Secrets to developing <i>Wnt</i>‐age melanoma revealed

The Wnt family includes 19 secreted glycoproteins that are involved in cell-fate determination, differentiation and proliferation during development and homeostasis. Wnt molecules were defined by sequence homologies and transforming activities on C57MG murine mammary epithelial cells. Basically (and simplistically), two classes can be defined. The first class comprises Wnt3A, Wnt1 and Wnt8. These factors induce axis duplication in Xenopus and are able to transform C57MG cells. They activate the so-called canonical Wnt/β-catenin signalling pathway. The second class includes Wnt5A, Wnt4 and Wnt11 that do not display the two activities described for the first class. They activate non-canonical Wnt pathways and do not increase the level of β-catenin in C57MG cells. Predictably, however, the situation is more complex and the precise pathways activated by Wnt ligands depend also on the receptors present at the membrane and the Wnt family members expressed in a given cell. Furthermore, there is significant cross-talk between the Wnt pathways, and between Wnt and other pathways, adding further complexity to the full signalling network active at any particular stage/period of development/tumorigenesis. Analysis of melanocytic tumours reveals that Wnt3A and Wnt5A are secreted, and activate the Wnt/β-catenin canonical and Wnt/PKC non-canonical pathways respectively. The canonical Wnt/β-catenin signalling pathway is implicated in a vast number of cancers, so it was not surprising to find hyperactivation/dysregulation of this pathway in melanoma (Rimm et al., 1999). The involvement of Wnt5A in transformation was discovered in the mid-1990s, even though it was not considered to be a ‘transforming Wnt’ according to its initial definition. The involvement of the non-canonical Wnt5A in motility and invasion of melanoma was revealed subsequently (Weeraratna et al., 2002). Two recent articles, by Chien et al. (2009) and Dissanayake et al. (2008) clearly describe the specific roles of Wnt3A and Wnt5A, and highlight the importance of multiple Wnt signalling pathways in melanomagenesis. Canonical Wnt signalling triggers a cascade of events, leading ultimately to an increase of nuclear β-catenin; the latter may also happen as a consequence of activating mutations of β-catenin itself. The detection of β-catenin in the nucleus indicates that the Wnt/β-catenin signalling pathway is activated, and this occurs in about 30% of melanomas (Rimm et al., 1999). However, only 3.3% of melanoma biopsies and 8.5% of melanoma cell lines were found to carry β-catenin mutations (Larue and Delmas, 2006). The mechanism responsible for the localization and/or the maintenance of β-catenin in the nucleus remains largely unknown. Nuclear β-catenin can potentially activate a large number of target genes, some of them being ubiquitously expressed, such as cyclin D1 and c-Myc, some being cell lineage-restricted genes, such as Brn2 and other melanocyte-specific genes including Mitf-M and Dct. In their recent study, Chien et al. (2009) re-evaluated the importance of β-catenin signalling in melanoma. First, they analysed a tissue microarray of about 350 human melanoma tumour cores composed of 100 primary tumours and 250 recurrences/metastases. They show that elevated levels of β-catenin in the nucleus of primary and metastatic melanoma correlate with a better prognosis. Moreover, they used Ki-67 and PCNA to estimate the percentage of cycling cells in all specimens. They concluded from this correlative analysis that the presence of β-catenin in the nucleus is associated with decreased proliferation. The relevance of this study depends, in large part, on the long (30 yr) clinical follow-up of a cohort of melanoma patients. It is a remarkable achievement that shows the insight of the pathologists involved in this work who understood, as early as the 1970s, the importance of creating a collection of primary and metastatic melanoma samples. It is noted that melanoma presents as nodular, lentigal, acral, mucosal and other forms, and the prognosis differs according to the type of melanoma. It would be certainly of interest to determine if any of these forms of melanoma exhibits nuclear β-catenin more frequently than others. In the second part of their work, Chien et al. recapitulate in vitro the action of β-catenin on proliferation, using the B16-F1 murine melanoma cell line. They transduced melanoma cells with lentivirus expressing Wnt3A, Wnt5A and GFP control, and evaluated the consequences of expression of these proteins on proliferation and transcription profiles. They concluded that proliferation, evaluated using MTT and cell counts, is specifically inhibited by the presence of Wnt3A. At the molecular level, classical β-catenin targets are transcriptionally activated, as are proteins found to be specifically expressed in melanocytes and neural crest cells. The low activation of M-Mitf appears not to contribute to the inhibition of proliferation (Carreira et al., 2005), therefore the molecular mechanism of the reduction of proliferation in this model cell system remains unclear. It should be emphasized that the anti-proliferative effect of nuclear β-catenin, recently reported in transformed cells (Chien et al., 2009), was initially shown to occur during development (Delmas et al., 2007). β-catenin targets can be pro-proliferative, such as Myc, cyclinD1 or Brn2 and anti-proliferative such as M-Mitf. Therefore, there may be a pro- or anti-proliferative function according to the specific β-catenin target that is activated. Few months ago, Dissanayake et al. (2008) noticed that Wnt5a is more strongly expressed in motile melanoma samples than in others. This observation fits perfectly well with their previous studies showing that Wnt5a expression is associated with more aggressive and metastatic behaviour (Weeraratna et al., 2002). In addition, they found that the expression of various tumour antigens, such as TYRP-1, DCT, gp100 and MART-1, is inversely correlated with Wnt5a expression in various studies performed. Dissanayake et al. first used melanoma biopsies to confirm these results, obtained with melanoma cell lines; the ratio of Wnt5A-positive to MART-1-negative tumours increased dramatically as tumours progressed with the exception of lymph node metastasis. The correlation of Wnt5A and MART-1 expression in lymph node melanoma metastasis has not yet been explained and will certainly be investigated in the near future. The authors deciphered the molecular pathway linking Wnt5A and MART-1; they found that Wnt5A activates PKC, which in turn activates STAT3 by phosphorylation. STAT3, by an unknown mechanism, downregulates PAX3 which leads to a direct downregulation of MITF and finally MART-1. This pathway linking WNT5A to MART-1 is not fully characterized but it can occur in vitro in some contexts. These various findings collectively may explain the limited success of immunotherapy: melanoma cells express a high level of Wnt5A and in consequence low level of tumour antigens, leading to poor CTL response. The possible cross-talk between Wnt5A and the Wnt canonical pathway reinforces the downregulation of tumour antigens because Wnt5A downregulates β-catenin, and this leads to the downregulation of M-MITF, and therefore tumour antigens. Perhaps, as a tumour antigen, the regulation of MART-1 by Wnt5A could be used therapeutically to induce a better immune anti-tumour response. The authors propose that novel therapies including downregulation of Wnt5A before immunotherapy may lead to the enhancement of targeted immunotherapy for patients with melanoma metastasis. Both groups suggest therapies based on the induction of differentiation, leading to the production of melanocytic antigens and inhibition of proliferation. It is certainly important to control proliferation during melanomagenesis but it is not the only cellular mechanism involved. Moreover, there is another side to the proliferation coin. Pushing the cells onwards to differentiation may lead to a cytostatic effect on these cells and to the induction of an increased resistance to cytotoxic therapies as non-cycling cells are generally more resistant to treatment. Another approach could be to force the cells back into the division cycle so as to render them sensitive to therapy in a cytotoxic manner. Will we occasionally have to flip the coin as we flip bottles to activate maturation of vintage wine?

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.