Windorphen: A New Wnt Signaling Inhibitor
By: Carol A. Rouzer, VICB Communications
Published: September 23, 2013
A high-throughput screen based on zebrafish development reveals a new Wnt signaling inhibitor that acts by blocking the interaction of β-catenin with p300.
The Wnt family of secreted proteins plays an important signaling role during embryonic development across a diverse range of species. Abnormal Wnt signaling also contributes to the malignant behavior of many cancers. These two key roles are reflected in the first two Wnt family proteins to be discovered, Wingless (Wg), which was identified through its role in fruit fly development and integrase-1 (int1), recognized as a proto-oncogene in the mouse mammary tumor virus. The name Wnt, a combination of Wg and Int1, recognizes these early discoveries. Due to its role in tumorigenesis, the Wnt pathway is of great interest as a potential target for new anti-tumor agents, leading Vanderbilt Institute of Chemical Biology (VICB) members Chaz Hong and Ethan Lee to search for novel inhibitors of Wnt signaling. Their approach, which exploits the importance of the Wnt pathway in zebrafish development, has led to the discovery of windorphen (WD), a new inhibitor of Wnt signaling with anti-tumor activity [J. Hao et al. (2013) Cell Reports, 4, 898].
Figure 1. Simplified diagram of the Wnt signaling pathway. (Top) In the absence of Wnt, β-catenin is found in a complex with GSK-3 (glycogen synthase kinase-3), APC (adenomatous polyposis coli), and axin. Phosphorylation of β-catenin by GSK-3 results in its degradation by the proteosome. (Bottom) Wnt binds to its receptor frizzled (Fz) and coreceptor LRP (low density lipoprotein receptor related protein). As a result, axin is recruited to the membrane with the aid of Dvl (disheveled), and the complex that phosphorylates β-catenin is disrupted. Freed from the destruction complex, β-catenin is no longer phosphorylated and degraded. It translocates to the nucleus, where it interacts with Tcf/Lef (T cell factor/lymphoid enhancing factor) to stimulate transcription of target genes. Proteins such as CBP (CREB binding protein), and p300 may also interact with β-catenin to stimulate transcription.
Wnt signaling occurs through a number of different pathways, but one of the best characterized, the canonical pathway, involves the dual function protein β-catenin. Originally discovered for its role in cell-cell adhesion, β-catenin also serves as a regulator of transcription, and it is this function that is modulated by Wnt signaling. In the absence of Wnt, β-catenin is bound in a cytosolic complex containing multiple proteins, including axin, glycogen synthase kinase-3 (GSK-3), and adenomatous polyposis coli (APC) (Figure 1, top). Association of β-catenin with this complex subjects it to phosphorylation by GSK-3, marking it for degradation by the proteasome. In contrast, when Wnt proteins are present, they bind to their receptor, frizzled (Fz) and co-receptor low density lipoprotein receptor-related protein (LRP). This leads to binding of axin to the receptor complex through the interaction of another protein, disheveled (Dvl). As a result, β-catenin is released from the destruction complex, escaping phosphorylation and degradation. It translocates to the nucleus and associates with transcription factors of the T-cell factor (Tcf)/lymphoid enhancing factor (Lef) family leading to transcription of target genes (Figure 1, bottom).
Figure 2. Structures of windorphen showing the inactive E-isomer and the active Z-isomer. Also shown is BAS, the inactive analog used as a negative control.
Zebrafish have two isoforms of β-catenin. The β-catenin-1 isoform shares considerable homology with the mammalian protein, while the sequence of β-catenin-2 is more divergent from that of other species. Prior work had shown that β-catenin-2 is required for correct development of the dorsal side of a zebrafish embryo, while β-catenin-1 directs the development of the ventral and posterior regions. To find modulators of Wnt signaling, the investigators used a high-throughput screen designed to detect modulators of dorsoventral patterning in zebrafish embryos. Their screen of a 30,000 compound library yielded WD (Figure 2), a compound that caused “dorsalization” of the embryos, meaning that dorsal development appeared to occur normally, while development of ventral structures was stunted (Figure 3).
Figure 3. Zebrafish embryos showing normal development (top), and severe dorsalization (bottom). Image kindly provided by the Hong lab.
To verify that WD acted through modulation of Wnt signaling, the investigators tested the compound in STF293 cells, which express a Tcf/Lef-responsive luciferase reporter gene. They found that WD inhibited Wnt3-dependent luciferase expression in these cells in a dose-dependent fashion. In contrast, a structural analog of WD, BAS (Figure 2) had no activity in this assay. The investigators were puzzled by the finding that not all preparations of WD showed activity in the Wnt signaling assay. Evaluation by mass spectrometry and nuclear magnetic resonance revealed that WD exists in two isomers based on the configuration around the double bond (Figure 2). The investigators discovered that only the less abundant Z-isomer is active. The specificity of WD for the Wnt pathway was confirmed by its lack of effect on the Hedgehog, tumor necrosis factor-alpha (TNF-α), nuclear factor-kappa B (NF-κB), or bone morphogenic protein (BMP) signaling pathways.
The investigators used an inhibitor of GSK-3 and a mutation in the gene for axin to artificially stimulate Wnt signaling in STF293 cells and zebrafish embryos, respectively. WD was able to suppress this signaling in both cases. However, WD had no effect on the translocation of β-catenin to the nucleus in Wnt3a-stimulated colon carcinoma cells. These findings suggested that WD acts to inhibit Wnt signaling at a step after β-catenin translocation. Studies using STF293 cells expressing zebrafish β-catenin-1 or β-catenin-2 indicated that WD could only inhibit Wnt signaling mediated by β-catenin-1. The investigators constructed a fusion protein from the DNA-binding C-terminal region of the Lef transcription factor and the C-terminal transactivation (TA) domain of mouse β-catenin and showed that this protein could stimulate transcription of a Tcf/Lef-modulated luciferase gene. WD blocked transcription stimulated by this fusion protein, strongly suggesting that the site of action of WD was the TA domain of β-catenin.
The TA domain of β-catenin interacts with the transcriptional modulators CREB binding protein (CBP) and p300, both of which have histone acetyltransferase (HAT) activity. Co-immunoprecipitation studies showed that both of these proteins could be immunoprecipitated with β-catenin from lysates of human colon carcinoma cells. In the presence of WD, the interaction of CBP with β-catenin was unaffected, while that of p300 was disrupted. WD also inhibited the HAT activity of p300 in these cell lysates, while having no effect on the activity of CBP.
Together, the results suggest that WD suppresses Wnt signaling by interfering with the interaction of β-catenin with p300. In zebrafish, only β-catenin-1 is affected, explaining why ventral development is disrupted in WD-treated embryos. However, the similarity of mammalian β-catenin to zebrafish β-catenin-1 also renders the mammalian protein susceptible to WD inhibition. The Hong and Lee labs further confirmed the relevance of WD to mammalian Wnt signaling by showing that the compound induces apoptosis in human colon adenocarcinoma cells (SW480 and RKO), and prostate cancer cells (DU145 and PC3), all of which exhibit aberrant Wnt signaling. In contrast, WD had no effect on H460 lung cancer cells, which do not depend on the Wnt pathway.
Clearly, WD provides a valuable new tool with which to explore the role of Wnt signaling in both development and carcinogenesis. It may also serve as an important lead compound for the development of novel anti-cancer therapeutics.