Emerging Insights into Wnt/β-catenin Signaling in Head and Neck Cancer

Abstract

Head and neck cancer presents primarily as head and neck squamous cell carcinoma (HNSCC), a debilitating malignancy fraught with high morbidity, poor survival rates, and limited treatment options. Mounting evidence indicates that the Wnt/β-catenin signaling pathway plays important roles in the pathobiology of HNSCC. Wnt/β-catenin signaling affects multiple cellular processes that endow cancer cells with the ability to maintain and expand immature stem-like phenotypes, proliferate, extend survival, and acquire aggressive characteristics by adopting mesenchymal traits. A central component of canonical Wnt signaling is β-catenin, which balances its role as a structural component of E-cadherin junctions with its function as a transcriptional coactivator of numerous target genes. Recent genomic characterization of head and neck cancer revealed that while β-catenin is not frequently mutated in HNSCC, its activity is unchecked by more common mutations in genes encoding upstream regulators of β-catenin, NOTCH1, FAT1, and AJUBA. Wnt/β-catenin signaling affects a wide range epigenetic and transcriptional activities, mediated by the interaction of β-catenin with different transcription factors and transcriptional coactivators and corepressors. Furthermore, Wnt/β-catenin functions in a network with many signaling and metabolic pathways that modulate its activity. In addition to its effects on tumor epithelia, β-catenin activity regulates the tumor microenvironment by regulating extracellular matrix remodeling, fibrotic processes, and immune response. These multifunctional oncogenic effects of β-catenin make it an attractive bona fide target for HNSCC therapy.

Introduction

The Wnt signaling pathway plays critical roles in development, adult homeostasis, and cancer (reviewed in Zhan et al. 2017). This signaling cascade comprises distinct branches: the Wnt/β-catenin or the canonical Wnt signaling pathway, the planar cell polarity (Wnt-PCP) pathway, and the Wnt-calcium signaling pathway (Gordon and Nusse 2006). To date, most studies have focused on the Wnt/β-catenin branch, revealing its prominent roles in diverse cellular processes that underlie human cancers (reviewed in Nusse and Clevers 2017).

Deregulated Wnt/β-catenin activity has been associated with diverse cancer types (reviewed in Nusse and Clevers 2017). Mounting evidence from cellular and animal models coupled with genomic and molecular characterization of head and neck cancers has added this malignancy to the list of tumors affected by Wnt/β-catenin signaling. To date, several excellent reviews have described the involvement of the Wnt/β-catenin pathway in the pathobiology of head and neck cancer (reviewed in Castilho and Gutkind 2014; Cancer Genome Atlas 2015; Beck and Golemis 2016). This review summarizes seminal discoveries and recent advances in Wnt/β-catenin signaling in cancer while highlighting their relevance to head and neck squamous cell carcinoma (HNSCC), and it describes current strategies to inhibit Wnt/β-catenin signaling.

Wnt/β-catenin Signaling in Head and Neck Cancer

Head and neck cancer is a complex malignancy involving multiple tissue sites with distinct biology, extensive intra- and intertumor heterogeneity, and poorly understood intercellular communication (Hedberg et al. 2016; Puram et al. 2017). On the genomic level, HNSCC is characterized by somatic alterations grouped into human papillomavirus (HPV)–positive and HPV-negative tumors with distinct epidemiologic, molecular, and clinical features (reviewed in Cancer Genome Atlas 2015; Beck and Golemis 2016).

Mounting evidence indicates that Wnt/β-catenin signaling plays pivotal roles in the pathobiology of HPV-negative and HPV-positive HNSCC (Yang et al. 2006; Chang et al. 2013; reviewed in Castilho and Gutkind 2014; Gonzalez-Moles et al. 2014; Cancer Genome Atlas 2015). Increased basal cell expression of Wnt1 and Wnt pathway activation by epigenetic alterations of Wnt inhibitory factors, SRFP, WIF, and DKK3, has been reported (Pannone et al. 2010). Augmented tissue levels of β-catenin in HNSCC have been aligned with its increased transcriptional activity (Rampias et al. 2010; Hu et al. 2015), and inappropriate stabilization of β-catenin has been correlated with de-differentiation and poor prognosis (Hu et al. 2015; Padhi et al. 2015). Notably, HNSCC has been shown to rely on high Wnt/β-catenin activity to promote epigenetic changes associated with open chromatin structure and induction of stem cell gene signatures (Wend et al. 2013).

While mutations in β-catenin are infrequent in HNSCC, alterations in genes encoding upstream regulators, NOTCH1, FAT1, and AJUBA, converge on β-catenin signaling and enhance its nuclear activity (Cancer Genome Atlas 2015; Beck and Golemis 2016). To date, mutations in AJUBA and FAT1 have been almost exclusively associated with HPV-negative tumors and linked to the loss of epithelial differentiation programs. In HPV-positive disease, deregulated Wnt/β-catenin signaling has been linked to viral oncoproteins E6- and E7- (E6/E7)–mediated activation of β-catenin (reviewed in Castilho and Gutkind 2014; Rampias and Psyrri 2014). Specifically, oncoprotein E6 has been shown to drive nuclear translocation of β-catenin through the epidermal growth factor receptor (EGFR)–dependent mechanism in oropharyngeal squamous cell carcinoma (OPSCC) (Hu et al. 2015). Inhibition of E6 expression with small interfering RNA and downregulation of EGFR activity with erlotinib abrogated nuclear localization of β-catenin and EGFR phosphorylation concomitant with reduced invasive properties of HNSCC-positive cell lines in vitro (Hu et al. 2015). Also, E6/E7 have been reported to induce nuclear translocation of β-catenin by inhibiting the Siah E3 ubiquitin ligase protein (Rampias et al. 2010). Nonetheless, more studies are needed to better understand the role of Wnt/β-catenin signaling in HPV-positive HNSCC, as they have lagged behind HPV-negative disease due to limited tumor specimens and associated clinical data (Cancer Genome Atlas 2015; Beck and Golemis 2016).

Wnt/β-catenin Signaling: Components and Regulation

The Wnt/β-catenin pathway is pivotal in controlling cell fate determination, cell proliferation, cell survival, and differentiation (reviewed in Nusse and Clevers 2017). A central effector of Wnt signaling is β-catenin that, under conditions of epithelial tissue homeostasis, is maintained at low levels in the cytoplasm while being prominently localized to E-cadherin junctional complexes or adherens junctions (AJs). Steady-state levels of cytoplasmic β-catenin are controlled by a complex of proteins that include adenomatous polyposis coli (APC), the scaffold protein Axin1, glycogen synthase kinase 3β (GSK3β), casein kinase 1α (CK1α), protein phosphatase 2A (PP2A), and additional components that sequester excess cytoplasmic β-catenin and target it for ubiquitination and proteosomal degradation via the β-TrCP, an E3 ubiquitin ligase subunit (Fig. 1). The Wnt/β-catenin pathway is activated by the binding of the Wnt ligand to a 7-pass transmembrane Frizzled (Fzd) receptor and its coreceptor, the low-density lipoprotein receptor-related protein 6 (Lrp6) or Lrp5. This results in a conformational change and phosphorylation of Lrp5/6, leading to the sequestration of Axin 1 and inactivation of the destruction complex. In the absence of the destruction complex, the newly synthesized unphosphorylated β-catenin accumulates in the cytoplasm and translocates to the nucleus, where it displaces repressors from the promoters of target genes and interacts with transcription factors to drive their expression. The most prominent transcriptional partner of β-catenin is the T-cell factor (TCF) and lymphoid enhancer-binding factor (LEF1) family of DNA-binding proteins that together induce the expression genes encoding myc, cyclin D1, axin 2, and survivin, among others.

Figure 1.

Schematic of the Wnt/β-catenin signaling cascade. Wnt signaling inactive: in the absence of Wnt ligand, cytoplasmic β-catenin is maintained at low levels by the destruction complex composed of the scaffold protein Axin in complex with APC, GSK3β, and CK1α that phosphorylates the N-terminal region of β-catenin and targets it for proteosomal degradation via βTrCP E3 ubiquitin ligase. Wnt signaling active: the binding of Wnt ligand to the Fzd receptor-Lrp6/5 coreceptor complex inactivates the destruction complex and leads to the accumulation of cytoplasmic β-catenin and its subsequent translocation to the nucleus, where it interacts with TCF-LEF transcription factor and activates target genes. In the nucleus, β-catenin partners with different transcriptional coactivators, such as CBP and p300, through which it affects distinct subsets of genes. Additional effectors of Wnt/β-catenin signaling include R-spondins and their receptors LGR4/5, which potentiate Wnt signaling, and RNF43/ZRNF3 E3 ubiquitin ligases, which antagonize it. In addition, E-cadherin functions as a tumor suppressor that interferes with Wnt/β-catenin activity.

Transcriptional activity of β-catenin is determined by its cellular abundance and nuclear translocation and by interactions with transcription factors and transcriptional repressors and coactivators. β-Catenin affects diverse cellular functions through its complex protein structure that consists of 12 repeats of approximately 40 amino acids long, named armadillo repeats that together form a single, rigid protein domain, the armadillo (ARM) domain. Phosphorylation of the N-terminal region of β-catenin facilitates its binding to β-TrCP E3 ubiquitin ligase that targets β-catenin for degradation by the proteasome (Fig. 1). In contrast, the C-terminal region interacts with different binding partners and is critical for Wnt signaling. For some of these transcriptional partners, interactions have been mapped to specific domains of β-catenin in functional studies (Fig. 2) (reviewed in Valenta et al. 2012).

Figure 2.

Structure of β-catenin facilitates interactions with different nuclear partners. The β-catenin protein is comprised of 12 armadillo repeats (numbered boxes), an amino-terminal domain (NTD) and carboxy-terminal domain (CTD), and the conserved helix C as the flexible component of CTD. (A) At least 4 different transcription factors (purple bars) and 3 transcriptional inhibitors (gray bars) have been shown to interact with β-catenin and mapped to its specific structural domains. (B) Transcriptional coactivators bind to defined regions of β-catenin to mediate its diverse signaling outcomes (green bars).

In addition to binding Tcf/Lef, β-catenin directly interacts with other transcription factors to affect gene expression and tumor cell behavior (Fig. 2A). Under limiting oxygen conditions, hypoxia-induced factor 1α (HIF1α) is upregulated in abundance and competes with Tcf for β-catenin binding. Furthermore, increased oxidative stress is accompanied by enhanced levels of reactive oxygen species (ROS), leading to upregulation of forkhead transcription factors FOXO that also compete with Tcf for β-catenin. The significance of these Lef/Tcf-independent transcription partners of β-catenin has not been extensively explored in HNSCC, although hypoxia plays an important role in the pathobiology of this malignancy and in treatment resistance (Gammon and Mackenzie 2016; reviewed in Valenta et al. 2012).

Nuclear β-catenin has been shown to interact with numerous coactivators for which its binding domains have been mapped and give rise to distinct functional outcomes (Fig. 2B). Additional key binding partners include Bcl9, shown to promote early stages of intestinal cancer by controlling a subset of β-catenin target genes with roles in epithelial-to-mesenchymal transition (EMT) and invasion (Brembeck et al. 2011). Also, Kindlin2 binds to β-catenin and promotes Axin2-Snail target gene expression critical for cancer stem cells (CSCs), EMT, invasion, and metastases (Yu et al. 2012). Moreover, β-catenin interacts with Klf4 and c-myc to regulate telomerase reverse transcriptase (TERT) gene expression required for telomere maintenance in cancer (Hoffmeyer et al. 2012). Nuclear β-catenin also interacts with the effectors of the Hippo pathway YAP1 and TAZ to regulate the expression of genes associated with cancer, including SOX2, SNAI2, and BIRC5 (Rosenbluh et al. 2012). Notably, β-catenin has a central role in the epigenetic modification of chromatin (reviewed in Mosimann et al. 2009). Nuclear β-catenin interacts with the histone acetyltransferase cAMP-responsive element binding protein (CBP) to maintain a pluripotent state in embryonic tissues (Ma et al. 2005). In HNSCC, the β-catenin/CBP complex has been shown to recruit histone methyltransferase and myeloid/lymphoid or mixed-lineage leukemia 1 (MLL1) to methylate H3K4me3 and activate stem cell–related genes (Wend et al. 2013). Another coactivating partner of β-catenin is p300, a histone acetyltransferase closely related to CBP but with distinct patterns of gene activation (Ma et al. 2005) (Figs. 1 and 2B).

Because of its central roles in diverse cellular processes, levels of β-catenin are highly regulated. Besides the destruction complex, cytoplasmic levels of β-catenin are regulated by YAP1 and TAZ that target it for degradation by β-TrCP E3 ubiquitin ligase (Azzolin et al. 2014). Another level of regulation of β-catenin is imposed by E-cadherin via recycling of E-cadherin/β-catenin complexes from the cell membrane, a process likely to augment the cytoplasmic pool of β-catenin (Heuberger and Birchmeier 2010) (Fig. 1). Furthermore, β-catenin regulates its own abundance by affecting the expression Wnt/β-catenin pathway components, including its antagonists Dickkopf-1 (Dkk-1) and E3 ubiquitin ligases ZNRF3 and RNF43 that induce endocytosis of Wnt receptors (Gonzalez-Sancho et al. 2005; Hao et al. 2012; Koo et al. 2012).

Wnt/β-catenin signaling is dynamically regulated through posttranslational modifications, such as palmitoylation and glycosylation, which affect conformation, secretion, stability, and activity of its components (Komekado et al. 2007). Wnt secretion is also regulated by Wntless, a G protein–coupled receptor GPR177 of the retromer complex, through Wntless-containing exosomes (McGough and Vincent 2016). Also, the activity of the Wnt/β-catenin pathway is modulated through the function of its inhibitors and agonists, such as its antagonist, Dickkopf-1 (Dkk-1), Wnt inhibitory factor (WIF1), and secreted Frizzled-related proteins (SFRP) (Gonzalez-Sancho et al. 2005; Katoh 2005). In addition, R-spondins (RSPO1–4) synergize with Wnt to activate β-catenin-mediated signaling by forming complexes with their receptors LGR4–6 that promote clearance of the RNF43/ZNRF3 E3 ligases to increase Wnt receptor levels to potentiate Wnt signaling (Carmon et al. 2011; Hao et al. 2012; Koo et al. 2012).

As a master regulatory pathway, Wnt/β-catenin signaling cross-talks with other signaling pathways. In mammary and intestinal tumors, Wnt/β-catenin collaborates with TGFβ on a transcriptional level, promoting EMT and fibrosis (Labbe et al. 2007). Also, Wnt/β-catenin collaborates with Notch during development and tumorigenesis (Kwon et al. 2011). Many HNSCC HPV-negative tumors harbor inactivating mutations in NOTCH1, which are associated with enhanced β-catenin activity (Duan et al. 2006; Cancer Genome Atlas 2015; Beck and Golemis 2016). Moreover, β-catenin interacts with the Hippo pathway and its effectors, transcriptional coactivators YAP/TAZ. The Hippo tumor suppressor pathway inhibits the oncogenic activities of YAP/TAZ through their retention in the cytosol, and hyperactivation of YAP/TAZ contributes to aggressive carcinomas, including HNSCC (Hiemer et al. 2015). Another prominent signaling pathway that interacts with Wnt/β-catenin signaling is the EGFR signaling cascade (Hu and Li 2010). The Wnt/β-catenin and EGFR signaling pathways have the ability to transactivate one another through Fzd receptors and via the PI3K/Akt or the ERK pathways (Hu and Li 2010). Furthermore, EGFR has been shown to form a complex with β-catenin in the cytoplasm (Schroeder et al. 2002) and to regulate its localization and stability in HNSCC (Lee et al. 2010). Last, PIK3CA is frequently amplified or harbors activating mutations in HNSCC (Lui et al. 2013). In colon cancer, PI3K and Wnt/β-catenin pathways converge at the level of the Akt substrate FOXO3a to induce IQ motif-containing GTPase activating protein IQGAP2 that inhibits E-cadherin adhesion and promotes cell migration and metastatic potential (Tenbaum et al. 2012).

Similar to other carcinomas, HNSCC cells undergo dramatic changes in their metabolic programs (reviewed in Cairns et al. 2011). During tumor progression, metabolic reprogramming is critical for the maintenance of CSCs and for communication between the tumor epithelia and the stroma (Porporato et al. 2014; Luo and Wicha 2015). As oxygen becomes limiting, cells stabilize HIF1α to upregulate glucose transporters and glycolytic enzymes to promote glycolysis, and β-catenin is diverted from Tcf to enhance transcriptional activity of HIF1α (Mazumdar et al. 2010).

One of the key metabolic processes that interacts with Wnt/β-catenin signaling is the protein N-glycosylation pathway, which modifies protumorigenic proteins with N-glycans (Liu et al. 2013) (reviewed in Varelas et al. 2014). Wnt/β-catenin signaling controls protein N-glycosylation through transcriptional regulation of the DPAGT1 gene at the first committed step in the lipid-linked assembly pathway in the endoplasmic reticulum (ER). Aberrant induction of DPAGT1 by β-catenin promotes a positive feedback loop with Wnt/β-catenin signaling that represses E-cadherin adhesion in HNSCC.

Wnt/β-catenin Signaling in Cancer Stem Cells

Maintenance of tissue homeostasis and repair depends on the activity of tissue-specific stem cells that display the capacity for long-term self-renewal and ability to give rise to 1 or more differentiated cell lineages. Cancer tissues, including HNSCC, also contain small subpopulations of cells with stem-like properties, or CSCs that give rise to tumors with hierarchal organization (Prince et al. 2007; Wend et al. 2013).

An acknowledged function of β-catenin is in the regulation of cell fates in development and cancer (reviewed in Holland et al. 2013). β-Catenin controls asymmetric cell division, a mechanism that generates cells capable of renewing themselves and committed progenitor cells that differentiate into a specified cell type by promoting unequal distribution of Dvl, Fzd, Axin, and APC in the cytoplasm of the mother cell (Lien and Fuchs 2014). The decision to divide or differentiate is controlled by a stem cell niche formed by extrinsic factors, including Wnt ligands, and by cells that produce them. The niche provides signals that promote stem cell phenotypes in multiple epithelial stem cell compartments during development (Yang et al. 2017). In HNSCC, endothelial cells have been shown to secrete EGF that promotes stem cell–like phenotypes, indicating that such a scenario is likely at play in this malignancy (Zhang et al. 2014).

In HNSCC, CSCs were first characterized by the expression of the CD44 surface marker and the Bmi1 oncogene (Prince et al. 2007). These cells were also shown to be marked by high aldehyde dehydrogenase activity, by the expression of c-Met and SOX2, and by the ability to efflux vital dyes and grow under nonadherent conditions as tumor spheres. Moreover, HNSCC CSCs had the ability to seed tumors at low numbers in nude mice and to drive cell expansion (Prince et al. 2007; Clay et al. 2010; Krishnamurthy et al. 2010; Wend et al. 2013). Also, HNSCC CSCs have been aligned with mesenchymal properties, resistance to treatments, and tumor recurrence (Nör et al. 2014; Chen et al. 2017).

To date, multiple components of Wnt/β-catenin signaling have been associated with CSCs. For instance, the R-spondin receptor Lgr5 is a potential CSC marker in intestinal stem cells that can promote tumor growth when APC is deleted (Schepers et al. 2012). In colon adenoma, Lgr5-positive cells can give rise to Lgr5-positive cells and other cell types. Furthermore, RAC1 is required for expansion of the Lgr5 population after APC loss (Myant et al. 2013). RAC1 activation drives ROS production and activates NF-κB signaling, which then enhances Wnt signaling (Myant et al. 2013).

Significantly, HNSCC CSCs were also characterized by an aberrant activation of Wnt/β-catenin activity (Wend et al. 2013; Lee et al. 2014). Recent studies provide evidence that a central role of β-catenin signaling in CSCs derives from its control of epigenetic changes that define distinct chromatin states (reviewed in Mosimann et al. 2009). Specifically, β-catenin promotes H3K4me3 by recruitment of CBP (Parker et al. 2008) and histone methyltransferase, MLL, to induce active chromatin and maintain immature cell phenotypes (Sierra et al. 2006). Indeed, increased interaction of β-catenin with CBP underlies HNSCC CSCs and disease progression (Zhou et al. 2012; Wend et al. 2013; Lenz and Kahn 2014; Chan et al. 2015). This ability to promote aggressive properties is specific to CBP, as recruitment of the p300 histone acetyltransferase by β-catenin drives cellular differentiation (Li et al. 2007).

Wnt/β-catenin Signaling in E-cadherin Adhesion and EMT

In addition to its signaling function, β-catenin serves as a major structural component of E-cadherin–mediated multiprotein complexes, or AJs that function in maintaining epithelial cell-cell adhesion and cell polarity. AJs are powerful tumor suppressors, in part, by inhibiting Wnt/β-catenin signaling (Birchmeier 1994). The major components of AJs include members of the catenin family: β-catenin, α-catenin, and p120 catenin. The C-terminal region of E-cadherin binds β-catenin, and E-cadherin/β-catenin complexes recruit α-catenin to facilitate interaction with the actin cytoskeleton, while the juxtamembrane domain of E-cadherin binds p120 catenin that regulates its turnover. Loss of either α-catenin or p120 from AJs leads to reduced cell-cell adhesion and the development of high-grade cancers (Benjamin and Nelson 2008; Kourtidis et al. 2015). In addition, in HNSCC, β-catenin indirectly inhibits AJs by inducing the expression of the N-glycosylation gene, DPAGT1, leading to their extensive N-glycosylation and destabilization (reviewed in Varelas et al. 2014). Moreover, aberrant activation of β-catenin in HNSCC is associated with mutations in FAT1, a member of the cadherin superfamily (reviewed in Cancer Genome Atlas 2015; Beck and Golemis 2016).

Inappropriate downregulation of E-cadherin is associated with EMT, a developmental program that involves loss of intercellular adhesion and apical-basal polarity, reorganization of cytoskeleton, and increased expression of vimentin, N-cadherin, and fibronectin concomitant with acquisition of mesenchymal traits (Kalluri and Weinberg 2009). Similar to other carcinomas, HNSCC cells adopt EMT to drive progression to advanced disease (Pectasides et al. 2014). Wnt/β-catenin signaling regulates key transcription factors responsible for EMT, including SNAI2 and the zinc finger E-box binding homeobox 1 protein (ZEB1), among others (Baum et al. 2008). Our own studies identified β-catenin–dependent increased expression of collagen triple helix containing 1 (CTHRC1) to be associated with enhanced HNSCC cell migration through collaboration with protein N-glycosylation and induction of the Wnt/PCP pathway (Liu et al. 2013). In addition, important regulators of EMT are microRNAs (miRNAs) that are frequently dysregulated via Wnt/β-catenin signaling (reviewed in Ghahhari and Babashah 2015). In HNSCC, deregulation of miRNA200b and miRNA15b has been linked to EMT and advanced disease (Sun et al. 2012).

Wnt/β-catenin Signaling in the Tumor Microenvironment

The tumor microenvironment (TME) undergoes dynamic alterations with cancer progression to advanced disease and contributes to cancer cell plasticity. Numerous studies highlight contributions of distinct components of the TME to HNSCC tumorigenesis (reviewed in Koontongkaew 2013). The TME includes the extracellular matrix (ECM), a noncellular structure that provides physical support and elasticity to tissues and transmits signals to epithelia by serving as ligands for receptors and regulates availability of growth factors and morphogens, such as Wnts. Many downstream targets of β-catenin are components of the ECM, including laminin, a key protein in the basement membrane, and lysyl oxidase and fibronectin that reside in the interstitial matrix, as well as invasion-associated genes such as MMP-9, MMP-74, and CD44 ligands (reviewed in Bonnans et al. 2014).

The TME also includes the stroma with fibroblasts and cells of the innate and adaptive immune systems with important roles in cancer progression. Wnt/β-catenin signaling is recognized for its inhibitory effects on innate and adaptive immune systems in cancer (Swafford and Manicassamy 2015), which are associated with immune tolerance, a major mechanism underlying failed host antitumor immune responses (reviewed in Pai et al. 2017). β-Catenin inhibits dendritic cells via activation of the ATF3 repressor activity, which downregulates the CCL4 chemokine required for the recruitment of CD103+ dendritic cells to the tumor (Swafford and Manicassamy 2015). Impaired recruitment of dendritic cells leads to diminished activation of CD8+ T cells. Indeed, β-catenin signaling has been aligned with non–T cell–inflamed tumors by promoting Treg persistence and activity, thus suppressing adaptive responses by reducing CD8+ T-cell proliferation, activation, and effector functions (Spranger and Gajewski 2015).

Wnt/β-catenin Signaling as a Therapeutic Target for HNSCC

Due to the important roles of β-catenin as a structural component of AJs and as a regulator of CSCs in cell renewal during homeostasis, the benefits of targeting Wnt/β-catenin signaling have been controversial (Kahn 2014). Nonetheless, Wnt/β-catenin signaling has been increasingly recognized as an attractive target for HNSCC therapy (Aminuddin and Ng 2016). To date, a number of Wnt inhibitors have been developed and tested for antitumor properties in preclinical models, with some moving on to clinical trials (Table). A key member of the Wnt/β-catenin cascade is the acyl-transferase Porcupine, for which inhibitors have been developed, including IWP (Dodge et al. 2012), LGK974 (Kulak et al. 2015), ETC-159 (Madan et al. 2016), and C59 (Proffitt et al. 2013). LGK974 has been shown to inhibit HNSCC growth and metastasis using the chick chorioallantoic membrane (CAM) assay (Rudy et al. 2016). In addition, compounds targeting extracellular Wnt ligands and their receptors are undergoing evaluation in clinical trials (reviewed in Zhan et al. 2017). They include OMP-54F28, a fusion Fzd8-Fc decoy receptor, and antibodies against Fzd-7 (OMP18R5) and R-spondin3 (OMP131R10), as well as Wnt5a mimetic (Foxy-5), among others (Table). Also, inhibitors of tankyrase-mediated adenosine diphosphate (ADP)–ribosylation that increase Axin levels are in development, including XAV939 (Huang et al. 2009) and IWR (Kulak et al. 2015), although they have not moved on to clinical trials.

Table.

Inhibitors of Wnt/β-catenin Signaling and Clinical Trials Targeting the Wnt/β-catenin Pathway.

Compound	Target	Reference	Trial Phase; Identifier	Tumor Sites
IWP	Porcupine	Dodge et al. 2012	None	NA
LGK974	Porcupine	Kulak et al. 2015	1/2; NCT02649530	Metastatic colorectal cancer with Wnt pathway mutations; HNSCC with Notch receptor mutations
ETC-159	Porcupine	Madan et al. 2016	1; NCT02521844	R-spondin fusion-positive colorectal carcinoma
C59	Porcupine	Proffitt et al. 2013	None	NA
OMP-54F28 (Ipafricept)	FZD8-Fc Decoy receptor Targets CSCs	OncoMed 2014 (Clinicaltrials.gov)	1; NCT02092363 NCT02050178	Hepatocellular and pancreatic ductal carcinoma, ovarian cancer
OMP18R5	Anti-FZD7 antibody	Gurney et al. 2012	1; NCT01957007 NCT02005315 NCT01973309	Non–small lung cancer, pancreatic ductal adenocarcinoma, breast cancer
OMP131R10	Anti–R-spondin 3 antibody	OncoMed 2015 (Clinicaltrials.gov)	1; NCT02482441	Metastatic colorectal cancer (RSPO3 positive)
Foxy-5	Wnt5a mimetic	WntResearch 2016 (Clinicaltrials.gov)	1; NCT02655952	Breast, colorectal, and prostate cancer
XAV939	Tankyrase	Huang et al. 2009	None	NA
IWR	Tankyrase	Kulak et al. 2015	None	NA
ICG-001	β-Catenin–CBP interaction	Emami et al. 2004 Wend et al. 2013	None	NA
PRI-724, second-generation ICG-001 derivative	β-catenin–CBP interaction	Prism Pharma 2012 (Clinicaltrials.gov)	1/2; NCT01606579 NCT02413853 NCT01764477	Myelogenous leukemia; colorectal and pancreatic adenocarcinoma
E7386, third-generation ICG-001 derivative	β-Catenin–CBP interaction	Eisai, Inc. 2017 (Clinicaltrials.gov)	1; NCT032664	Advanced neoplasms; planned expansion to HNSCC

CBP, cAMP-responsive element binding protein; CSC, cancer stem cell; HNSCC, head and neck squamous cell carcinoma; NA, not applicable.

Increased insights into the mechanisms of β-catenin function in the nucleus have provided support for its potential to serve as a druggable target (Aminuddin and Ng 2016). Recently, a small-molecule inhibitor of nuclear β-catenin activity, axitinib, has been shown to downregulate nuclear β-catenin abundance through stabilization of ubiquitin ligase SHPRH (SNF2, histone linker, PHD, and RNIG finger domain-containing helicase) (Qu et al. 2016). Our own ongoing investigations have shown that a small-molecule inhibitor of the interaction between β-catenin and CBP in the nucleus, ICG-001, suppresses HNSCC CSCs and tumor growth in preclinical models (Kartha et al., unpublished data, 2018). Furthermore, an active metabolite of a closely related compound to ICG-001, PRI-724, is undergoing clinical trials in patients with leukemia and colorectal and pancreatic cancer. A third-generation derivative of ICG-001, E7386, is currently in phase I clinical trials in patients with HNSCC (Table). Importantly, inhibitors of the β-catenin/CBP axis are likely not only to target CSCs but also to intercept oncogenic activities in the tumor microenvironment, as β-catenin activity has been associated with the fibrotic process and immune exhaustion. Thus, it is likely that targeting the β-catenin–CBP interaction will have intratumoral inhibitory effects, as proposed in Figure 3. Last, inhibiting oncogenic pathways in combination with Wnt/β-catenin signaling offers improvement over current monotherapies and standard radiation therapy/chemotherapy for the treatment of HNSCC. In particular, combination therapies targeting the β-catenin/CBP axis and an immune checkpoint blockade may enhance the deletion of aggressive CSCs while increasing anticancer immune cell function (Fig. 3).

Figure 3.

Targeting the β-catenin/cAMP-responsive element binding protein (CBP) axis in head and neck squamous cell carcinoma (HNSCC). Inhibition of the β-catenin–CBP interaction is proposed to abrogate subpopulations of tumor cancer stem cells (CSCs) and promote epithelial cell differentiation. In addition, reduced β-catenin/CBP activity is expected to induce antitumor immunity by depleting exhausted T cells and decreasing the frequency of immature heterogeneous cells of the myeloid lineage, or myeloid-derived suppressor cells (MDSCs) involved in immune evasion.

Future Directions

Mounting evidence indicates that Wnt/β-catenin signaling is at the nexus of both tumor epithelial and stromal deregulation, including the ECM and immune dysfunction that collectively contribute to this malignancy. The complexity and dynamic nature of Wnt/β-catenin signaling represent a challenge to the design of therapeutic strategies that target the oncogenic functions of β-catenin while not affecting its structural contributions. Future studies focused on decoding the biology of stem cells and their lineages and evolving niches, including control of pluripotency and epigenetic regulation, as well as understanding how cellular and metabolic plasticity dictate interpretation of signals in the context of Wnt/β-catenin activities, will improve our understanding of this pathway in HNSCC. Indeed, recent single-cell sequencing of HNSCC has revealed a complex ecosystem with diverse cell populations exhibiting distinct gene signatures with different tumor-promoting potential (Puram et al. 2017). Future studies will continue to leverage computational and system biology methodologies to map signaling networks and cellular metabolism in different tumor cell populations and apply emerging technologies in genome and epigenome editing to elucidate causal relationships between chromatin organization and regulatory processes across the genome that are deregulated by β-catenin and its colluding partners in HNSCC. Also, comprehensive characterization of gene expression changes at the protein level, including alterations in posttranslational modifications (i.e., lipidation, glycosylation) and protein-protein interactions, will contribute new insights into how to interfere with unchecked Wnt/β-catenin transcriptional signals to improve HNSCC therapy.

Author Contributions

K.A. Alamoud, contributed to data acquisition and analysis, drafted the manuscript; M.A. Kukuruzinska, contributed to conception and design, critically revised the manuscript. Both authors gave final approval and agree to be accountable for all aspects of the work.