MEKK1 plays a critical role in activating the transcription factor C/EBP-beta-dependent gene expression in response to IFN-gamma

Abstract

IFN-gamma induces a number of genes to up-regulate cellular responses by using specific transcription factors and the cognate elements. We recently discovered that CCAAT/enhancer-binding protein-beta (C/EBP-beta) induces gene transcription through an IFN-response element called gamma-IFN-activated transcriptional element (GATE). Using mutant cells, chemical inhibitors, and specific dominant negative inhibitors, we show that induction of GATE-driven gene expression depends on MEK1 (mitogen-activated protein kinase kinase/extracellular signal-regulated protein kinase kinase) and ERKs (extracellular signal-regulated protein kinases) but is independent of Raf-1. Interestingly in cells lacking the MEKK1 gene or expressing the dominant negative MEKK1, ERK activation, and GATE dependent gene expression is inhibited. A dominant negative MEKK1 blocks C/EBP-beta-driven gene expression stimulated by IFN-gamma. These studies describe an IFN-gamma-stimulated pathway that involves MEKK1-MEK1-ERK1/2 kinases to regulate C/EBP-beta-dependent gene expression.

Figure 1

Effect of MAPK pathway inhibitors on IFN-γ-induced gene expression. The indicated cells were transfected with the P4-luicferase construct (0.4 μg) and treated with murine IFN-γ (500 units/ml). Transfection was performed and luciferase activity was determined [relative luciferase units (RLU)] after treatment with the indicated agents for 16 h. Before stimulating with IFN-γ, cells were exposed to the DMSO (D), U0126 (U, 10 μM), and SB202190 (S, 25 μM) for 30 min. A plus (+) sign indicates IFN-γ treatment. Bars represent the mean ± SE of triplicates. C is similar to A except different cells were used. (B and D) The effects of MEK1 and p38 kinase inhibitors on the IFN-γ-induced luciferase gene expression from pIRE, a STAT1 binding element. (E) The effect of EGF (10 ng/ml) on AP1-RE-driven luciferase gene expression. (F) Western blots of cell extracts (50 μg) probed with antibodies specific for C/EBP-β and actin.

Figure 2

MEKK1 is required for GATE-dependent gene expression. GATE-Luc plasmid was transfected into the indicated cells, and gene expression was analyzed as described in Fig. 1. (A) Cells were transfected with the indicated effector plasmids in addition to GATE-Luc. (B) Effect of MEKK1 mutation on GATE-Luc. (C) A CA-MEKK1 rescues GATE-dependent gene expression in MEKK1^−/− cells. Cells were transfected with expression vector (pCMV), DN-MEKK1, or CA-MEKK1 along with the P4 reporter, and gene expression was analyzed. A Western blot (Bottom) shows the expression of MEKK1 mutants as analyzed with specific antibodies. (D) Expression of endogenous ISGF3γ. Cell extracts (60 μg) Western blotted by using the indicated antibodies. RLU, relative luciferase units.

Figure 3

MEK1 is critical for MEKK1-initiated GATE-dependent gene expression. (A) The effects of a CA-MEK1 on GATE-dependent gene expression in MEKK1^+/+ and MEKK1^−/− cells. After transfection with the indicated plasmids, cells were treated with IFN-γ and analyzed for P4-driven luciferase activity. RLU, relative luciferase units. (B) A catalytically inactive MEK1 inhibits GATE-dependent gene induction in MEKK1^+/+ cells. A Western blot showing expression of endogenous and DN-MEKK1 is shown also.

Figure 4

ERK1/2 activation in c-Raf^−/− and MEKK1^−/− cells. N, no treatment; E, EGF (10 ng/ml); I, IFN-γ (500 units/ml); Ei, 11,15,17-eicosatetranoic acid (10 μM). (A and C) The blots were probed with phospho-ERK-specific antibodies. (B and D) The blots shown in A and C were stripped and probed with ERK2-specific antibodies to demonstrate the presence of an equal amount of ERK in each lane. Whole-cell protein (50 μg) was used from each treatment.

Figure 5

ERK1/2 are critical for GATE-driven gene expression in MEKK1^+/+ cells. (A) Mutant ERKs block IFN-γ-stimulated gene expression through GATE. The P4 reporter was transfected with expression vector carrying DN-ERK1, ERK2 mutants, or both. The total amount of DNA transfected was kept constant by adding pCEP4 vector where necessary. V, vector, pCEP4; E1, ERK1 (K71R) mutant; E2, ERK2 (K52R) mutant. Luciferase activity was determined as described above. A similar experiment was repeated in ERK1^−/− cells (B). RLU, relative luciferase units.

Figure 6

MEKK1 is necessary for C/EBP-β-dependent gene expression. The indicated cell lines were transfected with the specific plasmids along with the P4 reporter, and luciferase activity was measured. (A) A mutant C/EBP-β lacking ERK consensus motif blocks IFN-γ-stimulated gene expression. C/EBP-W, wild-type C/EBP-β; C/EBP-M, a mutant C/EBP-β with a T-A conversion of the ERK consensus GTPS motif. (B) DN-MEKK1 blocks IFN-γ-stimulated C/EBP-β-dependent gene expression in RAW macrophage cells. (C) IFN-γ-stimulated C/EBP-β-dependent gene expression through GATE is blocked in MEKK1^−/− cells. (D) DN-MEKK1 blocks gene induction by the transfected C/EBP-β in response to IFN-γ in C/EBP-β^−/− cells. (E) Effect of C/EBP-β on GATE mutant-driven expression. Expression of the transfected C/EBP-β is shown for the luciferase data of D and E under the respective panels. (F) Effect of C/EBP-β on GATE-driven gene expression in Raf^−/− cells. RLU, relative luciferase units.