The Wnt signalling pathway balances the opposing activities of two proteins to transmit signals within cells. An inhibitor that stabilizes one of these proteins reveals a new target for anticancer drug development.
Cancer is often caused by unwanted activation of the Wnt signalling pathway, a group of proteins that work together to transmit signals from the outside of a cell to its nucleus. This is especially true in the colon, where mutations in a Wnt-pathway component called APC (adenomatous polyposis coli) activate signalling and are associated with the majority of colon cancers1. Compounds that prevent Wnt signals from reaching the cell nucleus are therefore attractive candidates for treating colon cancer and other Wnt-dependent cancers.
Few promising candidates have been discovered, however, largely because the known components of the Wnt pathway are difficult drug targets. On page 614 of this issue, Huang et al.2 reveal a compound, known as XAV939, that potently inhibits Wnt signalling through a hitherto-unknown component of the Wnt pathway. The finding of a 'druggable' member of the pathway could change our fundamental understanding of Wnt signalling and provide an entrance point for finding drugs that target Wnt-dependent cancers.
The Wnt signalling pathway is fascinating, not only because of the biological functions it controls, but also because of its unusual mechanism for transmitting information to the cell nucleus. Most pathways do this through a chain of enzymes known as kinases, which work by transferring phosphate groups to molecules, and which are relatively easy to inhibit. By contrast, the Wnt pathway transmits signals by controlling the relative stabilities of two proteins3. The first of these is axin, a scaffold protein on which is built a 'destruction complex' that destabilizes the second protein, β-catenin. When a cell comes into contact with Wnt proteins, axin is destabilized, the destruction complex is inactivated, and so β-catenin is stabilized. The stabilized β-catenin can then migrate to the nucleus, where it turns on a specific set of Wnt target genes. Axin and β-catenin thus exist in opposition to each other, with stabilization of one typically linked to destabilization of the other3.
Given the dearth of kinases and other obvious drug targets along the Wnt pathway, Wnt inhibitors have been hard to come by. Nevertheless, compounds that disrupt interactions between Wnt-pathway proteins have been identified, and some anti-inflammatory drugs, known as COX inhibitors, have been shown to diminish Wnt signalling, albeit by poorly understood mechanisms4. Compounds that affect the production of Wnt proteins and axin stability were also reported earlier this year5. Each of these classes of compound has its own strengths and shortcomings as potential therapeutic agents, but perhaps most impressively, COX inhibitors have been shown to reverse the growth of colorectal polyps (precursors of colorectal cancer). This seems to validate the Wnt pathway as an important anticancer target, and provides an impetus for discovering compounds that act elsewhere in the pathway.
To identify new inhibitors of Wnt signalling, Huang et al.2 used a human cell line that was engineered to glow when Wnt signalling is activated. This enabled the authors to screen a large collection of drug-like molecules to identify those that could block Wnt signalling. One of the compounds to do so was XAV939, which seemed to be a potent stabilizer of axin. But how exactly does it work?
To determine the stabilization mechanism, Huang et al. performed an impressive proteomics analysis of 700 proteins that bound to XAV939-coated beads. They found that the binding of 18 of these proteins to the beads could be disrupted by excess XAV939 in solution, indicating specific binding of these proteins to the inhibitor. Of these eighteen, two — the related proteins tankyrase 1 and 2 — bound directly to axin. When Huang and colleagues knocked down tankyrase function in cells using small interfering RNAs, this stabilized axin. Taken together, these results suggest that the tankyrases might be the targets of XAV939, and could thus represent previously unknown components of the Wnt signalling machinery.
Tankyrases are enzymes that modify proteins by attaching several ADP nucleotides to them, a process called PARsylation. These enzymes had previously been implicated6 in such cellular processes as protecting the ends of chromosomes (telomeres), insulin responsiveness and spindle assembly during mitosis, but no prior link to the Wnt pathway had been reported. In the case of telomere protection, the PARsylation of tankyrase substrates leads to their ubiquitination — tagging with the protein ubiquitin, which marks the substrate for subsequent degradation — hinting that tankyrases might also regulate axin stability by PARsylation and ubiquitination. Consistent with this idea, Huang et al.2 demonstrated that tankyrase 2 directly PARsylates axin in vitro, and that axin PARsylation occurs in vivo. They also found that treatment of cells with XAV939 reduced the PARsylation, ubiquitination and degradation of axin. These findings led the authors to propose that tankyrases are essential components of the Wnt signalling pathway, and that they promote axin degradation by PARsylating it directly (Fig. 1).
Naturally, questions remain about the details of tankyrase involvement in Wnt signalling. For example, is tankyrase activity constitutive — does it simply remove axin from an active pool at a constant rate — or is it regulated? If regulated, what factors influence tankyrase activity, and how? It will also be crucial to determine how widespread the effects of tankyrase inhibition are in cells. The answer to this question will probably influence the usefulness of tankyrase inhibitors as drugs: if tankyrase activity is widespread and affects several biological processes, it may be difficult to achieve specific, therapeutic inhibition of Wnt signalling without causing undesirable side effects. In fact, a growing list of tankyrase substrates indicates the enzymes' involvement well beyond the Wnt pathway, as does the fact that the embryos of genetically engineered mice that lack both tankyrase 1 and tankyrase 2 die before birth7.
Conversely, several of the reported functions of tankyrases — such as their involvement in the elongation of telomeres — favour cancer progression, so inhibition of non-Wnt tankyrase activities could theoretically complement Wnt inhibition in cancer chemotherapy8. The results of future animal tests of tankyrase inhibitors will therefore be of great interest. In the meantime, Huang and colleagues' demonstration2 that XAV939 can inhibit proliferation of APC-deficient colorectal cancer cells serves as an encouraging hint of the therapeutic possibilities of tankyrase inhibitors.
Powell, S. M. et al. Nature 359, 235–237 (1992).
Huang, S.-M. A. et al. Nature 462, 614–620 (2009).
Tolwinski, N. S. & Wieschaus, E. Trends Genet. 20, 177–181 (2004).
Barker, N. & Clevers, H. Nature Rev. Drug Discov. 5, 997–1014 (2006).
Chen, B. et al. Nature Chem. Biol. 5, 100–107 (2009).
Hsiao, S. J. & Smith, S. Biochimie 90, 83–92 (2008).
Chiang, Y. J. et al. PLoS ONE 3, e2639 (2008).
Seimiya, H. Br. J. Cancer 94, 341–345 (2006).
About this article
Journal of Tissue Engineering and Regenerative Medicine (2018)
Inhibition of Wnt/β-catenin signaling promotes epithelial differentiation of mesenchymal stem cells and repairs bleomycin-induced lung injury
American Journal of Physiology-Cell Physiology (2014)
Synthesis and biological evaluation of imidazo[1,2-a]pyrimidines and imidazo[1,2-a]pyridines as new inhibitors of the Wnt/β-catenin signaling
European Journal of Medicinal Chemistry (2014)
Journal of Medicinal Chemistry (2013)
Patterns of avoidance behaviours in the light/dark preference test in young juvenile zebrafish: A pharmacological study
Behavioural Brain Research (2011)