USP21 regulates Hippo pathway activity by mediating MARK protein turnover

Abstract

The Hippo pathway, which acts to repress the activity of YAP and TAZ trancriptional co-activators, serve as a barrier for oncogenic transformation. Unlike other oncoproteins, YAP and TAZ are rarely activated by mutations or amplified in cancer. However, elevated YAP/TAZ activity is frequently observed in cancer and often correlates with worse survival. The activity and stability of Hippo pathway components, including YAP/TAZ, AMOT and LATS1/2, are regulated by ubiquitin-mediated protein degradation. Aberrant expression of ubiquitin ligase complexes that regulate the turnover of Hippo components and deubiquitylating enzymes that counteract these ubiquitin ligases have been implicated in human cancer. Here we identify the USP21 deubiquitylating enzyme as a novel regulator of Hippo pathway activity. We provide evidence that USP21 regulates YAP/TAZ activity by controlling the stability of MARK kinases, which promote Hippo signaling. Low expression of USP21 in early stage renal clear cell carcinoma suggests that USP21 may be a useful biomarker.

Keywords: Hippo pathway; LATS; MARK; YAP; ubiquitin.

Figure 1. USP21 regulates YAP/TAZ activity

(A) Luciferase reporter assays showing USP21 shRNAs increases YAP/TAZ activity. HEK293T cells were co-transfected to express the YAP/TAZ reporters together with USP21 shRNAs or with a control vector. YAP/TAZ activity was calculated as a ratio of firefly to Renilla activity. Data were normalized to the relevant control and represent the mean of three independent transfection experiments ± SD. (B) Luciferase reporter assays showing YAP/TAZ activity. HEK293T cells were co-transfected with reporter plasmids together with USP21 siRNAs or with a control siRNA. Data represent the mean of three independent transfection experiments ± SD. (C) Quantitative PCR for the expression of YAP transcriptional targets. HEK293T cells were transfected to express a scrambled control or independent siRNAs targeting USP21. Data represent the average of 3 independent experiments ± SD. (D) Luciferase reporter assays showing YAP/TAZ activity. HEK293T cells were co-transfected to express the reporters together with a plasmid expressing USP21 wild type or mutant (C221S) or a control vector. Data represent the mean of three independent replicates ± SD. (E) YAP localization was scored as previously described [24]. Representative images are provided in Supplementary Figure 2. Data represent the average of three independent experiments ± SD.

Figure 2. USP21 regulates the stability of MARK proteins and MARK1 ubiquitylation

(A-C) Immunoblots showing effects of USP21 depletion on Hippo pathway components in HEK293T cells. Blots were probed with the indicated antibodies. Anti-Actin was used to control for loading. Data in panels A-C are from the same blots, so the same anti-Actin controls were used. (D) Immunoblots showing co-immunoprecipitation of MARK1 and USP21. Lysates from MG132-treated HEK293T cells were immunoprecipitated with agarose-conjugated antibodies against USP21 or MARK1 or IgG control antibodies overnight. Separate blots were probed with antibodies against USP21 and MARK1. (E) Immunoblots of HEK293T cells co-transfected to express myc-MARK1 (or myc-MARK1K768A)/HA-tagged ubiquitin and a scrambled or USP21-specific siRNA to deplete USP21 for 36h. Lysates were immunoprecipitated with anti-myc to recover MARK1 and blots were probed with anti-HA (for ubiquitin detection), anti-myc (MARK1), anti-USP21 and anti-Actin.

Figure 3. USP21 inhibition promotes oncogenic transformation in BJ and cancer cell lines

Assays for colony formation in soft agar. (A, B) BJT/p53kd/p16kd/RasG12V/YAP^wtcells, (C) A549 lung cancer cells, (D) MDA-MB-231 breast cancer cells. Cells were transduced to express shRNAs against USP21 or a control. After antibiotic selection, cells were plated in soft agar for 10 days (A549 cells), 2 weeks (BJ cells) and 3 weeks (MDA-MB-231 cells). Representative images of colonies formed from BJ cells after 2 weeks (A). Average colony number per well ± SD from three independent experiments of BJ (B), A549 (C) and MDA-MB-231 (D) cells. p values were determined using the student's t test (2-tailed, unequal variance).

Figure 4. USP21 expression in kidney tumors and adjacent matched normal tissues

(A) Immunohistochemical staining of kidney tissues (in pairs of tumor and adjacent normal samples) with anti-USP21. (B) Plot showing histopathology scores for anti-USP21 staining of RCC tumors and matched adjacent normal tissue. The slides were examined by experienced pathologists to confirm the tissue identity and assigned a score: 0 (no staining), 1 (weak staining of less than 10% of tissue), 2 (weak staining of 10–25% of tissue), 3 (weak to moderate staining of up to 50% tissue), 4 (moderate to strong staining of 50–75% of tissue), 5 (moderate to strong staining of more than 75% of tissue). Data show the IHC score for the tumor minus that of adjacent normal tissue. (C, E) IHC scores for anti-USP21 staining of individual RCC samples separated by histological grade. Data show the average of the tumor-normal scores. * indicates a statistically significant difference using the Mann-Whitney test (p=0.006 in C and = 0.02 in E). (D, F) Histopathology scores for anti-USP21 staining of RCC samples separated by tumor stage. Data show the average of the tumor-normal scores. Student's t-test (2-tailed, unequal variance) was used to determine the significance of differences among groups. * indicates p=0.002 (D) and =0.01 (F).