Emerging roles and biomarker potential of WNT6 in human cancers

The WNT6 ligand is a well-known activator of the WNT signaling pathway, considered a vital player in several important physiologic processes during embryonic development and maintaining homeostasis throughout life, regulating the proliferation and differentiation of multiple stem/progenitor cell types. More recently, as it is the case for many key molecular regulators of embryonic development, dysregulation of WNT6 has been implicated in cancer development and progression in multiple studies. In this review, we overview the most significant recent findings regarding WNT6 in the context of human malignancies, exploring its influence on multiple dimensions of tumor pathophysiology and highlighting the putative underlying WNT6-associated molecular mechanisms. We also discuss the potential clinical implications of WNT6 as a prognostic and therapeutic biomarker. This critical review highlights the emerging relevance of WNT6 in multiple human cancers, and its potential as a clinically-useful biomarker, addressing key unanswered questions that could lead to new opportunities in patient diagnosis, stratification, and the development of rationally-designed precision therapies.

WNT ligands comprise a family of secreted molecules that are able to activate the WNT signaling pathway, which have raised a multidisciplinary interest due to their wide range of physiological functions [1]. The human genome contains 19 WNT genes, conserved among mammals and also found in several simpler multicellular organisms. WNT ligands are conserved cysteine-rich proteins of approximately 40 kDa, which present similar structural motifs and general physiological functions [1].

In the endoplasmic reticulum, WNTs are translated and undergo post-transcriptional glycosylation and PORCN-mediated palmitoylation [2,3,4]. These lipid-modified WNT ligands are then transported to the cell membrane by the Wntless protein and secreted into the extracellular space, ultimately interacting with various receptor complexes [5].

Several receptors and co-receptors interact with WNT ligands to initiate signal transduction [6,7,8,9,10,11,12,13]. These interactions can activate different signaling pathways, depending on the specific ligand-receptor combination, in a cell- and stage-dependent manner [5]. It has been suggested that rather than linear and distinct signaling pathways, multiple WNT signaling variations may be activated simultaneously, in a complex integration of various inputs [14,15,16].

WNT signaling has been historically divided into canonical (β-catenin-dependent) and non-canonical (β-catenin-independent) WNT pathways [17, 18]. The β-catenin-dependent WNT pathway, the most widely studied [17], involves the stabilization and nuclear translocation of β-catenin, which then modulates gene expression via TCF/LEF family of transcription factors [19,20,21,22]. Non-canonical pathways are a diverse and less-characterized group of pathways that act independently of β-catenin [18, 23], of which the most well-defined are planar cell polarity (PCP) and the WNT-Ca2 + pathways [5].

The WNT signaling pathways play a multitude of roles in various stages of development and throughout life, including embryogenesis, organogenesis and tissue homeostasis [24,25,26,27,28,29,30,31]. Remarkably, recent research continues to uncover novel roles to these pathways in diverse physiological processes, such as neural development, osteoblast metabolic regulation, and cardiac physiology [32,33,34,35].

Conversely, dysregulation of WNT signaling often leads to disease, including cancer [36,37,38]. Aberrant β-catenin expression, along with the dysregulation of other WNT pathway components, has been implicated in the oncogenic process of several human tumors [39,40,41,42,43,44,45,46,47,48,49,50]. Notably, the first mammalian WNT gene, WNT1, was identified in the context of a virally-induced mouse breast tumor [51,52,53,54], and later recognized as an oncogene [55]. Since then, numerous WNT genes have been implicated in various cancers [56,57,58,59].

Until recently, WNT6 was among the less-explored members of the WNT ligand family. However, it has emerged as a significant molecule in both physiological and pathological contexts (as reviewed in Wei et al. [60]). WNT6 activates WNT signaling pathways and plays pivotal roles in various human physiological processes, including embryonic development (e.g., neural crest induction [61]), organogenesis of heart muscle and kidneys [62, 63], maintenance and differentiation of adult progenitor cells [64, 65], and in immune contexts (e.g., macrophage M2-like polarization and proliferation [66]). Understandably, as a crucial orchestrator of several homeostatic processes, its dysregulation has been recently linked to various oncogenic processes.

In recent years, studies have increasingly described WNT6 as a key player in the development, progression, and aggressiveness of several tumor types. The first studies exploring WNT6 in cancer reported its upregulation in colorectal and gastric cancer in vitro models [67, 68]. Since then, WNT6 has been studied across a wide range of cancers, where it has been shown to activate multiple WNT signaling variations (Fig. 1). Significant findings have been reported at the molecular level, including gene expression and epigenetic alterations, as well as on dysregulation of downstream pathways. These studies also drew connections between WNT6 and key clinical features of cancer patients (Fig. 2 and Table 1). Below are summarized the major findings related to the relevance of WNT6 in various human cancers, and discussed the future challenges for the translation of this body of knowledge to the clinical settings.

WNT signaling pathways affected by WNT6 in cancer. Both β-catenin-dependent (canonical [69,70,71,72,73,74,75,76]) and -independent (non-canonical [74, 75]) WNT signaling pathways have been linked to WNT6 in cancer, including ROR2, RYK, PTK7, and planar cell polarity (PCP). Upon engagement with its receptor molecules, WNT6 may trigger multiple signaling cascades, which ultimately result in the activation and upregulation of specific target genes

Full size image

Summary of key findings regarding novel roles and clinical relevance of WNT6 in cancer, highlighting functional effects affecting tumor aggressiveness, molecular regulatory and effector mechanisms of WNT6 in each tumor type, and its relevance in the clinical setting. Multiple studies described WNT6 as an oncogenic factor in various tumor types, being associated with increased tumor aggressiveness features and worse patient prognosis. The regulation of WNT6 expression by different molecules and its ability to activate WNT/β-catenin signaling, as well as other relevant mechanisms, underscore its multiple molecular partners and identify potential therapeutic opportunities

Full size image

Colorectal cancer

The majority of colorectal cancer cases present alterations in the WNT signaling pathway, mostly through APC inactivating mutations [37], compromising the negative regulation of this pathway. This association has led to extensive research on this pathway’s influence in colorectal cancer, with numerous studies describing the relevance of abnormal WNT ligand expression in this context [98,99,100,101].

Initial reports over 20 years ago identified WNT6 and WNT10A to be clustered in close proximity in the human genome, and highly co-expressed in the SW480 colorectal tumor cell line [67]. Subsequently, SNPs within the WNT6 gene were linked to colorectal adenoma risk, namely the rs6747776 variant, where individuals with the CC genotype had an increased risk compared to those with the GG genotype [102]. Moreover, the risk associated with this polymorphism was found to be further influenced by dietary factors, particularly the proportion of calories derived from fat, with the greatest risk observed for those with one or more minor alleles that consumed less than 30% of daily calories from fat [102].

In clinical samples from patients with sporadic colorectal cancer, WNT6 was found to be hypermethylated along with several other genes related to β-catenin-dependent WNT signaling [103]. Posteriorly, WNT6 expression was linked to increased features of tumor aggressiveness, demonstrated by increased in vitro cell proliferation, viability, migration, and decreased cell cycle arrest [69,70,71, 77]. In vivo, it was implicated in mechanisms that promote tumor subcutaneous growth [69] and orthotopical tumorigenesis [70]. Concordantly, a study exploiting samples from patients in clinical settings have also established a prognostic value for WNT6, being associated with reduced disease-free survival [70]. Curiously, another study reported a prevalent deleterious WNT6 mutation in locally recurrent rectal cancer, and suggested that WNT6 protein and mRNA levels were decreased in tumor samples compared to adjacent non-tumor tissues in rectal adenocarcinoma and colorectal cancer patients [104].

WNT6 has been associated to oncogenic molecules in colorectal cancer [69, 70]. Specifically, the transcription factor PLAGL2, a zinc-finger protein, was found to interact with WNT6 promoter region, promoting its expression. This interaction ultimately, activates the β-catenin-dependent WNT signaling (Fig. 2), through which it may be inciting tumor aggressiveness [69]. Additionally, nuclear NR4A2, induced by prostaglandin E2 (PGE2) – an inflammatory molecule recurrently implicated in colorectal cancer and often linked to WNT signaling [105, 106] – has also been shown to interact with WNT6 promoter region, inducing its expression and, consequently, activating β-catenin-dependent WNT signaling (Fig. 2) [70]. This mechanism can be suppressed by aspirin treatment, which inhibits cyclooxygenase signaling, decreasing PGE2, NR4A2 and WNT6, thus reducing colorectal cancer tumorigenesis [70]. In colorectal cancer, PLAGL2 was also reported to promote tumor aggressiveness in vitro and in vivo, via a β-catenin-dependent mechanism, in an interplay with AKT/GSK-3β signaling [107]. Similarly, NR4A2, induced by PGE2, has been shown to promote chemotherapy resistance in vitro and is considered an independent predictor of unfavorable prognosis in colorectal cancer [108]. More recently, the long non-coding RNA (lncRNA) PVT1 was shown to interact with the hsa-miR-1207-5p microRNA, which in turn can bind WNT6, activating the β-catenin-dependent WNT pathway (Fig. 2) [71]. These ncRNAs were reported to be increased in patient samples and associated with increased tumor aggressiveness in vitro [71]. Nevertheless, complementary molecular and functional assays are still necessary to more precisely define the molecular mechanisms through which WNT6 influences colorectal cancer, and the phenotypic impact of its manipulation.

Moreover, in vitro, WNT6 expression has been associated to a decreased expression of the pro-apoptotic molecule Bax and an increase in the expression of caspase-3 precursor, indicating a possible inhibition of cell apoptosis [77]. WNT6 expression has been also linked to an increased MMP2 expression, a molecule often involved in epithelial-mesenchymal transition, a crucial event in metastasis, frequently reported to be promoted by β-catenin-dependent WNT signaling [77].

Colorectal cancer metastasis prevails as a significant colorectal cancer-driven cause of death, with the liver being the most usual long distance metastatic site [109]. Interestingly, WNT6 was found to be highly expressed in colorectal liver metastases, primarily localized in the cytoplasm, and considered relevant in patients’ prognosis [79]. High WNT6 expression was associated to a higher mortality rate upon liver resection, and it was defined as an independent negative predictor of patients’ 5-year overall survival, more so in patients with low-risk of recurrence [79]. In addition, WNT6 expression was significantly lower in colorectal cancer patients with liver metastasis who had received preoperative chemotherapy than in those who had not received it [79].

Globally, this body of data positions WNT6 as a novel key molecule with biomarker potential in colorectal cancer, warranting further studies to better define its underlying molecular mechanisms, which may be amenable to rational targeted therapeutic interventions.

Gastric cancer

The aberrant expression of WNT ligands, including WNT6, and the subsequent overactivation of WNT signaling, have been frequently reported in gastric cancer [72, 110,111,112,113].

WNT6 was found to be expressed in both gastric cancer cell lines derived from primary tumors and from distant metastasis [68, 72]. Moreover, in a subset of patient tumor samples, WNT6 mRNA was upregulated in comparison to non-tumoral gastric tissue and, interestingly, WNT6 protein was upregulated as well in the tumor tissue lysates of 3 out of 4 patients [72]. These findings would benefit from validation in larger cohorts to confirm this putative differential expression. WNT6 was identified in multiple subcellular locations within these cell lines, namely in the membrane, cytoplasm, and nucleus, and was also present in conditioned media from gastric cancer cell lines, confirming its secretion [72]. In normal gastric tissue, WNT6 was observed in the apical region of the foveolar epithelium and in areas of intestinal metaplasia, as well as in macrophages and plasma cells present in the lamina propria [72]. Interestingly, in a retrospective analysis, it was reported that the majority of gastric cancer patients non-responsive to preoperative standard chemotherapy regimen (epirubicin, cisplatin and 5-fluorouracil (5-FU)) exhibited high levels of WNT6 within the tumor area, whereas an elevated number of responsive patients were essentially WNT6-negative [72]. Complementarily, WNT6 was also indirectly associated with a worse patient prognosis, associating positively with tumor stage and nodal metastatic status, both important prognostic factors for poor gastric cancer patient survival [72].

WNT6 expression was also associated to caveolin-1 (CAV1), their levels varying accordingly, both in parental cell lines and upon CAV1 manipulation [72]. Interestingly, WNT6 expression and β-catenin-dependent WNT signaling activation were induced upon anthracycline treatment, which potentiated β-catenin binding to the distal promoter region of WNT6, ultimately leading to decreased chemotherapy efficacy, exclusively in cells expressing CAV1 [72]. Importantly, these effects were recapitulated in a spontaneous gastric cancer mouse model, in which epirubicin exposure upregulated both WNT6 and CAV1 expression [72]. It is suggested that chemotherapy-induced DNA damage may trigger a specific cell response involving p53 [72], a well-established regulator of CAV1 [114] and of various WNT-related genes [115, 116], leading to the upregulation of CAV1 and WNT6. Ultimately, activating β-catenin-dependent WNT signaling (Fig. 2) and inducing pro-survival genes [72].

These findings highlight an eminent role of WNT6 as an oncogenic molecule in gastric cancer, and underline its potential therapeutic interest, together with CAV1, given their upregulation upon anthracycline treatment and association with decreased sensitivity to this therapy.

Breast cancer

WNT6 was initially identified to be strongly expressed in patient-derived infiltrating ductal carcinoma samples, and in corresponding non-tumor breast epithelium samples [80], while posterior studies reported WNT6 to be upregulated in invasive ductal carcinoma tissues compared to non-tumor tissues [81, 117]. Regarding breast cancer cell lines, WNT6 was found to be strongly expressed in both estrogen-receptor positive and negative models [80, 81, 83].

It has been demonstrated that the microRNA hsa-miR-566, which has tumor suppressive functions in breast cancer, is able to downregulate WNT6 and β-catenin by targeting WNT6 3’-UTR, ultimately reducing cell viability, proliferation, and migration (Fig. 2) [81]. In addition, WNT6 expression, along with β-catenin and DVL1, were found to be increased in doxorubicin resistant breast cancer cell lines [75]. This abnormal WNT6 expression may be induced by PLAGL2, a transcription factor with known oncogenic functions in breast cancer. PLAGL2 is able to strongly bind the WNT6 promoter region, inducing its expression, and resulting in increased cell proliferation, invasion and migration (Fig. 2) [75]. In fact, in PLAGL2 knockdown doxorubicin-resistant breast cancer cells, BML-284 (a pharmacological activator of β-catenin dependent WNT signaling) was able to rescue cell viability and decrease apoptosis, supporting the notion that the WNT pathway promotes aggressiveness features [75].

Interestingly, WNT6 has also recently been implicated in predicting the 1-, 3- and 5-year survival status of breast cancer patients with bone metastasis [82]. In this context, WNT6 was also positively correlated to KLK6, GJB3, FBN3, and GABBR2 [82]. Relevantly, KLK6 upregulation has been described to promote oncogenic behavior in breast cancer [118], and reported to induce nuclear accumulation of β-catenin in a mouse keratinocyte cell line, corroborating a potential link between KLK6 and WNT signaling, as well as with tumor promotion. Additionally, WNT6 expression was also positively correlated with the extent of tumor infiltrating immune cells, namely regulatory T-cells [82]. This raises the interesting hypothesis that WNT6 can, as part of the WNT signaling cascade, influence the immune landscape and, potentially impact the response to newly-developed anti-cancer immunotherapies.

These relevant studies denote WNT6 roles as a promoter of tumor aggressiveness, and identify important relevant interactions that influence its expression in breast cancer.

Bladder cancer

In bladder cancer, WNT6 expression was firstly found to be upregulated following the overexpression of the lncRNA UCA1 in vitro [85]. Later, in patient tissues, WNT6 mRNA levels were positively correlated to UCA1 [73]. UCA1 has been described to be upregulated in bladder cancer tissues and to sustain oncogenic functions in vitro and in vivo [73, 85]. Interestingly, UCA1 expression was found to increase β-catenin-dependent WNT signaling activation as well (Fig. 2) [73]. This role of WNT6 and β-catenin-dependent WNT signaling in bladder tumor aggressiveness was also demonstrated by manipulating β-catenin-dependent WNT signaling with the pharmacological inhibitor IWR-1, which led to a decrease in cell viability of UCA1-overexpressing cells. Conversely, genetically overexpressing WNT6 in UCA1-knockdown cells led to an increase in cell viability [73].

While follow-up studies are necessary to explore the direct roles of WNT6 in bladder cancer aggressiveness, these data from two independent studies strongly suggest WNT6 may be a key molecular partner for UCA1 in sustaining its oncogenic effects in bladder cancer.

Brain cancer

In the context of malignant brain tumors, particularly gliomas, tumor aggressiveness has been often linked to the aberrant activation of the WNT signaling pathway [119,120,121,122,123], particularly through the abnormal expression of WNT ligands [78, 124,125,126]. Namely, WNT3A [124, 125] and WNT5A [126,127,128], which have been both associated to tumor aggressiveness in glioblastoma, the most aggressive and lethal form of glioma. However, the description of WNT6 in sustaining oncogenic properties in gliomas, particularly glioblastoma, has only recently been recognized [74].

WNT6 expression has been shown to increase alongside glioma malignancy grade, regardless of IDH mutations and 1p/19q co-deletion status, with glioblastoma presenting the highest WNT6 levels among gliomas [74, 87]. In glioma patient samples, WNT6 expression was predominantly cytoplasmic in tumor cells, either presenting a diffuse or more scattered pattern, while tumor infiltrating lymphocytes were considered negative for this molecule, as well as endothelial cells, presenting low to undetectable WNT6 [74]. Although its expression does not seem to be modulated by copy number alterations in lower grade gliomas or glioblastoma, DNA methylation may play an important part in the regulation of WNT6 expression in various glioma cell lines and patients (Fig. 2) [87]. In fact, in glioblastoma, the methylation levels of two specific regions, one downstream of the WNT6 promoter and another within the gene body, have been negatively and positively correlated with WNT6 expression, respectively [87]. Concordantly, in another study, WNT6 was found to be methylated in several glioblastoma cell lines, as well as in a glioblastoma and two astrocytoma patient samples [129]. However, the specific regions of CpG methylation were not disclosed, and no associations with WNT6 expression levels were assessed.

Consistent with findings in other tumor contexts, increased WNT6 levels in gliomas were associated with higher tumor aggressiveness in vitro, reflected in increased cell viability, proliferation, migration, invasion, chemotherapy resistance, and maintenance of glioma stem cell features [74]. Interestingly, the association between WNT6 and increased chemoresistance was later corroborated in independent studies, with evidence suggesting that the p53-hsa-miR-34a axis sensitizes glioblastoma cells to the chemotherapeutic temozolomide by reducing WNT6 expression and β-catenin-dependent WNT signaling [89].

Additionally, in in vivo orthotopic glioblastoma models, WNT6 expression promoted accelerated tumor-related death, and the WNT6-positive tumors showed increased expression of proliferation, stem cell and anti-apoptosis markers (Fig. 2) [74].

Molecularly, in vitro phospho-proteomic assays identified that high WNT6 levels in glioblastoma are associated with the activation of several cancer-related pathways, such as, SFK, STAT, AKT, and RTK (Fig. 2) – which, ultimately contribute to the increased aggressiveness profile of this tumor subtype [74]. These results were further corroborated by patient data, which showed WNT6 positively correlated genes were enriched for gene sets upregulated by WNT, SRC, MAPK, AKT, MYC, and JNK signaling [74]. WNT6 expression was also associated to an increased activation of the β-catenin-dependent signaling (Fig. 2), suggesting this ligand could be exerting its effects on glioblastoma through the direct activation of this pathway [74].

Moreover, HOXA9, a key mediator of glioblastoma aggressiveness [130,131,132], was described as an inducer of WNT6’s expression, interacting directly with its promoter region and activating the β-catenin-dependent WNT signaling pathway (Fig. 2) [87]. In glioblastoma patients, WNT6 and HOXA9 expression levels were positively correlated [87]. Interestingly, high WNT6 expression was associated with shorter overall survival of glioblastoma patients (Fig. 2), independently of HOXA9 expression, as well as other major prognostic variables, such as age, gender, IDH1 mutation status, treatment regimen, and the Karnosfky Performance Score [74, 87]. Importantly, glioblastoma patients with WNT6- and HOXA9-high tumors showed a particularly dismal prognosis when compared to all other glioblastoma patients. Moreover, among HOXA9-low patients, those with high WNT6 expression had a significantly shorter overall survival compared to those with low WNT6 and HOXA9 expression [87].

Curiously, in IDH1-mutant lower-grade glioma patients, WNT6 expression, along with four other genes, was part of a hypoxia-related risk score, which was associated with shorter survival, independently of 1p/19q codeletion, age, and WHO malignancy grade (II and III) [88]. Moreover, WNT6 has also been studied in other non-glioma brain tumors, particularly in high-risk neuroblastoma without MYCN amplification, where WNT6 was among the most highly expressed WNT ligands, possibly contributing to the aberrant activation of β-catenin-dependent WNT signaling in these tumors [86].

These data strongly indicate that WNT6 may have a pertinent role in brain cancers, with particularly compelling evidence in glioblastoma. Additional research is needed to explore the link between WNT6 and HOXA9, as correlations between these two key genes were also identified in several other tumors, including leukemia, melanoma, testicular germ cell tumor, and cholangiocarcinoma [87]. Finally, exploring the therapeutic inhibition of WNT6, through pharmacological or viral-mediated genetic inhibitors (e.g. adeno-associated virus (AAV)-based therapies), either as monotherapy or in combination approaches (e.g., with the standard-of-care temozolomide-based therapy, or with novel experimental therapies), could be invaluable in the context of glioblastoma, a particularly dramatic cancer in urgent need of better therapies.

Esophageal cancer

In esophageal squamous cell carcinoma patients, WNT6 was detected in the plasma of tumor cells, and nearly half the patients presented high WNT6 expression [90]. In addition, its expression associated with various clinicopathologic characteristics, including patient age, gender, tumor stage and histopathological type. Critically, WNT6 was considered an independent prognostic factor, as high levels of WNT6 associated with shorter overall survival and disease-free survival of esophageal carcinoma patients (Fig. 2) [90].

Interestingly, Wnt6 was found to be upregulated in the esophagi of rats treated with (S)-N’-nitrosonornicotine ((S)-NNN), as well as in immortalized oral keratinocytes after treatment with (S)-NNN and (R)-N’-nitrosonornicotine ((R)-NNN) (Fig. 2) [91]. These two carcinogenic compounds are found in tobacco, a major environmental risk factor for esophageal cancer [91], further supporting the hypothesis that WNT6 may be relevant to the pathophysiology of this tumor. Nevertheless, the differences in WNT6 expression reported in the rat model were not recapitulated in human esophageal carcinoma patient data. However, it should be noted that this patient data analysis compared tumor and non-tumor samples, not discriminating between smokers and non-smokers, thus not fully accounting for the effects of (S)-NNN on gene expression, complicating the analysis of their exposure to (S)-NNN [91].

These studies highlight that WNT6 may be altered in a subgroup of esophageal cancer patients, and that its upregulation may identify a subset of patients with worse prognosis. Naturally, additional approaches are needed to unveil the molecular intricacies underlying this possible WNT6-driven tumor aggressiveness.

Bone cancer

In osteosarcoma, WNT6 has incited interest both in the context of infant and adult patients [92, 93]. In a pediatric osteosarcoma in vitro model, WNT6 was shown to be upregulated in comparison to a human osteoblast cell line, presenting also significantly lower overall methylation levels [92]. This was corroborated in pediatric patient tumors, in which WNT6 was found to be expressed in a large percentage of the samples and was reported to be largely unmethylated in the majority of cases [92]. Interestingly, patients with unmethylated WNT6 presented a significantly shorter 5-year survival rate compared to those with methylated WNT6 [92], suggesting that WNT6 may have potential as a prognostic biomarker in these cancers (Fig. 2). Unfortunately, in this study, the specific CpG sites whose methylation was assessed in the WNT6 gene were not disclosed, which is crucial information as DNA methylation in different gene regions can have diverse effects on gene expression.

Concordantly, recent findings showed that adult patient tissue samples present higher WNT6 mRNA and protein levels than non-tumoral samples [93]. Additionally, WNT6 levels were significantly higher in the serum of osteosarcoma patients compared to those with Ewing’s sarcoma, osteomyelitis, and cancer-free controls [93]. The detection of WNT6 in liquid biopsies of bone cancer patients was proposed to be a reliable diagnostic tool [93]. Moreover, WNT6 expression was associated with patient’s age, tumor grade, and presence of distant metastasis [93]. High WNT6 expression was found to be associated with shorter overall survival, and it was considered an independent factor for worse patient prognosis [93].

While these studies underline the potential of WNT6 as a valuable biomarker with diagnostic potential in osteosarcoma, further studies in independent and larger cohorts of patients are necessary. Additionally, gaining deeper insights into the molecular mechanisms underlying these associations will be key to identify putative novel therapeutic targets.

Smooth muscle cancer

A recent study identified WNT6 was highly expressed in post-transplant smooth muscle tumors positive for Epstein-barr virus, while visceral leiomyomas, leiomyosarcomas, angioleiomyomas, and endothelial haemangiomas presented low levels or no significant differences in WNT6 expression [94]. Interestingly, in individual cases of post-transplant smooth muscle tumors, WNT6 and WNT10A were co-expressed at similar levels [94]. However, several elements of the β-catenin-dependent and -independent WNT signaling were not altered in this tumor type, with only CCND2 upregulated compared to angioleiomyomas and endothelial haemangiomas, and MYC compared to angioleiomyomas and visceral leiomyomas [94]. Thus, while the concomitant upregulation of WNT6 and WNT10A in post-transplant smooth muscle tumors is interesting, whether these genes sustain effects on tumor behavior and patient clinical outcome remains to be explored.

Lung cancer

In non-small cell lung cancer, a study found WNT6 was one of the genes upregulated upon HOXC6 overexpression in vitro [96]. Considering HOXC6 is a transcription factor with known oncogenic roles in non-small cell lung cancer, promoting cell proliferation and migration, these preliminary findings raise the hypothesis that HOXC6 may enhance the expression of other oncogenes, such as WNT6, ultimately increasing tumor aggressiveness (Fig. 2) [96]. Nonetheless, it is critical to firstly validate this putative HOXC6-WNT6 molecular link in additional preclinical models of non-small cell lung cancer, both in vitro and in vivo, and in human tumor specimens.

In small cell lung cancer, in a microarray analysis, WNT6 was reported to be downregulated in cisplatin-resistant in vitro models and its expression was inversely correlated to the lncRNA NCRNA00173 [95]. This lncRNA has been considered a promoter of aggressiveness in several tumors [133,134,135,136], being also associated with chemoresistance in small cell lung cancer [137]. However, WNT6 expression did not differ significantly between tumor and non-tumor adjacent tissues in a small cohort of these patients [95]. The results herein reported are particularly interesting, since WNT signaling had been previously described to induce resistance to platinum-based chemotherapy in this tumor type [138]. Hence, it would be important to validate this association in a larger cohort, and to compare cisplatin-resistant and -sensitive tumor data, to understand the exact role of the WNT pathway and WNT6 in small cell lung cancer chemoresistance.

While these results indicate that WNT6 may be of interest in both non-small cell and small cell lung cancer, to ascertain its pertinence, it is necessary to further study the phenotypical and molecular implications of its aberrant expression.

Reproductive cancers

The WNT signaling pathway has also been recurrently linked to the onset and progression of ovarian cancer, with certain WNT ligands implicated in such processes [139,140,141]. Recently, WNT6 was found to be increased in ovarian cancer patient samples in comparison to non-tumor tissues, and this upregulation was also observed in ovarian cancer in vitro models relative to non-tumor ovarian cells [76]. Furthermore, WNT6 expression was associated to increased cell proliferation, invasion, as well as decreased cell cycle arrest, and inhibition of apoptosis in vitro [76]. Interestingly, high WNT6 expression was also associated with decreased progression-free survival in ovarian cancer patients [76]. Molecularly, WNT6 associated with β-catenin and NOTCH1 expression in ovarian cancer cells, suggesting WNT6 may exert its oncogenic effects through the β-catenin-dependent WNT and Notch pathways (Fig. 2) [76], whose interconnection has been often reported as important for the aggressiveness of different tumor types [142,143,144,145]. This important study revealed a biomarker potential of WNT6 in ovarian cancer patients, identifying as well possible mechanisms through which it may be inciting its oncogenic effects. Future efforts should focus on better defining the molecules and pathways that act upstream and downstream of WNT6, possibly opening new therapeutic opportunities.

Additionally, in testicular germ cell tumors, WNT6 was identified to be downregulated in a small number of cisplatin-resistant cell lines when compared to cisplatin-sensitive cells [97]. While this study suggested that WNT6 could be of interest in testicular tumors, it would be noteworthy to complement this data with functional and molecular analyses to provide mechanistic and causality insights.

Emerging pro-tumoral roles of WNT6 in multiple cancers

WNT6 exerts its physiological functions through the activation of WNT signaling pathways, encompassing both β-catenin-dependent and independent cascades, in a manner specific to cell type and developmental stage [61, 62, 64,65,66]. Emerging evidence has highlighted WNT6 aberrant expression as a prevalent alteration in various tumor types [67, 70, 72,73,74, 76, 77, 79,80,81, 83, 84, 86,87,88, 90,91,92,93,94, 96]. More significantly, extensive evidence emphasizes its pertinent role in tumor aggressiveness across several malignancies (Fig. 2 and Table 1) [69,70,71,72,73,74,75,76,77, 79, 81, 82, 85, 87,88,89,90, 92, 93, 102]. Nonetheless, curiously, contrasting findings in a subset of tumor types suggest a possible dual role for WNT6 in those contexts [95, 97]. At the light of this increasing body of data, it is critical to understand the putative relevance of WNT6 in multiple aspects of cancer, namely cancer cell functions, sensitivity to anticancer therapies, and patient prognosis. Equally relevant is to dissect whether WNT6 may be a novel therapeutic target in specific cancer types.

In vitro, WNT6 has been classified has a cancer-promoting molecule (Fig. 2), sustaining multiple tumorigenic capabilities across a range of cancer types [69,70,71,72,73,74,75,76,77, 81, 85]. Although a limited number of studies have reported in vivo approaches, they corroborate an association between higher WNT6 expression and increased tumor aggressiveness [69, 70, 72, 74]. Of note, clinical data from several tumors reveal that high WNT6 may be a predictor of unfavorable patient outcome [74, 76, 79, 82, 88, 90, 93]. Complementarily, data from multiple tumors suggests WNT6 may be involved in increased chemoresistance to anthracyclines [72, 75] and to alkylating agents [73, 74, 89]. Interestingly, patients with colorectal cancer liver metastasis who received preoperative chemotherapy presented significantly lower WNT6 expression, in comparison to those who did not receive it [79]. This raises the hypothesis that preoperative treatment may reduce WNT6 expression in colorectal cancer liver metastasis, and/or that WNT6-positive cells could be more susceptible to chemotherapy-induced cell death.

The apparent dual effects of WNT6 expression in anticancer therapy response warrant further investigation, namely clarifying whether this is dependent on the mechanism of action of the particular therapeutic agent used in each study, or if it is tumor-type/subtype related. To further unravel this, the use of ever-increasing bioinformatic tools (e.g. Connectivity map and multi-omics data combined with clinical data), and drug screening assays could be interesting approaches. Additionally, it remains to be studied whether WNT6 may affect sensitivity to radiotherapy, which would be particularly critical given its wide clinical use in multiple cancer types.

Molecular regulators and interactors of WNT6 in cancer

Mechanistically, WNT6 may be promoting tumor aggressiveness through the activation of both β-catenin-dependent and -independent WNT pathways (Fig. 1). Although WNT6 is more frequently associated with β-catenin-dependent signaling activation [69,70,71,72,73,74,75,76], this may be due to the limited analysis of other possible signaling cascades in which WNT6 could be involved. Indeed, when this was analyzed, as, for example, in glioblastoma [74] and post-transplant smooth muscle tumors [94], WNT6 was also associated with the activation of β-catenin-independent WNT signaling pathways.

Curiously, WNT6 and WNT10A co-expression was reported in colorectal [67] and post-transplant smooth muscle tumors [94], but whether this WNT6-WNT10A association also occurs in other malignancies remains to be explored.

Depending on the tumor type, various molecules have been identified as regulators of WNT6 (Table 2), many of which are known oncogenes [107, 108, 132], such as PLAGL2 [69, 75], HOXA9 [87], and NR4A2 [70]. These molecules function as transcription factors directly interacting with the WNT6 promoter, inducing its expression, and consequently promoting tumor aggressiveness [69, 70, 75, 87]. Notably, PLAGL2 has been reported in both colorectal and breast cancers [69, 75], indicating that it could be valuable to study this association beyond the in vitro models, particularly exploring whether these effects can be replicated in vivo, and ultimately validated in primary tumor patient samples.

Additionally, WNT6 has been described to interact with and be regulated by non-coding RNAs, such as lncRNA UCA1, hsa-miR-566 and hsa-miR-1207-5p [71, 73, 81], as well as other molecular players whose interactions with WNT6 remains incompletely understood [72, 82, 84, 96], such as CAV1 [72]. Importantly, further molecular assays are required to unveil causative associations, assess functional implications, and understand the nature of these interactions. It would be interesting to assess whether these molecules may be associated with WNT6 in other tumor types, as it has been done for glioblastoma, where it was found that molecules such as CAV1, PLAGL2, and UCA1 have no significant correlation to WNT6 in this context [87].

Of note, some putative associations were also revealed through transcriptomic analyses (e.g., KLK6 in breast [82] and HOXC6 in non-small cell lung cancer [96]), but further investigations must be conducted for comprehensive validation of such molecular relations.

WNT6 as a putative novel therapeutic target in cancer

Considering WNT6 was shown to promote tumor aggressiveness and progression in most cancer types in which it was studied [77, 78, 87, 93, 104, 105, 116, 146], it would be interesting to explore its relevance as a novel therapeutic target. Unfortunately, structural similarities and functional redundancy of WNT ligands pose additional challenges for the development of specific inhibitors of WNT6. Nonetheless, considering some of the already-known signaling pathways activated by WNT6 in particular cancers, a legitimate therapeutic strategy would be to test targeted therapies already developed to inhibit those WNT6-related (upstream or downstream) pathways (Fig. 3). For example, PORCN inhibitors [147,148,149,150,151] prevent WNT ligand extrusion (e.g., IWP2 [147], IWP-L6 [148], ETC-159 [149], C59 [150], or LGK974 [151]), of which LGK974 has shown great promise in clinical studies for safety and tolerability in various tumor types, both in monotherapy and in combination with other therapeutics [146, 152, 153]. Currently, a phase I clinical trial is testing LGK974 in patients whose tumors present alterations in the WNT signaling pathway (NCT01351103). The safety and tolerability of ECT159 has also been clinically tested in patients with advanced/unresponsive solid tumors (NCT02521844).

Compounds that target components of WNT pathway. Schematic representation of the β-catenin-dependent WNT signaling pathway alongside pharmaceutical modulators. These are strategically designed to inhibit WNT pathway activity, acting at various stages, including inhibition of WNT ligand maturation and secretion, of WNT ligand-receptor complex interactions, and of key molecular players within the β-catenin-dependent intracellular WNT signaling cascade. These include small molecule inhibitors, peptide antagonists, and monoclonal antibodies

Full size image

Other therapeutic strategies include the inhibition of WNT interactions with receptor complexes using peptide antagonists (e.g., Fz7-21 [154]), monoclonal antibodies (e.g., OMP-18R5 [155], OTSA-10 [156]) or decoy receptors (e.g., OMP-54F28 [157]). OMP-18R5 [158] and OMP-54F28 [157, 159, 160] were already tested in multiple early-phase clinical trials, and while the efficacy results of many are still unpublished, general good tolerability was shown, despite some reports of bone toxicity when combined with certain chemotherapies [157,158,159,160,161]. Of note, as the full spectrum of WNT6 receptors remains to be identified, a single drug affecting only a particular ligand/receptor interaction may not be able to inhibit all the WNT6-mediated signaling cascades.

Other putative approaches consist of targeting the dishevelled protein, a critical cytoplasmic partner that interacts with the majority of WNT co-receptors (NSC668036 [162]), or inhibiting specifically β-catenin-dependent WNT signaling [69,70,71,72,73,74,75,76]. Several compounds have been developed to target various intermediates of this pathway, aiming to stabilize the β-catenin destruction complex [147, 163, 164] or to inhibit the β-catenin-TCF/LEF complex [165,166,167,168,169]. Particularly, BC2059 (tegavivint [169], an inhibitor of β-catenin and TBL interaction; NCT03459469, NCT04851119, NCT04874480, NCT04780568) and PRI-724 (an inhibitor of β-catenin-CBP interaction [170]; NCT01302405, NCT01606579, NCT01764477) have been under clinical investigation for their efficacy in various malignancies. The full completion of these clinical trials may provide invaluable findings, potentially impacting human cancers with WNT6-driven activation of WNT signaling.

Nonetheless, while specific inhibitors of WNT6 are not yet available, the landscape of cancer treatment offers multiple potential strategies that could be adapted to target WNT6. As above-described for many other molecules that take part in WNT signaling, the development of small-molecule inhibitors and antibody-based therapies offer promising avenues. A particularly intriguing approach is the use of bispecific antibodies, which may simultaneously target WNT6 together with other key tumor oncogenic biomarkers or immune-system activating molecules. This strategy may not only interfere with the ability of WNT6 to bind its receptor complex, but may also enhance the immune system’s response against tumor cells. Interestingly, a recent study explored the effects of targeting LRP6 along with cell-type specific antigens, to inhibit WNT signaling in a cell-type specific manner [171]. Additionally, recent advances in gene-based therapies may hold significant potential by precisely targeting WNT6 expression in tumor tissues, such as those based on CRISPR-Cas9 editing and optimized viral delivery systems. For example, gene therapies are already FDA-approved for rare, non-oncogenic conditions, such as dystrophic epidermolysis bullosa [172], and haemophilia B [173], and similar approaches may be exploited in malignant contexts.

Thus, in the challenging quest to successfully inhibit the pro-tumoral effects of WNT6 in cancer, several possible courses of action can and should be explored, possibly rationally adapted to the specificities of each particular tumor type. Given some of these human cancers are particularly prevalent, these novel therapeutic tools hold great clinical potential. In a time where precision medicine holds immense promise for improved cancer treatment outcomes, it would also be important that both the scientific community and pharmaceutical industry join efforts and prioritize the development of novel molecules capable of directly targeting WNT6, paving the way for more effective and precise interventions in cancer therapy.

Future directions

Despite the growing body of evidence supporting the relevance of WNT6 in the promotion of oncogenic processes, and of the putative molecular interactions in which it may be involved, there are still many unanswered questions, mostly in what regards the causative or associative nature of particular molecular links with WNT6, the identification of upstream regulatory mechanisms, as well as the dissection of its receptors, downstream effector pathways and molecular partners. These insights would be highly relevant, as they could contribute to the development and selection of more precise targeted-therapies. This could be achieved by performing, for example, systematic and complementary bioinformatic analyses, which are becoming widely available and of increased power, such as through gene set enrichment analysis (GSEA; https://www.gsea-msigdb.org) [174], and it would be particularly useful to validate some of the reported findings in well-characterized cohorts (e.g., querying data from The Cancer Genome Atlas), establishing important correlations between the expression of WNT6 and its potential molecular partners, as well as further assessing their impact in the survival outcome of patients with different tumor types. Additionally, complementing this knowledge with more integrative approaches, exploiting comprehensive transcriptomic analysis, such as RNA-sequencing, ChIP-sequencing, as well as epigenomics, and proteomics analyses, would be particularly informative, allowing the concomitant characterization of multiple layers of molecular data, potentially contributing to a more integrated understanding of how they interact and relate to WNT6 expression and function. Indeed, for some tumors, such as glioblastoma, upregulation of WNT6 expression was already shown to depend, at least partly, on specific gene methylation patterns, and the presence of a particular transcription factor [87]. Importantly, more advanced methodologies should also be explored, such as single-cell sequencing and spatial transcriptomics, since these consider tumor heterogeneity and offer a more in-depth comprehensive analysis of the different cell populations present in the tumor microenvironment, possibly revealing specific populations exhibiting altered WNT6 levels. Moreover, these would be extremely valuable to explore the effects of WNT6 on cancer stem cell (CSC) populations and stem-cell like behavior of tumor cells, considering the significance of WNT signaling and WNT6 in the regulation of progenitor and stem cell fate under physiological conditions, and the recent reported influence of WNT6 expression in a stem-cell like phenotype in glioblastoma cells [74]. Additionally, given the relevance of CSCs in therapy resistance, it is also important to assess how modulation of WNT6 in these cells could affect their malignant phenotype and investigate how it may be combined with particular molecularly-targeted therapies.

Despite a few exceptions with studies reporting in vivo findings [69, 70, 72, 74, 91], several articles present only in vitro data, highlighting a need for validation in more refined and relevant preclinical in vivo models that would undoubtedly strengthen the validity and translational impact of the findings. For example, using more sophisticated animal models, particularly orthotopic models, or xenograft models, would be immensely informative in what regards the impact of WNT6 in tumor progression, molecular interactions, and specificities in particular niches of the tumor microenvironment. Of these, syngeneic and/or humanized mouse models would be of particular relevance, given their immunocompetent contexts, considering the described role of WNT6 in immune response, specifically in macrophage differentiation and proliferation [66], as well as its reported correlation to regulatory T-cell infiltration in breast cancer patients with bone metastasis [82], it is reasonable to hypothesize that WNT6 may also affect the immune system.

Furthermore, posterior to these validations, it may be pertinent to explore the implications of WNT6 modulation in the response of tumor cells to particular anti-cancer immunotherapies, such as, for example, immune checkpoint inhibitors. In this context, it would be relevant to perform analyses in large patient cohorts, associating WNT6 expression with patient responsiveness to immunotherapies.

Of note, given that most of these studies collectively suggest WNT6 as a promoter of cancer aggressiveness, it would be invaluable to deploy innovative strategies to investigate its potential role in tumor development. One approach could involve manipulating non-tumor cells in vitro to overexpress WNT6, and assess how it may influence transformation and tumorigenesis. Complementarily, a more sophisticated and informative methodology could use syngeneic genetically-modified mice presenting modulated expression levels of human WNT6, taking advantage of inducible temporal- and tissue-specific conditional models. Exploring WNT6-driven tumorigenesis in a more complex living organism would offer key insights at multiple levels. For example, it could reveal how WNT6 may affect intricate intercellular interactions within the various niches of the tumor microenvironment, and help uncover whether WNT6 effects are tissue-specific or, as a secreted ligand, if it may also be broader at systemic levels. Additionally, tracking tumor-initiating cells could provide crucial information on WNT6’s role in tumorigenesis. For example, studying clonal evolution and lineage tracing may also reveal whether WNT6 can drive the expansion of specific tumor subpopulations, particularly those associated with aggressive cancer behavior. These methods would help determine whether WNT6 is among the key drivers of tumorigenesis in multiple tumor types.

In conclusion, WNT6 has recently emerged as a key player in multiple oncogenic processes of a significant variety of tumors. Its elevated expression has been associated with tumor aggressiveness, sensitivity to chemotherapeutics, and patient prognosis. While frequently being reported to activate the β-catenin-dependent WNT cascade, various other molecular inter-players have been found to interact with and modulate the expression of WNT6. Yet, there is still much to unveil for a comprehensive understanding of its role in cancer, as well as its therapeutic potential. Undoubtedly a clearer understanding will contribute with invaluable insights, paving the way for new breakthroughs in the field.

No datasets were generated or analysed during the current study.

5-FU:

5-Fluorouacil

(S)-NNN:

(S)-N’-nitrosonornicotine

APC:

Adenomatous polyposis coli

CAV1:

Caveolin-1

CK1:

Casein kinase 1

CSC:

Cancer stem cell

FZD:

Frizzled

GSK3:

Glycogen synthase kinase 3

lncRNA:

Long non-coding RNA

MuSK:

Muscle skeletal receptor Tyr kinase

PGE2:

Protaglandin E2

PORCN:

Porcupine

PTK7:

Protein Tyr kinase 7

ROR:

Tyr kinase-like orphan receptor

RYK:

Tyr kinase receptor

Not applicable.

Financial support was provided by projects and grants from FCT – Foundation for Science and Technology (2022.04859.PTDC, https://doi.org/10.54499/2022.04859.PTDC to B.M.C.; 2021.02600.CEECIND/CP1664/CT0015, https://doi.org/10.54499/2021.02600.CEECIND/CP1664/CT0015 and 2022.05143.PTDC, https://doi.org/10.54499/2022.05143.PTDC to C.S.G.); and by National funds, through the Foundation for Science and Technology (FCT) – project UIDB/50026/2020, https://doi.org/10.54499/UIDB/50026/2020 and LA/P/0050/2020, https://doi.org/10.54499/LA/P/0050/2020 and by the project NORTE-01–0145-FEDER-000055, supported by Norte Portugal Regional Operational Programme (NORTE 2020), under the PORTUGAL 2020 Partnership Agreement, through the European Regional Development Fund (ERDF).

Authors and Affiliations

1. Life and Health Sciences Research Institute (ICVS), School of Medicine, University of Minho, Campus Gualtar, Braga, 4710-057, Portugal

   Joana M. Ferreira, Céline S. Gonçalves & Bruno M. Costa

2. ICVS/3B’s - PT Government Associate Laboratory, Braga/Guimarães, Portugal

   Joana M. Ferreira, Céline S. Gonçalves & Bruno M. Costa

Contributions

J.M.F., C.S.G., and B.M.C. contributed to the conceptualization and writing of the manuscript.

Corresponding author

Correspondence to Bruno M. Costa.

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions