Urinary bladder cancer is the fourth most frequent cancer in men in the world today
[1–3]. Due to its great plasticity and biological variability with respect to multifocality, recurrence, and invasion comportment, it is difficult to predict the behavior of an initially noninvasive UC
[2, 3]. This feature makes the diagnosis of UC challenging, not only for its early detection, but also for accurate prediction of tumor recurrence and progression.
Multiple biomarkers used for other cancers have been described with potential diagnostic or prognostic value for UC as well, including p53, cytokeratin 20, E-cadherin, Ki67, CD44 and survivin
[27, 28]. Currently a combination of these immunohistochemical biomarkers assayed within panel is the best way to characterize subtypes and predict the behavior/aggresiveness of UC
. More recently, other promising biomarkers for UC have been reported. In a large retrospective exploratory study, expression of microRNA-100 was negatively associated with stage, recurrence, prognosis and death of patients with bladder cancer
. In another large study, the cell surface zinc-dependent metalloprotease CD10 was strongly correlated with tumor grade and stage and possibly associated with tumor progression in bladder cancer pathogenesis
. In contrast to initial studies correlating increased COX-2 expression and recurrence of urinary bladder cancer, a large study focused on non-muscle invasive bladder cancer found that lower levels of COX-2 and tumor-infiltrating lymphocytes were predictive of recurrence, suggesting the need of close follow-up and adjuvant therapy for these patients
. In the current study Wnt5a protein expression by malignant cells was observed in all 33 samples included in the study; however, intensities, and hence protein levels, varied, correlating with both grade and stage and suggesting that Wnt5a may also serve as a biomarker of UC. To prove the predictive nature of Wnt5a as a diagnostic or prognostic biomarker of UC, further investigation into the relationship between Wnt5a expression by the tumor and clinical progression of the disease in a large number of patients, is required.
Expression of Wnt5a by UC tumor cells may also be a predictor of an aggressive subgroup of UC. We found that weak Wnt5a expression was predominantly seen in low grade UC samples and all were in a noninvasive (pTa) stage. This suggests that low Wnt5a expression may be indicative of non-aggressive tumors. In contrast, moderate to strong Wnt5a expression was seen in all three pathological stages, but was usually associated with high grade, high stage tumors. Thus, moderate to strong expression of Wnt5a is likely to indicate an aggressive tumor or transformation to an invasive phenotype. In support of this hypothesis previous studies have reported the role of Wnt5a as a pro-angiogenic gene
[32, 33] and also the importance of the Wnt signaling pathway in epithelial-mesenchymal transition (EMT) of cancer
. In addition, Huang CL et al. reported the association of Wnt5a with tumor proliferation in non-small-cell lung cancer (NSCLC); the authors concluded that overexpression of Wnt5a could increase aggressiveness in NSCLC
The present study also assessed Wnt5a mRNA expression in two UC cell lines, both from high grade, poorly differentiated tumors. Interestingly, the expression differed significantly between them, with one of the cancer cell lines expressing much higher levels and the other expressing much lower levels of Wnt5a mRNA in comparison to the normal urothelial cell line. This result is similar to the result published many years ago by Bui et al. in which they described different Wnt5a mRNA expression levels for different UC cell lines
. In that study the authors suggested that Wnt5a may modulate cell shape and cell migration
In support of a role of Wnt5a in cell migration, Wnt5a has been associated with migration ability in melanoma
. Our group previously reported constitutive Wnt5a expression in human pancreatic cancer and malignant melanoma cell lines and suggested the role of Wnt5a in growth and migration in both of these cancer cell lines
. In this study we investigated migration response to exogenously applied Wnt5a in the two UC cell lines. We found that the T24 cell line, which has a low level of Wnt5a transcription, showed a significant increase in migration rate while J82, with high Wnt5a transcription, did not significantly respond to the application of Wnt5a. It is possible that J82, with high Wnt5a RNA, is already maximally saturated by Wnt5a and cannot respond further, while T24, with low Wnt5a RNA, is sub-maximally saturated and thus capable of responding to exogenous Wnt5a. Alternatively, it is possible that different cell lines have different responses to Wnt5a due to expression of different Wnt5a receptors. It is well known that the interaction between Wnt ligands and their receptors determine the signaling that will be activated, and multiple receptors may respond to Wnt5a signaling, each leading to activation of different downstream signaling pathways and effects
[17, 37, 38]. Therefore, different responses might be associated with the presence of different receptors for Wnt5a ligand in each tumor cell line.