Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Aug 4;63(3):498-513.
doi: 10.1016/j.molcel.2016.06.019.

Tankyrase Requires SAM Domain-Dependent Polymerization to Support Wnt-β-Catenin Signaling

Affiliations
Free PMC article

Tankyrase Requires SAM Domain-Dependent Polymerization to Support Wnt-β-Catenin Signaling

Laura Mariotti et al. Mol Cell. .
Free PMC article

Abstract

The poly(ADP-ribose) polymerase (PARP) Tankyrase (TNKS and TNKS2) is paramount to Wnt-β-catenin signaling and a promising therapeutic target in Wnt-dependent cancers. The pool of active β-catenin is normally limited by destruction complexes, whose assembly depends on the polymeric master scaffolding protein AXIN. Tankyrase, which poly(ADP-ribosyl)ates and thereby destabilizes AXIN, also can polymerize, but the relevance of these polymers has remained unclear. We report crystal structures of the polymerizing TNKS and TNKS2 sterile alpha motif (SAM) domains, revealing versatile head-to-tail interactions. Biochemical studies informed by these structures demonstrate that polymerization is required for Tankyrase to drive β-catenin-dependent transcription. We show that the polymeric state supports PARP activity and allows Tankyrase to effectively access destruction complexes through enabling avidity-dependent AXIN binding. This study provides an example for regulated signal transduction in non-membrane-enclosed compartments (signalosomes), and it points to novel potential strategies to inhibit Tankyrase function in oncogenic Wnt signaling.

Figures

None
Figure 1
Figure 1
Requirement of ARCs and SAM Domains for Tankyrase-Driven Wnt Signaling (A) Domains of human TNKS and TNKS2 are shown. (B) Activation of β-catenin/TCF/LEF-dependent transcription by MYC2-Tankyrases in unstimulated HEK293T cells, assayed by TOPFlash and control FOPFlash reporters. Data are expressed relative to mean reporter activities obtained without MYC2 construct (seven samples in set; n = 3 duplicate experiments; error bars, SEM). (C) Transcription reporter assay as in (B), using 16 ng of MYC2-Tankyrase constructs. Fold activation is relative to vector only (n = 6 duplicate experiments; error bars, SEM). (D) Transcription reporter assay as in (C). Cells were treated with 9.8 nM to 10 μM XAV939 in a 2-fold dilution series. Data are expressed relative to reporter activity in the vector control in the absence of XAV939 (n = 3 duplicate experiments; error bars, SEM). See Figure S2A for TNKS2 PARylation assessment. (E) Transcription reporter assay as in (C) (n = 3 duplicate experiments; error bars, SEM). See Figure S1 for Tankyrase expression levels in luciferase reporter assays. (F) In vitro PARylation assay for the indicated immunoprecipitated MYC2-tagged Tankyrases. Top: western blot analysis of immunoprecipitates is shown; and bottom: autoradiograph is shown.
Figure 2
Figure 2
Polymerization of the TNKS and TNKS2 SAM Domains (A) Ultracentrifugation sedimentation assay. Purified SAM domains (25 μM) were centrifuged and total samples (T), supernatants (S), and pellets (P) were analyzed by SDS-PAGE and Coomassie staining. The diagram illustrates the assay principle. (B) Electron micrographs of SAM domains at the indicated concentrations are shown. Scale bars, 50 nm. (C) SEC-MALS. Chromatograms show one experiment with differential refractive index (dRI), light scattering (LS), and calculated molecular weight per slice i (Mi). Weight-average molecular weights (Mw) and dispersity (Ð) ± SD over peaks are indicated. See Figures S3B and S3C for eluate analyses by SDS-PAGE. The atypically delayed elution of the long TNKS2 SAM filaments likely reflects an interaction/entanglement with the column solid phase. Scatterplots combine data from two experiments with Mw, Ð, and associated SD indicated. Plotted data points with mean and error bars (SD) refer to Mi. See Figures S2 and S3 for further data.
Figure 3
Figure 3
Crystal Structure of the TNKS2 SAM Domain (A) A structural representation of the TNKS2 DH902/924RET2 SAM domain filament is shown. (B) A pair of WT-rendered TNKS2 SAM domains from the filament, colored by Coulombic surface electrostatic potential, is shown. (C) Ultracentrifugation sedimentation assay as for Figure 2A at increasing [NaCl] is shown. (D) Detailed representation of a TNKS2 DH902/924RET2 SAM domain pair. Interface residues are in stick representation with orange lines indicating contacts. Mutations required for crystallization are indicated.
Figure 4
Figure 4
Crystal Structures of the TNKS SAM Domain and Comparison with TNKS2 (A and B) Structural representations of TNKS D1055RT1 SAM domain filaments are shown. Subscript numbers of chain identifiers denote the corresponding asymmetric units. See Figure S4 for a contact analysis. (C) EH-presenting SAM domains from unique SAM domain pairs were superimposed over residues 1,030–1,068T1/877–933T2, and average Cα root-mean-square deviation (RMSD) values for both protomers were calculated. (D) Multiple sequence alignment of SAM domains from representative Tankyrase orthologs. Circles denote interface residues (by solvent inaccessibility); filled circles indicate explicit contacts in crystal structures. X, mutated residues. See Supplemental Experimental Procedures for sequence accession numbers. (E) Conservation of the SAM-SAM interface. Top: interface residues observed in any of the crystal structures are in green, and bottom: residues identical in ≥80% of the orthologs shown in (D) are in red.
Figure 5
Figure 5
Characterization of Tankyrase SAM Domain Mutants (A) Ultracentrifugation sedimentation assays as for Figure 2A. Color coding indicates the degree of abrogated sedimentation. H924T2, K928T2, and E906T2 charge reversals were more severe than changes to alanine. (B) SEC-MALS of Tankyrase SAM domains, as in Figure 2C (Mw ± SD, Р± SD, n = 2). Color coding is as in (A). TNKS2 WT reference data, from the same experiment, are identical to Figure 2C. See Figure S3B for eluate analyses by SDS-PAGE and Figures S5A and S5B for CD spectroscopy. (C) EM of TNKS2 SAM domains. Color coding is as in (A). See Figure S3F for further mutants. Scale bars, 50 nm. (D) ITC analysis for the indicated SAM domain pairs. Mutated surfaces are indicated by the star in the schematics. See Figure S5C for a second experiment.
Figure 6
Figure 6
Tankyrase Requires Polymerization to Drive Wnt-β-Catenin Signaling (A) Homo- and heterotypic interactions of TNKS and TNKS2 in HEK293T cells. FLAG3-Tankyrases were immunoprecipitated, and co-precipitation of MYC2-Tankyrases was assessed by SDS-PAGE and western blotting. TNKS2 expression is lower than TNKS, accounting for the weaker apparent TNKS2 self-association (Figure S1E). See Figures S6A and S6B for cell lysate fractionations and additional co-immunoprecipitations. (B) Tankyrase polymerization controls localization. Serum-starved HeLa cells expressing the indicated mCherry- and mCitrine-tagged Tankyrases were vehicle or XAV939 treated. See Figures S6C and S6D for controls and additional experiments. Scale bar, 10 μm. (C and D) Tankyrase polymerization controls Wnt-β-catenin signaling. Transcription reporter assays for selected TNKS2 and TNKS SAM domain mutants, as for Figure 1C. Reporter activity was normalized to WT TNKS2 or TNKS (100%). Color coding reflects polymerization defects of the corresponding SAM domains as assessed by sedimentation, SEC-MALS, and EM (see Figure 5; n = 3 duplicate experiments; error bars, SEM). See Figure S1 for expression levels and Figure S7 for further data.
Figure 7
Figure 7
Tankyrase Polymerization Supports PARP Activity and Interaction with AXIN (A) In vitro PARylation by immunoprecipitated MYC2-TNKS2. Top: autoradiograph with quantitation is shown; middle: corresponding Coomassie-stained SDS-PAGE gel is shown, and bottom: western blot analysis of immunoprecipitates prior to in vitro PARylation is shown. (B) PAR was released from samples analyzed in (A) and equal amounts of PAR, or all available sample for vector and TNKS2 G1032W, analyzed by PAGE and autoradiography. Origin (O), the xylene cyanol (XC) and bromophenol blue (BPB) markers and PAR chain length are indicated. (C) SW480 cells expressing the indicated MYC2-tagged TNKS2 constructs were XAV939 treated, fixed, and stained for MYC2-TNKS2, endogenous AXIN2, and DNA. Yellow arrows denote degradasomes with AXIN2-TNKS2 colocalization; red arrows denote degradasomes containing AXIN2, but not TNKS2. Scale bar, 10 μm. (D) Endogenous AXIN1 was immunoprecipitated from HEK293T cells expressing the indicated MYC2-TNKS2 constructs. Samples were analyzed by SDS-PAGE and western blotting. (E) A model for the role of polymers and multivalency in the Tankyrase-AXIN system. See the Discussion for details. Red arrows, interactions; black arrow, regulation.

Similar articles

See all similar articles

Cited by 21 articles

See all "Cited by" articles

References

    1. Alvarez-Gonzalez R., Jacobson M.K. Characterization of polymers of adenosine diphosphate ribose generated in vitro and in vivo. Biochemistry. 1987;26:3218–3224. - PubMed
    1. Bernardes de Jesus B., Blasco M.A. Telomerase at the intersection of cancer and aging. Trends Genet. 2013;29:513–520. - PMC - PubMed
    1. Bienz M. Signalosome assembly by domains undergoing dynamic head-to-tail polymerization. Trends Biochem. Sci. 2014;39:487–495. - PubMed
    1. Busch A.M., Johnson K.C., Stan R.V., Sanglikar A., Ahmed Y., Dmitrovsky E., Freemantle S.J. Evidence for tankyrases as antineoplastic targets in lung cancer. BMC Cancer. 2013;13:211. - PMC - PubMed
    1. Callow M.G., Tran H., Phu L., Lau T., Lee J., Sandoval W.N., Liu P.S., Bheddah S., Tao J., Lill J.R. Ubiquitin ligase RNF146 regulates tankyrase and Axin to promote Wnt signaling. PLoS ONE. 2011;6:e22595. - PMC - PubMed

Publication types

MeSH terms

Feedback