The many ways of Wnt in cancer
Introduction
Mutations in the Adenomatous polyposis coli (APC) gene were identified as the basis for familial adenomatous polyposis coli (FAP), a heritable predisposition to colorectal cancer [1, 2]. Mutations in APC were subsequently identified in the majority of sporadic colorectal tumors, where they appear early in the progression to cancer. The finding that wild type APC facilitated the destabilization of the β-catenin protein, whereas Wnt had the opposing effect, positioned APC as a negative regulator of Wnt signaling [3, 4]. The oncogenic relationship between β-catenin and APC was further fortified by the discovery of mutations in β-catenin that prevented its regulation by APC [5, 6, 7]. Axin was later revealed to be an additional member of the β-catenin regulatory complex, and it too was found to be mutated in human cancers [8, 9]. Thus, the foundation of a new human cancer pathway was established, complete with extracellular ligands, cell surface receptors, intracellular transducers, and transcription factors that activate and repress genes that contribute to neoplasia (Figure 1).
Pathway activation involves interaction of a Wnt ligand with a seven-transmembrane Frizzled molecule and an LRP5 or LRP6 co-receptor. The ensuing phosphorylation of the LRP co-receptor generates direct binding site for Axin, which is thereby recruited to the plasma membrane and ultimately degraded. The APC–Axin complex is disrupted along with its ability to mark β-catenin for destruction through its phosphorylation and targeting to the proteosome. The stabilized β-catenin interacts with the T-cell factor (TCF) and lymphoid enhancer factor (LEF) transcription factors in a manner that displaces genetic repressors and recruits coactivators. Activation of Frizzled and its co-receptors LRP5 and LRP6 by Wnt ligands leads to inhibition of the β-catenin regulatory complex by what remains an arcane mechanism. There are many ways for cancer to co-opt this program, wresting control from extracellular cues and recklessly driving the expression of genes governing growth, survival and cell fate determination. Here, I examine individual components of the Wnt pathway with respect to the genetic defects that affect them.
Section snippets
APC
Mutations in the APC gene typically truncate the protein in a manner consistent with selection against negative regulation of β-catenin [10]. Also apparent from this mutational spectrum is a positive selection for the retention of an APC fragment that extends about half way through the reading frame into the mutational cluster region (MCR) [11, 12]. The advantage conferred by these partial APC products remains unclear. Interference with wild type APC seems unlikely, because MCR mutations are
β-catenin
Although the APC gene is mutated in the majority of sporadic colorectal cancers, this rarely occurs in any other cancers originating outside of the gastrointestinal tract. The discovery of APC mutations in sporadic lung, ovarian and breast cancers, as well as extracolonic tumors associated with FAP, attests to their oncogenic ability in these tissues [16, 17, 18, 19], but their rarity suggests that they are not a preferred genetic route to cancer. Mutations in β-catenin that abrogate its
Axin
The fidelity of Axin2 as a marker for Wnt signaling probably relates to its function as a dedicated negative regulator of the Wnt pathway. Accordingly, both Axin1 and Axin2 are mutated in a variety of human cancers [32]. Clearly, the polypeptide chain-terminating mutations that ablate essential structure are inactivating, because elimination of just the DIX domain from the extreme C-terminus is sufficient to cripple Axin1 function [33]. In the absence of experimental support, it is more
T-cell factor 4
Although the Axins, APC and β-catenin are now widely recognized as the central culprits in the genetic activation of Wnt signaling, mutations in additional pathway components have also been reported. TCF4, one of the β-catenin binding transcription factors, is mutated in nearly half of the MSI+ colorectal cancers [38, 39, 40]. Consistent with the mutator mechanism characteristic of MSI+ cancers, the TCF4 mutations are all frameshift alterations targeting a polyA tract in the final exon. The
sFRPs
Additional but rare cases of genetic activation in the Wnt pathway include inactivating mutations in the secreted Frizzled-related protein sFRP1, which inhibits receptor activation by competitively binding to Wnt proteins. An analysis of 10 advanced colorectal cancers, selected on the basis of genomic interstitial deletions in the sFRP1 region, yielded three truncating mutations in exon 1 of the remaining allele [42]. The nature of these mutations would be expected to ablate the inhibitory
Glycogen synthase kinase 3
Glycogen synthase kinase 3 (GSK3) is a well-established negative regulator of Wnt signaling, yet mutations have never been identified in cancer. Recent knockout studies from the Woodgett laboratory have demonstrated that constitutive activation of Wnt signaling, in the resulting mouse embryonic fibroblasts, requires inactivation of both alleles of GSK3α and GSK3β (JR Woodgett, personal communication). The odds of this occurring spontaneously in a single cell seem quite remote. Moreover, GSK3 is
ICAT
ICAT, a molecule that binds directly to β-catenin and thereby displaces the TCF and LEF transcription factors and the co-activator CBP, could also qualify as a potential tumor suppressor. A mutation that altered the ICAT initiation codon was identified in 1 of 37 melanomas [45], a cancer in which ICAT is frequently downregulated. By contrast, no mutations where detected in a separate analysis of 178 melanomas [46].
PP2A
PP2A is family of heterotrimeric enzymes with phosphatase activity. The PP2A phosphatase complex is another component of the Wnt pathway that is complicated by its participation in numerous other signaling processes. The PP2A regulatory subunit B56 binds to APC, and the catalytic subunit binds to Axin [47, 48], but whether PP2A is a positive or negative regulator of Wnt depends on the experiment and the interpretation thereof. That B56 associated with APC, downregulated β-catenin and
Wnt receptors
The Wnt co-receptors LRP5 and LRP6 seem like ripe candidates for genetic activation of the pathway, yet sporadic mutations have not been reported for human cancers. Hyperactivating mutations in LRP5 have been identified but are associated with a familial autosomal dominant syndrome characterized by high bone-density [55, 56, 57]. These LRP5 mutations attenuate Dickkopf1-mediated inhibition of Wnt signaling [58]. Nevertheless, cancer susceptibility in affected members of these kindreds has not
Conclusion
In addition to the Wnt pathway genes discussed above, there exist additional components, such as Pygopus, Bcl-9 and others that might activate the pathway if mutated. Many of these remain to be investigated, and perhaps some new mutations will be discovered. However, recent experiences with high-throughput sequencing suggest that it is unlikely that highly prevalent mutations, such as those in APC in colorectal tumors, remain undiscovered [63•]. The constellation of mutations in human cancers
References and recommended reading
Papers of particular interest, published within the period of review, have been highlighted as:
• of special interest
•• of outstanding interest
References (65)
- et al.
Identification and characterization of the familial adenomatous polyposis coli gene
Cell
(1991) - et al.
The mouse Fused locus encodes Axin, an inhibitor of the Wnt signaling pathway that regulates embryonic axis formation
Cell
(1997) The adenomatous polyposis coli (APC) tumor suppressor
Biochim Biophys Acta
(1997)- et al.
APC mutations are infrequent but present in human lung cancer
Cancer Lett
(2004) - et al.
Molecular analysis of the APC gene in 205 families: extended genotype–phenotype correlations in FAP and evidence for the role of APC amino acid changes in colorectal cancer predisposition
J Med Genet
(1999) - et al.
Exon 3 β-catenin mutations are specifically associated with colorectal carcinomas in hereditary non-polyposis colorectal cancer syndrome
Gut
(2005) - et al.
High frequency of β-catenin mutations in borderline endometrioid tumours of the ovary
J Pathol
(2006) - et al.
Wnt/β-catenin/Tcf signaling induces the transcription of Axin2, a negative regulator of the signaling pathway
Mol Cell Biol
(2002) - et al.
Mutations and elevated transcriptional activity of conductin (AXIN2) in hepatoblastomas
J Pathol
(2004) - et al.
Axin prevents Wnt-3a-induced accumulation of β-catenin
Oncogene
(1999)