Tnik co-immunoprecipitates with Tcf4 in the mouse small intestinal crypt. (A) Biochemical purification of crypt and villus fractions from mouse small intestine used in Mass Spec experiment. Western blot analysis using antibodies directed against genes expressed specifically in the crypt (c-Myc, Ephb2 and Ephb3), villus (Keratin 20), as well as tubulin as control were used to assess the quality of the purified fractions. (B) Silverstained gel of Tcf4 containing complexes immunoprecipitated from purified crypt and villus fractions using antibody directed against Tcf4. The Tcf4-interacting proteins are indicated by arrows and were identified by MS (see Material and methods for details). (C) Peptide coverage of Tnik in MS experiment. Amino-acid sequences of Tnik peptides detected in the Tcf4 immunoprecipitate from crypt lysates are highlighted in yellow. (D) Cell lysates from purified crypt and villus fractions were immunoprecipitated with antibodies directed against endogenous Tcf4, β-catenin and Tnik as indicated and analysed by western blotting with the indicated antibodies.

Nuclear localization of Tnik in Wnt-activated intestinal crypt. IHC of Tnik in mouse small intestine (A, B), mouse colon (C, D), human normal colon (E, F) and human colorectal tissues (G, H). (B) Nuclear localization of Tnik in mouse intestinal crypt is indicated by arrows. Corresponding magnification is shown on the bottom left of figures.

Tnik is recruited to Wnt target genes in mouse crypts and CRC cells in β-catenin-dependent manner. (A) Purified crypt and villus fractions from mouse small intestine were subjected to ChIP using antibodies directed against Tcf4, β-catenin and Tnik. Formaldehyde cross-linked chromatin was immunoprecipitated with the specified antibodies followed by qPCR using primers specific for the Axin2 and c-Myc proximal promoters as well as Axin2 downstream and c-Myc upstream control regions as indicated. The results are presented as relative enrichment over the non-bound downstream SP5 control region and are representative of three independent experiments. (B) TNIK interaction with TCF4 is mediated by β-catenin. Western blot analysis of β-catenin depletion in Ls174T cells expressing doxycycline (Dox) inducible β-catenin shRNA (left panel). Immunoprecipitated TNIK–protein complexes from untreated or doxycycline-treated cells were resolved by SDS–PAGE followed by western blotting using antibodies directed against TNIK and TCF4 (top right panel), and the result was further confirmed by the reverse IP (bottom right panel). (C) ChIP experiments in Ls174T cells uninduced or induced with Dox using antibodies specific for TCF4, β-catenin, and TNIK. The immunoprecipitated DNA was analysed by qPCR using primer pairs specific for the Axin2 and c-Myc promoters, and Axin2 upstream and c-Myc downstream control regions. The results are presented as relative enrichment over the non-bound second exon of the myoglobin gene and are representative of three independent experiments.

TNIK kinase activity is required for TCF/LEF-mediated transcription. (A) siRNA-mediated depletion of TNIK reduces β-catenin/TCF-driven transcription in Ls174T CRCs. Activity of the TOPFlash (black bars) and FOPFlash (green bars) 96 h post siRNA transfection is shown. Error bars represent standard deviation from three independent experiments. Expression of TNIK, β-catenin and control tubulin was analysed by western blotting after depletion of TNIK (left panel). (B) Amino-acid alignment of the N-terminal kinase domain of human TNIK with Ser/Thr kinases SLK and PRKACA. Identical amino acids are shown in black and conserved amino acids are highlighted in grey. Asterisks indicate the conserved amino acids mutated to generate TNIK mutants used in (C). (C) Expression of TNIK kinase mutants (second, third and fourth panels) abrogates β-catenin/TCF-driven transcription, whereas over expression of WT TNIK (last panel) specifically increases β-catenin/TCF-driven transcription in Ls174T CRCs. Black bars indicate TOPFlash activity, whereas green bars indicate FOPFlash activity 24 h post transfection of expression vectors. Error bars represent standard deviation from three independent experiments. Western blot analysis indicates expression of Flag-tagged WT and mutant TNIK proteins using M2 Flag antisera (bottom panel). ^*P-value<0.05, ^**P-value<0.01. (D) Phosphorylation of TCF4 by TNIK is shown by in vitro kinase assay using immunoprecipitated Flag-tagged WT and mutant TNIK and GST–TCF4 and GST–β-catenin as substrate. WT TNIK is able to phosphorylate TCF4, but not β-catenin, whereas the phosphorylation is largely suppressed in all the TNIK kinase mutants. Autophosphorylation of WT TNIK is also detected. GST-tagged protein input is shown at the bottom.

TNIK is interacting with TCF4 in β-catenin-dependent manner. (A) TNIK interacts with TCF4 and β-catenin on Wnt stimulation in HEK293T cells. Lysates from cells stimulated with either Wnt3A- or control-conditioned media as indicated for 8 h were used to immunoprecipitate TCF4 (bottom panel) or TNIK (top panel) and their associated proteins. Immunoprecipitated protein complexes were resolved by SDS–PAGE followed by western blotting using antibodies directed against TNIK, β-catenin and TCF4 as indicated. (B) siRNA-mediated depletion of TNIK reduces β-catenin/TCF-driven transcription in HEK293T cells. Expression of TNIK and control tubulin was analysed by western blotting after depletion of TNIK (top panel). Activity of the TOPFlash and FOPFlash 12 h post Wnt stimulation is shown (bottom panel). Error bars represent standard deviation from three independent experiments. *P-value=0.006. (C) Schematic summary of different TNIK deletion mutants (top panel). All five TNIK deletion mutants were in vitro translated with ³⁵S-labelled methionine and were incubated with GST–TCF4 or GST–β-catenin to examine the direct interaction. GST protein alone was used as control (bottom left panel). Only TNIK kinase and ΔC domains bind to GST–TCF4 (bottom middle panel, arrows), whereas TNIK intermediate, ΔK and ΔC domains bind to GST–β-catenin (bottom right panel, arrows).

TNIK is an essential activator and specifically regulates Wnt target genes in HEK293T cells. (A) Expression pattern of 172 Wnt-induced genes selected against overlapping probes with 1.5-fold change in either 7 or 9 h Wnt induction. Representative Wnt-induced genes were listed. (B) Differential expression pattern of genes after TNIK suppression in HEK293T cells on 4 and 7 h Wnt induction with >1.5-fold variation at 7 h time point. The corresponding expression pattern on Wnt induction without TNIK suppression is shown on the left. Green, down-regulated after TNIK suppression; red, up-regulated after TNIK suppression; grey, missing data. (C) Significant negative correlation of the overlapping probes between 7 h WNT induction and TNIK suppression selected by 1.5-fold change after TNIK suppression. x axis: fold change (log2) after 7 h Wnt induction; y axis: fold change (log2) after TNIK suppression. (D) qRT–PCR validation of four selected candidate genes. ^*P-value<0.05, ^**P-value<0.01, ^***P-value<0.001.