1. Introduction
CRC is considered to be a highly metastatic disease that occurs in the colon, rectum and appendix (1). Globally, CRC is one of the most common carcinomas and has remained as the second leading cause of carcinoma-associated mortality since 2008 (2). Generally, patients with CRC in stages 1 and 2 have a relatively favorable prognosis and 5-year survival rate, whereas the prognosis of those in stage 4 is very poor (3). Only 39% of patients with CRC are diagnosed in the early stages (4). Increased data have shown that the 5-year survival rate of patients with metastatic CRC is 10-15%, indicating a poor prognosis (5). Additionally, invasion and metastasis contribute to 90% of CRC-related mortalities (6). For advanced or metastatic CRC, surgical removal alone fails to be curative. Accordingly, understanding the underlying molecular mechanisms of CRC progression and metastasis has become increasingly important. It has been demonstrated that the Wnt/β-catenin signaling pathway is implicated in the proliferation, invasion and metastasis of CRC cells (7,8). The Wnt signaling pathway is divided into two main categories, the non-canonical (β-catenin-independent) and canonical (β-catenin-dependent) pathways (9,10). Wnt ligands are an evolutionary conserved family of 19 secreted glycoproteins produced in the endoplasmic reticulum (ER) of mammalians (11). β-catenin is one of the Armadillo repeat protein superfamily and a primary transducer of Wnt signals (12). In the present review, the potential underlying mechanism of the canonical Wnt pathway in CRC and possible therapeutic strategies were explored to further understand the mode of action of this canonical pathway on CRC.
In the absence of Wnt ligands, cytoplasmic β-catenin is targeted for proteasome-mediated degradation via the association with a destruction complex (13), mainly including adenomatous polyposis coli (APC), axis inhibition protein 1 (Axin1), glycogen synthase kinase-3β (GSK-3β), casein kinase 1 (CK1) and E3-ubiquitinligase β-transducin repeats-containing protein (β-TrCP) (14). In the presence of Wnt ligands, the biding of Wnt ligands to their cognate receptors [frizzled (FZD)/low-density lipoprotein receptor-related protein 5 or 6 receptor (LRP5/6)] triggers the recruitment of Axin1 to the Axin-binding molecule Dishevelled (DVL), leading to the disassembly of the destruction complex (14,15). Following the disassembly, the stabled β-catenin accumulates in the cytoplasm, and subsequently translocates into the nucleus where it binds to TCF/LEF, together with the recruitment of co-activator complexes to activate the downstream target genes of Wnt/β-catenin signaling (16). These downstream genes mainly include cyclin D1, c-Myc, matrix metalloproteinase (MMP)-7, leucine-rich repeat-containing G protein-coupled receptor 5 (Lgr5), CD44, Axin2 (17,18) and surviving (19). A schematic diagram depicting the molecular mechanism of the members of the Wnt/β-catenin signaling pathway involved in inducing the expression of its downstream target genes is presented in Fig. 1. To develop effective targeted therapies, it is necessary to understand the molecular mechanisms of Wnt/β-catenin-mediated CRC proliferation, invasion and metastasis. This review focused on the relationship between the Wnt/β-catenin signaling pathway and the promising therapeutic targets for CRC. The present review included 178 references. Databases including 'Pubmed', 'SpringerLink' and 'Clinical key' were used to search for the cited studies. The key words searched included 'Wnt/β-catenin', 'colorectal cancer', 'CRC', 'therapy', and 'treatment'.
2. Role of Wnt/β-catenin signaling in colorectal cancer (CRC)
A previous study has demonstrated that activation of the Wnt/β-catenin signaling pathway and overexpression of its downstream genes, such as c-Myc and cyclin D1, promote CRC tumorigenesis (8). Additionally, several mechanisms by which the Wnt/β-catenin signaling pathway is aberrantly activated are presented below.
Aberrant accumulation of β-catenin
Loss of APC contributes to the aberrant accumulation of nuclear β-catenin, which is a hallmark of CRC progression and activation of Wnt/β-catenin signaling (20). β-catenin knockdown effectively promotes the apoptosis of CRC cells (21). The cooperation of nuclear β-catenin with TCF/LEF induces the transcription of c-MYC oncogene, which is upregulated in CRC (22). Increased cyclin D1 transcription as a result of high levels of β-catenin occurs in colon cancer cells (23). Furthermore, c-Myc and cyclin D1 together with APC gene mutations contribute to autonomous proliferation of colon cancer cells (24). Therefore, aberrant activation of Wnt/β-catenin signaling due to the accumulation of β-catenin may contribute to the initiation and progression of CRC via the upregulation of cyclin D1 and c-Myc (25). Additionally, high levels of nuclear β-catenin expression occur at the invasion front of CRC, and they are associated with poor prognosis (26). These findings indicate that the aberrant accumulation of nuclear β-catenin and the failure of β-catenin degradation contribute to CRC development.
Mutations
Investigation of the underlying mechanism of how Wnt/β-catenin-related mutations affect CRC has demonstrated that mutations aberrantly activate this pathway leading to the onset of CRC tumorigenesis. It is reported that mutations in the Wnt/β-catenin signaling pathway contribute to initiating the majority of CRC (27,28) and 90% of sporadic patients with CRC harbor mutations in the core components of the Wnt/β-catenin pathway (29). The development and progression of CRC due to mutations in this pathway involve a cascade of genetic events. A previous study suggested significantly aberrant methylation in mutated genes, including Wnt2, Wnt3A, Wnt6, FZD5, LRP5 and TCF7L1 in sporadic CRC (30). Wnt ligands, such as Wnt4, are reported to be upregulated in the serum of patients with CRC and CRC tissues (31). Extensive evidence indicates that mutations in APC, β-catenin, Axin1, Axin2 and TCF-4 initiate the process of CRC tumorigenesis (32-38). These mutations are identified in the earliest neoplasms showing that this pathway plays a vital role in colorectal tumorigenesis (39). Although a study has suggested that mutations in multiple components of this pathway occur in the hyper-mutated neoplasms, most neoplasms harbor a mutation in a single component of the Wnt/β-catenin pathway (14,40).
CRC cell stemness and epithelial-mesenchymal transition (EMT)
Cancer stem-like cells (CSCs) and EMT are major factors contributing to the metastasis of cancer cells (6,41). Patients with CRC develop recurrence and metastasis owing to the survival of CSCs. Numerous studies have provided evidence for the role of EMT in the acquisition of the CSC phenotype in CRC and have suggested that EMT may facilitate CSC formation (42,43). It is currently not understood how EMT enables CSCs to reside in the CRC microenvironment to escape the effects of chemotherapy. Recently, accumulating evidence has shown the pivotal role of Wnt/β-catenin signaling in CRC stem-like cells and EMT (41,44-46).
In CRC, aberrant activation of Wnt/β-catenin signaling disrupts the Wnt/β-catenin gradient, leading to the hyperplasia of intestinal stem cells along the crypt-villus axis and contributing to tumor initiation via CRC stem-like cells (47). To clarify the underlying molecular mechanisms of CRC stem-like cells in Wnt/β-catenin signaling, an increasing number of studies have focused on identifying the markers of CRC stem cells. For example, Lgr5 has been identified as a surface marker of CRC stem cells, and its high expression is involved in the development and progression of CRC (21). In 2007, CD133 was first identified as a potential specific marker of colon CSCs, although the role of CD133 has been questioned (42). A recent meta-analysis study suggested that CD133 is strongly associated with poor prognosis and vascular invasion in CRC (48). Additionally, great progress has been achieved in identifying other CRC stem-like cell markers, such as aldehydedehydrogenase-1, homeobox protein NANOG, achaete-scute family bhlh transcription factor 2, SRY-Box 2 and octamer-binding protein 3/4, as well as CRC stem-like cell surface markers, such as CD24, CD29, CD44 and CD166 (45,49,50).
Wnt/β-catenin signaling plays a critical role in EMT via inducing the expression of EMT-associated nuclear factor-κB (51). Wnt/β-catenin signaling may promote EMT, in which β-catenin and GSK-3β are two key players (46). Ectopic expression of β-catenin, which is strongly associated with EMT in CRC cells, contributes to the dedifferentiation of post-mitotic intestinal epithelial cells, resulting in the generation of tumor-initiating cells (52). In the Wnt/β-catenin signaling pathway, the E-cadherin/β-catenin complex usually exists at the membrane, contributing to tight cell-cell interactions and inhibition of cell mobility (53). The disturbance of this complex may promote the nuclear translocation of β-catenin and induce the transcription of EMT-promoting genes (54). EMT markers mainly contain the epithelial markers [E-cadherin and tight junction protein ZO-1 (ZO-1)] and the mesenchymal markers [N-cadherin, vimentin, Snail, MMP-9, Smad2, fibronectin and Zinc finger E-box binding homeobox 1 (ZEB1)] (1,6,55-57). TCF-4, cyclinD1, c-Myc, Twist-related protein 1 (Twist) and MMP-14 are active factors in the EMT process (58). Snail, a direct target of the β-catenin/TCF complex and also an EMT-initiating gene, is upregulated in the process of EMT (59) and is known to activate EMT during cancer metastasis (12). Recently, certain results suggested that β-catenin-mediated expression of MMPs also induces EMT and increases the migratory and invasive capabilities of CRC cells via MMP-mediated degradation of the extracellular matrix (1,60).
Abnormality of non-coding RNAs (ncRNAs)
NcRNAs constitute more than 90% of RNAs from the human genome, and according to their size ncRNAs can be divided into different classes including small ncRNAs [such as microRNA (miRNA), piRNAs and tsRNAs], long non-coding RNAs (lncRNAs) and subclasses of lncRNAs (such as circRNAs and pseudogenes) (61). Aberrantly expressed ncRNAs have been demonstrated to be associated with CRC tumorigenesis (25,62-67). Additionally, the aberrant expression of ncRNAs is correlated to the clinicopathological characteristics of CRC, such as prognosis (68), survival (69,70) and metastasis (1,7,8,62,71-75). In the present review, the aberrant expression of ncRNAs and their effects on CRC in the Wnt/β-catenin signaling pathway are presented in Table I (5,7,8,25,36,62-70,72-94).
miRNAs, with a length of 20-24 nucleotides, have been demonstrated to be post-transcriptional regulators in tumorigenesis (5,7). Among the dysregulated miRNAs, some are upregulated (69,70), however others are downregulated (76,80) in CRC cell lines or tissues. miRNAs are the most extensively studied in cancers compared with other small ncRNA species, such as piRNAs and tsRNAs (61). As revealed in Table I, certain miRNAs as oncogenes, promote CRC tumorigenesis and drug resistance combined with other regulators or pathways, including: i) Negatively regulating DKK1 by binding to Wnt-receptor complexes (LRP5 and LRP6) (5); ii) upregulating β-catenin (5,78) and phosphorylated GSK-3β (5,78), and downregulating GSK-3β (69,78) and phosphorylated-β-catenin (78); iii) promoting the expression of Wnt/β-catenin signaling responsive target genes, such as Lgr5 (70), CD44 (70) and Axin2 (70); iv) upregulating ABCG2 through Wnt/β-catenin signaling (ABCG2 is a major multidrug resistance pump that enables cancer cells to resist chemotherapeutic drugs. Wnt/β-catenin signaling is one of the upstream mechanisms to regulate the expression of ABCG2 in CRC stem-like cells. The upregulation of miRNA-199a/b contributes to the drug resistance of CRC stem-like cells via upregulating ABCG2 due to the downregulation of GSK-3β and the upregulation of β-catenin.) (69); v) inhibiting the HBP1-mediated inactivation of Wnt/β-catenin signaling (70); vi) inducing EMT via the upregulation of vimentin (78) and N-cadherin (78), and downregulation of E-cadherin (78); and vii) targeting FOXL1 leading to activation ofWnt/β-catenin signaling (79).
LncRNAs, with a length >200 nucleotides (nt), may act as oncogenes that promote CRC growth and are often upregulated in CRC (25,62-67,73,84,86). The activation of the Wnt/β-catenin signaling pathway is closely associated with the abnormal expression of lncRNAs in CRC (25) and this pathway is involved in lncRNA-mediated CRC progression (63). Accumulating evidence has shown that the aberrant expression of lncRNAs is strongly associated with the β-catenin-mediated upregulation of Wnt target genes. For example, lncRNAs promote the expression of cyclin D1, c-Myc (25,62,73,82,85,88,95), CD44 (85), MMP-7 (62,73), Wnt3a (64), Oct3/4 (85) and Survivin (64). Apart from the upregulation of these Wnt target genes, certain lncRNAs are capable of increasing the expression levels of TCF/LEF transcription factor (73) and CDK1 (88). Additionally, the aberrant expression of lncRNAsis also able to downregulate GSK-3β (62,88). Certain lncRNAs promote CRC progression via sponging miRNAs, for example, lncRNA-HCG18 sponging miR-1271 (66) and lncRNA-UCID sponging miR-152-3p (67). LncRNAsact as competing endogenous RNAs (ceRNAs) to be involved inWnt/β-catenin signaling via sponging miRNAs (65,96), for example, lncRNA-PART1 sponging miR-150-5p/miR-520h (86) and lncRNA-NEAT1 sponging miR-486-5p (89). Selectively binding miRNAs by lncRNAs contributes to inhibiting their transcription and promoting their degradation and induces EMT of CRC cells (63,72,90). In addition, those lncRNAs also promote the expression of nuclear β-catenin, cyclin D1, c-Myc, N-cadherin, vimentin, CD44, Jun and TCF1 via activating transcription factor and targeting miRNA, such as the lncRNA-ARAP1-AS1/YY1 (63) and lncRNA-AC010789.1/miR-432-3p/ZEB1 axes (90).
CircRNAs exhibit specific roles as the sponges of miRNAs (74). Additionally, numerous studies have reported that circRNAs as 'miRNA sponges' play a vital role in promoting CRC initiation and progression and inhibiting apoptosis (74,94,97), for example, circRNA-MBOAT2 sponging miR-519d-3p (91), circRNA-0082182 sponging miR-1205 and miR-411 (74) and circRNA-100290 sponging miR-516b to promote FZD4 expression (93). The exact mechanisms of the functional roles of circRNA in CRC remain elusive. Additionally, the upregulation of circRNA-PDSS1 promotes migratory ability and angiogenesis in CRC cells via the upregulation of β-catenin, c-Myc, cyclin D1 and MMP-9, leading to lymphatic metastasis and distant metastasis (75).
Other abnormalities
The high levels of nuclear β-catenin and the defect of E-cadherin are founded in CRC (98). E-cadherin localized with β-catenin in the nucleus suppresses the LEF-1-mediated activation of the Wnt downstream target genes via recruiting β-catenin and prevents its nuclear localization, leading to inhibiting the interaction between β-catenin and LEF-1 (99). Another study also reported that the loss of E-cadherin leads to the accumulation of free β-catenin in the cytoplasm and promotes its interaction with LEF-1 (100). Additionally, other abnormalities in Wnt/β-catenin signaling also aberrantly activate this pathway to promote CRC growth and metastasis. The aberrant expression of FZDs is associated with tumorigenesis, particularly in CRC (3). The upregulation of DVL has also been reported to activate Wnt/β-catenin signaling in CRC (15).
3. Therapeutic strategies targeting Wnt/β-catenin signaling for CRC
Currently, common therapies for CRC mainly include surgical removal of the carcinoma, chemotherapy and radiotherapy. The conventional treatment strategy for patients with CRC is the surgical removal of the CRC tumor combined with chemotherapy (92). The main chemotherapeutic drugs for CRC include oxaliplatin, irinotecan and 5-fluorouracil (5-FU) (101). 5-FU is the most commonly used standard drug for CRC chemotherapy (92). Radiotherapy remains a major adjuvant therapy for patients with advanced CRC. Therefore, improved understanding of the molecular mechanism of the development and progression of CRC may contribute towards the development of drugs to prevent the migration and invasion of CRC cells and aid in the identification of novel therapeutic targets and effective diagnostic and prognostic indicators. CRC is associated with the aberrant activation of the Wnt/β-catenin signaling pathway due to a combination of factors, therefore, in the present review, therapeutic targets for CRC are mainly divided into four categories, as summarized in Fig. 2.
Biological factor-associated targets
CRC stemness-associated therapeutic targets
A number of researchers have studied the interaction between CRC stem cell markers and Wnt/β-catenin signaling. Chen et al found that phenethyl isothiocyanate (PEITC) is an effective inhibitor of CRC stem cells by targeting Wnt/β-catenin signaling (44). In CRC stem cells treated with PEITC, decreased phosphorylated-GSK-3β, nuclear β-catenin, and CD133 were observed. Diallyl trisulfide inhibits CRC stem cells by promoting the expression of GSK-3β and inhibiting the expression of β-catenin and CRC stem cell markers in CRC stem cells (49). IC-2, a derivative of Wnt/β-catenin signaling inhibitor ICG-001, inhibits the expression of CRC stem cell marker proteins, as well as the sphere formation activity of CRC cells (102). ICG-001 promotes cell differentiation, suppresses cell proliferation, and induces apoptosis via binding to CREB-binding protein in CRC cells (103,104), suggesting that IC-2 may also bind to this protein to suppress the sphere formation of CRC cells via inhibition of the expression of CRC stem cell markers.
However, some proteins drive CRC cell stemness by activating this pathway. For example, transmembrane 4 L6 family member 1, a tumor-associated antigen, promotes invasion, tumorigenesis, recurrence and metastasis via the upregulation of stemness markers, N-cadherin, MMP-7, MMP-9, Smad2, vimentin, c-Myc, Axin2 and TCF7, and the downregulation of E-cadherin (6,105). A recent study reported that human anterior gradient-2 (AGR2, an oncogene) activates the non-canonical Wnt signaling pathway via the upregulation of Wnt 11, and subsequently promotes the migration of CRC cells (106). Later, Dahal Lamichane et al studied the effect of AGR2 on spheroids containing CRC stem-like cells via the activated canonical Wnt signaling pathway (50). Their study revealed that AGR2 is upregulated in CRC, and promotes the expression of CRC stem cell markers. Another oncogene, inhibitor of differentiation/DNA binding 1 (Id-1) is upregulated in CRC, and patients with CRC with lymph node metastasis have a higher level of Id-1 expression than those without lymph node metastasis (107). Id-1 expression is strongly associated with CD133 expression, and Id-1 promotes CRC stem cell-like capacity and induces EMT, partially via activating the Wnt/β-catenin and sonic hedgehog protein signaling pathways (42). Anterior gradient 3 (AGR3) is also upregulated in CRC, and silencing of AGR3 decreases nuclear β-catenin and inhibits the expression of CRC stem cell markers in CRC cells (45).
EMT-associated therapeutic targets
Qi et al identified three novel Wnt/β-catenin/TCF4 target genes, nuclear factor interleukin-3-regulated protein, cyclin-dependent kinases regulatory subunit 2, and ribosome biogenesis protein BOP1, which inhibit E-cadherin expression and increase the expression levels of vimentin, fibronectin and N-cadherin (56). Additionally, accumulating evidence has shown that certain genes that promote the expression of EMT markers are upregulated in CRC, and their dysregulation is strongly associated with CRC via activating Wnt/β-catenin signaling, such as tumor suppressor candidate 3 (57), cell migration-inducing hyaluronan binding protein (12), human rhomboid family-1 gene (58), ubiquitin-like modifier activating enzyme 2 (108), stromal cell-derived factor (109) and runt-related transcription factor 1 (71). Their overexpression promotes the expression of β-catenin (57,71,108,109), Snail (12,71), CD44 (71), c-Myc (58,71), MMP-2 (108), Met (71), MMP-7 (108), c-Jun (71), MMP-9 (108), MMP-14 (58), TCF-1/7 (71,108), vimentin (12,57,71,108,109), ZO-1 (12), LEF1 (71), N-cadherin (12,71,109), phosphorylated-GSK-3β (58,108), Twist (58), ZEB1 (71) and cyclinD1 (58,71), and inhibits the expression of GSK-3β (57,58), E-cadherin (12,57,71,108,109), APC (58) and Axin-2 (58).
Compared with the aforementioned EMT-associated therapeutic targets, two lncRNAs exhibit an inhibitory effect on CRC EMT by targeting Wnt/β-catenin signaling (55,110). LncRNA-CTD903 overexpression inhibits the expression of Snail, Twist and vimentin, and upregulates ZO-1 in CRC cells (55). LncRNA-KIAA0125 overexpression inhibits β-catenin at both the mRNA and protein levels, as well as cyclin D1 and myc at the protein level in CRC cells (110). Similarly, GTPase-activating protein for Rac and SRY-box transcription factor 15 also suppress CRC EMT and tumor growth via the upregulation of E-cadherin and the downregulation of vimentin, Snail, Twist1, ZEB1, free β-catenin, cyclin D1, c-Myc, MMP-2, MMP-7 and MMP-9 in CRC cells (1,17). Among these negative regulators, they exhibit a similar ability to inactivate the Wnt/β-catenin signaling pathway. Notably, compared with other negative regulators, lncRNA-KIAA0125 inactivates this pathway independently of E-cadherin expression (110). These findings indicated that the loss of E-cadherin is not essential for the activation of the Wnt/β-catenin signaling pathway, since other components in this pathway, such as Axin or APC, also promote β-catenin nuclear translocation to induce EMT (54). This result provided additional evidence that the activation of Wnt/β-catenin signaling may be independent of the loss of E-cadherin. However, the underlying molecular mechanisms remain unclear.
Apoptosis-associated therapeutic targets
ICG-001, a small molecule inhibitor, induces the apoptosis of the colon carcinoma cells and inhibits the growth of the colon cancer cells via the downregulation of TCF/β-catenin (103). Survivin, an inhibitor of apoptosis, is highly expressed in cancers and it is reported that TCF/β-catenin participates in the regulation of the survivin transcription and expression in colon cancers (104). Therefore, the disruption of TCF/β-catenin signaling is a promising means for early chemopreventive intervention to CRC. ICG-001-mediated inhibition of the survivin expression increases caspase-3 (a key enzyme of apoptosis) activity, promotes cell differentiation, suppresses cell proliferation and induces apoptosis via enhancing p300 (histone acetylase)/β-catenin interaction at the expense of CBP/β-catenin (103,104,111). Trials with the second generation CBP/β-catenin inhibitor PRI-724 are underway for patients with refractory CRC (112). PRI-724 is still in clinical trials (27) and it may become a therapy for CRC. The treatment of the cancer stem cells with CBP/β-catenin inhibitors promotes them to undergo differentiative divisions and subsequently clears cancer stem cells from the niche (112). Collectively, these studies indicated that Wnt/β-catenin signaling is related to CBP/β-catenin signaling and p300/β-catenin signaling, with CBP/β-catenin interaction promoting CRC cell proliferation, however p300/β-catenin interaction facilitates CRC cell differentiation and, possibly, apoptosis.
Due to the biological characteristics of two histone acetylases (CBP and p300) in CRC, an increasing number of experts have focused on the effects of histone deacetylases (HDAC) on CRC. Histone acetylases and HDACs are involved in the maintenance of the balance between histone acetylation and deacetylation. The interaction between β-catenin and histone acetylases (CBP and p300) influences Wnt/β-catenin signaling and colorectal cell physiology (28). Therefore, HDAC inhibitors may affect CBP and p300 to induce apoptosis via regulating Wnt/β-catenin signaling (27).
Butyrate (a HDAC inhibitor and short-chain fatty acid) is produced by the bacterial fermentation of the dietary fiber (113). In populations with low dietary fiber intake, a doubling of the total dietary fiber intake may reduce the risk of CRC by 40% (114). Therefore, certain researchers extended their studies to butyrate and investigated whether butyrate also induces apoptosis. Studies found that butyrate induces CRC apoptosis, cell cycle arrest and differentiation and regulates Wnt/β-catenin signaling activity in CRC cells (27,28,113,115-117). In addition, trichostatin A, a more selective inhibitor of HDAC, is also capable of inducing apoptosis in CRC cells (118). HDAC inhibitor-induced Wnt/β-catenin hyperactivation and CRC cell apoptosis may partly explain the preventive function of the dietary fiber against CRC (27,116). Similarly, other studies also reported that the dysregulated Wnt/β-catenin signaling pathway in moderate activity levels induces CRC cell proliferation and carcinogenesis, however relatively high levels of this pathway lead to increasing cell apoptosis and decreasing cell proliferation in CRC (27,115,119). Additionally, accumulating evidence indicates that there is a linear relationship between the increase in β-catenin transcriptional activity and the level of the apoptosis in CRC cells treated with butyrate (114,119). Another study proposed that butyrate inhibits HDAC leading to chromatin relaxation and the changes of the chromatin structure, which promotes the activation of caspase-3 by mitochondria and triggers the apoptosis of CRC cells (118).
Angiogenesis-associated therapeutic targets
The dual actions of angiogenic and anti-apoptotic processes are critical for CRC tumor growth (87). Yang et al reported the effect of the oncogene tissue transglutaminase (TGM2) on the angiogenesis and apoptosis of CRC via activating Wnt/β-catenin signaling (120). It was revealed that the downregulation of TGM2 inhibits angiogenesis and the expression of β-catenin, Wnt3a, MMP-2, MMP-9 and cyclin D1, whereas its downregulation promotes the expression of TIMP-1 and caspase-3 and apoptosis in TGM2-siRNA-transfected CRC cells. Caspase-3 is a key enzyme of apoptosis (111), and invasion-associated proteins (MMP-2 and MMP-9) (2) and TIMP-1 are involved in tumor invasion and metastasis via angiogenesis (121). These results indicated that TGM2 is able to effectively promote the angiogenesis and inhibits apoptosis via activating the Wnt/β-catenin signaling pathway in CRC cells.
Exogenous modifier-associated targets
Natural compounds as therapeutic targets
Numerous phytochemicals derived from edible plants have been reported to inhibit tumorigenesis. A number of mechanisms have been shown to contribute to the anti-carcinogenic actions of phytochemicals (122), and more attention has been focused on the Wnt/β-catenin signaling pathway as a target for phytochemicals. In recentyears, natural compounds have been intensively screened for this pathway sincethese compounds are involved in the regulation of Wnt signaling and they are in our diet (123). Vitamin D3 plays its physiological role via the active metabolite 1, 25-(OH)2D3, and vitamin D deficiency is associated with CRC (124). Ligand-activated vitamin D receptors compete with TCF-4 for β-catenin binding, accordingly, vitamin D3 inhibits β-catenin/TCF-4 transcriptional activity, leading to the downregulation of β-catenin/TCF-4 target genes, including c-Myc, Tcf-1 and CD44 (98). These contribute to promoting the differentiation of colon cancer cells. 24-Hydroxylase is upregulated in CRC cells, and its inhibition promotes the antitumor effects of 1, 25-(OH)2D3 via inducing the translocation of β-catenin from the nucleus to the plasma membrane and enhancing the inhibitory effect of 1, 25-(OH)2D3 on c-Myc (124). At present, a phase III study (trial no. NCT03389659) exploring the effect of vitamin D3 in combination with XELOX/mFOLFOX as first-line chemotherapy in untreated advanced or metastatic CRC is in progress (15).
Flavonoids are polyphenolic compounds with low toxic effects and are divided into four subclasses: i) chalcones; ii) flavanones; iii) flavones; and iv) flavonols (11). Evidence from case-control studies has demonstrated that flavonoids have a protective role against CRC (122). A previous study has indicated that flavonoids are capable of acting on different components of the Wnt/β-catenin signaling pathway, such as β-catenin/TCF/LEF, segment polarity protein disheveled homolog DVL-1 (DVL1) and GSK-3β (123). Among the various types of flavonoids, quercetin is one of the most studied flavonoids with Wnt-inhibitory properties and has been submitted to clinical trials (123). Quercetin has been highlighted as a potential anti-CRC drug, and the effect of quercetin on CRC growth is strongly associated with Wnt/β-catenin signaling (11). Quercetin may interact with β-catenin and inhibit the binding of β-catenin to TCF, leading to the inactivation of the Wnt/β-catenin signaling pathway in CRC cells in vitro (125,126). Another extensively studied anticancer flavonoid is the green tea compound epigallocatechin 3-gallate (EGCG), which inhibits the Wnt/β-catenin signaling pathway in CRC cells (126). The underlying mechanism by which EGCG functions on this pathway is that EGCG targets Wnt inhibitory factor 1 (WIF-1) and promotes its expression via demethylation, leading to the inactivation of Wnt/β-catenin signaling via the direct binding of WIF-1 to Wnt molecules (11). Other studies suggested that EGCG inhibits CRC cell proliferation in the APCmin/+ mouse model for intestinal tumorigenesis via suppressing the expression of nuclear β-catenin, indicating the inhibitory effect of EGCG on the translocation of β-catenin to the nucleus (127,128). Silibinin, another natural polyphenolic flavonoid, inhibits CRC cell proliferation, induces apoptosis and reduces nuclear levels of β-catenin in SW480 cells harboring a mutation in APC, but not in HCT 116 cells harboring a β-catenin mutation and wild-type APC, indicating the selective effect of silibinin on Wnt/β-catenin signaling (129). Additionally, silibinin also inhibits the expression of cyclin D1 via the downregulation of cyclin-dependent kinase 8 (CDK8, a CRC oncoprotein that promotes β-catenin activity), leading to the inhibition of the growth of SW480 tumors (129). Wogonin, a naturally occurring mono-flavonoid isolated from Scutellaria radix, inhibits the expression of Wnt3A, c-Myc, LRP6 and cyclin D1, suppresses the kinase activity of CDK8, and promotes the expression of Axin1 in HCT116 cells, leading to the inhibition of CRC cell proliferation (130). Genistein, an isoflavonoid derived from soy products, is capable of inhibiting the transcriptional activity of β-catenin/TCF in SW480 cells via promoting the phosphorylation of β-catenin by GSK-3β (11). Genistein has been tested in a phase I-II clinical trial in combination with FOLFOX or FOLFOX-Avastin to reduce chemotherapy resistance and improve response rates for stage IV CRC (trail no. NCT01985763) (15). Lonchocarpin, a chalcone isolated from Lonchocarpus sericeus, blocks the nuclear localization of β-catenin, and constitutively inhibits TCF4 activity, thus suppressing CRC cell proliferation and migration in SW480, HCT116 and DLD-1 cells (131).
N-3 polyunsaturated fatty acids (N-3 PUFAs), which regulate several signaling pathways, potentiate the effects of radiotherapy, however the molecular mechanisms through which N-3 PUFAs enhance the anti-carcinogenic effects of this therapy in CRC treatment remain unclear (19). Radiotherapy is an adjuvant therapy for patients with advanced CRC, although the fractionation schedules for the treatment of CRC frequently result in the repopulation of surviving CRC cells, progression and metastasis (132). The acquired radioresistance of surviving CRC cells shows an adaptive status in which CRC progression and evasion of cell death are increased. Docosahexaenoic acid (DHA, one of the main N-3 PUFA family members) promotes cell cycle arrest, and inhibits the effects of irradiation on cell proliferation in CRC cells exposed to γ-radiation via reducing GSK-3β phosphorylation and inhibiting irradiation-triggered nuclear translocation of β-catenin and subsequently the expression of cyclin D1 (19).
Sanguisorba officinalis has been used as a herbal medicine for a number of years in China, Japan and Korea, and has broad biological activities, including antioxidant, anti-inflammatory and anticancer activities (133). Sanguisorba officinalis exhibits strong anti-proliferative activities against CRC cells (33). Liu et al further studied the inhibitory effect of tannins extracted from Sanguisorba officinalis on CRC cells (134). Their study revealed that Sanguisorba officinalis and its tannins inhibit the expression of β-catenin, cyclinD1, c-Myc and Axin2, while decreased mRNA levels of lysophosphatidic acid receptor 6 may be associated with the inhibitory effects of Sanguisorba officinalis and its tannins on CRC cells. Later, Li et al investigated the underlying molecular mechanism of Sanguisorba officinalis against Wnt/β-catenin signaling leading to apoptosis in CRC cells (33). They discovered that 1, 4, 6-Tri-O-galloyl-β-D-glucopyranose (a tannin compound from Sanguisorba officinalis) downregulates the expression of β-catenin, Survivin, and c-Myc, and upregulates the expression of caspase-3 in CRC cells, indicating the inactivation of the Wnt/β-catenin signaling pathway. Survivin is an inhibitor of apoptosis (19) and caspase-3 is a key enzyme of apoptosis (111), which may explain the apoptosis of CRC cells via inhibition of Wnt/β-catenin signaling by Sanguisorba officinalis.
In addition to the aforementioned natural compounds, certain other extractions from natural products also inhibit CRC growth and induce CRC cell apoptosis via inactivation of the Wnt/β-catenin signaling pathway. Their inhibitory mechanisms in this pathway mainly include: i) Downregulating the levels of β-catenin or inhibiting nuclear translocation of β-catenin, such as TKP extracted from a medicinal plant Trichosanthes kirilowii (2), magnolol derived from Magnolia obovata (135), ethanol extract of Scutellaria barbata D. Don (EESB) (136), jatrorrhizine (a main component of Rhizoma coptidis) (46) and pristimerin (a natural triterpenoid) (137); ii) downregulating phosphosylated-GSK-3β or upregulating GSK-3β, for example TKP (138); iii) inhibiting the expression of Wnt downstream target genes, including magnolol (135), EESB (136), sulforaphane extracted from cruciferous vegetables (18) and pristimerin (137); iv) blocking the interaction between β-catenin and TCF4 complexes, such as magnolol (135) and 11α, 12α-epoxyleukamenin E isolated from Salvia cavaleriei (37); v) downregulating invasion-associated proteins MMP-2 and MMP-9, such as TKP (138); vi) inhibiting EMT via increasing E-cadherin and decreasing N-cadherin, such as jatrorrhizine (46); and vii) exerting a synergistic effect on the inhibition of CRC, for example, the combination of cucurbitacin E and 5-FU inhibits the expression of β-catenin and chemosensitizes CRC cells via negatively regulating the transcription factor AP4/Wnt/β-catenin axis (139).
Aloe vera has been used as a herbal remedy to treat constipation and promote digestion (140,141). However, several studies have suggested an association between Aloe vera and the development of CRC (140,142,143). Among the components of aqueous extract from Aloe vera, aloesin and aloin may have potential functions in promoting CRC tumorigenesis via activating Wnt/β-catenin signaling (144,145). Dai et al found that the aqueous extract of Aloe vera at concentrations between 0.98-12.50 µg ml−1 promotes the expression of active β-catenin, DKK1 and Axin2, indicating that Aloe verais capable of activating this signaling pathway at a low concentration (145). The crosstalk between the Wnt/β-catenin and the Notch signaling pathways has been known to be strongly associated with CRC tumorigenesis and cell proliferation (146). Therefore, the effects of Aloe vera on these two pathways and the proliferation of CRC cells have been studied. In the presence of Wnt3a, the aqueous extract of Aloe vera at a concentration of 33.3 µg/ml promotes CRC cell proliferation, and aloin activates Wnt/β-catenin signaling and inhibits Notch signaling (144). However, in the absence of Wnt3a, aloesin directly activates the Wnt/β-catenin signaling pathway and suppresses Notch signaling (144). These results do not support the possibility that there is crosstalk between the Wnt/β-catenin and Notch pathways in CRC cell proliferation (146). Recently, a study explored the potential mechanism of the carcinogenic activity of Aloe verain CRC at the post-transcriptional level, and found that Aloe veramay contribute to the regulation of the alternative splicing of Notch-related genes and Wnt/β-catenin-related genes, such as β-catenin, FZD6, LEF1, DVL1, Axin2, Myc and TCF7L2 (141).
Drug-associated therapeutic targets
5-FU is the commonly used standard drug for CRC chemotherapy (92,147). Recently, the synergistic effects of 5-FU with other drugs on the Wnt/β-catenin pathway in CRC cells have been studied. Studies revealed that although 5-FU initially reduces the size of the CRC tumor, resistance (101) and recurrence (47) following chemotherapy prevent favorable clinical outcomes for patients with CRC. Additionally, typical patient response rates for 5-FU are between 10 and 30% (148). Therefore, numerous studies have investigated the interaction between 5-FU resistance and the Wnt/β-catenin pathway in 5-FU-resistant CRC cells. CircRNA-PRKDC promotes 5-FU resistance and CRC cell invasion via activating this pathway (92). A previous study reported that 5-FU activates tumor suppressor p53 via RNA damage or DNA damage of tumor cells (149). Adenovirus-mediated transfer of wild-type p53 gene markedly reverses 5-FU resistance and inhibits the high expression of multidrug resistance gene-1 in multidrug-resistant human CRC LoVo/5-FU cells (150). However, compared with these prior results, another study suggested that 5-FU activates the Wnt/β-catenin pathway and enriches CRC stem cells in tumor organoids derived from APCmin/+/Lgr5EGFP mice, leading to recurrence following treatment with 5-FU, via promoting Wnt3 transcription by p53 (47). Therefore, the promoting effect of 5-FU on CRC stemness via the activation of p53-mediated Wnt/β-catenin signaling may contribute to recurrence following initial chemical debulking of CRC tumor cells treated with 5-FU.
Cisplatin (DDP) is another common drug for CRC chemotherapy (148). Similar to 5-FU, DDP chemoresistance frequently occurs in patients with CRC (151). Fibroblast growth factor 9 (FGF9) is upregulated in CRC and is associated with poor clinical outcome in patients with CRC (152). FGF9 upregulation contributes to the development of DDP resistance via the inhibition of APC expression and promotion of β-catenin expression in CRC cells (151). Tankyrase 1 inhibitor XAV939, a small molecule inhibitor of the Wnt/β-catenin pathway, increases chemosensitivity in CRC cells treated with 5-FU/DDP, combined with the upregulation of Axin and the downregulation of β-catenin (148). A study reported that the overexpression of Golgi phosphorylated protein 3 (GOLPH3) in CRC tissue could promote CRC proliferation and inhibit apoptosis, combined with the upregulation of β-catenin (153). GOLPH3 overexpression also promotes Wnt/β-catenin-mediated DDP resistance in CRC cells treated with 10 µM DDP (147). Therefore, GOLPH3 and FGF9 should be the focus of future research on DDP resistance. Additionally, pimozide has also been considered as a potential anticancer drug and has exhibited satisfactory results in CRC. Furthermore, it is able to inhibit EMT via the inactivation of the Wnt/β-catenin signaling pathway (154). Another study demonstrated that pimozide downregulates the expression of β-catenin, Axin 2, cyclin D1 and c-Myc, and subsequently inhibits EMT through the upregulation of E-cadherin and the downregulation of N-cadherin, Snail and vimentin (59). These studies indicated that the inhibitory effect of pimozide on CRC results from the suppression of EMT.
Bacterium-associated therapeutic targets
Certain studies have reported that Fusobacterium nucleatum, an anaerobic Gram-negative oral commensal bacterium, is important in CRC tumorigenesis and is often associated with metastasis, chemo-resistance and poor prognosis (53,155-157). A previous study has suggested that Fusobacterium nucleatum is present in CRC and adheres to and invades intestinal epithelial cells mainly via virulence factors, such as Fusobacterium adhesin A (FadA) (53). Rubinstein et al investigated the mechanism by which Fusobacterium nucleatum promotes CRC tumorigenesis via its FadA adhesin (155). It was revealed that FadA activates Wnt/β-catenin signaling via inducing β-catenin-mediated transcription and regulates oncogenic and inflammatory responses, leading to CRC tumorigenesis due to Fusobacterium nucleatum. Later, Rubinstein et al further researched the effect of the interaction between FadA and Annexin A1 (a Wnt/β-catenin signaling modulator) on tumor growth in CRC cells infected by Fusobacterium nucleatum (156). It was revealed that Fusobacterium nucleatum combined with Annexin A1 and β-catenin preferentially binds to CRC cells leading to the upregulation of cyclin D1. However, Ma et al suggested that Fusobacterium nucleatum interacts with E-cadherin leading to EMT instead of β-catenin, which subsequently promotes the malignant phenotype of CRC cells (53). Another study showed that Fusobacterium nucleatum promotes CRC tumor growth via cyclin-dependent kinase 5 (Cdk5)-activated Wnt/β-catenin signaling and the Cdk5-mediated upregulation of β-catenin, cyclin D1 and c-Myc (157). Lactobacillus species, another bacterium, inhibits CRC progression via the upregulation of GSK-3β, as well as the downregulation of the Wnt/β-catenin-related genes, such as Axin-2, CTNNB1, cyclin D1, pygopus family PHD finger 2 and inhibitor of DNA binding 2 (158). The results may provide insights into the development of the novel antimicrobial therapy targets for CRC.
Endogenous modifier-associated targets
miRNAs as therapeutic targets
miRNAs may function as tumor suppressors. As revealed in Table I, overexpression of miR-377-3p suppresses CRC cell proliferation, migration and chemoresistance via targeting two positive regulators of the Wnt/β-catenin signaling pathway, ZEB2 and E3 ubiquitin-protein ligase XIAP (80). miR-144-3p inhibits CRC cell proliferation and progression via the targeting of B cell lymphoma 6 protein to LRP6, FZD4, FZD8, LEF1 and TCF7L2 (81,159). miR-490-3p inhibits the expression ofproto-oncogeneFRAT1 (a GSK-3-binding protein), CRC progression and EMT (76). miR-532-3p directly targets the protein ETS1/TGM2 axis and activates apoptosis signaling, and upregulates apoptosis-related proteins, such as cleaved (c)-caspase-3, c-caspase-7 and c-caspase-9 (77). miR-93 inhibits Smad7 expression, which contributes to preventing the nuclear accumulation of β-catenin and subsequently suppresses CRC cell viability, colony formation and migration (36). Among miRNAsas tumor suppressors, some also inhibit the expression of β-catenin (36,77,81), Smad7 (36) and Wnt target genes, such as CD44 (77), c-Myc (36,77,81), MET (77), cyclin D1 (36,81) and c-Jun (77), while promoting the expression of Axin (36). The downregulation of Wnt target genes and β-catenin, and the upregulation of Axin contribute to the inactivation of Wnt/β-catenin signaling and subsequently inhibit CRC tumorigenesis and metastasis. Notably, miR-124 inhibits CRC cell proliferation and invasion, and inducesapoptosisvia the interaction with 1-phosphatidylinositol 4,5-bisphosphate phosphodiesterase b-1 and upregulation of Wnt4 (7). However, another study suggested that Wnt4 secreted by CRC tissues promotes CRC progression via EMTandangiogenesis (31). Therefore, under different circumstances, the effect of Wnt4 on CRC may be inhibitory or promoting. The aforementioned studies indicate that miRNAs acting as tumor suppressors broadly interact with other regulators and reveals the complexity of gene regulation.
LncRNAs as therapeutic targets
Certain lncRNAs also function as safeguards against CRC growth via inhibiting CRC cell proliferation, inducing apoptosis and promoting the expression of tumor suppressors. Wang et al reported that apoptosis was detected after lincRNA-p21-overexpressing CRC cells were treated with X-rays, which indicated that the overexpression of lincRNA-p21 promotes the sensitivity of CRC radiotherapy by inducing the apoptosis of CRC cells (83). The overexpression of lncRNA-00675 inhibits the proliferation, invasion and migration of CRC cells via suppressing the expression of miR-942, which is upregulated in clinical CRC tissues (8). Another lncRNA, lncRNA-GAS5, suppresses CRC cell invasion, migration and tumor growth, as well as angiogenesis by lowering microvascular density (87). LincRNA-p21 decreases the expression levels of β-catenin and c-Myc, which leads to the inactivation of the Wnt/β-catenin signaling pathway. Similar regulatory mechanisms also exist for lncRNA-00675 and lncRNA-GAS5. These results indicated that lncRNAs acting as tumor suppressors inhibit CRC growth via inactivating this pathway.
CircRNAs as therapeutic targets
CircRNAs also act as tumor suppressors and inhibit CRC growth and invasion via inactivating Wnt/β-catenin signaling. For example, circRNA-MTO1 inhibits the growth and invasion of CRC cells via the downregulation of β-catenin, cyclin D1 and c-Myc (94). However, compared with miRNAs, the proportion of circRNAs acting as suppressors is relatively low. As revealed in Table I, for miRNAs, suppressors account for 54.55% (6/11), lncRNAs, 16.67% (3/18) and circRNAs, only 14.29% (1/7). Therefore, compared with miRNAs, lncRNAs and circRNAs mainly exert promoting effects on CRC via activating Wnt/β-catenin signaling.
Other targets for CRC therapy
In addition to the aforementioned studies, researchers have found several other Wnt/β-catenin signaling-associated targets for CRC therapy. These targets include functional oncogenes (160-162), epidermal growth factor-like proteins (163), adaptor proteins (4), stomatin-like proteins (164), chromatin organizers (165), antioxidants (166), and architectural transcription factors (105), which promote CRC growth via activating Wnt/β-catenin signaling. Additionally, negative regulatory targets inhibit CRC progression via inactivating Wnt/β-catenin signaling, and these targets include tumor suppressor genes (167,168), metastasis-suppressor genes (169), ubiquitin-like proteins (170) and anticancer bioactive peptides (171). In the present review, the effects of these therapy targets on CRC in the Wnt/β-catenin signaling pathway are presented in Table II (3,4,21,35,52,60,105,160-178).
4. Challenges in targeting Wnt/β-catenin signaling in CRC
Aberrant Wnt/β-catenin signaling is known to drive CRC tumorigenesis. However, it remains unclear whether this pathway can be effectively targeted for CRC therapy. Additionally, the important clinical implications of targeting Wnt/β-catenin signaling have been recognized for decades following the discovery of this pathway. Nevertheless, it has not been translated into the clinic. Three factors have hindered the development in this field: i) The complexity of the Wnt/β-catenin signaling pathway is the main reason. There are 19 Wnt ligands in humans, which indicates the complexity of Wnt/β-catenin signaling and biological functions, and knowledge of their roles in CRC tumorigenesis remains limited. The complexity of Wnt receptors parallels the complexity observed in the diversity of Wnt ligands. Wnt receptors mainly contain FZD family, LRP-5/6 and receptor tyrosine kinase-like orphan receptor-1/2. Its complexity is also reflected in the number of components involved in this pathway, including β-catenin, Axin, APC, GSK-3β, DVL, CK1 and the TCF/LEF family. ii) The activity levels of the Wnt/β-catenin signaling pathway differentially affect the regulatory functions of components in this pathway. For example, moderate activity levels of this pathway induce CRC proliferation and carcinogenesis, whereas relatively high levels lead to decreased CRC proliferation and increased apoptosis. Therefore, it is difficult to accurately know at which activity level of this pathway Wnt/β-catenin signaling is effectively inhibited. iii) Approximately 90% of sporadic CRC cases harbor mutations in components of the Wnt/β-catenin signaling pathway, such as APC, β-catenin, Axin1, Axin2 and TCF-4. Certain mutations are strongly associated with this pathway and contribute to constitutively active β-catenin, which is difficult to inhibit by antagonists of the upstream proteins of this pathway (15).
Collectively, these are the main challenges related to finding the optimaltherapeutic strategy and developing effective therapeutic agents via targeting the Wnt/β-catenin signaling pathway in CRC.
5. Conclusions and future perspectives
In the present review, Wnt/β-catenin-associated therapy targets for CRC have been described, and it has been revealed that they act as positive or negative regulators and play a pivotal role in therapeutic strategies against CRC. The carcinogenic effects of positive regulators are achieved by aberrantly activating the following pathway: Wnt → FZD → LRP5/6 → DVL → disassembly of β-catenin destruction complex → aberrant accumulation of β-catenin → nuclear translocation of β-catenin → TCF/LEF → transcription of Wnt downstream target genes. Whereas, the anticarcinogenic effects of the negative regulators are achieved by inactivating this pathway. In these regulatory networks, once one of the components is abnormally expressed, a chain reaction ultimately affects the functions of targets for CRC in the Wnt/β-catenin signaling pathway.
Despite numerous years of research on the canonical Wnt/β-catenin signaling pathway, the underlying mechanism of how the activation of the extracellular Wnt signal leads to aberrant accumulation of intracellular β-catenin remains controversial. Therefore, further studies regarding this pathway in CRC are required, with focus on the following: i) Given the inhibitory effects of Traditional Chinese medicine on CRC, the strategy of integrated Chinese-Western medicine for CRC may be more effective. ii) Synergistic effects of multitarget therapy for cancers have been demonstrated. CRC heterogeneity may render targeted therapies ineffective. Therefore, certain multitarget therapeuticstrategies concurrently targeting Wnt/β-catenin signaling should improve treatment. iii) Due to the complexity of Wnt/β-catenin signaling, focus should be on further understanding the crosstalk among oncogenes. iv) More research will be required to identify the highly selective inhibitor for CRC to avoid unnecessary side effects. v) Identifying more inhibitors for downstream targets of the Wnt/β-catenin signaling pathway may be a promising therapeutic strategy for CRC.
In the Wnt/β-catenin signaling pathway, the roles of therapeutic targets for CRC have been previously established. However, these extensive studies suggest that the regulation of Wnt/β-catenin signaling is more complex than previously hypothesized, and thus various mechanisms still need to be further elucidated. Further insight into these underlying mechanisms may enable deeper understanding of CRC tumorigenesis and metastasis, and identify novel diagnostic and prognostic markers, as well as potential therapeutic targets.