EGF-induced ERK activation promotes CK2-mediated disassociation of alpha-Catenin from beta-Catenin and transactivation of beta-Catenin

Abstract

Increased transcriptional activity of beta-catenin resulting from Wnt/Wingless-dependent or -independent signaling has been detected in many types of human cancer, but the underlying mechanism of Wnt-independent regulation remains unclear. We demonstrate here that EGFR activation results in disruption of the complex of beta-catenin and alpha-catenin, thereby abrogating the inhibitory effect of alpha-catenin on beta-catenin transactivation via CK2alpha-dependent phosphorylation of alpha-catenin at S641. ERK2, which is activated by EGFR signaling, directly binds to CK2alpha via the ERK2 docking groove and phosphorylates CK2alpha primarily at T360/S362, subsequently enhancing CK2alpha activity toward alpha-catenin phosphorylation. In addition, levels of alpha-catenin S641 phosphorylation correlate with levels of ERK1/2 activity in human glioblastoma specimens and with grades of glioma malignancy. This EGFR-ERK-CK2-mediated phosphorylation of alpha-catenin promotes beta-catenin transactivation and tumor cell invasion. These findings highlight the importance of the crosstalk between EGFR and Wnt pathways in tumor development.

Figure 1. EGFR Activation Results in Disruption of β-catenin and α-catenin Complex and Promotes β-catenin Transactivation

(A, B) Immunoprecipitation of α-catenin was followed by immunoblotting with a β-catenin antibody. (A) The indicated cell lines were treated with EGF (100 ng/ml) for the indicated time. (B) The subcellular fractionation was prepared from A431 cells treated with EGF (100 ng/ml) for 6 hr. (C, D) A431 cells without (C) or with (D) transfection of FLAG-α-catenin were treated with EGF (100 ng/ml) for 10 or 24 hr (C) or 6 hr (D) and immunostained with the indicated antibodies. (E) The nuclear fractions (left panel) or total cell lysate (right panel) were prepared from 293T cells expressing EGFR with or without FLAG-α-catenin. The cells were treated with EGF (100 ng/ml) for 6 hr and immunoblotting was performed with the indicated antibodies. (F, H) The luciferase activity was determined after cells were treated with EGF (100 ng/ml) for 8 hr. The relative levels of luciferase activity were normalized to the levels of untreated cells and to the levels of luciferase activity of the Renilla control plasmid. Data represent the mean ± standard deviation of three independent experiments. (F) pCep4-EGFR with either TOP-FLASH or FOP-FLASH were co-transfected with or without FLAG-α-catenin into 293T cells. (G) A431 cells were stably expressed with control shRNA or α-catenin shRNA. Immunoblotting analyses were performed with the indicated antibodies. (H) TOP-FLASH or FOP-FLASH was transfected into A431 cells that were stably expressed with control shRNA or α-catenin shRNA.

Figure 2. CK2α Interacts with and Phosphorylates α-catenin at S641

(B-E) Immunoblotting analyses were performed with the indicated antibodies. (A) α-catenin immunoprecipitated from EGF-stimulated A431 cells was analyzed with mass spectrometry. Mass spectrometric analysis of a tryptic fragment 634-TPEELDDSDFETEDFDVR-651 indicates S641 was phosphorylated. The m/z difference between y-11 and y-10 matched with phospho-Ser. (B) In vitro kinase assays were performed with purified bacterially expressed WT His-α-catenin or His-α-catenin S641A with or without CK2α. (C) pRc/CMV2-HA-CK2α was co-transfected with pEGFP N1-FLAG-tagged WT α-catenin or α-catenin S641A mutant into 293T cells. Immunoblotting of immunoprecipitated FLAG-tagged α-catenins with the anti–α-catenin Ser641 antibody was performed. (D) The subcellular fractionation was prepared from A431 cells treated with EGF (100 ng/ml) for 30 min. (E) A431 cells expressing FLAG-tagged WT α-catenin were treated with or without EGF (100 ng/ml) for 1 or 3 hr. CK2α was immunoprecipitated. (F) A431 cells with or without expression of CK2α shRNA or a control shRNA were treated with or without EGF (100 ng/ml) for 30 min.

Figure 3. CK2 α-mediated Phosphorylation of α-catenin at S641 Releases α-catenin from Binding to α-catenin

Immunoprecipitation and immunoblotting analyses were performed with the indicated antibodies. (A) In vitro kinase assays were performed with purified bacterially expressed WT His-α-catenin with or without CK2α, which is followed by GST pull-down analyses with GST (control) or GST-α-catenin glutathione-agarose beads. (B) FLAG-tagged α-catenin was co-transfected with pRc/CMV2-HA-CK2α or pRc/CMV2-HA-CK2α K68M into 293T cells. (C) A431 cells expressing with or without CK2α shRNA were treated with or without EGF (100 ng/ml) for 2 hr. (D) FLAG-tagged WT α-catenin, α-catenin S641A, or α-catenin S641D mutant were transfected into 293T cells. Immunoprecipitation of the total cell lysate with an anti-FLAG antibody was followed by immunoblotting analyses with the indicated antibodies. (E) FLAG-tagged WT α-catenin, α-catenin S641A, or α-catenin S641D mutant were transfected into 293T cells. Immunoprecipitation of the membrane or nucleus fraction was performed with the indicated antibodies.

Figure 4. ERK1/2 Phosphorylates CK2α and Enhances CK2α Activity Toward α-catenin in Response to EGF Stimulation

Immunoprecipitation and immunoblotting analyses were performed with the indicated antibodies. (A) In vitro kinase assays were performed by mixing purified ERK2 or ERK5 with boiled CK2α. (B) In vitro kinase assays were performed by mixing purified ERK2 with boiled CK2α, CK2α T360A, CK2α S362A, or T360/S362A mutants. (C) 293T cells expressing MEK1 Q56P and ERK2 with FLAG-tagged WT CK2α or CK2α T360/S362A were metabolically labeled with ³²P-phosphate for 12 hr. Immunoprecipitation with an anti-FLAG antibody was performed. (D) Purified CK2α with or without purified His-α-catenin (left panel) or HDAC3 (right panel) was mixed with purified ERK2 in kinase assays. The imagines were quantified by scanning densitometry. (E) In vitro kinase assays were performed by mixing non-boiled WT CK2α or CK2α K68M with or without purified ERK2. (F, G) MEK1 Q56P was co-transfected with WT ERK2 or ERK2 K52R kinase-dead mutant together with WT α-catenin or α-catenin S641A into 293T cells. The cells were treated without (F) or with TBB (40 μM) (G) for 6 hr before harvesting. (H) A431 or U373 cells were pretreated with U0126 (25 μM) or TBB (40 μM) for 30 min before EGF (100 ng/ml) stimulation for 30 min.

Fig. 5. ERK2 directly binds to CK2α through its docking groove

Immunoprecipitation and immunoblotting analyses were performed with the indicated antibodies. (A) Immobilized purified His-CK2α was mixed with purified ERK2 followed by washing three times with cell lysis buffer and immunoblotting analysis. Control, nickel-agarose beads. (B) 293T cells were transfected with a control vector or vectors expressing MEK1 Q56P and FLAG-ERK2. (C) A vector expressing MEK1 Q56P were co-transfected with FLAG-WT ERK2, FLAG-ERK2 T/E (T157/158E), FLAG-ERK2 D/N (D316/319N), or FLAG-ERK2 T/E-D/N mutant into 293T cells.

Figure 6. EGF-induced ERK2 Activation Promotes β-catenin Transactivation and Tumor Cell Invasion by Enhancing CK2-phosphorylated α-catenin at S641

(A-D) The relative levels of luciferase activity were normalized to the levels of untreated cells and to the levels of luciferase activity of the Renilla control plasmid. Data represent the mean ± standard deviation of three independent experiments. (A) TOP-FLASH or FOP-FLASH was co-transfected with or without the indicated plasmids into 293T cells. (B) TOP-FLASH or FOP-FLASH was co-transfected with or without vectors expressing MEK1 Q56P and ERK2 into 293T cells. The cells were pretreated with TBB (40 μM) for 6 hr before harvesting. (C) TOP-FLASH or FOP-FLASH was co-transfected with a vector expressing EGFR into 293T cells. The cells were pretreated with TBB (40 μM), apigenin (40 μM), U0126 (25 μM), or PD98059 (30 μM) for 30 min before EGF (100 ng/ml) for 6 hr. (D) TOP-FLASH or FOP-FLASH was co-transfected with vectors expressing FLAG-tagged WT α-catenin, α-catenin S641A, or α-catenin S641D into A431 cells. The cells were treated with EGF (100 ng/ml) for 8 hr. Immunoblotting with indicated antibodies was performed (left upper panel). (E, F) A pool of A431 cells expressing with a control shRNA or CK2α shRNA (E) or stably transfected with a vector, FLAG-tagged WT α-catenin, α-catenin S641A, or α-catenin S641D (F) was plated at the top surface of the Matrigel in the absence or presence of EGF (100 ng/ml). One day after plating, cells that migrated to the opposite side of the insert were stained with crystal violet. Representative microphotographs are shown (left panel). The membranes with invaded cells were dissolved in 4% deoxycholic acid and read colorimetrically at 590 nm for quantification of invasion. Data represent the mean ± standard deviation of three independent experiments (right panel). Immunoblotting with indicated antibodies was performed (right upper panel).

Figure 7. Levels of α-Catenin S641 Phosphorylation Correlate with Levels of ERK1/2 Activity in Human GBM and with Grades of Glioma Malignancy

(A) Immunohistochemical staining with anti–phospho-ERK1/2 and anti–phospho-α-catenin S641A antibodies was performed on 46 GBM specimens. Representative photos of four tumors were shown. (B) Immunohistochemical staining on the tissue sections was scored semi-quantitatively as described in Materials and Methods (Pearson product moment correlation test r = 0.76; p < 0.0001). Note that some of the dots on the graphs represented more than one specimen (some scores overlapped). (C) Immunohistochemical staining of 23 diffuse astrocytoma specimens with an anti–phospho-α-catenin S641 antibody was performed and analyzed by comparing it with the staining of 46 GBM specimens (The student's t test, two tailed, p < 0.000001). Data represent the mean ± standard deviation of sample scores.