Tumor necrosis factor-alpha-mediated suppression of adipocyte apolipoprotein E gene transcription: primary role for the nuclear factor (NF)-kappaB pathway and NFkappaB p50

Abstract

The adipose tissue inflammation accompanying obesity has important consequences for adipocyte lipid metabolism, and increased adipose tissue TNFalpha plays an important role for mediating the effect of inflammation on adipocyte function. Recent studies have shown that apolipoprotein E (apoE) is highly expressed in adipose tissue where it plays an important role in modulating adipocyte triglyceride metabolism, triglyceride mass, and adipocyte size. We have previously reported that TNFalpha reduces adipocyte apoE, and the current studies were undertaken to evaluate the molecular mechanism for this regulation. TNFalpha repression of adipocyte apoE gene expression required an intact nuclear factor (NF)-kappaB binding site at -43 in the apoE promoter. Site-directed mutagenesis at this site completely eliminated TNFalpha regulation of an apoE gene reporter. TNFalpha treatment activated binding of NFkappaB p50, isolated from adipocyte nuclei, to the apoE promoter. Two structurally distinct inhibitors of NFkappaB complex activation or translocation abrogated the TNFalpha effect on the apoE gene. Using chromatin immunoprecipitation assays, we demonstrated that treatment of adipocytes with TNFalpha led to increased binding of NFkappaB p50, and decreased binding of p65 and Sp1, to this region of the apoE promoter in living cells. The key role played by increased p50 binding was confirmed by p50 knockdown experiments. Reduction of p50 expression using small interference RNA completely eliminated TNFalpha-mediated reduction of endogenous adipocyte apoE gene expression. These results establish the molecular link between adipose tissue inflammation and apoE gene expression in adipocytes. The suppression of adipocyte apoE by the proinflammatory adipose tissue milieu associated with obesity will have important downstream effects on adipocyte triglyceride turnover and content.

Figure 1

Time- and dose-dependent response of adipocyte apoE to TNFα. A, 3T3-L1 adipocytes were treated with TNFα (5 ng/ml) in complete medium for the times shown. Total RNA was extracted and apoE mRNA was quantitated by RT-PCR as described in Materials and Methods. Abundance of apoE mRNA was compared with untreated control cells harvested at the same time. B, 3T3-L1 cells were transfected to express a luciferase reporter gene under the control of the −623/+86 bp of the apoE promoter and the 620-bp enhancer. After 36 h, the cells were treated with TNFα at the doses indicated for 24 h. Relative luciferase activity was quantitated after normalization for β-galactosidase expression as described in Materials and Methods. All points shown are the mean ± sd from triplicate samples.

Figure 2

Evaluation of apoE gene promoter sequences responsive to TNFα in adipocytes. A, The reporter constructs used to evaluate apoE gene elements responsive to TNFα are shown. Base pairs are numbered relative to the transcription start site, and en represents inclusion of the 620-bp ME1. The hatched box represents the first apoE exon. B, 3T3-L1 adipocytes were transfected with the indicated apoE promoter-luciferase constructs. TNFα (5 ng/ml) was added for 24 h in complete growth medium before harvesting cells for measurement of luciferase and β-galactosidase activities. *, P < 0.01 for comparing TNFα-treated to untreated control. Results shown are mean ± sd from triplicate samples. C, Percent change in relative luciferase units (RLU) after treatment with TNFα is shown for each construct after normalization to the untreated control. *, P < 0.01 for comparing TNFα-treated to untreated control. The results shown are mean ± sd from three independent experiments, each performed using triplicate samples.

Figure 3

Importance of −43 to −34 apoE gene promoter sequence for mediating the response to TNFα in adipocytes. A, NFκB binding sites at −43 bp in the 5′-flanking region of the human and mouse apoE genes are compared with the sequences in other NFκB responsive genes. B, The presumptive sequence for NFκB binding site at −43 of the apoE gene was mutated as described in Materials and Methods. Lowercase letters indicate mutated nucleotides. C, pGL300 and the pGL300mut were transfected into 3T3-L1 cells, and cells were treated with 5 ng/ml TNFα for 8 or 16 h. The results shown are the mean ± sd of three independent experiments, each done in triplicate. *, P < 0.001 for comparison with untreated control. RLU, Relative luciferase units.

Figure 4

Involvement of NFκB protein complex for the apoE gene response to TNFα. A, 3T3-L1 adipocytes were treated with TNFα (5 ng/ml) for 24 h, and then 10 μg nuclear protein was isolated and tested for binding to a consensus NFκB binding sequence or to the putative NFκB binding site on the apoE gene proximal promoter. The results shown are mean ± sd from three independent experiments, each done in triplicate. B, Binding of nuclear extract to the NFκB consensus sequence was evaluated with no competitor or in the presence of a 10-fold molar excess of the native apoE promoter sequence or of the apoE promoter mutant shown in Fig. 3B. Results shown are from three independent experiments, each done in triplicate. C, 3T3-L1 cells were pretreated with 50 μg/ml SN50 (an inhibitor of NFκB complex translocation to the nucleus) or with 20 nm QNZ (an inhibitor of NFκB complex activation) for 1 h, and then TNFα (5 ng/ml) was added for 8 h to cells previously transfected to express pGL623en. The results shown are mean ± sd from triplicate samples and are representative of three independent experiments, each done in triplicate. D, Representative Western blot for the effect of NFκB pathway inhibitors on TNFα (5 ng/ml for 16 h) suppression of apoE protein. The apoE signal densities corrected for β-actin (mean ± sd) are shown. *, P < 0.01 for comparison with untreated control (A, C, and D) or for comparison with binding in the absence of the competitor oligonucleotide (B). RLU, Relative luciferase units.

Figure 5

ChIP assay. A, The location of the primers used and the target sequence of the apoE promoter are shown. 3T3-L1 cells were treated with or without TNFα (5 ng/ml) for 1 h (B) or times shown (C) in complete medium. Real-time PCR quantitation of fragments precipitated in the ChIP assay using antibodies specific for Sp1, p50, or p65 were analyzed in triplicate as described in Materials and Methods and data expressed as fold change relative to untreated control samples. Comparing treated and untreated samples: *, P < 0.05; **, P < 0.001. Values shown are mean ± sd of three independent experiments each done in triplicate.

Figure 6

Effect of silencing NFκB p50 on TNFα-mediated suppression of the apoE gene in adipocytes. 3T3-L1 cells were transfected with 100 nm p50 siRNA or a nontarget control siRNA for 24 h as described in Materials and Methods. Forty-eight hours after transfection of the siRNA, cells were incubated with or without TNFα (5 ng/ml) for 8 h. Total RNA was extracted for quantitation of apoE mRNA by RT-PCR as described in Materials and Methods. Results shown are mean ± sd from three independent experiments, each done in triplicate. A representative agarose gel is shown. *, P < 0.01 for the effect of TNFα on apoE mRNA level.