DUB3 Deubiquitylating Enzymes Regulate Hippo Pathway Activity by Regulating the Stability of ITCH, LATS and AMOT Proteins

Abstract

The YAP and TAZ transcriptional coactivators promote oncogenic transformation. Elevated YAP/TAZ activity has been documented in human tumors. YAP and TAZ are negatively regulated by the Hippo tumor suppressor pathway. The activity and stability of several Hippo pathway components, including YAP/TAZ, is regulated by ubiquitin mediated protein turnover and several ubiquitin ligase complexes have been implicated in human cancer. However, little is known about the deubiquitylating enzymes that counteract these ubiquitin ligases in regulation of the Hippo pathway. Here we identify the DUB3 family deubiquitylating enzymes as regulators of Hippo pathway activity. We provide evidence that DUB3 proteins regulate YAP/TAZ activity by controlling the stability of the E3 ligase ITCH, the LATS kinases and the AMOT family proteins. As a novel Hippo pathway regulator, DUB3 has the potential to act a tumor suppressor by limiting YAP activity.

Fig 1. DUB3 regulates Hippo activity by mediating YAP turnover.

(A) Luciferase reporter assays showing the effects of DUB3 shRNAs on YAP/TAZ activity. HEK293T cells were transfected to express the 8XGTIIC_luc YAP/TAZ reporter, which contains 8 TEAD binding sites to control the expression of firefly luciferase and a vector expressing CMV-Renilla luciferase to normalize for transfection efficiency, together with independent shRNA vectors to deplete DUB3 or with a control shRNA. shRNA to deplete LATS2 was used as a positive control. Data represent the average of three independent transfection experiments ± SD. P values were determined using Student’s T-test (2-tailed, unequal variance). (B) Luciferase reporter assays showing the effects of DUB3 siRNAs on YAP/TAZ activity. HEK293T cells were transfected to express the luciferase reporters together with independent siRNAs to deplete DUB3 or with a control scrambled siRNA. Data represent the average of three independent transfection experiments ± SD. P values were determined using Student’s T-test (2-tailed, unequal variance). (C) Luciferase reporter assays showing the effect of DUB3 on YAP/TAZ activity. HEK293T cells were transfected to express the luciferase reporters together with a vector expressing Flag-DUB3, the C89S mutant form of DUB3 or relevant controls vectors. shRNA to deplete LATS2 was used as a positive control. Data represent the average of three independent replicates ± SD. P values were determined using Student’s T-test (2-tailed, unequal variance). (D) Immunoblots showing the effect of DUB3 expression on YAP protein. HEK293T cells were transfected with a vector expressing Flag-DUB3, the C89S mutant form of DUB3 or a control vector. Blots were probed with anti-YAP antibody and anti-Flag. Anti-Actin was used to control for loading. (E) Immunoblots showing the effect of DUB3 depletion on the YAP expression level. HEK293T cells were transfected with independent DUB3 siRNAs or a scrambled siRNA control. Blots were probed with anti-YAP antibody and anti-Flag. Anti-Actin was used to control for loading. (F) Effect of DUB3 siRNAs on the expression of YAP transcriptional targets. HEK293T cells were transfected with DUB3 or control siRNAs. mRNA expression of DUB3, Cyr61, ANKRD1 was measured by RT-PCR. GAPDH mRNA was used for normalization and TBP was used as an additional control gene. Data represent the average of 3 independent experiments ± SD. * indicates p <0.01, compared to the relevant controls (Student’s T-test; 2-tailed, unequal variance). (G) Effect of DUB3 on cell growth. Human primary fibroblast BJ cells were engineered to express hTert, H-Ras^G12V and to deplete p53 and p16, or with the addition or YAP^S127A/S397A. BJ^{p53kd/p16kd/HRas} and BJ^{p53kd/p16kd/HRas/YAPS127A/S397A} cells were virally transduced and selected to stably express DUB3, its inactive C89S mutant form or with an empty vector as a control. Cells were counted at 24-hour intervals for a period of 72h. The assay was performed in triplicate. Data represent the average ± SD. * indicate p <0.05, compared to the relevant controls (Student’s T-test; 2-tailed, unequal variance).

Fig 2. DUB3 interacts with ITCH and mediates its stability.

(A) Immunoblots showing the effect of DUB3 expression on ITCH. HEK293T cells were transfected with a vector expressing Flag-DUB3, its C89S DUB3 or a control vector. Blots were probed with anti-Flag, anti-ITCH, anti-NEDD4 and anti-SMURF1 antibodies. Anti-Actin was used to control for loading. (B) Immunoblots showing the effect of DUB3 siRNAs on ITCH. HEK293T cells were transfected with independent siRNAs against DUB3 or a scrambled siRNA. Blots were probed with antibodies against ITCH or Actin for loading control. (C) Immunoprecipitation assays showing interaction between DUB3 and ITCH. HEK293T cells were transfected to express Flag-tagged DUB3 and Myc-tagged ITCH as indicated. Transfected cells were treated with MG132 5μM overnight before being harvested for immunoprecipitation with anti-Flag or anti-Myc-conjugated beads. Blots were probed with anti- Flag to detect DUB3 or anti-Myc to detect ITCH. (D) Ubiquitylation assay showing the effect of DUB3 on ITCH ubiquitylation. HEK293T cells were co-transfected with a vector expressing Myc-ITCH and a vector expressing Flag-tagged DUB3, Flag-tagged DUB3 C89S or a control vector. Transfected cells were treated with MG132 5μM overnight before being harvested for immunoprecipitation with anti-Myc-or isotype IgG-conjugated beads in PLC buffer freshly supplemented with 10mM of NEM. Immunoblots were probed with antibodies against HA, Flag and Myc.

Fig 3. LATS and AMOT proteins are required for DUB3-mediated regulation of Hippo signaling.

(A) Immunoblots showing the effect of DUB3 expression on LATS kinases, AMOT and YAP. HEK293T cells were transfected with a vector expressing Flag-DUB3 or a control vector together with a mixture of siRNAs targeting AMOT, AMOTL1 and AMOTL2 or a scrambled siRNA control. Blots were probed with antibodies against Flag, AMOT, AMOTL1, LATS1, LATS2 and YAP. Anti-Actin was used to control for loading. Samples were run on the same SDS-acrylamide gels with intervening lanes removed. (B) Immunoblots showing the effect of DUB3 depletion on LATS kinases, AMOT and YAP. HEK293T cells were transfected with a vector expressing Flag-DUB3 or a control vector together with a mixture of siRNAs targeting AMOT, AMOTL1 and AMOTL2 or a scrambled siRNA control. Blots were probed with antibodies against DUB3, AMOT, LATS1, LATS2 and YAP. Anti-Actin was used to control for loading. Samples were run on the same gels with intervening lanes removed. (C) Immunoblots showing the effect of DUB3 depletion on ITCH, LATS kinases and AMOT. HEK293T cells were transfected with a control or DUB3 siRNA in the presence of LATS1 and LATS2 siRNAs or a scrambled siRNA control. Blots were probed with antibodies against DUB3, ITCH, AMOT, LATS1, LATS2 and YAP. Anti-Actin was used to control for loading. (D) Luciferase reporter assays showing the effects of DUB3 siRNAs on YAP/TAZ activity. HEK293T cells were transfected to express the luciferase reporters together with a control or DUB3 siRNA in the presence of a mixture of siRNAs targeting AMOT, AMOTL1 and AMOTL2, a mixture of siRNAs targeting LATS1 and LATS2 or a scrambled siRNA control. Data represent the average of three independent transfection experiments ± SD. P values were determined using Student’s T-test (2-tailed, unequal variance).

Fig 4. DUB3 mediates ITCH, LATS1/2 and AMOT proteins to regulate Hippo activity.

(A) A schematic view of DUB3-mediated regulation of Hippo signaling. DUB3 de-ubiquitylates ITCH, LATS1/2 and AMOT to promote their stability. In the presence of stabilized AMOT, ITCH promotes YAP degradation. (B) Luciferase reporter assays showing the effects of ITCH overexpression on YAP/TAZ activity. HEK293T cells were transfected to express the luciferase reporters together with a control or ITCH expression vector in the presence of a mixture of siRNAs targeting AMOT, AMOTL1 and AMOTL2, a mixture siRNAs targeting LATS1 and LATS2 or a scrambled siRNA control. Data represent the average of three independent transfection experiments ± SD. (C) Luciferase reporter assays showing the effects of DUB3 siRNAs on YAP/TAZ activity. HEK293T cells were transfected to express the luciferase reporters together with a control or DUB3 siRNA in the presence of ITCH siRNA, a mixture of siRNAs targeting ITCH and NEDD4 or a scrambled siRNA control. Data represent the average of three independent transfection experiments ± SD.