Regulation of inflammatory responses by neuregulin-1 in brain ischemia and microglial cells in vitro involves the NF-kappa B pathway

Abstract

Background: We previously demonstrated that neuregulin-1 (NRG-1) was neuroprotective in rats following ischemic stroke. Neuroprotection by NRG-1 was associated with the suppression of pro-inflammatory gene expression in brain tissues. Over-activation of brain microglia can induce pro-inflammatory gene expression by activation of transcriptional regulators following stroke. Here, we examined how NRG-1 transcriptionally regulates inflammatory gene expression by computational bioinformatics and in vitro using microglial cells.

Methods: To identify transcriptional regulators involved in ischemia-induced inflammatory gene expression, rats were sacrificed 24 h after middle cerebral artery occlusion (MCAO) and NRG-1 treatment. Gene expression profiles of brain tissues following ischemia and NRG-1 treatment were examined by microarray technology. The Conserved Transcription Factor-Binding Site Finder (CONFAC) bioinformatics software package was used to predict transcription factors associated with inflammatory genes induced following stroke and suppressed by NRG-1 treatment. NF-kappa B (NF-kB) was identified as a potential transcriptional regulator of NRG-1-suppressed genes following ischemia. The involvement of specific NF-kB subunits in NRG-1-mediated inflammatory responses was examined using N9 microglial cells pre-treated with NRG-1 (100 ng/ml) followed by lipopolysaccharide (LPS; 10 μg/ml) stimulation. The effects of NRG-1 on cytokine production were investigated using Luminex technology. The levels of the p65, p52, and RelB subunits of NF-kB and IkB-α were determined by western blot analysis and ELISA. Phosphorylation of IkB-α was investigated by ELISA.

Results: CONFAC identified 12 statistically over-represented transcription factor-binding sites (TFBS) in our dataset, including NF-kBP65. Using N9 microglial cells, we observed that NRG-1 significantly inhibited LPS-induced TNFα and IL-6 release. LPS increased the phosphorylation and degradation of IkB-α which was blocked by NRG-1. NRG-1 also prevented the nuclear translocation of the NF-kB p65 subunit following LPS administration. However, NRG-1 increased production of the neuroprotective cytokine granulocyte colony-stimulating factor (G-CSF) and the nuclear translocation of the NF-kB p52 subunit, which is associated with the induction of anti-apoptotic and suppression of pro-inflammatory gene expression.

Conclusions: Neuroprotective and anti-inflammatory effects of NRG-1 are associated with the differential regulation of NF-kB signaling pathways in microglia. Taken together, these findings suggest that NRG-1 may be a potential therapeutic treatment for treating stroke and other neuroinflammatory disorders.

Keywords: Bioinformatics; Gene expression; Inflammation; Ischemia; Microarray; Neuregulin; Stroke; Transcription factor-binding site (TFBS).

Fig. 1

Neuregulin-1 administration reduces MCAO/reperfusion-induced brain infarction. Representative TTC-stained coronal brain sections are shown. Rats were administered vehicle (a) or NRG-1 before MCAO (b). The white area indicates damaged neuronal cells (arrows) and red staining indicates normally functioning cells

Fig. 2

Predicted transcription factor-binding site (TFBS) activity for gene promoters using CONFAC analysis. CONFAC compared our gene list to seven random control datasets to identify statistically over-represented TFBS in genes altered by stroke and reversed by NRG-1. CONFAC identified 12 TFBS that were statistically over-represented. Blue bars represent the average number of TFBS/promoter for each transcription factor in our data set. Red bars are the average number of TFBS/promoter for each transcription factor in the control datasets. p < 0.05 for all transcription factors in the graph

Fig. 3

NRG-1 suppresses TNF-α and IL-6 concentrations in LPS-stimulated N9 microglia cells. N9 microglia cells were pre-treated with NRG-1 (100 ng/ml) for 24 h with or without LPS stimulation (10 μg/ml) for the indicated time points. Supernatants were collected, and TNF-α (a) and IL-6 (b) levels were determined by Luminex. Results are expressed as the mean +/− SD. Asterisk denotes a significant difference compared to cells treated with only LPS (p < 0.05)

Fig. 4

NRG-1 suppresses the phosphorylation of IkB-α in LPS-stimulated N9 microglia cells. N9 microglial cells were pre-treated with 100 ng/ml NRG-1 for 24 h followed by the absence or presence of LPS (10 μg/ml) for indicated time points. a, b Cell lysates were taken and were assayed using ELISA. Results were expressed as the mean +/− SD. Asterisk denotes significant difference compared to cells treated with only LPS (p < 0.05)

Fig. 5

NRG-1 inhibits the degradation of IkB-α in LPS-stimulated N9 microglia cells. N9 microglial cells were pre-treated with 100 ng/ml NRG-1 for 24 h followed by the absence or presence of LPS (10 μg/ml) for 1 h (a) and 3 h (b). Whole cell extracts were taken from untreated cells or cells pre-treated NRG-1 alone or the absence or presence of LPS (10 μg/ml) and were assayed using western blot. The band intensity was quantified using studio lite imager and is presented relative to the level of β-actin. Data are presented for three independent experiments. Results were expressed as the mean +/− SD. Asterisk denotes significant difference compared to cells treated with only LPS (p < 0.05)

Fig. 6

NRG-1 reduces nuclear translocation of P65. N9 microglial cells were pre-treated with 100 ng/ml NRG-1 for 24 h followed by the absence or presence of LPS (10 μg/ml) for 3 h. Nuclear (a) and cytoplasmic (b) extracts were assayed using western blot. Nuclear extracts were assayed using ELISA (c). Data are presented from three independent experiments. Results were expressed as the mean +/− SD. Asterisk denotes significant difference compared to cells treated with only LPS (p < 0.05)

Fig. 7

NRG-1 increased nuclear translocation of P52. N9 microglial cells were pre-treated with 100 ng/ml NRG-1 for 24 h followed by the absence or presence of LPS (10 μg/ml) for 3–24 h. Cell lysates were taken and were assayed using ELISA measuring levels of p52 (a) and RelB (b). Results were expressed as the mean +/− SD. Asterisk denotes significant difference compared to cells treated with only LPS (p < 0.05)

Fig. 8

NRG-1 increases G-CSF and IL-9 concentrations in N9 microglia cells. N9 microglia cells were pre-treated with NRG-1 (100 ng/ml) for 24 h with or without LPS stimulation (10 μg/ml) for the indicated time points. Conditioned medium was collected and G-CSF (a) and IL-9 (b) levels were determined by Luminex. Results are expressed as the mean +/− SD. Asterisk denotes a significant difference compared to cells treated with only LPS (p < 0.05)