Fatty acid-induced NLRP3-ASC inflammasome activation interferes with insulin signaling

Haitao Wen, Denis Gris, Yu Lei, Sushmita Jha, Lu Zhang, Max Tze-Han Huang, Willie June Brickey, Jenny P-Y Ting, Haitao Wen, Denis Gris, Yu Lei, Sushmita Jha, Lu Zhang, Max Tze-Han Huang, Willie June Brickey, Jenny P-Y Ting

Abstract

High-fat diet (HFD) and inflammation are key contributors to insulin resistance and type 2 diabetes (T2D). Interleukin (IL)-1β plays a role in insulin resistance, yet how IL-1β is induced by the fatty acids in an HFD, and how this alters insulin signaling, is unclear. We show that the saturated fatty acid palmitate, but not unsaturated oleate, induces the activation of the NLRP3-ASC inflammasome, causing caspase-1, IL-1β and IL-18 production. This pathway involves mitochondrial reactive oxygen species and the AMP-activated protein kinase and unc-51-like kinase-1 (ULK1) autophagy signaling cascade. Inflammasome activation in hematopoietic cells impairs insulin signaling in several target tissues to reduce glucose tolerance and insulin sensitivity. Furthermore, IL-1β affects insulin sensitivity through tumor necrosis factor-independent and dependent pathways. These findings provide insights into the association of inflammation, diet and T2D.

Figures

Figure 1
Figure 1
Palmitate activates NLRP3-PYCARD inflammasome. (a-d) Resting or LPS-primed bone marrow-derived macrophages (BMMs) were stimulated with palmitate conjugated with BSA (PA-BSA) or BSA control as indicated. ELISA was performed for IL-1β (b, c), IL-18 (d) and IL-6 (e) in supernatants. (e, f) Resting or LPS-primed BMM generated from WT, Nlrp3−/−, Pycard−/− or Nlrc4−/− mice were stimulated with PA-BSA as indicated. (g) IL-1β ELISA of supernatants from LPS-primed BMM stimulated with PA-BSA in the absence or presence of the pan-caspase inhibitor zVAD (10 μM) or caspase-1 inhibitor zYVAD (10 μM). (h) Resting or LPS-primed BMM generated from WT or Casp1−/− mice were stimulated with PA-BSA, BSA control, ATP (2 mM) or nigericin (5 μM). (i) Resting or LPS-primed BMM generated from WT, Nlrp3−/−, Pycard−/− or Casp1−/− mice were stimulated with PA-BSA as indicated. Values in a-i are expressed as mean ± s.d., and the results are representative of three independent experiments.
Figure 2
Figure 2
Palmitate induces IL-1β and caspase-1 processing, which is dependent on NLRP3 and PYCARD. Resting or LPS-primed BMM generated from WT, Nlrp3−/−, Pycard−/− (a,) or Nlrc4−/− (b) mice were stimulated with PA-BSA (0.5 mM) as indicated. Immunoblotting for caspase-1 and IL-1β were performed in supernatants (Sup) and cell lysates (Lys) as indicated. Data are representative of two independent experiments.
Figure 3
Figure 3
Palmitate-induced inflammasome activation requires ROS. (a) LPS-primed BMMs were stimulated with PA-BSA (0.5 mM) in the absence or presence of ROS inhibitor APDC (50 μM). ROS production was determined by flow cytometry using the fluoroprobe dihydrorhodamine 123 (DHR). (b, c) LPS-primed BMM were stimulated with PA-BSA in the absence or presence of increasing concentrations of APDC as indicated. ELISA for IL-1β (b) and immunoblotting for caspase-1 p10 and IL-1β p17 (c) were performed. (d) Resting or LPS-primed BMM were stimulated with PA-BSA (0.5 mM) in the absence or presence of ROS inhibitors such as APDC (10 or 50 μM) and NAC (5 or 25 mM), or NAPDH oxidase inhibitor DPI (5 or 25 μM). ELISA for IL-1β was performed. In a, one of two independent experiments is shown. In each of b-d, the results are representative of three independent experiments.
Figure 4
Figure 4
Palmitate-induced inflammasome activation involves AMPK. (a) Resting or LPS-primed BMM were untreated or stimulated with PA-BSA (0.5 mM) for 24 hrs. Phospho- (Thr172) and total AMPK α subunit was determined by immunoblotting. (b-e) LPS-primed BMM were stimulated with PA-BSA (0.5 mM) in the absence or presence of AMPK agonist AICAR (100 μM). ROS production was determined using the fluoroprobe dihydrorhodamine 123 (DHR) (b). IL-1β (c) and IL-6 (d) in supernatants were measured by ELISA. Immunoblotting for caspase-1, IL-1β and AMPK signaling molecules were performed (e). (f, g) BMM were transfected with empty vector (EV), or constitutively active (CA) AMPK α1 subunit. Sixteen hours later, the transfection efficiency was determined by flow cytometric analysis of GFP expression in macrophages cotransfected with CA α1 AMPK and pmaxGFP (f). Cells were stimulated with PA-BSA (0.5 mM) for 24 hrs. ELISA was performed for IL-1β in supernatants (g). In each of a, b and f, one of two independent experiments is shown. In each of c-e and g, the results are representative of three independent experiments. * P < 0.05, versus controls.
Figure 5
Figure 5
Palmitate-induced AMPK inactivation leads to defective autophagy and the generation of mitochondrial ROS. (a) BMMs were pretreated with LPS (200 ng/ml) for 3 hrs, followed by PA-BSA (0.5 mM) treatment for 24 hrs in the absence or presence of chloroquine (50 μM). Immunoblotting for AMPK activation (AMPKα pThr172) and LC3B were performed. Densitometric analysis was performed to quantify LC3B-II to actin ratio. (b, c) BMMs were pretreated with LPS for 3 hrs, followed by PA-BSA treatment for 24 hrs in the absence or presence of AICAR (100 μM). Cells were fixed and stained for LC3B (b). Quantitation of autophagosomes was performed by counting LC3B puncta in 100 cells. Cells were fixed and examined by transmission electron microscopy for autophagosmes (c). Quantitation is based on counting autophagosomes in 10 cells per treatment. (d, e) BMMs were stimulated as indicated. Immunoblotting were performed. Densitometric analysis was performed to quantify LC3B-II/actin and ULK1 species (d). Mitochondrial ROS production was determined using the MitoSOX fluorescence indicator (e). In each of a and e, one of three independent experiments is shown. In each of b and d, one of two independent experiments is shown. * P < 0.05, versus controls.
Figure 6
Figure 6
Inflammasome-generated IL-1β inhibits insulin signaling in vitro. (a, b) FL83B mouse liver cells were pretreated with mouse recombinant IL-1β (2 ng/ml) (a) or TNF (2 ng/ml) (b) for 24 hrs, then stimulated with insulin (200 nM) for 10 min. Phospho-Akt (Ser473) was determined by flow cytometry with quantification shown in the graph at left (a). Phospho-Akt (Ser473) was determined by immunoblotting (b). (c) FL83B mouse liver cells were treated with either IL-1β or TNF for 24 hrs. Phospho-IRS1 (Ser307) was determined by immunoblotting. (d-f) FL83B cells (d, e) or WT primary hepatocytes (f) were pretreated with conditioned medium (CM) generated from WT, Pycard−/− (d), Nlrp3−/− or Casp1−/− (e) macrophages for 24 hrs, then stimulated with insulin. (g) FL83B cells were pretreated with WT or Pycard−/− CM in the absence or presence of IL-1R antagonist anakinra (1 μg/ml) for 24 hrs, then stimulated with insulin. (h) FL83B cells were pretreated with WT CM in the absence or presence of either IL-1R antagonist anakinra or neutralizing anti-TNF antibody. Phospho-Akt (Ser473) was analyzed by immunoblotting and quantified by densitometric analysis. The results shown are representative of three independent experiments and are expressed as mean ± s.d..
Figure 7
Figure 7
IL-1β and TNF cooperatively mediate insulin resistance in vivo. (a) Eight-week old male C57BL/6 mice were intraperitoneally injected with saline or mouse recombinant IL-1β (1 μg/kg). Insulin tolerance test (ITT) was performed at 2 hrs after IL-1β injection. (b) ITT in WT mice on regular diet (RD) or high-fat diet (HFD) for 12 weeks, or Il1b−/− mice on HFD for 12 weeks. (c, d) Recombinant mouse IL-1β was administrated into WT and Tnfa−/− mice (n=5 for each group). ITT was performed at 2 hrs after IL-1β injection (c). Serum levels of IL-1β (d, left) and TNF (d, right) at 2 hrs after IL-1β injection were determined by ELISA. One of two independent experiments is shown (mean ± s.d.). * P < 0.05, versus controls.
Figure 8
Figure 8
The NLRP3-PYCARD inflammasome promotes insulin resistance in vivo. (a) Blood glucose and insulin levels were measured in WT (n=6), Nlrp3−/− (n=5) or Pycard−/− (n=6) mice under fasting or re-fed conditions after HFD feeding for 12 weeks. (b-e) Glucose tolerance test (GTT) (b, d) and insulin tolerance test (ITT) (c, e) were performed for WT, Nlrp3−/− (b, c) and Pycard−/− (d, e) mice on HFD for 12 weeks. (f) GTT was conducted on the indicated bone marrow chimeric mice on HFD for 12 weeks. (g, h) Insulin-stimulated phosphorylation of IRβ, IRS1 and Akt (Ser473) in liver tissues of individual WT and Pycard−/− mice (g), and phospho-Akt (Ser473) in liver, white adipose (WAT) and muscle tissues of individual WT and Nlrp3−/− mice (h) on HFD for 12 weeks after insulin (2 IU/kg body weight) infusion. Graphs at right of blots show the quantitation of each molecule. (i, j) Expression of Tnfa and Mcp1 mRNA relative to Actb in liver tissues of Nlrp3−/− mice (i) and Pycard−/− mice (j) on RD or HFD for 12 weeks. One of two independent experiments is shown (mean ± s.d.). * P < 0.05, versus controls.

Source: PubMed

3
Abonner