TANGO is a tumor suppressor of malignant melanoma

Abstract

The TANGO gene was originally identified as a new family member of the melanoma inhibitory activity gene family. The gene codes for a 14 kDa protein of so far unknown function. In our study we revealed that TANGO was downregulated or lost in 9 melanoma cell lines when compared to normal melanocytes and in most of the 8 tumor samples analyzed. The losses were associated with advanced stage of the disease. These results were confirmed in situ by immunohistochemistry on 10 paraffin-embedded sections of human malignant melanoma primary tumors and melanoma skin metastases. A small reduction of TANGO was also seen in different benign and atypical nevi when compared to normal skin. For functional analysis of TANGO we evaluated TANGO re-expressing melanoma cell clones and antisense TANGO cell clones with a complete loss of TANGO. Functional assays with TANGO transfected or treated cell lines revealed that TANGO expression reduces motility, whereas reduction of TANGO enhances migration. Our studies, therefore, indicate that reduction of TANGO expression contributes to tumor progression. These results taken together provide the first indications for a tumor suppressor role of TANGO gene in human malignant melanoma. © 2006 Wiley-Liss, Inc.

Malignant melanoma arises from melanocytic cells of the skin and belongs to the most aggressive types of cancer. Malignant melanoma develops when the melanocytes no longer respond to normal control mechanism of cellular growth and are capable of invasion locally or spreading to other organs in the body. Several processes involved in tumor development are known today but detailed molecular changes of this disease are still under investigation.

Recent data indicated that loss of cell–cell and cell–matrix contacts, induction of growth factor expression and deregulation of signaling pathways are involved in early development of the disease.1, 2, 3

A novel human gene (TANGO) encoding a melanoma inhibitory activity (MIA) homologous protein was identified by a gene bank search.4TANGO, together with the homologous genes MIA, OTOR (FPD, MIAL) and MIA 2, defines a novel gene family sharing important structural features, significant homology at both the nucleotide and protein level and similar genomic organization. TANGO encodes a mature protein of 103 amino acids in addition to a hydrophobic secretory signal sequence. Sequence homology confirms the highly conserved SH3 structure present also in MIA, OTOR and MIA2.5, 6 Interestingly, in situ hybridization, RT-PCR and Northern Blots revealed very broad TANGO expression patterns in contrast to the highly restricted expression patterns previously determined for the other members of the MIA gene family. High levels of TANGO expression were observed both during embryogenesis and in adult tissues. The only cells lacking TANGO expression are cells belonging to the hematopoietic system.4

The present study was performed to evaluate the role of TANGO in human tumor development and progression. Therefore, we screened the expression profiles of TANGO in 9 human melanoma cell lines, in different melanoma tumor samples in situ and in different benign and atypical nevi. In addition, functional assays with TANGO negative, TANGO re-expressing and TANGO-treated cells were performed to characterize the biological effects.

Material and methods

Cell culture

The melanoma cell lines Mel Im, Mel Ei, Mel Wei, Mel Ho, Mel Juso, SK Mel 28, SK Mel 3, HMB2 and HTZ19 were described previously.7 The cell lines Mel Ei, Mel Wei, Mel Ho, HMB2 and Mel Juso were derived from a primary cutaneous melanoma, Mel Im, SK Mel 3, SK Mel 28 and HTZ19 were derived from metastases of malignant melanomas. Cells were maintained in DMEM supplemented with penicillin (400 U/ml), streptomycine (50 μg/ml), L-glutamine (300 μg/ml) and 10% fetal calf serum (FCS; Sigma, Deisenhofen, Germany) and split at a 1:5 ratio every 3 days. Human primary melanocytes derived from normal skin were cultivated in melanocyte medium MGM-3 (Gibco, Eggenstein, Germany) under a humidified atmosphere of 5% CO₂ at 37°C. Cells were used between passages 3 and 6 and not later than 3 days after trypsinization. Cells were detached for subcultivation or assay with 0.05% trypsin and 0.04% EDTA in PBS.

RNA isolation and reverse transcription

Total cellular RNA was isolated from cultured cells or from microdissected tissues using the RNeasy kit (QIAGEN, Hilden, Germany) and cDNAs were generated by reverse transcriptase reaction performed in 20 μl reaction volume containing 2 μg of total cellular RNA, 4 μl of 5× first strand buffer (Invitrogen, Groningen, The Netherlands), 2 μl of 0.1 M DTT, 1 μl of dN₆-primer (10 mM), 1 μl of dNTPs (10 mM) and DEPC-water. The reaction mixture was incubated for 10 min at 70°C, 200 units of Superscript II reverse transcriptase (Invitrogen) were added and RNAs were transcribed for 1 hr at 37°C. Reverse transcriptase was inactivated at 70°C for 10 min and the RNA was degraded by digestion with 1 μl RNase A (10 mg/ml) at 37°C for 30 min.

Analysis of expression by quantitative PCR

Quantitative real time-PCR for TANGO was performed on a Lightcycler (Roche, Mannheim, Germany). cDNA template (2 μl), 2.4 μl (25 mM) MgCl₂, 0.5 μl (20 mM) of forward and reverse primers (hTANGO for: 5′-GGCTCTTGAAGATTTCAC-3′; hTANGO rev: 5′-ATCCGTCTCATCTGTTGG-3′) and 2 μl of SybrGreen LightCycler Fast-Start DNA Master SYBR Green Mix in a total of 20 μl were applied to the following PCR program: 10 min at 95°C (initial denaturation); 20°C/sec temperature transition rate up to 95°C for 15 sec, 60°C for 3 sec, 72°C for 5 sec, 81°C acquisition mode single, repeated for 40 times (amplification). The PCR reaction was evaluated by melting curve analysis and checking the PCR products on 2% agarose gels. β-Actin was amplified to ensure cDNA integrity and to normalize expression. Each real time-PCR was repeated at least 3 times.

Western blot analysis

Approximately 7 million trypsinized cells were incubated in 200 μl RIPA-buffer (Roche) for 15 min at 4°C. Insoluble fragments were removed by centrifugation at 13,000 rpm for 10 min and the supernatant lysate was stored at −20°C. RIPA-cell lysate was loaded and separated on 12% SDS-PAGE and subsequently blotted onto PVDF membrane. After blocking for 1 hr with 3% BSA/PBS, the membrane was incubated for 16 hr with a primary antibody against TANGO (generated by BioGenes GmbH, Berlin, 1:1,000) or β-actin (Sigma, MO, 1:2,500). Then, the membrane was washed 3 times in PBS, incubated for 1 hr with an alkaline phosphatase conjugated secondary antibody (Chemicon, CA, 1:5,000) and then washed again. Finally, immunoreactions were visualized by NBT/BCIP (Zytomed, San Francisco, CA) staining. All Western blots were repeated at least 3 times.

Immunohistochemistry

Paraffin-embedded sections of human malignant melanoma primary tumors and melanoma skin metastases and different benign (compound, junctional, dysplasic, blue) and atypical nevi were screened for TANGO protein expression by immunohistochemistry. The tissues were deparaffinized, rehydrated and subsequently incubated with rabbit anti-TANGO antibody (BioGenes GmbH, Berlin, 1:1,000). The secondary antibody (biotin-labeled anti-rabbit, anti-mouse, DAKO, Germany) was incubated for 30 min at room temperature, followed by incubation with streptavidin-POD (DAKO) for 30 min. Antibody binding was visualized using AEC-solution (DAKO). Finally, the tissues were counterstained by haemalaun solution (DAKO). The evaluation of the staining was performed semi quantitatively by light-microscopy.

Expression of recombinant biotinylated TANGO protein

A TANGO prokaryotic expression vector with a 15 amino acid Avi-tag peptide sequence including a FXa cleavage site was constructed by overlap extension PCR. The TANGO full length cDNA construct was cloned into the vector pIVEX2.3-MCS (Roche). The expression vector was used in the Rapid Translation System, a cell free E. coli based protein transcription/translation system (Roche). By adding biotin, ATP and the E. coli biotin protein ligase BirA during the procedure, the protein was biotinylated at the introduced Avi-tag at the N-terminus and called recombinant biotinylated TANGO.

Stable transfection of melanoma cells with TANGO or antisense TANGO

Mel Im cell clones re-expressing TANGO were established by stable transfection with sense, Mel Im cells with further reduction of TANGO expression by transfection with an antisense (as) TANGO expression plasmid. The sense TANGO expression plasmid was generated by cloning the full length TANGO nucleotide sequence (US patent, WO 00/12762) in sense orientation into the pCMX-PL1 vector. The antisense expression plasmid was produced by cloning the same TANGO sequence in antisense orientation into the pCMX-PL1 expression vector. Transfections were performed using lipofectamin plus (Invitrogen). Plasmids were co-transfected with pcDNA3 (Invitrogen) containing the selectable marker for neomycin resistance. Controls received pcDNA3 alone. One day after transfection, cells were placed into selection medium containing 50 μg/ml G418 (Sigma). After 25 days of selection, individual G418-resistant colonies were subcloned.

Migration and invasion assay

Migration and invasion assays were performed using Boyden Chambers containing polycarbonate filters with 8 μm pore size (Costar, Bodenheim, Germany), essentially as described previously.7 Filters were coated with gelatin or matrigel (diluted 1:3 in H₂O; Becton Dickinson, Heidelberg, Germany), respectively. The lower compartment was filled with fibroblast-conditioned medium, used as a chemo-attractant. Melanoma cells were harvested by trypsinization for 2 min, resuspended in DMEM without FCS at a density of 2 × 10⁴ cells/ml (migration) or 3 × 10⁵ cells/ml (invasion) and placed in the upper compartment of the chamber. After incubation at 37°C for 4 hr, the filters were collected and the cells adhering to the lower surface fixed, stained and counted. All assays were repeated at least 3 times.

Anchorage-independent growth assay

Cells were seeded into 6-well plates in DMEM, 0.36% agar (Sigma), supplemented with 10% FCS on top of a 0.72% agar bed in similar medium. The cultures were incubated for 10 days. After 10 days the number of colonies was not changed between the cell clones and the control (mock) or parental Mel Im cell line. Differences were only seen in the size of the colonies. Therefore, the diameter of the colonies was measured and photographed. Colony size was measured using a Carl Zeiss microscope (Carl Zeiss Vision GmbH, Hallbergmoos, Germany). For each cell clone the diameter of 20 colonies was determined and statistics was performed. Experiments were repeated at least twice.

Scratch assay

Migration of cells was assayed by scratch assays. For scratch assays (“wound-healing-assay”), cells were seeded in high density into 6-well plates and scratched by a pipette tip in a definite array. Migration into this array was documented and measured after 24 and 48 hr. Total migration of mock 2 or Mel Im control cells after 48 hr was measured using a Carl Zeiss microscope (Carl Zeiss Vision GmbH, Hallbergmoos, Germany) and set as 100%. Migration in percentage between the TANGO cell clones and the mock or Mel Im control was calculated. Each analysis was performed in duplicate and was repeated 3 times.

Additionally, we added 200 ng/ml recombinant TANGO into each well of a 6-well plate to observe differences in migration of the as TANGO cell clones after adding external TANGO to the cells.

Cell adhesion assay

For the determination of the relative attachment of Mel Im, mock control and antisense TANGO cell clones to extracellular matrix proteins such as human fibronectin (Cat. No. CBA011), human vitronectin (Cat. No. CBA012) and human collagen IV (Cat. No. CBA013), 15,000 cells were seeded onto the coated substrates, incubated for 15 min, followed by the determination of relative cell attachment using a fluorescent dye as described by the manufacturer (Calbiochem, EMD Biosciences, Darmstadt, Germany). Fluorescence was measured with the Fusion fluorescence reader from Packard Bio Science Company. Experiments were repeated at least twice.

Statistical analysis

Results are expressed as mean ± SD (range) or percent. Comparison between groups was made using the Student's paired t test. A p value < 0.05 was considered statistically significant. All calculations were performed using the GraphPad Prism 4 software (GraphPad software Inc., San Diego). *p < 0.05; **p < 0.01; ***p < 0.001.

Results

Reduction of TANGO transcription during tumor development

Nine human melanoma cell lines (Fig. 1a) were evaluated for the expression of TANGO using quantitative RT-PCR and compared to human primary melanocytes (NHEM). Strong reduction of TANGO expression was found in all melanoma cell lines when compared to NHEM. In cell lines derived from metastases of malignant melanomas (Mel Im, SK Mel 3, SK Mel 28 and HTZ19) expression of TANGO was more reduced than in primary melanoma cell lines (Mel Ei, HMB2, Mel Ho, Mel Wei and Mel Juso). Western blot analysis loading up to 80 μg of protein lysate per line detected no TANGO expression in all melanoma cell lines when compared to 4 NHEM cultures from different donors (Fig. 1b). The TANGO antibody is not highly sensitive and the amount of TANGO in the tumor samples is so low that it is not possible to detect TANGO in these samples on protein level.

Figure 1. TANGO expression in malignant melanoma cell lines and in situ. (a) By RT-PCR, the amount of TANGO mRNA expression was carefully quantified. All melanoma cell lines showed a strong reduction in TANGO expression. An increased reduction of TANGO was seen in all cell lines derived from metastases of malignant melanoma. (b) By Western blot, loading 80 μg of protein lysate, the amount of TANGO protein expression was quantified. TANGO protein expression could be detected in NHEMs from 4 different donors and no TANGO expression was detected in all melanoma cell lines. (c) Melanoma tissue samples were screened on TANGO mRNA expression in comparison to normal skin (basal layer). A reduction of TANGO transcription was observed in all melanoma tissues when compared to basal layer. Distant metastases showed an increased reduction of TANGO expression when compared to primary melanoma or skin metastases. (d) Immunohistological analysis of paraffin-embedded sections of human primary malignant melanoma, invasive front and skin metastases confirmed reduction of TANGO in primary tumor and invasive front and a loss of TANGO in melanoma metastases when compared to normal skin tissue. (e) Immunohistological analysis of paraffin-embedded sections of human benign and atypical nevi showed reduction of TANGO when compared to normal skin tissue, however, less dramatic than in melanoma.

To locate the loss of TANGO also in situ, RNA isolated after microdissection from melanoma tissue samples was screened by quantitative RT-PCR in comparison to normal skin (basal layer). All RT-PCR analyses were repeated at least 3 times. Reduction of TANGO transcription was observed in all melanoma tissues when compared to cells in the basal layer (Fig. 1c). Strong reduction was observed in distant metastases whereas in primary malignant melanoma TANGO was only weakly down regulated, suggesting that an increased reduction of TANGO expression correlates with metastasis. Immunhistological analysis of paraffin-embedded sections of human malignant melanoma primary tumor and metastasis confirmed a reduction of TANGO in primary tumor and showed a stronger reduction of TANGO at the invasive front. In melanoma skin metastasis, TANGO expression was completely lost when compared to normal skin tissue (Fig. 1d). Reduction of TANGO expression was also seen in different types of benign (dysplastic, compound, junctional, blue) and atypical nevi (Fig. 1e) when compared to normal skin, however, less dramatic than in melanoma. Interestingly, blue nevi were strongly positive for TANGO in epithelial and dermal areas and dysplastic nevi demonstrated a lower expression level of TANGO in the dermis (dashed arrow) than in the epidermis (black arrow).

Functional relevance of TANGO

To analyze the functional role of TANGO in tumor cells we restored expression of TANGO in the cell line Mel Im by stable transfection with a TANGO expression construct. Successful re-expression of TANGO in the cell clones K1, K3 and K5 was verified by quantitative RT-PCR (Fig. 2a), whereas no change of TANGO expression was seen in a control transfected cell clone (mock1). The expression level of TANGO in normal human epithelial melanocytes (NHEM) and normal keratinocytes was added to the diagram to see the difference of TANGO expression in normal cells in comparison to melanoma cells. On protein level, loading 40 μg protein lysate, TANGO was not detectable in either parental Mel Im or mock-transfected cells, whereas K1, K3 (data not shown) and K5 clones express high levels of the protein (Fig. 2b).

Figure 2. TANGO re-expressing cell clones. (a) Mel Im melanoma cells were stable transfected with a TANGO expression construct. Successful overexpression of TANGO in the cell clones K1, K3 and K5 was verified by quantitative RT-PCR, whereas no changes of TANGO expression were seen in a control transfected cell clone (mock 1). The TANGO expression of NHEM and keratinocytes was added to the diagram. ***p < 0.001. (b) 40 μg of cell lysate from TANGO re-expressing cell clones K1 and K5 were tested on their TANGO protein expression level by Western blotting. β-Actin was used for loading control. TANGO was not detectable in either parental Mel Im or mock-transfected cells, whereas K1 and K5 clones express high levels of the protein.

To study the functional relevance of TANGO expression, anchorage-independent growth, proliferation, invasion, migration and attachment was analyzed in the sense TANGO cell clones (Fig. 3). First, we tested the effect of TANGO on melanoma cell tumorigenicity in an anchorage-independent growth assay. As shown in Figure 3a, TANGO re-expression reduced colony formation in sense TANGO clones K1, K3 and K5, whereas the number of colonies was not changed between the cell clones and the control (mock) or parental Mel Im cell line after 10 days of incubation. Therefore, only the diameter of the colonies was measured, calculated in percentages and photographed. TANGO re-expressing clones were then tested for their proliferation capacity. Compared to the parental cell line Mel Im, sense TANGO cell clones did not exhibit any modification in their proliferative rate (data not shown). Additionally, the influence of TANGO re-expression in melanoma cells was examined on several aspects of tumor cell behavior in vitro. As shown in Figure 3b, TANGO re-expressing clones exhibited a strongly reduced capacity to penetrate Matrigel in a Boyden chamber assay, as compared to both untransfected and mock-transfected parental Mel Im cells. Migration performed in a gelatine-coated Boyden chamber assay showed a reduction in migration capacity of the cell clones (Fig. 3c). Cell migration, determined additionally in a scratch assay, was similarly reduced in TANGO re-expressing cell clones (Fig. 3d). In addition, we determined the relative attachment of Mel Im, mock control and sense TANGO cell clones to extracellular matrix proteins such as human fibronectin, human vitronectin and human collagen IV. Adhesion of sense TANGO cell clones to either fibronectin, collagen IV or vitronectin was not modified when compared to parental Mel Im or mock control (data not shown).

Figure 3. Functional assays with TANGO re-expressing cell clones. (a) Anchorage-independent growth assay. TANGO overexpression reduced colony formation for sense TANGO clones K1, K3 and K5. The diameter of 20 colonies was measured and photographed; the mean value was calculated and shown in an additional diagram. *p < 0.05; **p < 0.01; ***p < 0.001. The number of colonies between the mock control and the sense TANGO cell clones was not changed after an incubation of 10 days and therefore not shown. (b) Boyden chamber invasion assay. TANGO re-expressing clones exhibited a reduced capacity to penetrate Matrigel, as compared to both untransfected and mock-transfected parental Mel Im cells. ***p < 0.001. (c) Boyden chamber migration assay. Migration as estimated in a gelatine-coated Boyden chamber assay showed a significantly reduced migration capacity of the cell clones. ***p < 0.001. (d) Scratch assay. Cell migration, as estimated additionally in a scratch assay, was as well reduced by TANGO re-expressing cell clones. Migration capacity calculated in percentages compared to Mel Im (set as 100%) was added.

Additionally, we generated Mel Im cell clones with a reduced TANGO expression on mRNA level when compared to the parental Mel Im cell line and mock-transfected cells (mock 2) (Fig. 4). We used Mel Im cell line for generating as TANGO cell clones because we wanted cell clones with a complete loss of TANGO expression. Since the expression of TANGO in Mel Im cells is already very low, we decided to use this cell line for complete TANGO reduction. Additionally, Mel Im cells are well established in our laboratory for successful stable transfection experiments. The expression of TANGO in NHEM and keratinocytes was added again to the diagram. The antisense TANGO cell clones K2, K7, K8 and K10 were subsequently tested in the same functional in vitro assays as the sense TANGO clones before. We expected that a complete loss of TANGO could give us additional information about the functional relevance of TANGO.

Figure 4. TANGO nonexpressing cell clones. Loss of TANGO mRNA expression in the cell clones K2, K7, K8 and K10 was verified by quantitative RT-PCR, whereas no changes of TANGO expression was seen in a control transfected cell clone (mock 2) when compared to Mel Im parental cell line. ***p < 0.001. NHEM and keratinocyte expression of TANGO was added to the diagram.

Anchorage independent growth showed an increase in the growth of the antisense TANGO cell clones when compared to the parental control Mel Im and mock 2 cells, whereas the number of colonies was again not changed between the cell clones and the control (mock) or parental Mel Im cell line after 10 days of incubation (Fig. 5a).

Figure 5. Functional assays with TANGO nonexpressing cell clones. (a) Anchorage independent growth assay. A significant increase of colony formation of antisense TANGO cell clones K2, K7, K8 and K10 was measured when compared to the parental control Mel Im and mock 2 cells. The diameter of 20 colonies was measured and the mean value was calculated and added in an additional diagram. (b) Boyden chamber invasion assay. Mel Im as TANGO clones exhibited a strongly increased capacity to penetrate Matrigel, as compared to both untransfected and mock-transfected parental Mel Im cells. (c) Boyden chamber migration assay. Migration as estimated in a gelatine-coated Boyden chamber assay showed an increased migration capacity of the cell clones K7, K8 and K10. The migration capacity of K2 was not significant (ns). (d) Scratch assay. Cell migration, as estimated additionally in a scratch assay, was increased by loss of TANGO. Total migration of mock 2 after 48 h was set as 100%. Migration of the as Tango cell clones compared to mock 2 was added. (e) Scratch assay. Cell migration after scratching with a pipette tip in a definite array was measured after 24 and 48 hr. Total migration of mock 2 after 48 hr was set as 100%. Migration of the as Tango cell clones compared to mock 2 was added (black bars). Migration of these cell clones was reduced effectively when compared to the mock 2 control after adding external recombinant TANGO to the cells (grey bars).

Proliferation of antisense TANGO clones was measured by XTT method but no significant changes in proliferation were observed between the cell clones and Mel Im parental cells (data not shown). Both the sense and antisense TANGO cell clones did not exhibit any modification in their proliferative rate, suggesting that TANGO has no direct effect on cell proliferation.

On the other hand, the invasive and migratory capacity of these antisense TANGO cell clones was even stronger than in the original Mel Im cell line or mock 2 controls (Figs. 5b and 5c). Wound healing assay indicated a higher ability to migrate of these antisense TANGO cell clones when compared to mock 2 control cells (Fig. 5d). Additionally, we treated the antisense TANGO cell clones with recombinant TANGO. Wound healing in these cell clones was reduced effectively when compared to the mock 2 control after adding external recombinant TANGO to the cells (Fig. 5e). All experiments have been repeated independently for 3 times.

Discussion

In our study, we investigated the expression pattern of TANGO in malignant melanoma. We initiated our study to explore the possibility that loss or downregulation of TANGO expression may play a role in tumor development or progression. Specifically, we were interested to know whether TANGO transcription is altered during tumor development and progression and whether loss or re-expression of TANGO may change the migratory activity of the tumor cells.

Our data indicated that loss of TANGO expression has an impact on tumor development and progression. We revealed that TANGO expression on mRNA level was reduced in all 9 analyzed melanoma cell lines. On protein level we detected expression of TANGO in none of the 9 melanoma cell lines. We have repeated the Western blot loading 80 μg protein lysate on the SDS-gel and incubating the membrane for more than 5 hr but it was not possible to detect TANGO in the melanoma samples. We think that the TANGO antibody is not highly sensitive and the amount of TANGO in the tumor cell lines is so low that it is not possible to detect TANGO in these samples. On mRNA level it was possible to detect TANGO using quantitative RT-PCR. However, this method is very sensitive and very small amounts of mRNA can be detectable.

In cell lines derived from metastases of malignant melanomas, expression of TANGO was more reduced than in primary melanoma cell lines. Similar reduction or loss of TANGO expression on mRNA and protein level was also found in situ in malignant melanoma when compared to normal tissue samples. The process of TANGO reduction correlates with advanced stage of the disease. Primary melanoma tumors show an initial reduction of TANGO expression and strong reduction or loss of TANGO was detected in skin and distant metastases. Studies on epigenetic modifications have shown that silencing of tumor-related genes due to methylation plays an important role in melanoma development.8 However, our analysis revealed that loss of TANGO expression not correlates with epigenetic modification of the TANGO promoter through methylation-mediated transcriptional silencing (data not shown). Another possibility of regulating TANGO expression is on transcriptional level. Usually, changes in transcription factor expression or activity can lead to more than just 1 downstream modification, as transcription factors are higher, thinking in a hierarchical way of expression control.9 AP-1, AP-2-α, CREB, CtBP, PAX3, SKI, Snail and STAT are transcription factors, which are often deregulated in melanoma development and progression.9 Further studies have to reveal how TANGO is regulated and what transcription factors are involved in this process.

In nevi already reduction of TANGO expression was seen. However, in dysplastic nevi one gets the impression that reduction of TANGO is influenced by microenvironment. Interestingly, downregulation of TANGO expression presents to be associated with localization of the melanocytic cells. In dermal areas of dysplastic nevi and in invasive melanoma, downregulation of TANGO occurred more frequently than in epidermal areas of dysplastic nevi or primary tumors, respectively. Phenomena like this have been published before e.g. for IL8 where the expression level was influenced by the microenvironment.10 It is known that factors produced by fibroblasts can influence gene expression in melanoma cells, which may lead to downregulation of TANGO expression.

In functional assays we could show that induction of expression of TANGO protein resulted in a significant decrease in motility and invasive potential and that a reduction of TANGO expression promotes invasion and migration. A detailed molecular explanation of how TANGO is involved in melanoma migration and invasion process is not possible so far. Recently, we published that MIA, a homologous gene to TANGO, can directly bind to integrin α4 β1 and α5 β1 and modulates integrin activity negatively.11 This mechanism can additionally support and regulate the detachment process. It is possible that TANGO as a homologous protein to MIA can modulate integrin activity which in turn promotes migration and invasion of the cells. However, experiments must be performed to confirm these hypotheses.

The function of TANGO in melanoma seems to be in clear contrast to the function of the family member MIA. Here, several studies revealed strong upregulation of MIA expression in melanoma.12 It was shown in previous studies that MIA expression in vivo correlates with progressive malignancy of melanocytic tumors.12, 13 Additionally, in recent studies enhanced MIA protein levels were measured specifically in the serum of patients with metastatic melanomas.14In vitro studies revealed a role of MIA in supporting the invasive and migratory potential of melanoma cells. In vivo studies in 2 animal model systems supported the importance of MIA in melanoma metastasis. MIA expression levels parallel closely the capability of melanoma cells to form metastases in syngeneic animals.15, 16

Taken together, we present the first time that TANGO is downregulated or even lost in cancer and that the process of TANGO reduction correlates with advanced stage of the disease. Therefore, we suggest a tumor suppressor role for TANGO. All together, loss of TANGO expression is frequent in advanced melanomas and metastases of melanoma and could be a prognostic marker of progression in melanoma patients. TANGO is less useful as a prognostic marker in patients with different benign or atypical nevi because the reduction of TANGO in these tissue sections was less dramatic and indicated a strong variability when compared to melanoma. Subsequent analysis is needed to further clarify the function and regulation of the TANGO protein.

Acknowledgements

We are indebted to Dr. J. Johnson (University of Munich, Germany) for providing melanoma cell lines and to Susanne Wallner and Sibylla Lodermeyer for excellent technical assistance.