Identification of ubiquitin-specific protease 34 (USP34) as an axin-interacting protein. (A) Human AXIN1 and AXIN2 protein interaction network. Lines represent interactions found in AXIN1 and AXIN2 (red circles) pulldown experiments by using LC-MS/MS. Green circles are previously described associated proteins, yellow circles are new associations, and the blue circle represents USP34. (B) Confirmation of the axin-USP34 interaction using co-affinity purification. In HEK293T cells, endogenous USP34 associates with AXIN1 (lane 3) but not with the unrelated protein RADIL (lane 4).

USP34 confers ubiquitin protease activity to the axin protein complex. (A) Cleavage of K48-linked ubiquitin by recombinant USP2 and USP34 core domains and by purified AXIN1 protein complexes but not by AXIN1 protein complexes isolated from cells where USP34 expression was knocked down using shRNA. Cleavage efficiency is monitored with the appearance of monoubiquitin from the polyubiquitin chains. (B) Quantification of ubiquitin protease activity using the ubiquitin-PLA2 assay. Purified AXIN1 complexes from SBP-HA-CBP-AXIN1 cells but not from cells expressing a USP34 shRNA exhibited USP activity. A similar amount of the unrelated RADIL protein complex showed no activity. Recombinant USP2 and USP34 were used as positive controls in this assay. (C) Western blot verification of endogenous USP34 knockdown in HEK293T SBP-HA-CBP-AXIN1 cells stably expressing USP34 shRNA. (D) Cleavage specificity of the USP34 core domains. UB-, SUMO3-, ISG15-, and NEDD8-PLA₂ assays were used to demonstrate that the USP34 core domain preferentially cleaves ubiquitin. (E) Ubiquitinated axin is sensitive to USP activity. Cells stably expressing SBP-HA-CBP-AXIN1 were transfected with FLAG-ubiquitin. Input lysates prepared for streptavidin affinity purification were left untreated (lane 1) or treated with the nonspecific cysteine protease inhibitor NEM (lane 2). Ubiquitin-linked axin conjugates were resolved by SDS-PAGE and detected by using anti-FLAG antibodies (top panel). Equivalent pulldown of axin was monitored using anti-HA antibodies (bottom panel). The inhibition of USP activity robustly increased the amount of ubiquitinated axin (compare lane 2 to lane 1). (F) The USP34 core domain deubiquitinates axin in vitro. FLAG-ubiquitin-axin conjugates, wild-type (WT), and catalytically inactive (C1903S) core domains of USP34 were separately purified from transfected cells using affinity purification. FLAG-UB-axin proteins were then incubated alone (lane 1) or with WT (lane 2) or catalytically inactive USP34 core domain (lane 3) proteins for 1 h. The reaction was stopped by addition of 2× sample buffer, and samples were run on an 8% SDS-PAGE. UB-axin conjugates were detected by using FLAG antibodies and AXIN1 and the core domains of USP34 with HA antibodies.

USP34 regulates the stability of axin. (A) A HEK293 cell line stably expressing Venus-AXIN1 was derived. These cells were transfected with control, AXIN1, or USP34 siRNAs. At 48 h posttransfection the cells were fixed or lysed, and the levels of Venus-AXIN1 were evaluated by fluorescence microscopy (left panels) or Western blotting with anti-GFP antibodies (top right panels). Where indicated, MG132 was added for the last 10 h to inhibit the proteasome. A similar experiment using WT-HEK293 cells and probing for endogenous AXIN1 with a monoclonal rabbit antibody was also performed (bottom right panels). (B) The same experiment performed in panel A was repeated but instead of MG132 the tankyrase inhibitor XAV939 was added for the last 12 h. AXIN1 levels were monitored by fluorescence microscopy (left panels) and Western blotting with anti-GFP antibodies (right panels). Both MG132 and XAV939 were able to rescue the degradation of AXIN1 resulting from the depletion of USP34. (C) Quantification of panels A and B. Regions of interest were drawn randomly around the periphery of cells taking the phase-contrast images as a template, and the pixel intensities of the fluorescence images were quantified by using ImageJ (n = 20 to 25 cells). Error bars represent the standard error of the mean (SEM); a single asterisk represents statistical significance compared to the control (Ctl) siRNA condition, whereas double asterisks represent statistical significance compared to the USP34 siRNA condition. (D) USP34 regulates the turnover of axin. HEK293 cells expressing control or USP34 shRNA were treated with 3 μM XAV939 for 12 h. Cells were washed twice and were switched to medium containing 10 μg of cycloheximide/ml for the indicated times. Lysates were collected and proteins resolved on by SDS-8% PAGE. Axin levels were determined by Western blotting with anti-AXIN1 antibodies (left), quantified by densitometry, and normalized to tubulin levels for each time point (right).

USP34 has a positive regulatory function in Wnt signaling. (A) Validation of USP34 siRNAs. Lysates from RKO cells treated with control siRNA (lanes 1&2), four different USP34 siRNAs (lanes 3 to 6) or β-CATENIN siRNA (lane 7) were resolved by SDS-PAGE and probed with anti-USP34 antibodies to monitor USP34 knockdown and anti-ERK antibodies as loading controls (bottom panels). Each USP34 siRNA was able to reduce the Wnt3A stimulated activation of the TopFlash reporter (top panel). (B) TopFlash assays were performed in HEK293T and RKO cells treated with control (lanes 1, 4, 7, and 10), β-CATENIN (lanes 2, 5, 8, and 11), or USP34 (lanes 3, 6, 9, and 12) siRNA and stimulated with control conditioned medium (lanes 1, 2, 3, 7, 8, and 9) or Wnt3A conditioned medium (lanes 4, 5, 6, 10, 11, and 12). USP34 knockdown inhibited Wnt3A-mediated activation of the reporter in both cell lines (compare lanes 4 and 6 to lanes 10 and 12). (C) Epistasis analysis of USP34 function. TopFlash assays in HEK293T cells showed that USP34 and β-CATENIN siRNAs antagonized the activation of the pathway by overexpression of the stabilized form of β-catenin (compare lanes 2, 3, and 4) but not by the chimeric VP16-LEF1 protein (compare lanes 6, 7, and 8). (D) USP34 knockdown does not influence Wnt3A-induced stabilization of β-catenin. RKO cells treated with control or USP34 siRNA were stimulated with Wnt3A conditioned medium for the indicated durations, and lysates were probed for β-catenin levels by using Western blotting. (E) TopFlash assays in HCT116 and SW480 cells treated with USP34 siRNA showed that the constitutive activation of the β-catenin reporter in these cells requires USP34 function (compare lanes 1 and 2 to lanes 4 and 5). (F) Quantitative RT-PCR analysis of SW480 cells treated with USP34 siRNA shows reduced expression of Wnt target genes NKD1 and TNFRSF19. The levels are expressed as a percentage of the control siRNA.

Positive role of axin during Wnt signaling in colon cancer cells. TopFlash assays in HCT116 and SW480 colon cancer cells (A) or HEK293T and RKO cells treated with control, β-catenin, or AXIN siRNAs (B) were performed. In panel A, AXIN1 and AXIN2 siRNAs inhibit the constitutive Wnt pathway activation due to mutations in APC (SW480) or β-catenin (HCT116), while in panel B AXIN1 plus AXIN2 knockdowns potentiate the Wnt3A-stimulated activity of the reporter. (C) Activation of the Wnt pathway in HEK293T cells using a degradation resistant β-catenin mutant (pt.β-catenin) is inhibited by AXIN1 plus AXIN2 siRNAs. Figures are representative of at least three independent experiments performed in duplicates, where the error bars represent the standard errors. In panel A, TopFlash levels are expressed as the percent activation compared to the basal constitutive levels observed with control siRNA (100%).

USP34 controls the nuclear accumulation of axin. (A) AXIN1, detected by using polyclonal anti-AXIN1 antibodies, localizes to the nuclei of SW480 and HCT116 colon cancer cells (left panels). The specificity of the antibody was controlled using AXIN1 siRNA (middle panels). USP34 siRNA (right panels) inhibits the strong nuclear localization of axin seen in cells treated with control siRNA (left panels). Where indicated, 1 μM MG132 was added for the last 12 h. For quantification, regions of interest were drawn around the nuclear region, as indicated by counterstaining and the pixel intensity quantified using ImageJ (n > 80 cells). Error bars represent the SEM. An asterisk indicates statistical significance compared to the Ctl siRNA condition. (B) USP34 depletion also inhibits the nuclear accumulation of axin in HEK293T cells observed when the CRM1-dependent nuclear export is blocked with LMB.