Rac1 GTPase Regulates the βTrCP-Mediated Proteolysis of YAP Independently of the LATS1/2 Kinases

Abstract

Background: Oncogenic mutations in the KRAS gene are detected in >90% of pancreatic cancers (PC). In genetically engineered mouse models of PC, oncogenic KRAS drives the formation of precursor lesions and their progression to invasive PC. The Yes-associated Protein (YAP) is a transcriptional coactivator required for transformation by the RAS oncogenes and the development of PC. In Ras-driven tumors, YAP can also substitute for oncogenic KRAS to drive tumor survival after the repression of the oncogene. Ras oncoproteins exert their transforming properties through their downstream effectors, including the PI3K kinase, Rac1 GTPase, and MAPK pathways. Methods: To identify Ras effectors that regulate YAP, YAP levels were measured in PC cells exposed to inhibitors of oncogenic K-Ras and its effectors. Results: In PC cells, the inhibition of Rac1 leads to a time-dependent decline in YAP protein, which could be blocked by proteosome inhibitor MG132. This YAP degradation after Rac1 inhibition was observed in a range of cell lines using different Rac1 inhibitors, Rac1 siRNA, or expression of dominant negative Rac1^T17N mutant. Several E3 ubiquitin ligases, including SCF^βTrCP, regulate YAP protein stability. To be recognized by this ligase, the βTrCP degron of YAP (amino acid 383-388) requires its phosphorylation by casein kinase 1 at Ser384 and Ser387, but these events must first be primed by the phosphorylation of Ser381 by LATS1/2. Using Flag-tagged mutants of YAP, we show that YAP degradation after Rac1 inhibition requires the integrity of this degron and is blocked by the silencing of βTrCP1/2 and by the inhibition of casein kinase 1. Unexpectedly, YAP degradation after Rac1 inhibition was still observed after the silencing of LATS1/2 or in cells carrying a LATS1/2 double knockout. Conclusions: These results reveal Rac1 as an oncogenic KRAS effector that contributes to YAP stabilization in PC cells. They also show that this regulation of YAP by Rac1 requires the SCF^βTrCP ligase but occurs independently of the LATS1/2 kinases.

Keywords: GTPase; Hippo; Rac1; Ras; YAP; pancreatic cancer; ubiquitin ligases; βTrCP.

Figure 1

YAP levels are regulated by Rac1. (A) Impacts of oncogenic K-Ras^G12D inhibition on levels of YAP in PC cells. HPAF/CD18 cells, known to express oncogenic K-Ras^G12D, were treated with 50 nM MRTX1133 for the indicated times. The experiment was performed in medium supplemented with either 10% FBS or 0.3% FBS. Levels of pERK, YAP, and S127-phosphorylated YAP were measured by immunoblotting. Total ERK was used as a loading control. (B) YAP levels in PC cells treated with inhibitors of oncogenic Ras effectors. HPAF/CD18 cells were treated for 16 h with inhibitors of the MAPK pathway (50 μM U0126), PI3K kinase (20 μM LY294002), Rac1 GTPase (50 μM EHT-1864), and PAK1-3 kinases (5 μM FRAX597; 20 μM IPA-3). Levels of YAP and S127-phosphorylated YAP were measured by immunoblotting. Actin was used as a loading control. (C–F) YAP levels in a panel of human pancreatic cell lines treated with different Rac1 inhibitors. The indicated cell lines were treated with EHT-1864 (50 μM), NSC23766 (100 μM), and/or Ehop-016 (20 μM). Sixteen hours later, levels of YAP and S127-phosphorylated YAP were measured. Actin was used as an internal standard. (G) The siRNA-mediated knockdown of Rac1 reduces YAP levels in PC cells. In duplicate, HPAF/CD18 cells were transfected with Rac1 siRNA (Rac1) or with a non-targeting siRNA (NT). Two days later, levels of Rac1 and YAP were measured. GAPDH was used as an internal standard. Densitometry readings for the intensity ratios of Rac1/GAPDH and YAP/GAPDH, with the value of the first NT-transfected replicate arbitrarily set to 1. The dotted line indicates a lane that was spliced out of the raw imaging data. As shown in the supplement, the two sides of the dotted line are from the same exposure of the same blot. (H) Expression of dominant negative Rac1^T17N mutant reduces YAP levels in PC cells. HPAF/CD18 cells were infected for 24 h with adenoviral particles carrying no insert (Ad.CTR; 50 pfu/cell) or expressing Rac1^T17N (Ad.N17Rac1; 50 pfu/cell). Forty-eight hours post-infection, levels of YAP were quantified using western blotting. Actin was used as an internal standard. The right panel shows representative images of the infected cells at 48 h post-infection (100× magnification). The dotted line indicates a lane that was spliced out of the raw imaging data. As shown in the supplement, the two sides of the dotted line are from the same exposure of the same blot. For all panels, the relevant densitometry readings for the indicated intensity ratios (pYAP/ERK, YAP/ERK, pYAP/YAP, pYAP/Actin, YAP/Actin, Rac1/GAPDH, and YAP/GAPDH) are shown below the lanes of each western blot.

Figure 2

Activated Rac1 mutant cooperates with E6 and E7 to up-regulate YAP levels. (A) Levels of YAP and Rac1 in a panel of PC cell lines and HPNE cells. Levels of the two proteins were quantified by immunoblotting in the indicated cell lines. GAPDH was used as an internal standard. The graph on the right correlates the protein levels of YAP and Rac1. The Spearman’s correlation coefficient (r_s) is shown. Cell lines with wild-type KRAS are labeled with a blue circle and those with oncogenic KRAS have the black circles. (B) Transduction of HPNE and HPNE/E6/E7 cells with activated forms of Rac1 and BRAF. HPNE cells (Puro^R) and HPNE/E6/E7 cells (Puro^R, Neo^R) were infected with retroviral particles carrying no insert (pLXSH, Hygro^R) or expressing oncogenic mutants of either BRAF (pMSCV-BRAF^V600E, Zeocin^R) or Rac1 (pLXSH-Myc-Rac1^G12V, Hygro^R). After 10 days of selection for viral integration, cells were examined for differences in YAP levels and markers of senescence. (C) Oncogenic BRAF triggers oncogene-induced senescence in HPNE cells. In triplicates, selected cells were plated at low density, histochemically stained to reveal SA-β-galactosidase activity, and then counter-stained with eosin. Representative images of the stained cells are shown on the left. Scale bars are 200 μm in length. The right graph shows the percentage of senescent cells in each population expressed as a mean +/− S.D. (n = 3). * Statistically different from the other two populations in a Student’s t-test with p < 0.001. (D) Activated Rac1 cooperates with E6/E7 to elevate YAP levels in PC cells. Selected cell populations were analyzed by immunoblotting with the indicated antibodies. GAPDH was used as an internal standard. The dotted line indicates a lane that was spliced out of the raw imaging data. As shown in the supplement, the two sides of the dotted line are from the same exposure of the same blot. (E) YAP levels are slightly elevated in HPNE derivatives expressing oncogenic K-Ras^G12D. HPNE/E6/E7 and HPNE/E6/E7/st cells and the same expressing oncogenic K-Ras^G12D were analyzed for differences in levels of YAP and S127-phosphorylated YAP. Using a GST-PAK1 pulldown assay, levels of GTP-bound Rac1 (GTP-Rac1) were also measured, along with total Rac1 levels. Actin was used as an internal standard. For all panels, the relevant densitometry readings for the indicated intensity ratios (YAP/GAPDH, Rac1/GAPDH, Rac1/ERK, myc/ERK, p16/ERK, pYAP/ERK, YAP/ERK, pERK/ERK, pYAP/YAP, Rac1-GTP/Rac1, pYAP/Actin, and YAP/Actin) are shown below the lanes of each western blot.

Figure 3

YAP degradation after Rac1 inhibition requires the SCF^βTrCP E3 ubiquitin ligase. (A) Time course of YAP decline after Rac1 inhibition. HPAF/CD18 cells were harvested at the indicated time point after the addition of EHT-1864 (50 μM). Levels of YAP and S127-phosphorylated YAP were quantified by immunoblotting. GAPDH was used as an internal standard. (B) MG132 blocks the degradation of YAP elicited by the inhibition of Rac1. After 12 h of exposure to EHT-1864 (50 μM; +) or vehicle control (−), HPAF/CD18 cells were exposed or not to MG132 (20 μg/mL) for 4 h prior to harvesting. (C) Schematic description of YAP structure showing the position and sequence of critical degrons and phosphorylation sites. Upper drawing shows the structure of YAP, including its heterodimerization domain (TEAD), WD40 domains (WW), transactivation domain (TAD), putative degrons (βTrCP1/2, FBXW7), and LATS1/2 phosphorylation sites (S61, S109, S127, S128, S131, S163, S164, and S381). Lower left panel: Sequence of the putative FBXW7 phosphodegron highlighting its consensus and its potentially required phosphoserine group (S351). Changes introduced by the S351A/P352A mutation are also shown. Lower right panel: Sequence of the βTrCP degron of YAP highlighting its consensus and its required CK1 (S384, S387) and LATS1/2 (S381) phosphorylation sites. Changes introduced by the D383A/S384A mutation are also shown. (D) Detection of the Flag-YAP proteins in retrovirally infected Panc1 cells. Panc1 cells infected with pLXSH viruses carrying no insert (Empty), Flag-tagged YAP (WT), or its various mutant versions (5SA, D383A/S384A, S351A/P352A) were analyzed for the presence of Flag-YAP. GAPDH was used as an internal control. (E) The βTrCP degron is needed for YAP degradation after Rac1 inhibition, but not the S381 LATS1/2 phosphorylation site. In duplicate, Panc1 cells expressing the different mutants of Flag-YAP were exposed to EHT-1864 (50 μM). Sixteen hours later, Flag-tagged proteins were quantified using western blotting. Again, GAPDH was used as an internal control. (F) The siRNA-mediated knockdown of Skp1 blocks YAP degradation after Rac1 inhibition. Panc1 cells were transfected with skp1 siRNA or with a non-targeting siRNA. Forty-eight hours later, transfected cells were treated with EHT-1864 (50 μM; +) or vehicle control (−) for 16 h prior to western blot analysis. (G) The siRNA-mediated knockdown of the βTrCP1/2 proteins blocks YAP degradation after Rac1 inhibition. Panc1 cells were transfected with a non-targeting siRNA or with siRNA against βTrCP1, βTrCP2, or both proteins. Forty-eight hours later, transfected cells were treated with EHT-1864 (50 μM; +) or vehicle control (−) for 16 h prior to western blot analysis. GAPDH was again used as an internal standard. For all panels, densitometry readings for the indicated intensity ratios (pYAP/GAPDH, YAP/GAPDH, Skp1/GAPDH, and FLAG/GAPDH) are shown below the lanes of each western blot.

Figure 4

YAP degradation after Rac1 inhibition is LATS1/2-independent but requires CK1. (A) The silencing of LATS1/2 fails to prevent YAP degradation after Rac1 inhibition. Panc1 cells were transfected with a non-targeting siRNA or with siRNA against LATS1, LATS2, or both proteins. Forty-eight hours later, transfected cells were treated with EHT-1864 (50 μM; +) or the vehicle control (−) for 16 h prior to western blot analysis. GAPDH was again used as an internal standard. (B) Detection of LATS1 and LATS2 in the LATS1/2-proficient and -deficient HeLa cells. Cells were probed with the indicated antibodies. (C,D) LATS1/2 are not needed for YAP degradation after Rac1 inhibition. LATS1/2-proficient and -deficient cells HeLa cells were exposed to 50 μM EHT-1864 (C) or the DMSO vehicle (D) for 16 h, after which YAP levels were measured. (E,F) Panc1 cells expressing the Flag-YAP protein (E) or its 5SA mutant (F) were exposed or not to CK1 inhibitor IC-261 (10 μM), after which YAP levels were measured. GAPDH was used as an internal standard. For all panels, densitometry readings for the indicated intensity ratios (pYAP/GAPDH, YAP/GAPDH, LATS1/GAPDH, and FLAG/GAPDH) are shown below the lanes of each western blot.

Figure 5

A model for the regulation of YAP stability by Rac1 and Ras. Previous studies have revealed complex interactions between the Hippo and Ras pathways, including both positive and negative regulations of MST1/2 and LATS1/2 by the Ras oncogenes. Downstream of Ras, YAP can be activated by the AKT phosphorylation and inhibition of MST2 (Arrow #1) [54,55] or the MAPK-mediated phosphorylation of Ajuba, resulting in LATS1/2 inhibition (Arrow #2) [37]. However, oncogenic Ras can also cause the RASSF1A-dependent activation of the MST2-LATS1 complex, leading to YAP phosphorylation and inhibition (Arrow #3) [54,55,58]. These interactions are expected to impact the S381-phosphorylation of YAP, both positively and negatively. This S381 phosphorylation creates a priming site that allows the S384/S387 phosphorylation of YAP (in green) by casein kinase 1 (CK1). This phosphorylation activates a βTrCP degron (boxed) that promotes the polyubiquitination of YAP by the SCF^βTrCP ligase and its degradation by the 26S proteasome. In this article, we describe a novel Rac1-regulated pathway that controls the stability of YAP (Arrow #4). This pathway operated outside of the Hippo pathway and did not require the LATS1/2 kinases or the phosphorylation of S381 to regulate YAP stability. Like the Hippo pathway, this other pathway regulated YAP stability in manners that were still dependent on both the integrity of the βTrCP degron and the kinase activity of CK1. We propose that this new pathway activates the βTrCP degron by the direct phosphorylation of S384 by a so-called “DSG” kinase. At the actin cytoskeleton (gray shaded area), the activity of this “DSG” kinase would normally be inhibited by Rac1 and its downstream effectors. Upon Rac1 inhibition, this kinase would rapidly phosphorylate S384, resulting in the CK1 phosphorylation of S387 and full activation of the degron. The existence of a so-called “DSG” kinase has been proposed before to explain the βTrCP-dependent proteolysis of Claspin [118,119].