Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 29 (7), 1025-36

The Tpl2 Kinase Regulates the COX-2/Prostaglandin E2 Axis in Adipocytes in Inflammatory Conditions

Affiliations

The Tpl2 Kinase Regulates the COX-2/Prostaglandin E2 Axis in Adipocytes in Inflammatory Conditions

Flavien Berthou et al. Mol Endocrinol.

Abstract

Bioactive lipid mediators such as prostaglandin E2 (PGE2) have emerged as potent regulator of obese adipocyte inflammation and functions. PGE2 is produced by cyclooxygenases (COXs) from arachidonic acid, but inflammatory signaling pathways controlling COX-2 expression and PGE2 production in adipocytes remain ill-defined. Here, we demonstrated that the MAP kinase kinase kinase tumor progression locus 2 (Tpl2) controls COX-2 expression and PGE2 secretion in adipocytes in response to different inflammatory mediators. We found that pharmacological- or small interfering RNA-mediated Tpl2 inhibition in 3T3-L1 adipocytes decreased by 50% COX-2 induction in response to IL-1β, TNF-α, or a mix of the 2 cytokines. PGE2 secretion induced by the cytokine mix was also markedly blunted. At the molecular level, nuclear factor κB was required for Tpl2-induced COX-2 expression in response to IL-1β but was inhibitory for the TNF-α or cytokine mix response. In a coculture between adipocytes and macrophages, COX-2 was mainly increased in adipocytes and pharmacological inhibition of Tpl2 or its silencing in adipocytes markedly reduced COX-2 expression and PGE2 secretion. Further, Tpl2 inhibition in adipocytes reduces by 60% COX-2 expression induced by a conditioned medium from lipopolysaccharide (LPS)-treated macrophages. Importantly, LPS was less efficient to induce COX-2 mRNA in adipose tissue explants of Tpl2 null mice compared with wild-type and Tpl2 null mice displayed low COX-2 mRNA induction in adipose tissue in response to LPS injection. Collectively, these data established that activation of Tpl2 by inflammatory stimuli in adipocytes and adipose tissue contributes to increase COX-2 expression and production of PGE2 that could participate in the modulation of adipose tissue inflammation during obesity.

Figures

Figure 1.
Figure 1.. Tpl2 inhibition in 3T3-L1 adipocytes decreases the induction of COX-2 and the secretion of PGE2 in response to IL-1β, TNF-α, or to a cytokine mix.
A, COX-2 mRNA expression in 3T3-L1 adipocytes treated without (vehicle, DMSO) or with a pharmacological Tpl2-I (10μM) and then stimulated or not with IL-1β or TNF-α alone (10 ng/mL) or a mix of these 2 cytokines (MIX, 10 ng/mL each) for 4 hours (n = 4 experiments). B, Immunoblot analysis and quantification (n = 4 experiments) of COX-2 protein expression with Hsp90 as loading control in 3T3-L1 adipocytes treated as in A and stimulated for 16 hours with the indicated cytokines. C, PGE2 secretion in the cultured medium after 16 hours of incubation with the cytokines (n = 4 experiments). D, COX-2 mRNA expression in 3T3-L1 adipocytes treated with control siRNA (siC) or Tpl2 siRNA (siT) for 48 hours and then stimulated for 4 hours with the indicated cytokines (n = 4 experiments). E, Immunoblot analysis and quantification (n = 4 experiments) of COX-2 expression with Hsp90 as loading control in siRNA-treated 3T3-L1 adipocytes stimulated for 16 hours with indicated cytokines F, PGE2 secretion in the culture medium after 16 hours of treatment with the indicated cytokines (n = 4 experiments). Data in bar graphs represent mean ± SEM with ‡, P < .05; #, P < .01; ##, P < .001 vs effects in control cells treated with vehicle or siCtrl; *, P < .05; **, P < .01; ***, P < .001 vs IL-1β or MIX effects in vehicle- or siCtrl-treated cells.
Figure 2.
Figure 2.. Effect of IL-1β, TNF-α or a cytokine mix on 5-lipooxygenase mRNA expression and on caspase-3 activity in 3T3-L1 adipocytes.
A, Expression of Alox5 mRNA in 3T3-L1 adipocytes treated without (vehicle, DMSO) or with a pharmacological Tpl2-I (10μM) and then stimulated or not with IL-1β or TNF-α alone (10 ng/mL) or a mix of these 2 cytokines (MIX, 10 ng/mL each) for 4 hours (n = 3 experiments). B, Caspase-3/7 activity in 3T3-L1 adipocytes treated for 4 hours with the indicated cytokines at the same concentration as in A. C, Caspase-3/7 activity in 3T3-L1 adipocytes treated for 16 hours with the indicated cytokines at the same concentration as in A. D, COX-2 mRNA expression in 3T3-L1 adipocytes incubated without or with the pan-caspase inhibitor Z-VAD-FMK (100μM) for 1 hour and then stimulated or not with TNF-α (10 ng/mL) or a mix of TNF-α and IL-1β (MIX, 10 ng/mL each) for 4 hours. Bars are mean ± SEM with *, P < .05; **, P < .01; and ***, P < .001 vs Ctrl condition.
Figure 3.
Figure 3.. Tpl2 is involved in NF-κB but not CREB phosphorylation in response to IL-1β or the cytokine mix.
3T3-L1 adipocytes were incubated without or with a Tpl2-I (30μM) (A and B) or treated with control or Tpl2 siRNA (siC and siT, respectively) (C). Then, cells were stimulated or not with IL-1β (10 ng/mL) or IL-1β and TNF-α (MIX, 10 ng/mL each) for 20 minutes. Cell were lysed for Western blot analysis with the indicated antibodies. Hsp90 was used as loading control in B and C. Representative immunoblots are shown.
Figure 4.
Figure 4.. NF-κB is involved in Tpl2-induced COX-2 mRNA expression in response to IL-1β but not to TNF-α or the cytokine mix.
A–C, COX-2 mRNA expression in 3T3-L1 adipocytes treated for 48 hours with control siRNA (siCtrl) or p65 NF-κB siRNA (sip65) and stimulated or not for 4 hours with IL-1β (A), the cytokine mix (B), or TNF-α (C) in the absence or presence of a Tpl2-I (10μM). Data are expressed as a percentage of COX-2 mRNA in siCtrl-treated adipocytes stimulated with IL-1β, TNF-α, or the cytokine mix (n = 4 experiments). The inset in A shows a representative immunoblot of the down-regulation of p65 NF-kβ in sip65-treated adipocytes with Hsp90 as loading control. D, PGE2 secretion in the culture medium of 3T3-L1 adipocytes treated with siCtrl or sip65 for 48 hours before stimulation for 16 hours with the indicated cytokines. Data in bar graphs represent mean ± SEM with *, P < .05; **, P < .01; and ***, P < .001.
Figure 5.
Figure 5.. The pharmacological inhibition of Tpl2 or its silencing in adipocytes prevents the induction of COX-2 and the secretion of PGE2 induced by a coculture between adipocytes and macrophages.
A, COX-2 mRNA expression in 3T3-L1 adipocytes and RAW macrophages cultured separately (A+M) or cocultured in contact system (Coc) with DMSO (vehicle) or with a Tpl2-I (5μM) for 24 hours (n = 4 experiments). B, Immunoblot analysis and quantification (n = 4 experiments) of COX-2 protein expression with Hsp90 as loading control in the same experimental conditions as in A. C, COX-2 mRNA expression (n = 3 experiments) in 3T3-L1 adipocytes (left) or RAW macrophages (right) cocultured in the transwell system or cultured separately without (vehicle, DMSO) or with a Tpl2-I (5μM). D, COX-2 mRNA expression in 3T3-L1 adipocytes treated for 48 hours with control siRNA (Ad siC) or Tpl2 siRNA (Ad siT) and cocultured in the transwell system with RAW macrophages (n = 3 experiments). E, 3T3-L1 adipocytes treated with a control siRNA (Ad siC) or a Tpl2 siRNA (Ad siT) were cultured with RAW macrophages either separately (A+M) or in contact coculture (Coc) for 24 hours. The relative amount of COX-2 mRNA expression is presented (n = 4 experiments). F, Immunoblot and quantification (n = 5 experiments) of COX-2 protein expression with Hsp90 as loading control in the same experimental conditions as in E. G, PGE2 amount in the medium in the same experimental conditions as in E and F. Data from 4–5 independent experiments are presented. Data in bar graphs represent mean ± SEM with *, P < .05; **, P < .01; and ***, P < .001.
Figure 6.
Figure 6.. Tpl2 activity in adipocytes or in adipose tissue is required for LPS-induced COX-2 expression.
A, COX-2 mRNA expression (n = 3 experiments) in 3T3-L1 adipocytes incubated without (vehicle, DMSO) or with a Tpl2-I (5μM) and treated for 24 hours with a CM from RAW macrophages activated with LPS (0.5 ng/mL) or with cultured medium containing the same concentration of LPS (control medium, Ctrl). B, COX-2 mRNA expression (n = 4 experiments) in 3T3-L1 adipocytes treated with control siRNA (Adipo siCtrl) or Tpl2 siRNA (Adipo siTpl2) and incubated with CM or control medium (Ctrl) as described in A. C, COX-2 expression with Hsp90 as loading control in the same experimental conditions as in A. Immunoblots from 2 independent experiments are shown. D, Immunoblot analysis and quantification (n = 3 experiments) of COX-2 protein expression with Hsp90 as loading control in the same experimental conditions as in B. E, COX-2 mRNA expression in adipose tissue explants from WT (n = 4) and Tpl2 KO mice (n = 4) incubated with or without LPS (100 ng/mL) for 24 hours. F, COX-2 mRNA expression in epididymal adipose tissue of WT (n = 9) and Tpl2 KO mice (n = 11) ip injected for 5 hours with LPS (2 μg/g of body weight) or NaCl 0.9% as control. Data in bar graphs represent mean ± SEM with *, P < .05 and **, P < .01.

Similar articles

See all similar articles

Cited by 3 PubMed Central articles

Publication types

MeSH terms

Grant support

This work was supported by Inserm, the Université de Nice Sophia Antipolis, and Agence Nationale de la Recherche Grant 2010-BLAN-1117–01 (to J.-F.T.); LABEX SIGNALIFE Grant ANR-11-LABX-0028–01 and an Alfediam-Abbott grant (J.-F.T.). F.C. is supported by an Inserm/Région Provence Alpes-Cote d'Azur/FEDER doctoral fellowship and by a grant from the Société Francophone du Diabète. F.M. is supported by a postdoctoral fellowship from the Institut Thématique Multi-Organismes Cancer.
Feedback