ERK promotes tumorigenesis by inhibiting FOXO3a via MDM2-mediated degradation

Abstract

The RAS-ERK pathway is known to play a pivotal role in differentiation, proliferation and tumour progression. Here, we show that Erk downregulates Forkhead box O 3a (FOXO3a) by directly interacting with and phosphorylating FOXO3a at Ser 294, Ser 344 and Ser 425, which consequently promotes cell proliferation and tumorigenesis. The ERK-phosphorylated FOXO3a degrades via an MDM2-mediated ubiquitin-proteasome pathway. However, the non-phosphorylated FOXO3a mutant is resistant to the interaction and degradation by murine double minute 2 (MDM2), thereby resulting in a strong inhibition of cell proliferation and tumorigenicity. Taken together, our study elucidates a novel pathway in cell growth and tumorigenesis through negative regulation of FOXO3a by RAS-ERK and MDM2.

Figure 1

ERk suppresses FOXO3a stability and induces its nuclear exclusion. (a–d) Lysates of 293T cells were subjected to immunoblotting with the indicated antibodies after being transfected with ERk2 and MEk1^CA (a), control vector or ERk^DN (b), ERk2^DN and MEk1^CA (c), and control vector and ERK1 and ERK2 siRNA (d). (e–h) Lysates of the following cells were analysed by direct immunoblotting with the indicated antibodies: MDA-MB-453 cells were treated with DMSO or U0126 (2 μM) for 4 h (e), NIH3T3 cells and NIH3T3 RAS-transformed cells (f), Hep-3B and Hep-3BX (g), and Hep-3BX (h) cells were treated with increasing dosages of U0126. (i) MCF-7 cells were extracted at the indicated times after CHX (1 μg ml⁻¹) incubation before treatment with either DMSO (control) or U0126. (j–l) Lysates of MCF-7 cells (j) treated with (DMSO, U0126, or PD98059 (20 μM), NIH3T3 and NIH3T3 VRAS-transformed cells (k), and Hep-3B and Hep-3BX cells (l) were subjected to immunoblotting with the indicated antibodies. (m) Real-time PCR transcript of p27Kip1 and Bim were measured, and the relative fold of induction was calculated and compared between DMSO or U0126 treated cells. (n) Lysates of 293T cells cotransfected with control vectors were subjected to luciferase assays with a p27 promoter-driven luciferase reporter. In m and n, representative results s.d. from three experiments (n = 3) conducted in duplicates are shown. (o) Hep-3B cells were serum starved and then treated with EGF (50 ng ml⁻¹) for 30 min in the presence (+) or absence (−) of U0126. Nuclear and cytoplasmic fractions were then analysed by immunoblotting with the indicated antibodies. Equal amount of lysates was loaded in each lane. Similar results were also obtained by using MCF-7 cells (data not shown). Uncropped images of the scans in a, h, i and k are shown in the Supplementary Information, Fig. S5.

Figure 2

ERk interacts with and phosphorylates FOXO3a in vitro and in vivo. (a) Lysates of MCF-7 cells were subjected to immunoprecipitation and immunoblotting with anti-ERk and anti-FOXO3 antibodies, respectively. (b) Lysates of 293T cells transfected with FOXO3a–HA and ERk were subjected to immunoprecipitation with an anti-HA antibody and immunoblotting with an anti-ERk antibody. (c) The lysates from MCF-7 cells serum starved overnight and stimulated with EGF for 30 min were treated with CIP and subjected to immunoblotting. (d) The putative ERk consensus sequences for phosphorylation on FOXO3a were: P, proline; S, serine; T, threonine; and X, any amino acid. (e) Lysates of 293T cells transfected with GFP–FOXO3a and control or MEk1^CA plasmids were subjected to immunoprecipitation with an anti-FOXO3a antibody and immunoblotting with an anti-phospho-serine antibody or direct immunoblotting with either an anti-p-ERk or an anti-ERk antibody. (f) Cells treated with PD98059 for 4 h before the lysates were extracted and subjected to analysis as described in e. (g) Lysates from MCF-7 cells that had been serum starved overnight and stimulated with EGF for 30 min were subjected to immunoprecipitation with an anti-FOXO3a antibody, and the isolated FOXO3a bands were subjected to mass spectrometry. (h) In vitro kinase assays were conducted by incubating recombinant activated ERk2 with GST–FOXO3a (281–673) and mutant GST–FOXO3a^3A proteins. (i) Lysates of 293T cells cotransfected with FOXO3a or mutant FOXO3a^3A and MEk1^CA were subjected to immunoprecipitation with an anti-FOXO3a antibody and immunoblotting with an anti-phospho-serine antibody. An uncropped image of the blot is shown in the Supplementary Information, Fig. S5

Figure 3

MDM2 is required for ERk-mediated FOXO3a degradation. (a) MCF-7 cell lysates were subjected to immunoprecipitation with an anti-MDM2 antibody and immunoblotting with an anti-FOXO3a antibody. (b) Lysates of 293T cells transfected with MDM2 at different dosages and subjected to immunoblotting. (c) Lysates of cotransfected cells treated with the indicated vectors in the presence of MG-132 (20 μM, 4 h) were subjected to immunoprecipitation and immunoblotting with an anti-FOXO3a antibody. (d) Lysates of p53^−/− or p53^−/−MDM^−/− MEFs were subjected to immunoblotting using different antibodies, as indicated. (e) MEFs were treated with different inhibitors and analysed by immunoblotting. (f, g) MEFs were treated with or without CHX, harvested at the indicated times and subjected to immunoblotting. (h) MEFs were treated with methionine and cysteine-free medium overnight, pulsed with ³⁵S-Met-Cys for 30 min, and chased for the indicated times. Cell lysates were immunoprecipitated with an anti-FOXO3a antibody and subjected to SDS–PAGE, gels were fixed, dried,and subjected to autoradiography (with an intensifying screen). An uncropped image of the blot in e is shown in the Supplementary Information, Fig. S5.

Figure 4

Phosphorylation of FOXO3a by ERk facilitates MDM2-mediated FOXO3a degradation through an ubiquitin-proteasome pathway. (a) Lysates of 293T cells transfected with GFP–FOXO3a^3A or GFP–FOXO3a^3D harvested at different time points after treated with CHX or MG-132 were analysed by immunoblotting. (b) Lysates of MCF-7 cells were subjected to immunoprecipitation and immunoblotting under different conditions. (c, d) Lysates of 293T cells transfected with wild-type (WT) FOXO3a (c) and 293T cells transfected with wild-type FOXO3a or mutants FOXO3a^3A and FOXO3a^3D (d) were subjected to immunoprecipitation and immunoblotting. (e–g) Lysates of 293T cells cotransfected with FOXO3a^3D (e) or FOXO3a^3A (g), together with different dosages of wild-type MDM2 or deletion mutant of MDM2, MDM2^Δ9 (f), with or without MG-132, were analysed by immunoblotting. (h) Lysates of 293T cells cotransfected with FOXO3a^3A or FOXO3a^3D, together with wild-type MDM2 or mutant MDM2^464A were subjected to immunoblotting. (i) Lysates of 293T cells cotransfected with wild-type FOXO3a, FOXO3a^3D or FOXO3a^3A, together with MDM2 and ubiquitin in the presence of MG-132, were analysed by immunoblotting. (j) Lysates of 293T cells cotransfected with 6XDBE–Luc and pRL-Tk, and FOXO3a or FOXO3a^3A, together with control vector or MEk1^CA were subjected to luciferase assay. The results were from three experiments (n = 3) with the error bars representing s.d. from duplicates in each case. Uncropped images of the blots in e–g are shown in the Supplementary Information, Fig. S6.

Figure 5

FOXO3a^3A, but not FOXO3a^3D, inhibits tumorigenesis and induces apoptosis. (a) MDA-MB-435 cells were transfected with control vector, wild-type FOXO3a, FOXO3a^3A or FOXO3a^3D. After sorting with GFP marker, GFP-positive cells were subjected to the colony formation and MTT assays. (b) Cells obtained as described in a were measured for cell growth rate by MTT assay. All experiments were performed in triplicate (n = 3). A representative sample is shown in a and b. (c) Lysates from cells obtained as described in a and MDA-MB-435 cell lines, as indicated, were analysed by immunoblotting. (d) MDA-MB-435 cells were transfected with control vector, wild-type FOXO3a, FOXO3a^3A or FOXO3a^3D and, after propidium iodide (PI) staining, the cells were analysed by FACS, performed as described in the Methods. Only the GFP-positive cells were measured in the sub-G1 phase as apoptotic cells, and the apoptosis ratio was normalized with the control vector. (e) The MDA-MB-435-transfected cell lines were sorted for GFP-positive cells, which then were injected into the mammary fat pads of nude mice. (f) MDA-MB-435 cells were transfected with control vector, ERk1^DN and ERk2^DN (DN), and ERk1^DN, ERk2^DN and FOXO3a siRNA, and the cells were subjected to MTT assays. (g) Cells (2 × 10⁶) cells obtained as described in f were injected into the mammary fat pads of nude mice. The tumour volume was measured twice per week. The inset shows FOXO3a expression quantified by normalization with tubulin expression. An uncroppped image of the blot in c is shown in the Supplementary Information, Fig. S6.

Figure 6

ERk promotes tumorigenesis by inhibiting FOXO3a via MDM2-mediated degradation. Schematic representation showing that ERk interacts with and phosphorylates FOXO3a at Ser 294, Ser 344 and Ser 425; phosphorylation at these residues increases FOXO3a-MDM2 interaction and enhances FOXO3a degradation through a MDM2-dependent ubiquitin-proteasome pathway, leading to tumorigenesis.