A Wnt-er Migration: The Confusing Role of β-Catenin in Melanoma Metastasis

See allHide authors and affiliations

Science Signaling  26 Mar 2013:
Vol. 6, Issue 268, pp. pe11
DOI: 10.1126/scisignal.2004114


Wnt signaling in melanoma is complex, requiring the coordinate expression of multiple players. Depending on the context of receptors and co-receptors that are present, Wnt proteins may signal through either canonical or noncanonical pathways. The role of β-catenin in melanoma metastasis remains unclear; however, a new study points to the roles of Wnt5A and ARF6 in driving β-catenin expression and melanoma metastasis. Here, we discuss this finding and how it may help us define different subpopulations of melanoma cells that could have different outcomes, as well as different responses to therapy.

During development of the skin, the canonical Wnt signaling pathway guides the migration and expansion of neural crest-derived melanocytes (1). β-catenin signaling determines the fate of melanoblasts through the activation of the micropthalmia transcription factor, MITF (2, 3). During the transition to melanoma, β-catenin signaling is critical for several processes leading to the transformation of melanocytes, including the bypassing of melanocyte senescence, which occurs through the β-catenin–mediated suppression of the cell cycle protein p16 (4). In addition, β-catenin increases the proliferation and growth of melanoma cells through interactions with the POU domain transcription factor, Brn2 (5). However, the role of β-catenin in melanoma metastasis remains largely unclear. Some studies have indicated that β-catenin suppresses invasion of melanoma cells and that the loss of β-catenin predicts a poor rate of survival in melanoma patients (69). Other studies have shown that the increased abundance or stabilization of β-catenin leads to increased melanoma metastasis, both in vitro and in vivo (1012). Until the recent study by Grossmann et al. (13), these observations appeared to be largely independent of the noncanonical Wnt signaling pathway, specifically Wnt5A.

Wnt5A guides the migration and metastasis of melanoma and, unlike canonical Wnt ligands, signals through protein kinase C and calcium-dependent enzymes but not β-catenin (1418). In fact, Wnt5A has been shown to actively signal to decrease the activity of β-catenin. This is thought to occur in a manner independent of glycogen synthase kinase 3β, through the activation of the ubiquitin ligase SIAH2 (19, 20). However, in the aforementioned study, Grossmann et al. show that in a subset of melanoma cell lines, Wnt5A may partner with canonical Frizzled receptors (Fzd4) and co-receptors (LRP6) to increase β-catenin signaling. Thus, depending on the context of receptors and co-receptors available, Wnt proteins can be quite promiscuous and signal to activate both canonical and noncanonical pathways. Activation of canonical Wnt signaling by noncanonical Wnt5A has been shown in other systems (21), but not in melanoma and with endogenous amounts of proteins. This finding calls for a better understanding of the subpopulations of melanoma cells being studied and their Wnt receptor and co-receptor repertoires. For example, in cases in which ROR2 is the predominant co-receptor, it may be that Wnt5A signals to degrade β-catenin; when the LRP6 co-receptor is dominant, Wnt5A may signal instead to activate β-catenin (Fig. 1).

Fig. 1 Hypothetical model of Wnt5A degradation of β-catenin.

(Left) Several studies indicate that Wnt5A signals through the receptors Fzd2 or Fzd5 to degrade β-catenin. (Right) The study by Grossmann et al. indicates that Wnt5A can also pair with Fzd4 and the co-receptor LRP6 to stabilize β-catenin. Both outcomes result in an increase in melanoma cell invasion. This study highlights the need for a clearer understanding of the Wnt pathway in melanoma.


In addition to the Wnt receptors and co-receptors, the presence of other downstream molecules may be critical in determining the ability of noncanonical Wnts to activate β-catenin. Grossmann et al. demonstrate that the small guanosine triphosphatase ARF6 plays a critical role in the activation and nuclear translocation of β-catenin. In addition, Grossmann et al. show that ARF6 is necessary for invasion in these melanoma cells and that overexpressing stabilized β-catenin also promoted metastasis in LOX melanoma cells. This is consistent with the previous finding that stabilized β-catenin promotes metastasis of melanoma (10). One question that has not yet been answered is whether the knockdown of β-catenin in the presence of recombinant Wnt5A or ARF6 overexpression would inhibit invasion. Because ARF6 plays such critical roles in modulating cell-cell interactions, its effects likely extend beyond that of β-catenin.

Another intriguing possibility invoked by these data is that this particular population of melanoma cells, with high levels of Wnt5A and β-catenin, represents a very defined stage of melanoma progression. The phenotype-switching model of melanoma postulates that melanomas switch between proliferating or invading (“growing or going”), but they rarely do both simultaneously (22). This model was based on a study that analyzed gene expression in a large set of melanoma cells that the authors defined as proliferative or invasive based on their expression profiles (23). A follow-up study by this group confirmed that the expression data correlated to proliferative compared with invasive phenotypes in vitro and in vivo (22). In this model, highly proliferative melanoma cells typically had low WNT5A and high CTNNB1 expression, whereas more invasive melanomas expressed high WNT5A and low CTNNB1 (23). However, there was a third group that had moderately high abundance of both (23). Although this group was not as invasive as the high-WNT5A, low-CTNNB1 subset of melanoma cells, this group was more invasive than the low-WNT5A, high-CTNNB1 subset. It may be that this phenotype of high Wnt5A and β-catenin abundance represents a subset of cells that is moving from a proliferative to a metastatic state, at the initiation of metastasis. Alternatively, to colonize the site of metastasis, highly invasive cells may need to switch back to a proliferative phenotype, for which they would likely require β-catenin.

β-catenin is critical not only for tumor development and, potentially, metastasis but also because this protein has a role in the efficacy of targeted therapy for melanoma. The mutant form of the oncogene BRAF [BRAFV600E (V600E: Val600→Glu600)] is present in a large percentage of melanomas (24), a discovery that led to the development of drugs that target this mutation and revolutionized therapy in a disease that previously had few effective therapies. (25) Unfortunately, after dramatic remissions, patients often relapse, and understanding how this drug resistance develops is a key area of investigation. Melanoma cells with high β-catenin abundance are more sensitive to BRAF inhibition, and resistance to BRAF inhibition could be achieved through knockdown of β-catenin (26). These data suggest, therefore, that the presence of β-catenin may have a positive outcome for melanoma patients, further underscoring the complexity of the role of β-catenin in melanoma.

Wnt5A is a clear driver of metastasis in melanoma, whether or not it increases β-catenin levels. The data from Grossmann et al. highlight the necessity of accurately identifying the different subpopulations of melanoma cells, their Wnt and β-catenin status, as well the status of the different key co-receptors involved (especially ROR2 and LRP6). This finding is important because it may identify a signature of both invasive cells and cells that might be more resistant to therapy. Targeting Wnt5A in melanoma, regardless of β-catenin status, will likely have a positive effect because, in either case, metastasis will be reduced. The available data suggest that Wnt5A inhibition may also increase the efficacy of targeted therapy. The implications of inhibiting Wnt5A and how it affects therapy resistance are currently under investigation in our laboratory. Overall, understanding the complex interplay between noncanonical and canonical Wnt signaling in melanoma metastasis is essential to identify points of intervention that could both inhibit metastasis and increase the efficacy of targeted therapy.


View Abstract

Stay Connected to Science Signaling

Navigate This Article