Small molecule-mediated disruption of Wnt-dependent signaling in tissue regeneration and cancer

Abstract

The pervasive influence of secreted Wnt signaling proteins in tissue homeostasis and tumorigenesis has galvanized efforts to identify small molecules that target Wnt-mediated cellular responses. By screening a diverse synthetic chemical library, we have discovered two new classes of small molecules that disrupt Wnt pathway responses; whereas one class inhibits the activity of Porcupine, a membrane-bound acyltransferase that is essential to the production of Wnt proteins, the other abrogates destruction of Axin proteins, which are suppressors of Wnt/beta-catenin pathway activity. With these small molecules, we establish a chemical genetic approach for studying Wnt pathway responses and stem cell function in adult tissue. We achieve transient, reversible suppression of Wnt/beta-catenin pathway response in vivo, and we establish a mechanism-based approach to target cancerous cell growth. The signal transduction mechanisms shown here to be chemically tractable additionally contribute to Wnt-independent signal transduction pathways and thus could be broadly exploited for chemical genetics and therapeutic goals.

Figure 1. Chemical structure and potency of IWR and IWP compounds

(a) Structure and potency of IWR compounds. Two IWR compounds that differ by only a single methyl group and that share similar IC50's (as determined in L-Wnt-STF cells; upper right of graphs) were designated Class | compounds. The remaining three IWRs which share structural similarity (see also Supp. Fig. 3) were designated Class || compounds. (b) Structure and potency of IWP compounds. All IWP compounds share structural similarity and IC50's with IWP compounds 2-4 sharing the same core structure (IWP-2) and differing only by either the presence of an additional fluoro or methoxy adduct (IWP-3 and IWP-4, respectively).

Figure 2. Biochemical evidence for Wnt/β-catenin pathway inhibition by IWR and IWP compounds

L-Wnt-STF cells that exhibit constitutive Wnt pathway activation were incubated with IWR (10μM) and IWP (5μM) compounds for 24 hrs prior to lysis. Cellular lysates were subjected to Western blot analysis to determine levels of Lrp6 and Dvl2 phosphorylation, and β-catenin accumulation, all biochemical events associated with Wnt/β-catenin pathway activity. Predictably, IWPs blocked all three biochemical events whereas IWR compounds appear to block β-catenin accumulation without affecting Lrp6 and Dvl2 phosphorylation. Kif3A and tubulin serve as loading controls. Wild-type L-cells stimulated with exogenous Wnt3A protein provided in conditioned medium exhibit similar biochemical changes in Wnt pathway components as that observed in the L-Wnt-STF cells.

Figure 3. IWP compounds target the O-acyltransferase Porcn

(a) Overexpression Porcn but not the Wnt chaperone Evi counters the effects of IWP compounds on Wnt/β-catenin pathway activity as measured using the STF reporter. (b) Overexpression of Porcn reverses the block in Wnt protein secretion induced by IWP compounds as measured using a Wnt3A-Gaussia luciferase fusion protein (see Fig. 1 and Supp. Fig. 2b). (c) Expression of other MBOAT family members does not abrogate the effects of IWP on Wnt/β-catenin pathway response as measured using the STF reporter. (d) IWP compounds inhibit formation of detergent-soluble Wnt3A in a Porcn-dependent manner. Lipidated Wnt3A protein, found in the detergent fraction of TritonX-114 treated cells, is absent in IWP-treated cells but present in Porcn overexpressing cells. (e) IWP compounds inhibit palmitoylation of Wnt3A. Cells expressing Wnt3A with or without Porcn were treated with H³-palmitate. Palmitoylation of Wnt3A was then determined using autoradiography of immunoprecipitated Wnt3A. (f) Structure of a biotinylated IWP-2 derivative. A linker and biotin group (PEG-biotin; PB) were attached to IWP-2 at the para position in the phenyl group (IWP-PB; see Supp. Fig. 4b for synthetic scheme). (g) Porcn associates with IWP-2. IWP-PB or the control PB molecule (linker) bound to streptavidin-coated sepharose beads were incubated with cellular lysate containing Porcn-myc protein in the presence or absence of soluble IWP-2. IWP-PB-bound Porcn-myc binding was measured using Western blot analysis. (h) IWP compounds inhibit Porcn function thereby blocking palmitoylation of Wnt proteins. For (a-c), data represent mean values ± s.d.

Figure 4. Stabilization of the Axin2 destruction complex by IWR compounds

(a) IWR compounds induce accumulation of Axin2 protein as revealed by Western blot analysis of proteins involved in regulating β-catenin levels. (b) Levels of β-catenin not bound to the E-cadherin cell-cell adhesion receptor are decreased in the presence of IWR compounds. (c) IWR compounds induce phosphorylation of β-catenin as measured using an anti-phospho-β-catenin (Ser33/37/Thr41) antibody. Two different exposures of the phospho-β-catenin Western blot are shown. (d) Diastereomeric conformation influences IWR-1 activity. The “exo” form of IWR-1 has decreased Wnt pathway inhibitory activity as measured in L-Wnt cells. (e) The IWR-1-exo compound has reduced ability to stabilize Axin2 as compared to IWR-1-endo. (f) Identification of an Axin2 protein domain necessary for interaction with IWR. Cell lysates with various Axin proteins were incubated with or without sepharose-immobilized IWR-PB (top right) in the presence or absence of competing IWR-1. Bound Axin protein was detected by Western blotting. IWR-IS: IWR-interaction sequence. (g) IWR compounds bind directly to a portion of Axin2 IWR-IS expressed as a glutathione-S-transferase (GST) fusion protein in bacteria. (h) Expression of the IWR-IS protein abrogates the effects of IWR on Wnt/β-catenin pathway response. Inhibition of aberrant Wnt/β-catenin pathway, as measured using the STF reporter assay, by IWR-1 is reversed by expression of IWR-IS protein. Data represent mean values ± s.d. (i) IWR induces stabilization of Axin2 protein with consequential increase in β-catenin destruction.

Figure 5. Chemical inhibition of the Wnt/β-catenin pathway in regeneration of zebrafish tissue

(a) Specific inhibition of tailfin regeneration with IWR compounds. IWR-1, -2, and -3 but not the inactive IWR-1-exo and IWR-frag inhibit tailfin regeneration. Four animals in each group were analyzed. Scale bar: 2.5mm. (b) IWR-1 inhibits FGF20A expression in resected tailfins. Fish with resected tailfins were allowed to recover for 24hrs and then treated with small molecules for an additional 24hrs post amputation (hpa). Fins were probed for FGF20A expression by in situ hybridization. Arrows highlight a region that more readily reveals differences in FGF20a expression. (c) IWR-1 blocks normal homeostatic renewal of the GI tract. Representative histological sections of mid-intestinal tissue from fish treated with carrier or IWR-1 (10μM) for 8 or 14 days stained either with hematoxylin and eosin (H&E) or for BrdU incorporation. Loss of BrdU-labeled cells in the base of intestinal folds in IWR-1-treated fish (8 days; arrows) is followed by gross changes in intestinal tissue architecture after prolonged chemical exposure (14 days). Eight animals in each group were analyzed. Scale bar: 50μM. (d) Quantification of BrdU-labeled cells in the intestinal tract of control or IWR-1-treated fish. Histological sections as seen in b (middle column) were scored for the percentage of intestinal folds that contain BrdU-labeled cells. Four independent scorers analyzed sections from eight fish either from control or IWR-1 treated groups. Ratio: BrdU-labeled cells in the numerator and the number of intestinal folds scored in the denominator.

Figure 6. The effects of IWR-1 on caudal fin regeneration are reversible

Adult zebrafish with resected caudal fins were placed in water containing DMSO carrier or IWR-1 (10μM) for 7 days with replenishment of breeding water and compounds every day. Consistent with inhibition of Wnt/β-catenin pathway response by IWR-1, fish treated with IWR-1 but not DMSO failed to regenerate fin tissue. Nine days post-removal of chemicals, fish that were treated with IWR-1 display tissue regrowth suggesting the pluripotent cells required for regeneration are able to resume normal function. Numbers designate specific animals. Four fish were analyzed in each group. Scale bar: 2.5mm.

Figure 7. Chemical inhibition of cancerous Wnt/β-catenin pathway activity

(a) IWR-1 blocks β-catenin accumulation induced by loss of APC tumor suppressor. (b) IWRs block aberrant Wnt/βbcatenin pathway activity in the colorectal cancer (CRC) cells. Constitutive Wnt/β-catenin pathway activity in DLD-1 cells consequential to APC loss-of-function, is abrogated by IWR compounds as measured using the STF reporter. (c) β-catenin-dependent growth of several cancer cell lines. Cells from lung, colon, and prostate cancers transiently transfected with a β-catenin siRNA pool were seeded at clonal density and cell viability measured using Cell-Titer Glo assay 10 days later. (d) Growth-inhibitory effects of IWR compounds on cancerous cells. The same assay in (c) was performed except cells were treated with IWR-3 for 6 days. (e) Overexpression of β-catenin can rescue the growth-inhibitory effects of an IWR compound in DLD-1 cells as indicated by levels of Renilla luciferase (RL) activity in cells transfected with or without a β-catenin expression construct and the RL reporter DNA. (f) H460 cells lack aberrant Wnt/β-catenin pathway activity as measured by STF reporter assay. (g) IWR-3 induces Axin1 protein stabilization in H460 and DLD-1 cells. (h) IWR-1 stabilization of Axin2 results in decreased transcription of Axin2, a Wnt/β-catenin target gene as measured using RT-PCR. (i) The predicted utility of IWP and IWR compounds for inhibiting Wnt ligand-dependent and -independent pathway responses. For (b-f), data represent mean values ± s.d.