Discrimination of Melanocytic Tumors by cDNA Array Hybridization of Tissues Prepared by Laser Pressure Catapulting
2004; Elsevier BV; Volume: 122; Issue: 2 Linguagem: Inglês
10.1046/j.0022-202x.2004.22240.x
ISSN1523-1747
AutoresBernd Becker, Alexander Roesch, Christian Hafner, Wilhelm Stolz, Michael Landthaler, Thomas Vogt, Martin Dugas,
Tópico(s)Viral Infectious Diseases and Gene Expression in Insects
ResumoGene expression profiling by cDNA array analysis in melanoma is hampered by the need for large amounts of RNA to prepare reliable probes for array hybridization. On the other hand, for ex vivo analysis of malignant cells from melanocytic tumors laser pressure catapulting is an essential prerequisite to obtain noncontaminated melanocytic preparations; however, laser pressure catapulting prepared material provides only nanogram amounts of RNA. In this study we present an approach to overcome these limitations by combining laser pressure catapulting and real-time polymerase chain reaction based SMART cDNA amplification technology. Reproducible and reliable hybridization patterns from about 500 laser pressure catapulting prepared cell equivalents from 22 cases of melanocytic tumors were generated using array analysis. Univariate analysis revealed significant differences of the expression pattern of melanocytic nevi, melanomas, and melanoma metastases. Multivariate analysis with four genes being the best univariate discriminative features (tyrosinase related protein 2, translation initiation factor 2γ, ubiquitine conjugating enzyme E2I and one expressed sequence tag) allowed clustering of nevi, melanomas, and melanoma metastases with an accuracy of 82%. Data validation was performed by additional quantitative reverse transcription–polymerase chain reaction (TaqMan–reverse transcription–polymerase chain reaction). Taken together, this study shows, that (1) array analysis is feasible on tumors with rather low cell numbers, and (2) differences in expression profiles allow discrimination between benign and malignant lesions. Expression patterns of marker genes defined in unequivocal histopathologic entities may improve the diagnostic and prognostic assessment of difficult melanocytic lesions, which is still the hardest problem in dermatopathology. Gene expression profiling by cDNA array analysis in melanoma is hampered by the need for large amounts of RNA to prepare reliable probes for array hybridization. On the other hand, for ex vivo analysis of malignant cells from melanocytic tumors laser pressure catapulting is an essential prerequisite to obtain noncontaminated melanocytic preparations; however, laser pressure catapulting prepared material provides only nanogram amounts of RNA. In this study we present an approach to overcome these limitations by combining laser pressure catapulting and real-time polymerase chain reaction based SMART cDNA amplification technology. Reproducible and reliable hybridization patterns from about 500 laser pressure catapulting prepared cell equivalents from 22 cases of melanocytic tumors were generated using array analysis. Univariate analysis revealed significant differences of the expression pattern of melanocytic nevi, melanomas, and melanoma metastases. Multivariate analysis with four genes being the best univariate discriminative features (tyrosinase related protein 2, translation initiation factor 2γ, ubiquitine conjugating enzyme E2I and one expressed sequence tag) allowed clustering of nevi, melanomas, and melanoma metastases with an accuracy of 82%. Data validation was performed by additional quantitative reverse transcription–polymerase chain reaction (TaqMan–reverse transcription–polymerase chain reaction). Taken together, this study shows, that (1) array analysis is feasible on tumors with rather low cell numbers, and (2) differences in expression profiles allow discrimination between benign and malignant lesions. Expression patterns of marker genes defined in unequivocal histopathologic entities may improve the diagnostic and prognostic assessment of difficult melanocytic lesions, which is still the hardest problem in dermatopathology. The discrimination between benign and malignant melanocytic lesions is often conjectural. Furthermore, if a lesion is diagnosed as melanoma, it would be highly desirable to supplement the pathology with molecular markers discriminating potentially metastatic and nonmetastatic biologic phenotypes more precisely. Molecular markers, which were previously used (e.g., p53, Ki67, cathepsins, collagenase, c-fos, c-myc) have some prognostic value but due to conjectural results none of those has made it into clinical routine (Vogt et al., 1997Vogt T.M. Welsh J. Stolz W. Kullmann F. Jung B. Landthaler M. McClelland M. RNA fingerprinting displays UVB-specific disruption of transcriptional control in human melanocytes.Cancer Res. 1997; 57: 3554-3561PubMed Google Scholar). Therefore, establishing a technique enabling the analysis of multiple parameters simultaneously is crucial in order to expand the knowledge on known potential markers and add new ones. The development of melanoma is a process, in which the cellular homeostasis is dramatically changed by the deregulation of the expression of many genes (Herlyn et al., 2001Herlyn M. Ferrone S. Ronai Z. Finerty J. Pelroy R. Mohla S. Melanoma biology and progression.Cancer Res. 2001; 61: 4642-4643PubMed Google Scholar). These complex changes can now be analyzed by genome-wide profiling techniques, such as the cDNA array hybridization for generating the desired marker profiles. Cancer research, however, in this specific entity and in others with small amounts of tumor cells was hitherto hampered by the excessive amounts of RNA necessary to produce reliable probes for array hybridization (Bertucci et al., 1999Bertucci F. Bernard K. Loriod B. et al.Sensitivity issues in DNA array-based expression measurements and performance of nylon microarrays for small samples.Hum Mol Genet. 1999; 8: 1715-1722Crossref PubMed Scopus (156) Google Scholar). Therefore, the majority of tumors analyzed by array or chip analyses were either metastases (Bittner et al., 2000Bittner M. Meltzer P. Chen Y. et al.Molecular classification of cutaneous malignant melanoma by gene expression profiling.Nature. 2000; 406: 536-540Crossref PubMed Scopus (1660) Google Scholar) or tumor entities, which harbor enough tumor tissue for purification of sufficient amounts of RNA (Alizadeh et al., 2000Alizadeh A.A. Eisen M.B. Davis R.E. et al.Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling.Nature. 2000; 403: 503-511Crossref PubMed Scopus (7491) Google Scholar;Dhanasekaran et al., 2001Dhanasekaran S.M. Barrette T.R. Ghosh D. et al.Delineation of prognostic biomarkers in prostate cancer.Nature. 2001; 412: 822-826Crossref PubMed Scopus (1383) Google Scholar;Young et al., 2001Young A.N. Amin M.B. Moreno C.S. et al.Expression profiling of renal epithelial neoplasms: A method for tumor classification and discovery of diagnostic molecular markers.Am J Pathol. 2001; 158: 1639-1651Abstract Full Text Full Text PDF PubMed Scopus (275) Google Scholar;Ahr et al., 2002Ahr A. Karn T. Solbach C. Seiter T. Strebhardt K. Holtrich U. Kaufmann M. Identification of high risk breast-cancer patients by gene expression profiling.Lancet. 2002; 359: 131-132Abstract Full Text Full Text PDF PubMed Scopus (157) Google Scholar). Alternatively, cells from fresh tumor biopsies were taken in culture for several passages before preparing RNA from these cultured cells (Bittner et al., 2000Bittner M. Meltzer P. Chen Y. et al.Molecular classification of cutaneous malignant melanoma by gene expression profiling.Nature. 2000; 406: 536-540Crossref PubMed Scopus (1660) Google Scholar;Brem et al., 2001Brem R. Hildebrandt T. Jarsch M. Van Muijen G.N. Weidle U.H. Identification of metastasis-associated genes by transcriptional profiling of a metastasizing versus a non-metastasizing human melanoma cell line.Anticancer Res. 2001; 21: 1731-1740PubMed Google Scholar), but the molecular "proximity" to the original tumor is questionable. The tumor cells in nevi or early melanomas may constitute only a minor fraction of the total tissue volume excised during surgery. Isolation of the melanocytic cells from small early lesions by conventional microdissection techniques might include variable numbers of contaminating nonmelanoma cells. The Laser Pressure Catapulting (LPC) microscope allows to isolate directly and analyze small cell populations of interest virtually without any contaminating cells directly from hematoxylin and eosin stained histologic sections (Bohm et al., 1997Bohm M. Wieland I. Schutze K. Rubben H. Microbeam MOMeNT: Non-contact laser microdissection of membrane-mounted native tissue.Am J Pathol. 1997; 151: 63-67PubMed Google Scholar;Schutze and Lahr, 1998Schutze K. Lahr G. Identification of expressed genes by laser-mediated manipulation of single cells.Nat Biotechnol. 1998; 16: 737-742Crossref PubMed Scopus (365) Google Scholar;Westphal et al., 2002Westphal G. Burgemeister R. Friedemann G. et al.Noncontact laser catapulting: A basic procedure for functional genomics and proteomics.Methods Enzymol. 2002; 356: 80-99Crossref PubMed Scopus (35) Google Scholar). As only low amounts of RNA can be gained from LPC prepared cells, amplification of the RNA is a prerequisite for array analysis. Methods involving RNA amplification based on T7-primed in vitro transcription were shown to generate reliable results in studies of neuronal cells and early colorectal carcinomas (Luo et al., 1999Luo L. Salunga R.C. Guo H. et al.Gene expression profiles of laser-captured adjacent neuronal subtypes.Nat Med. 1999; 5: 117-122Crossref PubMed Scopus (630) Google Scholar;Kitahara et al., 2001Kitahara O. Furukawa Y. Tanaka T. et al.Alterations of gene expression during colorectal carcinogenesis revealed by cDNA microarrays after laser-capture microdissection of tumor tissues and normal epithelia.Cancer Res. 2001; 61: 3544-3549PubMed Google Scholar). The procedures involved, however, are rather time-consuming and expensive and may therefore be not applicable for routine purposes. We have recently shown a fast and inexpensive polymerase chain reaction (PCR)-based alternative approach that seems to be much more suitable for routine applications (Becker et al., 2001Becker B. Vogt T. Landthaler M. Stolz W. Detection of differentially regulated genes in keratinocytes by cDNA array hybridization: Hsp27 and other novel players in response to artificial ultraviolet radiation.J Invest Dermatol. 2001; 116: 983-988Crossref PubMed Scopus (33) Google Scholar) and applied this technique to RNA from LPC-prepared melanocytic cells. To demonstrate the reproducibility of our method, we performed the analysis of four of our cases as independent duplicate experiments (case nos 2, 9, 18, and 21; Table I). A typical result of one of the experiments is shown in Figure 1(b). For the pairwise comparison an expression threshold for a significant signal intensity was set as described (e.g., no. 2068-1 and no. 2068-2, Figure 1b: yellow lines). Genes below these thresholds are not taken into account because of an increased probability of giving false positive results because of low signal-to-background ratios (Figure 1b: yellow spots); 2.2% of the detected genes appeared as "regulated". These genes represent false positives (Figure 1b: blue spots above the red line and below the green line). The remaining signals (97.8%, 270 of 4467 genes displayed on the array) are evenly expressed in this duplicate experiment (Figure 1b: blue spots between the red and the green lines). The average rate of false positive regulated genes in four independent experiments was 6.9% (2.2% of 270 genes, 2.2% of 410, 10.3% of 77, and 13% of 84).Table IMelanocytic lesions included in this studyCase no.Histopathologic diagnosisClark levelTumor thickness (mm)1Melanocytic nevus2Dermal nevus3Dermal nevus4Compound nevus5Melanocytic nevus with lymph. infiltration6Melanocytic nevus with lymph. infiltration7Melanocytic nevus with lymph. infiltration8Compound nevus9NMMIII0.810NMMIII0.811SSMII0.2512NMMIV2.313SSMII0.214SSMIV0.815SSM, secondary NMMIV1.816SSMIII117SSMIII1.0518Metastasis amelanotic19Metastasis amelanotic20Metastasis21Metastasis melanotic22Metastasis melanotic23Melanocytic nevus24Melanocytic nevus25Melanocytic nevus26NMMV1827NMMV828NMMV4.729LMMV530LMMV7.531Metastasis amelanotic32Metastasis amelanotic33Metastasis melanotic34Metastasis amelanotic35Metastasis amelanoticCase nos 1 to 22 were used for array analysis; case nos 23 to 35 were used as further validation set for reverse transcription–PCR. SSM, superficial spreading melanoma; NMM, nodular melanoma; LMM, lentigo maligna melanoma. Open table in a new tab Case nos 1 to 22 were used for array analysis; case nos 23 to 35 were used as further validation set for reverse transcription–PCR. SSM, superficial spreading melanoma; NMM, nodular melanoma; LMM, lentigo maligna melanoma. For assessment of the proportions and sizes of classes of regulated genes during melanoma progression, all regulated genes were clustered into four regulative categories according to the pattern of regulation during progression from nevi versus melanoma versus melanoma metastases as described above. These major regulative categories were further split with regard to their functional context, based on the recent knowledge in the literature: (1) metabolism; (2) signaling/cell cycle; (3) receptors/attachment; (4) others with known function; and (5) unknown function. The absolute number of genes regulated in the different functional clusters are summarized in Figure 2. As expected, the most prominent differences in gene expression profiles of known genes was observed in the category "metabolism" showing upregulation in melanomas versus nevi (Figure 2, light gray). The average expression ratios of the genes included in the analysis and their categories are summarized in a detailed table, which is available as supplemental material. We analyzed the melanocytic tumors listed in Table I by array analysis. From 4467 spotted genes, 232 passed the filter an expression threshold of 5%, averaged over all cases. Univariate analysis revealed those genes that were statistically significantly regulated in different stages of progression. The genes with the highest F-values from univariate analysis are listed in Table II sorted according to their statistical relevance. Multivariate stepwise analysis of the cases using those four most significant univariate variables allowed classification of 82% of all cases correctly in accordance to the expert histopathologic diagnosis Table IIIa. The resulting clusters of lesions are displayed by plotting the canonical discriminant functions for each case in Figure 3. This plot visualizes the relative distances of the profiles using this set of four most discriminative variables. For those four genes the average expression value within each class of melanocytic lesion was calculated Table II.Table IIVariables selected by univariate analysisAverage expression valueAccessionGeneNeviMelanomaMetastasesF-valueAA292995Tyrosinase related protein 2461271847.2AA448301Translation initiation factor 2γ5.513.85.55.7AA487197Ubiquitine conjugating enzyme E2I1729662.9AA030013EST2954992.7 Open table in a new tab Table IIIClassification results based on a multivariate analysis of: (A) array dataa81.8% of all cases were classified correctly according to the genes with the highest F-values Table II, and (B) reverse transcription–PCR datab84.6% of all cases were classified correctly according to the expression values of eIF2γ, UBE2I, and one EST Table IIPredicted classificationClassTotalNevusMMMetastases(A)NumberNevus9810MM8260Meta5104%Nevus10088.911.10MM10025750Meta10020080(B)NumberNevus3210MM5050Meta5014%Nevus10066.733.30MM10001000Meta10002080a 81.8% of all cases were classified correctlyb 84.6% of all cases were classified correctly Open table in a new tab To verify the array data classification results, we analyzed the expression of the marker genes shown in Table II by quantitative real-time reverse transcription–PCR using TaqMan probes. In addition to the genes used for classification, we included in this analysis a further eight genes that reached the highest F-values in the univariate analysis. In order to assure, that the markers are valid independent of the set of lesions selected, the expression was analyzed in another set of melanocytic lesions (three nevi, five melanomas, and five melanoma metastases, Table I: case nos 23–35). The expression values were normalized to GAPDH expression. The normalized expression levels of three genes (eIF2γ, UBE2I, and the EST) were analyzed statistically by the same canonical discriminant analysis as used for the array data. According to the histopathologic diagnosis 85% of all cases could be classified correctly Table IIIb. The comparison of the expression values obtained by reverse transcription–PCR with the regulation observed in array analysis verifies the mode of regulation from benign towards malignant lesions in nine of 12 genes Table IV.Table IVMean expression valuesaMean expression ratios normalized to GAPDH. of genes analysed by Taqman reverse transcription–PCRGeneNeviMMMetastasesRegulation by array analysesbTendence of regulation observed by array analyses from nevi towards malignant cases (MM, metastases).SCAMP2cNo confirmation of regulation between reverse transcription–PCR and array results.4.883.063.44upMDA-71.4815.421.34upDesmin0.331.640.61upCD682.696.136.43upSyntaxincNo confirmation of regulation between reverse transcription–PCR and array results.3.070.471.97upProteoglycan0.811.711.94upHMOX11.905.013.31upRNPL 3cNo confirmation of regulation between reverse transcription–PCR and array results.5.583.816.11upUBE2I4.745.577.08upeIF2γ16.3111.7320.69upEST0.991.281.75upTrp226.2549.7212.07upa Mean expression ratios normalized to GAPDH.b Tendence of regulation observed by array analyses from nevi towards malignant cases (MM, metastases).c No confirmation of regulation between reverse transcription–PCR and array results. Open table in a new tab Gene expression profiles of melanoma cells were so far restricted to advanced tumors harboring microgram amounts of RNA or which were taken into culture from the tumors and passaged several times to get enough cells for RNA isolation (Bittner et al., 2000Bittner M. Meltzer P. Chen Y. et al.Molecular classification of cutaneous malignant melanoma by gene expression profiling.Nature. 2000; 406: 536-540Crossref PubMed Scopus (1660) Google Scholar;Brem et al., 2001Brem R. Hildebrandt T. Jarsch M. Van Muijen G.N. Weidle U.H. Identification of metastasis-associated genes by transcriptional profiling of a metastasizing versus a non-metastasizing human melanoma cell line.Anticancer Res. 2001; 21: 1731-1740PubMed Google Scholar). In contrast to the analysis of primary tumor material, the analysis of cultured tumor cells might bear misleading results due to adaptive effects under cell culture conditions. The LPC-microdissection real-time SMART–PCR approach presented herein, overcomes these limitations. We demonstrate that with this approach gene expression analysis of melanocytic cells isolated directly from routinely excised biopsy material is fast and reasonably inexpensive. By statistical analysis of the expression profiles of melanocytic lesions we obtained a set of potential markers Table II, which discriminates melanocytic tumors of different levels of malignancy with an accuracy of 82%, if compared with histopathology evaluation (Table III, Figure 3). The few falsely classified lesions might represent certain subsets of biologic phenotypes characterized by gene expression changes outside of the typical range, which are not recognizable by histopathology. This set of marker genes could be further validated by quantitative reverse transcription–PCR on a second set of melanocytic tumors, reaching the same accuracy of 85% correctly classified cases Table III, although for statistical significance reasons only three variables were included in the canonical discriminant analysis. In addition, we analyzed a further eight genes with the highest F-values from the univariate analysis in the same set of lesions by reverse transcription–PCR to validate the array results more stringently. Nine of 12 genes could be confirmed Table IV. This result is even more convincing, as the analyses were performed in two different sets of melanocytic lesions. The small amounts of RNA that are available from laser-microdissected cells require a preamplification before generating a probe for array hybridization. Recently, we could demonstrate that SMART amplification of RNA is a fast method for preparing reliable probes for cDNA array analysis (Becker et al., 2001Becker B. Vogt T. Landthaler M. Stolz W. Detection of differentially regulated genes in keratinocytes by cDNA array hybridization: Hsp27 and other novel players in response to artificial ultraviolet radiation.J Invest Dermatol. 2001; 116: 983-988Crossref PubMed Scopus (33) Google Scholar). This result is in good agreement with experiments from other groups (Gonzalez et al., 1999Gonzalez P. Zigler Jr., J.S. Epstein D.L. Borras T. Identification and isolation of differentially expressed genes from very small tissue samples.Biotechniques. 1999; 26: 888-892Google Scholar;Spirin et al., 1999Spirin K.S. Ljubimov A.V. Castellon R. et al.Analysis of gene expression in human bullous keratopathy corneas containing limiting amounts of RNA.Invest Ophthalmol Vis Sci. 1999; 40: 3108-3115PubMed Google Scholar;Vernon et al., 2000Vernon S.D. Unger E.R. Rajeevan M. Dimulescu I.M. Nisenbaum R. Campbell C.E. Reproducibility of alternative probe synthesis approaches for gene expression profiling with arrays.J Mol Diagn. 2000; 2: 124-127Abstract Full Text Full Text PDF PubMed Scopus (40) Google Scholar).Leethanakul et al., 2000Leethanakul C. Patel V. Gillespie J. et al.Distinct pattern of expression of differentiation and growth-related genes in squamous cell carcinomas of the head and neck revealed by the use of laser capture microdissection and cDNA arrays.Oncogene. 2000; 19: 3220-3224Crossref PubMed Scopus (256) Google Scholar andFink et al., 2002Fink L. Kohlhoff S. Stein M.M. et al.cDNA array hybridization after laser-assisted microdissection from nonneoplastic tissue.Am J Pathol. 2002; 160: 81-90Abstract Full Text Full Text PDF PubMed Scopus (67) Google Scholar applied this technique to material that was laser microdissected. We enhanced the previously published protocols by applying a real-time PCR-based amplification of the RNA. This way, we could directly monitor the quantity of the cDNA synthesized during the PCR process. In addition, this is not as RNA consuming as the original SMART protocol patented by Clontech, because we do not need to prepare two reactions ("tester and driver") for finding the log-phase of the SMART–PCR by sacrificing one of the two reactions for agarose gel electrophoresis. Accordingly, there is no need of a subsequent cycling of the second reaction to the optimal number of cycles, which further reduces the time spent performing the protocol. In order to demonstrate the reproducibility of our array hybridization method, we performed duplicate analyses of four cases. We could show a high reproducibility with only 6.9% (between 2.2% and 13%) of all spots above a certain threshold "falsely" indicating regulation. Similarly,Wang et al., 2001Wang X. Ghosh S. Guo S.W. Quantitative quality control in microarray image processing and data acquisition.Nucleic Acids Res. 2001; 29 (E75): E75Crossref PubMed Scopus (183) Google Scholar analyzed the relation between hybridization signal quality (i.e., signal/background ratio) and the consistency of array results. Depending on the quality of the hybridization signals they found up to 10% of low-quality hybridization signals, which is comparable with our data. The quality of array hybridization is influenced by many factors: printing of arrays, hybridization conditions, RNA quality, and RNA quantity. The last two are probably the most important factors for our experimental system, as the amount of RNA prepared from microdissected tissue is very limited and the material is exposed to conditions, which might be not optimal for RNA stability (e.g., hematoxylin and eosin staining). Although we control the amplification process by incorporation of SYBR Green, some amplification-born changes in the representation of single transcripts are possible and may affect the reproducibility (Becker et al., 2001Becker B. Vogt T. Landthaler M. Stolz W. Detection of differentially regulated genes in keratinocytes by cDNA array hybridization: Hsp27 and other novel players in response to artificial ultraviolet radiation.J Invest Dermatol. 2001; 116: 983-988Crossref PubMed Scopus (33) Google Scholar). Taken together, in this transcriptome profiling we found mainly genes, which were induced with increasing stage of malignancy. In the stringently filtered data we found four genes, which showed a reduction of expression by a factor of 2 from nevi to metastases and a further eight genes, which were downregulated from melanoma towards metastases. But this regulation turned out to be not significant. Genes that were downregulated from nevi versus melanoma might be underrepresented, because we included only those genes in the statistical analysis that exhibited a normalized expression ratio of 5% averaged over all cases. This strategy ensures that the remaining genes are more robust candidates. Using a threshold lower than 5% would lead to more genes entering the analysis but also to the excess of false positives. The average expression value of the four genes with the highest F-values in the univariate analysis of the array expression data within each class of melanocytic lesion was calculated Table III. The average expression level of Trp2 is induced in later stages of progression in our set of cases (Table I: nos 1–22). Interestingly, Trp2 has been found to be involved in the protection of melanoma cells against apoptosis (Nishioka et al., 1999Nishioka E. Funasaka Y. Kondoh H. Chakraborty A.K. Mishima Y. Ichihashi M. Expression of tyrosinase, TRP-1 and TRP-2 in ultraviolet-irradiated human melanomas and melanocytes: TRP-2 protects melanoma cells from ultraviolet B induced apoptosis.Melanoma Res. 1999; 9: 433-443Crossref PubMed Scopus (48) Google Scholar). In addition, it was detected to be abundantly expressed in glioblastoma lesions also deriving from cells of neuroectodermal offspring (Udono et al., 2001Udono T. Takahashi K. Yasumoto K. et al.Expression of tyrosinase-related protein 2/DOPAchrome tautomerase in the retinoblastoma.Exp Eye Res. 2001; 72: 225-234Crossref PubMed Scopus (13) Google Scholar). As cancer cells are much more resistant against apoptosis than their untransformed counterparts, this might be a further mechanism switched on during melanoma progression. On the other hand,Orlow et al., 1998Orlow S.J. Silvers W.K. Zhou B.K. Mintz B. Comparative decreases in tyrosinase, TRP-1, TRP-2, and Pmel 17/silver antigenic proteins from melanotic to amelanotic stages of syngeneic mouse cutaneous melanomas and metastases.Cancer Res. 1998; 58: 1521-1523PubMed Google Scholar, compared the expression of Trp2 in melanoma and melanoma metastases in a syngenic mouse model and found a trend of downregulation of Trp2 in lesions of more aggressive local growth. At later stages of progression melanoma lesions tend to become amelanotic and, therefore, one could expect that the expression of some of the genes of the melanogenesis pathway are lost or downregulated (Orlow et al., 1995Orlow S.J. Hearing V.J. Sakai C. Urabe K. Zhou B.K. Silvers W.K. Mintz B. Changes in expression of putative antigens encoded by pigment genes in mouse melanomas at different stages of malignant progression.Proc Natl Acad Sci USA. 1995; 92: 10152-10156Crossref PubMed Scopus (40) Google Scholar). Even if melanin production is impaired, however, some of the pigment genes are still expressed varying from case to case (Sarantou et al., 1997Sarantou T. Chi D.D. Garrison D.A. Conrad A.J. Schmid P. Morton D.L. Hoon D.S. Melanoma-associated antigens as messenger RNA detection markers for melanoma.Cancer Res. 1997; 57: 1371-1376PubMed Google Scholar). eIF2γ is a subunit of the eukaryotic translation initiation factor 2 (eIF2), which is a heterotrimeric G-protein required for GTP-dependent delivery of initiator tRNA to the ribosome. The eIF2γ seems to be transiently upregulated in melanomas in comparison with nevi and metastasis.Erickson and Hannig, 1996Erickson F.L. Hannig E.M. Ligand interactions with eukaryotic translation initiation factor 2: Role of the gamma-subunit.EMBO J. 1996; 15: 6311-6320Crossref PubMed Scopus (83) Google Scholar showed that a mutation in the eIF2γ gene leads to reduced growth rates. In addition they demonstrated that eIF2γ is involved in overriding the lethal effect of the tumor suppressor gene PKR (Erickson et al., 2001Erickson F.L. Nika J. Rippel S. Hannig E.M. Minimum requirements for the function of eukaryotic translation initiation factor 2.Genetics. 2001; 158: 123-132Crossref PubMed Google Scholar). These data suggest, that eIF2γ may be involved in an important phase of developing autonomous melanoma growth. Our observation of UBE2I being progressively upregulated (nevi<melanoma<metastases) underscores a possible influence of disrupted ubiquitine pathways in melanomas: Ubiquitinylation and subsequent degradation of the ubiquitinylated proteins by the proteasome machinery accounts for the regulation of proteins, such as cyclins, cyclin-dependent kinase inhibitors, p53, c-Jun and c-Fos (Ciechanover et al., 2000Ciechanover A. Orian A. Schwartz A.L. The ubiquitin-mediated proteolytic pathway: Mode of action and clinical implications.J Cell Biochem Suppl. 2000; 34: 40-51Crossref PubMed Scopus (240) Google Scholar). Loss of the fine-tuned regulation of the G1/S transition duri
Referência(s)