These included (i) the primary melanoma of patient 1622 and coexisting lymph node, liver, lung, brain, and skin metastases; (ii) liver and lung metastases of patient 1464; and (iii) lung, spleen, kidney, and thyroid gland metastases of patient 1214. In addition to surgery of the primary tumor, two patients (1464, 1214) had received one or more combination therapies with dacarbazine and fotemustine. Patient 1622 underwent palliative surgery only. A representative portion of each specimen was processed for histopathology that classified the primary tumor of patient 1622 as nodular melanoma (NM) and confirmed the malignant melanocytic nature of the remaining metastatic lesions. The rest of each specimen was snap-frozen in liquid nitrogen and stored at –70°C until further use. For control purposes, we isolated mononuclear cells from the heparinized blood (PBMC) of three healthy female donors by Ficoll-Hypaque (Pharmacia, Uppsala, Sweden) density gradient centrifugation. Cells recovered from the interface were resuspended in guanidinium isothiocyanate at a density of 5 × 106 cells per ml and stored at –70°C. HLA class-I typing of each patient was performed from normal and affected tissues (patient 1622, liver and skin metastases; patient 1214, normal liver; patient 1464, liver metastasis) using a cDNA-based multistep PCR and sequencing protocol. The procedure, including all PCR and sequencing primers, has been described previously (Faé et al., 1996Faé I. Petrasek M. Broer E. Mayr W.R. Fischer G.F. Nucleotide sequencing analysis of HLA-C alleles.in: Charron D. Genetic Diversity of HLA. Functional and Medical Implications, Paris, EDK,1996: 1-3Google Scholar). Total RNA was extracted from the frozen cell and tissue samples by the guanidinium isothiocyanate-cesium chloride procedure (Glisin et al., 1974Glisin V. Crkvenjakov R. Byus C. Ribonucleic acid isolated by caesium chlorid centrifugation.Biochemistry. 1974; 13: 2633-2637Crossref PubMed Scopus (1530) Google Scholar). First-strand cDNA was synthesized from 5 μg RNA in a final volume of 40 μl containing 8 μl 5 × reverse MoMLV reverse transcriptase reaction buffer (GIBCO/BRL, Gaithersburg, MD), 10 μl dNTP at 10 mM each, 2 μl of a 200 μM solution of oligo dT(15), 1 μl RNAsin at 40 U per μl (Promega, Madison, WI), and 5 μl of MoMLV reverse transcriptase at 200 U per μl (GIBCO/BRL). After 10 min at room temperature, the mixture was incubated at 42°C for 1 h, heated to 95°C for 5 min, chilled on ice, and stored at –20°C until further use. To determine the MAA (MAGE-1, MAGE-2, MAGE-3, MAGE-4, MAGE-6, BAGE, GAGE, MART-1/Melan-A, gp100) and TCRAV and TCRBV expression patterns, cDNA from the various samples were subjected to PCR amplification. For the MAA-specific PCR, one of 25 of the cDNA product was incubated with 10 mM Tris-HCl, pH 8.3, 50 mM KCl, 1.5 mM MgCl2, all four dNTP (each at 200 μM), 2.5 U AmpliTaq DNA polymerase (Perkin-Elmer Cetus, Emeryville, CA), and 0.4 μM of each sense and anti-sense primer in a final volume of 50 μl. Oligonucleotide primers were either devised from known cDNA sequences (Adema et al., 1994Adema G.J. de Boer A.J. Vogel A.M. Loenen W.A. Figdor C.G. Molecular characterization of the melanocyte lineage-specific antigen gp100.J Biol Chem. 1994; 269: 20126-20133Abstract Full Text PDF PubMed Google Scholar;Kawakami et al., 1994Kawakami Y. Eliyahu S. Delgado C.H. et al.Cloning of the gene coding for a shared human melanoma antigen recognized by autologous T cells infiltrating into tumors.Proc Natl Acad Sci USA. 1994; 91: 3515-3519Crossref PubMed Scopus (1007) Google Scholar) (gp100, 5′-TTACTGACCAGGTGCCTTTCT-3′ sense, 5′-TGTAGCCTCTGAGTTGACAT-3′ anti-sense;MART-1/Melan-A, 5′-CACTTCATCTATGGTTACCCC-3′ sense, 5′-TGAATAAGGTGGTGGTGACTG-3′ anti-sense) or taken from the literature (Brasseur et al., 1992Brasseur F. Marchand M. Vanwijck R. Herin M. Lethe B. Chomez P. Boon T. Human gene MAGE-1, which codes for a tumor-rejection antigen, is expressed by some breast tumors.Int J Cancer. 1992; 52: 839-841Crossref PubMed Scopus (163) Google Scholar;De Plaen et al., 1994De Plaen E. Arden K. Traversari C. et al.Structure, chromosomal localization, and expression of 12 genes of the MAGE family.Immunogenetics. 1994; 40: 360-369Crossref PubMed Scopus (570) Google Scholar;De Smet et al., 1994De Smet C. Lurquin C. van der Bruggen P. de Plaen E. Brasseur F. Boon T. Sequence and expression pattern of the human MAGE2 gene.Immunogenetics. 1994; 39: 121-129Crossref PubMed Scopus (133) Google Scholar;Boel et al., 1995Boel P. Wildmann C. Renauld J.C. Coulie P. Boon T. Bage van der Bruggen P. a new gene encoding an antigen recognized on human melanomas by cytolytic T lymphocytes.Immunity. 1995; 2: 167-175Abstract Full Text PDF PubMed Scopus (545) Google Scholar;Van den Eynde et al., 1995Van den Eynde B. Peeters O. de Backer O. Gaugler B. Lucas S. Boon T. A new family of genes coding for an antigen recognized by autologous cytolytic T lymphocytes on a human melanoma.J Exp Med. 1995; 182: 689-698Crossref PubMed Scopus (536) Google Scholar; MAGE-1, MAGE-2, MAGE-3, MAGE-4, MAGE-6, BAGE, GAGE – VDE18/VDE24), commercially synthesized and high performance liquid chromatography-purified (Pharmacia). Amplification was performed on a DNA thermal cycler (model 480, Perkin-Elmer). Before the first cycle, the reaction mixture was heated at 94°C for 4 min and after the last cycle the incubation was extended for another 15 min at 72°C. The PCR cycle profile was as follows: denaturation, 1 min at 94°C; annealing of primers, 2 min at 56°C (GAGE), 60°C (gp100, MART-1/Melan-A), 62°C (BAGE), 67°C (MAGE-2), 68°C (MAGE-4, MAGE-6), 72°C (MAGE-3), or 1 min at 72°C (MAGE-1); primer extension, 2 min at 72°C (gp100, MART-1, MAGE-1, MAGE-2, MAGE-3, MAGE-4, MAGE-6, GAGE) or 73°C (BAGE). After 30 cycles, 10 μl of the reaction mixture was fractionated on a 1.2% agarose gel and stained with ethidium bromide (4 μg per ml). To study the TCRV repertoire, a semiquantitative PCR was performed using sense primers complementary to each of the 22 TCRAV and 24 BV family sequences together with one TCRAC- and TCRBC-specific anti-sense primer, respectively. The TCRBC primer was constructed to detect and amplify both TCR β-chain constant regions. All primer sets were purchased from Clontech Laboratories (TCR Typing Amplimer Kit, Palo Alto, CA) and the PCR procedure was carried out according to protocols provided by the manufacturer. Briefly, a final 50 μl reaction mixture containing 2 μl cDNA, 10 mM Tris-HCl, pH 8.3, 50 mM KCl, 1.5 mM MgCl2, 200 μM dNTP, 2.5 U AmpliTaq DNA polymerase, and 0.5 μM of the respective TCR primers was subjected to 30 amplification cycles, each consisting of 1 min at 95°C, 1 min at 55°C, and 1 min at 72°C followed by a final 15 min extension at 72°C. To control for the occurrence of amplification artefacts by contaminating DNA, cDNA from EBV-transformed B cells or samples without cDNA were used as negative controls and checked by hybridization with the appropriate TCR or MAA probes. These experiments yielded negative results. Moreover, a 20-cycle PCR assay with primers specific for β-actin (β-Actin Control RT-PCR Amplimer Set, Clontech Laboratories) was initially carried out to ensure that each specimen contained intact RNA. A 10 μl aliquot of each PCR product was separated by electrophoresis on 1.2% agarose gels at 30 V overnight and blotted onto nylon membranes (Nytran-NY13N, Schleicher und Schuell, Dassel, Germany). Blotting, radioactive 3′ end-labeling of the oligonucleotide probes with (α32P) dATP (Amersham, Bucks, U.K.) and terminal deoxynucleotidyl transferase (Boehringer, Vienna, Austria), prehybridization, and hybridization were carried out according to standard protocols (Southern, 1975Southern E.M. Detection of specific sequences among DNA fragments separated by gel electrophoresis.J Mol Biol. 1975; 98: 503-517Crossref PubMed Scopus (21151) Google Scholar;Elbe et al., 1992Elbe A. Kilgus O. Strohal R. Payer E. Schreiber S. Stingl G. Fetal skin – a site of DETC development.J Immunol. 1992; 149: 1694-1701PubMed Google Scholar). Oligonucleotides used for hybridization either matched with the central part of the respective MAA genes (gp100, 5′-CTGACCAGGTGTAGTACCCACAACTTCTGT-3′;MART-1/Melan-A, 5′-CCGATGATCAAACCCTTCTTGTGGGCATCT-3′;MAGE-1, 5′-AGTGCAGACTCCTCTGCTCAAGAGACATGA-3′;MAGE-2,MAGE-3, andMAGE-6, 5′-AGTGCTGACTCCTCTGCTCAAGAGGCATGA-3′;MAGE-4, 5′-AGTGCTGACTCTTCTGCTCAGAAGACATGA-3′;BAGE, 5′-CCACAACCTCAGAAGATGAAGCACAGAGCT-3′;GAGE, 5′-TTGAGTTGCTGGTTCCCCTTCTTCAGGTGT-3′) or corresponded to 5′ TCR α/β constant region sequences (Cα, 5′-Gtacacggcagggtcagggttctggatat-3′;Cβ, 5′-CTTTtgggtgtgggagatctctgCTTCTGA-3′). Blots were air-dried and exposed to Kodak X-Omat/AR X-ray films (Kodak AG., Vienna, Austria) for 4–8 h at –70°C in the presence of an intensifying screen. In order to detect very weak signals, certain blots with amplified MAA cDNA were re-exposed for up to 96 h. For quantitation of TCRAV/BV expression, the respective autoradiographs were scanned on a Bio IMAGE System (BIO IMAGE, Ann Arbor, MI) computing densitometer. To generate area volumes from linear autoradiographic scans, individual bands were digitized and integrated using the VISAGE 4.6 M software package provided by the manufacturer. Because the relative frequency of each TCRAV/BV gene segment was expressed as a percentage of the sum of all TCRAV/BV signals detected, the individual volume of the particular TCRAV/BV band was divided by the total volume of all integrated bands and the results obtained were normalized to 100. According to the literature (Weidmann et al., 1993Weidmann E. Elder E.M. Trucco M. Lotze M.T. Whiteside T.L. Usage of T-cell receptor Vβ;-chain genes in fresh and cultured tumor-infiltrating lymphocytes from human melanoma.Int J Cancer. 1993; 54: 383-390Crossref PubMed Scopus (43) Google Scholar;Thor Straten et al., 1994Thor Straten P. Scholler J. Jensen K.H. Zeuthen J. Preferential usage of T-cell receptor α/β; variable regions among tumor-infiltrating lymphocytes in primary human malignant melanomas.Int J Cancer. 1994; 56: 78-86Crossref PubMed Scopus (33) Google Scholar;Sensi and Parmiani, 1995Sensi M. Parmiani G. Analysis of TCR usage in human tumors: a new tool for assessing tumor-specific immune responses.Immunol Today. 1995; 16: 588-595Abstract Full Text PDF PubMed Scopus (106) Google Scholar), overexpression of a certain TCRAV/BV gene family was defined as TCRV ratios ≥ 1, whereby the individual TCRAV/BV ratio was calculated as follows: relative percentage of TCRV expression in melanoma specimens/mean + 2 SD of relative percentage of TCRV expression in PBL of three healthy donors. We then established a PCR protocol that enabled us to amplify the dominating BV14 and BV13 gene segment transcripts within one series of PCR experiments. For this purpose, we lowered the annealing temperature and designed a new BV primer (5′-ACATGTCCTGGTATCGACAAG-3′), which represents a BV13S1-specific nucleotide sequence differing from BV14S1 and all other members of the amplified BV13 subfamilies in only one residue (5′end, position 2 or 7, respectively). PCR products were ligated into the T/A vector PCR II (TA cloning kit, Invitrogen, San Diego, CA) and used to transform INVαF′Escherichia coli strains. White colonies were screened for positive ligation using Eco-RI digests and electrophoresis of the purified DNA plasmids. Following the instructions provided by the company, plasmid DNA-containing inserts of appropriate length were sequenced by a modified dideoxy chain termination procedure using the PRISM Ready Reaction DyeDeoxy Terminator Cycle Sequencing Kit (Perkin-Elmer Cetus) and run on a DNA sequencer (model 373, Applied Biosystems, Weiterstadt, Germany). All TCR sequences were compared with GenBank entries (EMBL, Heidelberg, Germany) using the on-line software provided by the European Bioinformatics Institute (EBI, Cambridge, U.K. athttp://www.ebi.ac.uk) and classified according to family designations defined byArden et al., 1995Arden B. Clark S.P. Kabelitz D. Mak T.W. Human T-cell receptor variable gene segment families.Immunogenetics. 1995; 42: 455-500PubMed Google Scholar. We have adopted the new TCR nomenclature proposed by the International Union of Immunological Societies (Who, 1995Who -IuiS Nomenclature Sub-Committee on TCR Designation: Nomenclature for T-cell receptor (TCR) gene segments of the immune system.Immunogenetics. 1995; 42: 451-453PubMed Google Scholar). RNA was isolated from normal and affected tissues (Table 1) of the three patients and subjected to reverse transcription using HLA-A, -B, and -C locus-specific primers. The cDNA preparations obtained were separately amplified by PCR and the HLA genotypes determined by nucleotide sequencing of the PCR products. In patient 1464 the typing did not allow discrimination between the HLA-A*1101 and the *1102 allele, and between the HLA-B*35 subtypes B*3501, *03, *04, and *06. All other alleles were typed at high resolution. Whereas in patient 1622 all loci appeared heterozygous, only one allele could be detected at the HLA-B and -C loci of patient 1464 and at the HLA-A locus of patient 1214. This could be due either to homozygosity or to nonexpression of one allele.Table 1HLA class I genotypes and melanoma specimens of patients analyzedPatientsHLA-AHLA-BHLA-CMelanoma specimens1622*2402, *2605*3503, *27052*0401, *0202NM,aNodular malignant melanoma. LN,bbLymph node metastasis. liver, skin, lung, brain1464*0301, *11*35*0401lung, liver1214*24023501, *5101*0401, *02022lung, spleen, kidney, thyroid glanda Nodular malignant melanoma.b bLymph node metastasis. Open table in a new tab Although each of the patients has a unique pattern of HLA class I alleles, they share certain similarities: all three have alleles encoding the HLA-B35 and the HLA-Cw4 specificities. Although the HLA-Cw4 specificities are identical, HLA-B35 alleles differed in the subtype. RNA was purified from all lesional tissues, PCR amplified and subjected to Southern blot hybridization using the appropriate MAA primers and probes. Results obtained showed that all specimens transcribed MART-1/Melan-A and gp100 (Table 2 andTable 3), whereas BAGE and MAGE-4 were consistently absent (Table 2). MAGE-1, MAGE-2, MAGE-3, MAGE-6, and GAGE mRNA were unevenly distributed. GAGE and MAGE-1 were expressed in metastatic lesions of patient 1214 only; MAGE-2, MAGE-3, and MAGE-6 mRNA could be amplified from tumor specimens of patients 1214 and 1464 but not of patient 1622. In fact, MART-1/Melan-A and gp100 were the only MAA transcripts detected in patient 1622’s primary melanoma and in her metastatic lesions. Given the limitations of quantitating the relative amounts of MAA mRNA expression by means of conventional PCR, cDNA amplification with MART-1/Melan-A and gp100-specific primers revealed quite homogeneous results in all specimens tested, whereas, in the case of MAGE, GAGE, and BAGE, some differences in the specimen-specific amount of transcribed mRNA were seen between the different samples. Using a semiquantitative PCR/hybridization procedure with validated TCRAV/BV-specific primers, we comparatively delineated the relative frequency of TCRV gene expression within each of the 12 tumor lesions and correlated the relative amounts of TCRV family transcripts with the range of TCRBV gene usage in nonactivated PBL from three healthy donors. According to TCR studies byWeidmann et al., 1993Weidmann E. Elder E.M. Trucco M. Lotze M.T. Whiteside T.L. Usage of T-cell receptor Vβ;-chain genes in fresh and cultured tumor-infiltrating lymphocytes from human melanoma.Int J Cancer. 1993; 54: 383-390Crossref PubMed Scopus (43) Google Scholar andThor Straten et al., 1994Thor Straten P. Scholler J. Jensen K.H. Zeuthen J. Preferential usage of T-cell receptor α/β; variable regions among tumor-infiltrating lymphocytes in primary human malignant melanomas.Int J Cancer. 1994; 56: 78-86Crossref PubMed Scopus (33) Google Scholar, TCRV families were considered to be overexpressed when their individual lesional value exceeded or at least equalled the mean value + 2 SD of the same TCRV transcripts in healthy controls. As shown inTable 4 andTable 5, TIL of our patients exhibited a substantial, albeit limited, heterogeneity of TCRV specificities [number of overexpressed TCRAV/BV families: 10 of 15 (1622); four of six (1464); five of 10 (1214)]. Certain TCRV regions (e.g., AV8, AV17, BV2, BV20) predominated within only one, others (AV13, AV18, BV13, BV14) within the majority of metastases. When TCRV family transcripts were grouped according to their percentage of overexpression within all specimens analyzed, the overall predominance of only a few TCR α/β specificities became apparent. As seen inTable 6, two different TCRAV and three TCRBV gene segment families were particularly abundant with TCRAV13 and BV14 being the specificities most often identified. In this respect, it is interesting to note that these two specificities are also part of the indigenous skin-associated TCR repertoire (Dunn et al., 1993Dunn D.A. Gadenne A.S. Simha S. Lerner E.A. Bigby M. Bleicher P.A. T-cell receptor Vβ; expression in normal human skin.Proc Natl Acad Sci USA. 1993; 90: 1267-1271Crossref PubMed Scopus (74) Google Scholar;Strohal et al., 1994aStrohal R. Paucz L. Pehamberger H. Stingl G. T-cell receptor repertoire of lymphocytes infiltrating cutaneous melanoma is predominated by Vα specificities present in T-cells of normal human skin.Cancer Res. 1994 a; 54: 4734-4739PubMed Google Scholar) and often predominate cutaneous melanoma lesions (Strohal et al., 1994aStrohal R. Paucz L. Pehamberger H. Stingl G. T-cell receptor repertoire of lymphocytes infiltrating cutaneous melanoma is predominated by Vα specificities present in T-cells of normal human skin.Cancer Res. 1994 a; 54: 4734-4739PubMed Google Scholar). We determined nucleic acid sequence diversity of junctional and CDR3 (defined according toChothia et al., 1988Chothia C. Boswell D.R. Lesk A.M. The outline structure of the T-cell α/β; receptor.Embo J. 1988; 7: 3745-3755Crossref PubMed Scopus (624) Google Scholar) from overexpressed TCR transcripts to confirm semiquantitative PCR results and identify clonally expanded TCR specificities. For this purpose, we chose a protocol that enabled us to predominately amplify the overexpressed TCRBV14 and BV13 gene segment transcripts within one series of PCR experiments. Probing all the 12 tumor specimens, we found that 131 (95%) of the TCR transcripts analyzed (n = 138) showed fully coding in-frame rearrangements suggesting the functional expression of these transcripts. One hundred and nine (83%) of these 131 TCR β-chain transcripts contained the dominating BV13 and BV14 gene segment families. The remaining 22 TCR sequences showed BV regions of subgroup II (Schiffer et al., 1986Schiffer M. Wu T.T. Kabat E.A. Subgroups of variable region genes of beta chains of T-cell receptors for antigen.Proc Natl Acad Sci USA. 1986; 83: 4461-4463Crossref PubMed Scopus (40) Google Scholar). One hundred and four of 109 BV14/BV13-containing transcripts were found to belong to only three different subfamilies (BV14S1, BV13S1, BV13S2), from which 91% were associated with the TCRBJ regions 1S2, 2S1, and 2S7. The complete sequence analysis of 131 in frame TCR β-chain transcripts showed that 14 cDNA clones occurred more than twice in the 12 tumor lesions of the three patients analyzed (Table 7). The number of occurrences per patient and site varied considerably. Within patient 1622 (Table 6) four distinct cDNA clones accounted for 100% (seven of seven; BV13S1 J2S1), 100% (six of six; BV13S1 J2S7), 50% (six of 12; BV13S2 J1S5), and 40% (two of five; BV13S2 J2S1) of the functionally rearranged TCR β-chain sequences within the primary melanoma, brain, liver, and lung metastases, respectively. Lymph node and skin metastases did not show clonally expanded TCR β-chain transcripts. To our surprise, the clonally expanded BV13S1 J2S1 cDNA transcript within patient 1622’s primary tumor could not be found in any of her five metastases, although two other TCR cDNA clones were simultaneously expressed in liver and brain metastases, respectively. Clone 13S2 J2S1, which had been amplified from the lung metastasis, also comprised 42% (five of 12) of the cloned and sequenced TCR β-chains from the liver, and the brain-associated clone BV13S1 J2S7 was expressed once within the skin metastasis of this patient. The analysis of patient 1464’s lung and liver metastases yielded similar results (Table 6). Clone BV14S1 J2S7 accounted for 86% (18 of 21) of the in-frame β-chain transcripts within the lung specimen, but was not detected in the liver. In contrast, the liver metastasis of the same patient harboured three different cDNA clones at frequencies of 29% (six of 21; BV14S1 J1S2 and BV3S1 J1S2) and 24% (five of 21; BV12S2 J2S3). In patient 1214, we again did not find a cDNA clone ubiquitously expressed in all four metastases. Yet, clones BV13S1 J1S6 and BV15S1 J2S5 co-occurred at two different sites (kidney/thyroid gland and lung/thyroid gland) and clone BV13S1 J2S1 was simultaneously isolated from three tumor deposi
Referência(s)