LPS-Induced G-CSF Expression in Macrophages Is Mediated by ERK2, but Not ERK1

Abstract

Granulocyte colony-stimulating factor (G-CSF) selectively stimulates proliferation and differentiation of neutrophil progenitors which play important roles in host defense against infectious agents. However, persistent G-CSF production often leads to neutrophilia and excessive inflammatory reactions. There is therefore a need to understand the mechanism regulating G-CSF expression. In this study, we showed that U0126, a MEK1/2 inhibitor, decreases lipopolysaccharide (LPS)-stimulated G-CSF promoter activity, mRNA expression and protein secretion. Using short hairpin RNA knockdown, we demonstrated that ERK2, and not ERK1, involves in LPS-induced G-CSF expression, but not LPS-regulated expression of TNF-α. Reporter assays showed that ERK2 and C/EBPβ synergistically activate G-CSF promoter activity. Further chromatin immunoprecipitation (ChIP) assays revealed that U0126 inhibits LPS-induced binding of NF-κB (p50/p65) and C/EBPβ to the G-CSF promoter, but not their nuclear protein levels. Knockdown of ERK2 inhibits LPS-induced accessibility of the G-CSF promoter region to DNase I, suggesting that chromatin remodeling may occur. These findings clarify that ERK2, rather than ERK1, mediates LPS-induced G-CSF expression in macrophages by remodeling chromatin, and stimulates C/EBPβ-dependent activation of the G-CSF promoter. This study provides a potential target for regulating G-CSF expression.

Fig 1. U0126 inhibits ERK1/2 phosphorylation and G-CSF expression in LPS-stimulated RAW264.7 macrophages.

(A) Cells were left untreated (lane 1) or were incubated either with LPS (100 ng/ml) for 15 to 60 min (lanes 2–4) or with DMSO or U0126 (10 μM) for 30 min, followed by addition of same concentration of LPS and incubation for 0 to 60 min (lanes 5–8), then phosphorylated ERK1/2 and total ERK1/2 were analyzed by Western blotting. The results shown are representative of those obtained in three separate experiments. (B) Cells were incubated with DMSO or U0126 (10 μM) for 30 minutes, then with LPS (100 ng/ml) for the indicated time, then G-CSF levels in the medium were measured by ELISA. (C) Cells were incubated with DMSO or 0.01, 0.1, 1, or 10 μM of U0126 for 30 minutes, then with LPS (100 ng/ml) for 6 h and G-CSF protein levels in the medium were measured by ELISA. The results in B and C are the mean ± SD for three independent experiments (*p<0.01).

Fig 2. ERK2 knockdown in THP-1 macrophages blocks LPS-induced G-CSF expression.

THP-1 cells were infected with lentivirus carrying specific shRNA for ERK1 or ERK2 and induced to differentiate by incubation with PMA (160 nM) for 3 days; lentivirus carrying luciferase shRNA (Luc) was used as a control, then the following tests were performed. (A) Levels of ERK1/2 and β-actin in the cells were determined by Western blotting; the data shown are typical of the results of three experiments. (B) Cells were treated with LPS (100 ng/ml) for 8 h, then G-CSF levels in the medium were determined by ELISA; untreated Luc cells were used as the control. (C and D) Levels of G-CSF mRNA (C) and TNF-α mRNA (D) in cells were determined by RT-qPCR, normalized to the levels of GAPDH mRNA, and expressed relative to levels in the Luc control (relative value = 1). In (B-D), the results are the mean ± SD for five independent experiments. *p<0.05 compared to the corresponding cells not treated with LPS.

Fig 3. Threonine188 of C/EBPβ is important for the interaction of ERK2 with C/EBPβ that activates G-CSF promoter.

(A) RAW264.7 cells were co-transfected with a mixture of two reporter plasmids (pG-CSF(-283/+35)-Luc and phRLTK) and the CA-ERK2, p50, Oct-2, or C/EBPβ plasmid alone or the C/EBPβ plasmid plus the CA-ERK2, p50, or Oct-2 plasmid, then luciferase activity was measured at 24 h after transfection using the Dual-luciferase reporter assay system. (B) RAW264.7 cells were pretreated for 30 min with DMSO (D) or U0126 (U) (10 μM), then were incubated with LPS (100 ng/ml) (L) for another 4 h, after which a nuclear extract (N.E.) and cytosolic extract (C.E.) were isolated and levels of total and phosphorylated C/EBPβ, phosphorylated ERK1/2, β-actin (cytosol internal control), and laminin A/C (nuclear internal control) analyzed by Western blotting. The data shown are typical of those obtained in three experiments. (C) Cells were cotransfected with reporter plasmids (0.5 μg of pG-CSF (-283/+35)-Luc mixed with 0.1 μg of phRLTK), 0.15 μg of CA-ERK expression plasmid, and 0.15 μg of the WT, T188A, or S64A C/EBP expression plasmid or 0.15 μg of pcDNA3.1 (vector control), then the reporter luciferase activities were analyzed 24 h after transfection. (D) Cells were co-transfected with pG-CSF(-283/+35)-Luc carrying the wild type or mutant C/EBPβ binding site, 0, 0.15, 0.3, 0.6 μg of CA-ERK2 expression plasmid, and the internal control phRLTK, then luciferase activities were measured 24 h after transfection. All luciferase activities are shown relative to the vector control (relative value = 1). In A, C, and D, values are the mean ± SD for three independent experiments. *p<0.05 compared to wild-type cells.

Fig 4. U0126 reduces LPS-induced nuclear NF-κB and C/EBPβ binding to the G-CSF promoter in Raw264.7 macrophages.

Raw264.7 cells were incubated with DMSO or U0126 (10 μM) for 30 min, then with LPS (100 ng/ml) or PBS for 6 h, then the following tests were carried out. (A) Nuclear levels of p50, p65, C/EBPβ, and lamin B were measured by Western blotting assay. (B) Bindings of p50, p65, or C/EBPβ to the G-CSF promoter were analyzed by ChIP assay via precipitation with antibodies against the test protein or histone H3, used as the control, removal of the antibodies with proteinase K digestion for 12 h at 45ºC, and PCR amplification of a 179-bp G-CSF promoter fragment (-248 to -70 bp) or a 274 bp TNF-α promoter fragment (-270 to -4 bp). Ten per cent of the chromatin DNA used for immunoprecipitation was subjected to PCR and is indicated as ‘input’ (bottom row). In A and B, the data are representative of the results from three independent experiments.

Fig 5. U0126 decreases the LPS-induced increased accessibility of the G-CSF promoter to DNase I.

Raw264.7 macrophages were pretreated with DMSO or U0126 (10 μM), then incubated with LPS (100 ng/ml) or PBS for 4 h. Nuclei were then isolated and subjected to DNase I digestion for 2 min, then genomic DNA was isolated and quantitative real-time PCR was performed to measure the amount of undigested DNA in region -167/+12 of the G-CSF promoter (A) and in region -270/-4 of the TNF-α promoter (B), which is expressed as a percentage of that in the control cells. The results are the mean ± SD for three independent experiments. *p<0.05 compared to the control.

Fig 6. Knockdown of ERK2, but not ERK1, decreases LPS-induced DNase I accessibility of the G-CSF promoter.

THP-1 cells in which either ERK1 or ERK2 was knocked down using shRNA were induced to differentiate into macrophages. After incubation with DMSO or U0126 (10 μM), the cells were incubated with LPS (100 ng/ml) or PBS for 4 h, then nuclei were purified and subjected to DNase I digestion for 5 min. Genomic DNA was then isolated and the amount of undigested DNA within region -624/-450 of the G-CSF promoter (A) or within region -228/-71 of the TNF- α promoter (B) was determined by quantitative real-time PCR and expressed as a percentage of that in cells not treated with LPS and U0126. The results are the mean ± SD for three independent experiments. *p<0.05 compared to the corresponding control.