1. Biochemistry and Chemical Biology
  2. Cell Biology
Download icon

FAK/PYK2 promotes the Wnt/β-catenin pathway and intestinal tumorigenesis by phosphorylating GSK3β

  1. Chenxi Gao
  2. Guangming Chen
  3. Shih-Fan Kuan
  4. Dennis Han Zhang
  5. David D Schlaepfer
  6. Jing Hu  Is a corresponding author
  1. University of Pittsburgh School of Medicine, United States
  2. University of Pittsburgh Cancer Institute, United States
  3. Dietrich School of Arts and Sciences, United States
  4. University of California, San Diego, United States
Research Article
Cite this article as: eLife 2015;4:e10072 doi: 10.7554/eLife.10072
6 figures

Figures

β-TrCP-mediated GSK3β ubiquitination requires GSK3βY216 phosphorylation.

(A) The whole cells lysates from HEK293T cells transfected with empty vector or constructs expressing HA-tagged GSK3β or its phosphorylation mutants were used for immunoprecipitation (IP) and immunoblotting (IB) analysis with indicated antibodies. (B) Ubiquitination of HA-GSK3β was evaluated by IP with anti-FLAG using whole cell lysates derived from HEK293T cells transfected with indicated constructs followed by IB analysis with anti-HA. (C) In vitro synthesized wild type (WT) and mutant HA-GSK3β proteins were immunoprecipitated with anti-HA affinity gel. Left panel: in vitro ubiquitination assay was performed by incubating immunoprecipitated HA-GSK3β with E1, E2, Ub and SCFβ-TrCP1 in the presence of ATP at 37°C for 2 hr. The reaction mixtures were analyzed by IB. Right panel: the immunoprecipitated HA-GSK3β was incubated with ATP at 37°C for 2 hr before subjecting to IB using antibody recognizing phosphorylated GSK3βY216. (D) HEK293T cells were treated with Wnt3a-conditioned medium (CM) for the indicated time before harvesting for IB analysis. (E) HEK293T cells were transfected with indicated plasmids and treated with control-CM or Wnt3a-CM for 6 hr. Ubiquitination of HA-GSK3β WT and Y216F mutant protein were evaluated by IP with anti-FLAG followed by IB with anti-HA. (F) HEK293T cells were transfected with constructs expressing HA-GSK3β or its mutant. 24 hr post-transfection, the cells were harvested for IP with anti-HA followed by IB with anti-Axin1. (G) Upper panel: HA-tagged GSK3β or its mutants were ectopically expressed in HEK293T cells and immunoprecipitated with anti-HA. The precipitated proteins were incubated with purified CK1 and β-catenin in the presence of ATP at 37°C for 1 hr. Kinase activity was evaluated by IB with antibody recognizing phosphorylation of β-cateninS33/S37/T41 catalyzed by GSK3β. Lower panel: the whole cell lysates from WT MEFs, GSK3β knockout (KO) and GSK3α knockdown (KD) MEFs reconstituted with empty vector or indicated HA-GSK3β mutants were used for IB. Antibody recognizing both GSK3α and GSK3β was used for detecting endogenous GSK3 and HA-tagged GSK3β. This antibody recognized HA-GSK3βY216F but not HA-GSK3βY216E mutant protein.

https://doi.org/10.7554/eLife.10072.003
Activation of Wnt/β-Catenin signaling requires GSK3βY216 phosphorylation.

(A) IB analysis using whole cell lysates from indicated MEFs treated with control-CM or Wnt3a-CM for 6 hr. Anti-GSK3 was used to detect endogenous GSK3 and HA-GSK3β. This antibody recognized HA-GSK3βY216F but not HA-GSK3βY216E or GSK3βY216E/S4 mutant proteins. (B) The GSK3β KO/GSK3α KD MEFs were transfected with vectors expressing indicated proteins, treated with Wnt3a-CM and proteasome inhibitor MG132 (10 µM) for 4 hr as indicated, and then harvested for IP and IB analyses using indicated antibodies. (C) Left panel, SW480 cells and GSK3α/β KD SW480 cell lines expressing vector control, shRNA-resistant WT HA-GSK3β or its mutants were used for soft agar assays. Pictures were taken 16 days after plating at 40×. Right panel, quantification of soft agar assays. Data are presented as mean ±SD. Expressions of GSK3α/β (endogenous or mutants) and β-catenin were evaluated by IB at the time of plating. (D) Upper left panel, SW480 cell lines used in (C) were injected into both flanks of nude mice (5 mice in each cell group). Tumor sizes were measured at indicated time points. Tumor volumes were calculated and plotted. * denotes a statistically significant difference (p < 0.01) comparing with shRNA control group. Bottom left panel, tumor weighs from different groups at the end of the experiment. Values are means ±SD (n = 10). Right panel, representative tumors from different groups. (E) IB analysis using whole cell lysates from indicated MEFs treated with control-CM or Wnt3a-CM for 6 hr. Anti-GSK3 was used to detect endogenous GSK3 and Flag-GSK3α. This antibody recognized Flag-GSK3αY279F but not Flag-GSK3αY279E mutant protein. (F) Left panel: IB characterization of the SW480 cell lines. Middle panel: indicated SW480 cell lines were injected into both flanks of nude mice (5 mice in each cell group). Tumors from different groups were weighed at the end of the experiment. Values are means ±SD (n = 10). * denotes a statistically significant difference (p < 0.01) comparing with shRNA control group. Right panel, representative tumors from different groups.

https://doi.org/10.7554/eLife.10072.004
Phosphorylation of GSK3βY216 by FAK/PYK2.

(A) HEK293T cells were pre-treated with or without PF-562271 (4 µM) for 4 hr followed by incubation with control-CM or Wnt3a-CM containing PF-562271 (4 µM) for additional 4 hr as indicated. The whole cell lysates were subjected for IB with indicated antibodies. (B) FAK and PYK2 expressions were evaluated by IB analysis. (C) Parental, FAK KD and shRNA control HEK293T cells were treated with control-CM or Wnt3a-CM for 4 hr before harvesting for IB analysis using indicated antibodies. (DF) FAK KD, PYK2 KD or FAK/PYK2 KD cell lines and shRNA control cell lines were generated using lentiviral-based shRNA targeting FAK or PYK2 or control shRNA. The whole cell lysates were used to evaluate the levels of p-GSK3βY216 and β-catenin by IB with indicated antibodies. (G) GSK3β in SW480 and HT29 cells was immunoprecipitated with anti-GSK3β followed by IB with anti-PYK2 and anti-FAK, respectively. Normal IgG was used as IP control. (H) In vitro synthesized HA-GSK3β was immunoprecipitated with anti-HA. Beads-bound HA-GSK3β was then incubated with purified FAK or PYK2 recombined protein (or kinase buffer as a negative control) in the presence of ATP at 37°C for 2 hr. The reaction mixtures were subjected to IB analysis using indicated antibodies.

https://doi.org/10.7554/eLife.10072.005
FAK/PYK2 promotes the Wnt/β-catenin pathway through phosphorylating GSK3βY216.

(A) Time courses of PF-562271 treatment-induced inhibition of phosphorylation of PYK2, GSK3β and β-catenin in SW480 cells. The whole cell lysates were used for the IB analysis using indicated antibodies. (B) The whole cell lysates from indicated cells were used for IB analysis. (C) The WT MEFs, FAKR454/R454/PYK2 KD MEFs transfected with empty vector or vector expressing indicated GSK3β mutants were treated with Wnt-3a CM or control-CM for 6 hr before harvesting for IB analysis. (D) The WT MEFs and FAKR454/R454/PYK2 KD MEFs transfected with empty vector or vector expressing HA-GSK3βY216E were treated with Wnt3a-CM and proteasome inhibitor MG132 for 4 hr and then harvested for IP and IB analyses. KI/KD: FAKR454/R454/PYK2 KD MEFs expressing empty vector. KI/KD/Y216E: FAKR454/R454/PYK2 KD MEFs expressing HA-GSK3βY216E. (E) Parental SW480 cells and SW480 cell lines expressing vector control, HA-GSK3β or its mutants were treated with 2 µM PF-562271 for 4 days. Cell viability was determined by CCK-8 assay, using SW480 parental cell treated with DMSO as control. Data are presented as mean ±SD. * denotes a statistically significant difference (p < 0.01) between PF-562271-treated parental SW480 cells and cells overexpressing HA-GSK3βY216E. (F) Upper left panel: indicated HT-29 cell lines were injected into both flanks of nude mice (5 mice in each cell group). Animals were treated with either vehicle (5% Gelucire) or PF-562271 (33 mg/kg in vehicle) as indicated by oral gavage twice daily for 3 weeks. Tumor sizes were measured at indicated time points. Tumor volumes were calculated and plotted. * denotes a statistically significant difference (p < 0.01) comparing with vehicle-treated group. Upper right panel, tumor weighs from different groups at the end of the experiment. Values are means ±SD (n = 10). Bottom panel, representative tumors from different groups.

https://doi.org/10.7554/eLife.10072.006
Inhibition of FAK/PYK2 kinase activity suppresses adenoma formation in APCmin/+ mice.

(A) Representative staining results of ileum (distal) segments from APCmin/+ mice treated with vehicle or PF-562271 were shown. (B) Adenoma formation on entire small intestine was quantitatively assessed in APCmin/+ mice treated with vehicle (n = 5) or PF-562271 (n = 5). Data was presented as mean ±S.D. (C) Left panel: tissue lysates extracted from the scraped intestinal mucosa from indicated mice were used for IB analysis. Each lane represented a single mouse. Right panel: tissue lysates extracted from C57 control mouse, APCmin/+ mouse and polyps were used for IB. (D) Tissue lysates used in (C, left panel) were immunoprecipitated with anti-GSK3β. The beads-bound immunoprecipitates were resolved by SDS-PAGE and probed with indicated antibodies.

https://doi.org/10.7554/eLife.10072.007
FAK, PYK2, p-GSK3βY216 and β-catenin positively correlate in human colorectal cancer tissues.

(A) Hematoxylin and eosin (H&E) and immunohistochemical staining of the colorectal adenoma, carcinoma and adjacent normal tissues from a single patient. (B) 34 tumor samples on tissue microarray were immunohistochemical stained and scored. Score distribution and average score of each antibody staining were summarized. (C) Scoring differences between tumor and matched normal tissues from twenty-six patients. * Denotes p-GSK3βY216, FAK or PYK2 scores in different directions. _ highlights the numbers of patients whose TMA scores of p-GSK3βY216, FAK or PYK2, and β-catenin move in the same direction. (D) Hematoxylin and eosin (H&E) and immunohistochemical staining of colorectal carcinoma and adjacent normal tissues from a familial adenomatous polyposis patient. (E) Schematic diagram of the FKA/PYK2/p-GSK3βY216/β-catenin axis and its role in intestinal tumorigenesis.

https://doi.org/10.7554/eLife.10072.008

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Download citations (links to download the citations from this article in formats compatible with various reference manager tools)

Open citations (links to open the citations from this article in various online reference manager services)