The Oct DNA motif participates in the alcohol inhibition of the inducible nitric oxide synthase gene promoter in rat C6 glioma cells

Abstract

Induction of nitric oxide synthase-2 (iNOS) by cytokines and bacterial products is associated with protein binding at the proximal promoter and in an upstream enhancer region of the Nos2 gene. To clarify how ethanol suppresses rat iNOS activity, we constructed several deletion mutants of the Nos2 promoter fused to the luciferase gene and transfected the constructs into C6 glial cells. Acute ethanol exposure of stably transfected cells for 24 h inhibits induced activity of Nos2 promoter constructs containing deletions in the 5' flanking region, including a 94 bp promoter that lacks any known NF-kappaB site but which carries a C/EBPbeta and overlapping gamma-IRE, GAS and Oct motifs. Ethanol failed to inhibit the endogenous activity of a smaller, 78 bp promoter that lacks the C/EBPbeta and overlapping, gamma-IRE and GAS motifs and showed no inducible activity. As another approach, in vivo DNA footprinting was used and identified protein protections at five regions of the proximal Nos2 promoter in induced cells. Exposure to acute ethanol diminished protein occupation in the five promoter regions including the gamma-IRE/NF-kappaB and the overlapping gamma-IRE/GAS/Oct sites. Site-directed mutagenesis in the octamer domain of the gamma-IRE/GAS/Oct motifs was studied in a 1002 bp promoter to examine its role in ethanol inhibition of cytokine and lipopolysaccharide induced activity. The data indicate that ethanol failed to inhibit promoter activity when the Oct motif is missing. Electrophoretic mobility shift assays performed using a 22-mer probe containing the overlapping gamma-IRE/GAS/Oct sites showed three complexes with one of the complexes being competed by an octamer-1 antibody. These observations demonstrate the role of protein-DNA binding at the core promoter, and the likely involvement of the octamer motif, in ethanol modulation of cytokine and lipopolysaccharide induced iNOS expression.

Figure 1

Schematic showing the rat core promoter-luciferase expression vector with the identified five regions containing the relevant transcription factor binding sites (modified from Sanchez et al., 2003).

Figure 2

Ethanol (150 mM for 24 h) inhibits LPS+IFNγ-stimulated promoter activity including the piNOS-luc-94. Values are mean + SEM for pooled data samples from 5, 4, 5, 2 and 4 experiments (left to right) on at least 2 separate stable transfected C6 cell lines normalized to their own controls. *, P=0.008 versus control (paired t-test); **, P=0.0001 versus control (paired t-test); ***, P<0.0001 versus control (paired t-test).

Figure 3

Deletion of 16 bp from piNOS-luc-94 (−94 to −78 bp) abolished inducible promoter activity and ethanol suppression. Values are mean RLU/μg protein + SEM expressed as percent of the unstimulated no ethanol control. White bars are cells stimulated in the absence of ethanol and black bars represent cells stimulated in the presence of 150 mM ethanol for 24 h. One-way ANOVA indicated no significant differences (P=0.348) between unstimulated and stimulated groups for cells expressing the piNOS-luc-78 construct. *, P<0.05 (paired t-test) versus corresponding no ethanol condition; n=10 ***, P<0.0001 (paired t-test) versus corresponding no ethanol condition except for LPS+IFNγ where P=0.0006; n=6 a, P<0.0001; b. P<0.05 (Bonferroni post-hoc tests) versus corresponding piNOS-luc-94 condition

Figure 4

Effect of octamer site mutation on ethanol suppression of induced Nos2 promoter activity. The Oct mutation drastically reduced LPS+PMA induced promoter activity while enhancing cytokine mediated promoter activity. The Oct mutation also abolished ethanol suppression of induced Nos2 promoter activity. Repeated measures ANOVA indicated significant stimulated activity, with or without 150 mM ethanol, for cells expressing the piNOS-1002luc construct when exposed to cytokine mix (P=0.0276, n=6) or LPS+PMA (P=0.0082, n=4). Cells expressing the piNOS-1002Oct/M2 construct showed significant stimulated activity when exposed to cytokine mix (P=0.0002, n=6), LPS+IFNγ (P=0.0001, n=6), or LPS+PMA (P=0.0288, n=5). *, P<0.05 (paired t-test) versus corresponding no ethanol condition a, P<0.001; b. P<0.01 (Bonferroni post-hoc tests) versus corresponding piNOS-1002luc condition

Figure 5

Protein-DNA interactions using ESR-1 as probe (−86 to −65 bp). Arrows denote location of specific complexes repeatedly observed with C6 cell nuclear extracts. Essentially the same two major and one minor complex are formed when cells are stimulated with cytokine mixture or LPS+IFNγ.

Figure 6

Binding competition of the wild-type ESR-1 probe with indicated molar excess (1x–9x) of unlabeled probes, each containing point mutations within an overlapping binding site (see Table 1). Arrows denote location of specific complexes repeatedly observed with C6 cell nuclear extracts. The fastest migrating complex is absent because the nuclear extracts contained phosphatase inhibitors. Figure is representative of results obtained on 2 or 3 independent nuclear extracts, including some without phosphatase inhibitors. ‘free’ denotes lane contains radioactive γ-IRE/GAS/Oct probe only.

Figure 7

The overlapping wild-type γ-IRE/GAS/Oct sites in the proximal rat Nos2 promoter bind α-Oct-1 and Oct-1 related protein(s) to form 3 complexes. EMSA was performed using the 22-bp γ-IRE/GAS/Oct oligonucleotide probe incubated with nuclear extract from unstimulated C6 cells and cells stimulated with cytokine mixture for 3 h. Arrows denote location of specific complexes repeatedly observed with C6 cell nuclear extracts. 50x Oct-1 oligo denotes competition with a 50-molar excess of an octamer-1 consensus oligonucleotide (see Table 1). Figure is representative of results obtained on 3 or 4 independent nuclear extracts.

Figure 8

In vivo DNA footprinting by ligation-mediated PCR. Results show effect of 150 mM ethanol on sensitivity to DMS methylation of guanine residues in the rat Nos2 core promoter sequence in C6 wild type cells after indicated time of stimulation with LPS+IFNγ. Arrows denote bands that disappear over time following stimulation, indicating in vivo protein protection of guanine residues in the corresponding site on the 5′ flanking region of the endogenous rat core promoter region. Bracket denotes protection of guanine residues immediately upstream of the NF-κB site. Samples were run as side-by-side duplicate lanes on the electrophoresis separating gel.