Research ArticleMetabolism

TRIF-dependent Toll-like receptor signaling suppresses Scd1 transcription in hepatocytes and prevents diet-induced hepatic steatosis

See allHide authors and affiliations

Science Signaling  08 Aug 2017:
Vol. 10, Issue 491, eaal3336
DOI: 10.1126/scisignal.aal3336
  • Fig. 1 The aggravation of diet-induced hepatic steatosis in Trif−/− mice was mediated by nonmyeloid cells.

    (A to D) Six-week-old wild-type (WT) and Trif−/− mice were fed an NCD or HFD (60% fat kcal) for 10 weeks. (E to H) Irradiated eight-week-old WT and Trif−/− mice received bone marrow (BM) transplantation and were fed an HFD for 10 weeks as in (A) to (D). Livers were subjected to H&E (A and E) and Oil Red O staining (B and F). Scale bars, 100 μm (A and E) and 20 μm (B and F). Hepatic total triglyceride (TG) and total cholesterol (TC) content (C and G) and serum AST and ALT activities were measured (D and H). For (A) to (D), n = 8 to 10 mice per genotype and diet; for (E) to (H), n = 4 to 6 mice per genotype and treatment. *P < 0.05 and **P < 0.01. Representative images are shown.

  • Fig. 2 Activation of TRIF by poly(I:C) inhibited palmitic acid–induced hepatic SCD1 expression and lipid accumulation.

    (A and B) Six-week-old WT and Trif−/− mice were fed an NCD or HFD for 10 weeks. (A) Hepatic Chop, Acc1, Fasn, Scd1, and Dgat2 expression was examined by real-time polymerase chain reaction (PCR) and normalized to Actb. (B) Western immunoblotting was performed to detect hepatic SCD1 and β-actin proteins. (C to F) Primary hepatocytes from WT and Trif−/− mice were treated with or without palmitic acid (PA; 250 μM) (ctrl, control) and poly(I:C) (2.5 μg/ml) for (C) 6 hours or (D to F) 8 hours. Scd1 mRNA (C) and SCD1 protein abundance (D) were measured. Oil Red O staining (E) and quantification of hepatic TG and TC content (F) were performed. Scale bar, 40 μm. Representative images are shown. (G) The concentrations of serum RNA from WT mice fed an NCD or HFD for 10 weeks were measured. (H and I) WT and Trif−/− hepatocytes were treated with PA and RNA (2.4 μg/ml) isolated from adipose tissue and complexed with Lipofectamine as in (D). (H) SCD1 protein abundance and (I) intracellular TG content were measured. For (A), (B), and (G), n = 4 to 7 mice per genotype or diet; for (C) to (F) and (H) and (I), n = 3 independent experiments. *P < 0.05 and **P < 0.01.

  • Fig. 3 Ad-TRIF reversed the increased SCD1 and hepatic steatosis in Trif−/− mice.

    (A to C) WT and Trif−/− hepatocytes infected with adenovirus encoding GFP (Ad-GFP) or TRIF (Ad-TRIF) were treated with or without PA (250 μM) and poly(I:C) (2.5 μg/ml) for 8 hours. GFP was detected in cells transfected with Ad-GFP and Ad-TRIF (A). Scale bar, 40 μm. SCD1 protein (B) and intracellular TG content (C) were determined. (D to G) Six-week-old WT and Trif−/− mice were fed an HFD for 4 weeks, injected once with 109 viral particles (vp) per mouse of Ad-GFP or Ad-TRIF, and continued on the HFD for two more weeks. Livers were analyzed for GFP fluorescence (D), Scd1 mRNA expression (E), SCD1 protein abundance (F), and TG content (G). Scale bar, 20 μm. For (A) to (C), n = 3 independent experiments; for (D) to (G), n = 4 to 6 mice per genotype and/or infection condition. *P < 0.05 and **P < 0.01. Representative images are shown.

  • Fig. 4 Silencing SCD1 abolished lipid accumulation in Trif−/− hepatocytes.

    (A to C) Isolated WT and Trif−/− hepatocytes infected with lentivirus encoding either SCD1 siRNA (Lv-SCD1-siRNA) or scrambled RNA (Lv-Scr-RNA) were treated with or without PA (250 μM) and poly(I:C) (2.5 μg/ml) for 8 hours. Scd1 mRNA expression (A) and TG content (B) were measured, and Oil Red O staining (C) was performed. Scale bar, 40 μm. (D to F) Six-week-old Trif−/− mice were fed an HFD for 10 days, injected with either Lv-SCD1-siRNA or Lv-Scr-RNA, and continued on the HFD for another 3 weeks. Hepatic Scd1 mRNA (D), SCD1 protein (E), and TG content (F) were determined. A group of age-matched Trif−/− mice fed an NCD served as controls. n = 4 independent experiments for (A) to (C); n = 6 mice per genotype, diet, and/or injection condition for (D) to (F). *P < 0.05 and **P < 0.01. Representative images are shown.

  • Fig. 5 Suppressive action on the Scd1 promoter by TRIF was mediated by IRF3.

    (A) Six-week-old WT and Trif−/− mice fed an HFD for 10 weeks were hydrodynamically injected with the luciferase-expressing plasmid pGL3/–1537+155mSCD1. Luciferase activity was analyzed 1 hour after administration of d-luciferin. (B) HeLa cells expressing IRF3(WT) or IRF3(5D) were immunoblotted for IRF3 and SCD1 with β-tubulin as loading control. (C) β-Galactosidase activity was measured in HeLa cells expressing WT pcDNA3.1/–469+229mSCD1 and WT or mutated pcDNA3.1/–816+229mSCD1 that were treated with or without PA (250 μM) and poly(I:C) (2.5 μg/ml) for 8 hours. (D) HeLa cells transfected with WT or mutated pcDNA3.1/–816+229mSCD1 were treated with PA (250 μM) and poly(I:C) (2.5 μg/ml) for 2 hours. Chromatin immunoprecipitation with IRF3 antibody and PCR amplification using specific primers against the Scd1 promoter region were performed. IgG, immunoglobulin G. For (A), n = 4 mice per genotype; for (B) to (D), n = 3 to 4 independent experiments. *P < 0.05 and **P < 0.01. Representative images are shown.

  • Fig. 6 IRF3-mediated SCD1 suppression was validated in human hepatic cells.

    (A) β-Galactosidase activity was measured in HepG2 cells transfected with WT or mutated pcDNA3.1/–816+229mSCD1 followed by the treatment of PA and poly(I:C). (B) HepG2 cells expressing WT pcDNA3.1/–816+229mSCD1 and vector or IRF3(5D) plasmids were treated with or without PA (250 μM) and poly(I:C) (2.5 μg/ml) for 8 hours, and reporter activity was measured. (C and D) HepG2 cells were treated as in (B), and SCD1 protein with β-tubulin as loading control (C) and TG content (D) were measured. (E and F) HepG2 cells transfected with scrambled RNA or IRF3 siRNA were treated with or without PA and poly(I:C) for 8 hours. IRF3 and SCD1 protein abundance (E) and TG content (F) were measured. n = 3 to 4 independent experiments for (A) to (F). *P < 0.05 and **P < 0.01. Representative images are shown.

  • Fig. 7 Poly(I:C) ameliorated HFD-induced lipid accumulation.

    (A to D) Six-week-old WT and Trif−/− mice were injected with phosphate-buffered saline (PBS) or poly(I:C) (5 μg/g) intraperitoneally three times a week while being fed an HFD for 4 weeks. Hepatic Scd1 expression (A), SCD1 protein abundance (B), TG and TC content (C), and serum AST and ALT activities (D) were measured. n = 8 to 10 mice per genotype and treatment condition for (A) to (D). *P < 0.05 and **P < 0.01. (E) Working model. IRF3 in response to TLR/TRIF signaling serves as a transcriptional suppressor of Scd1 in hepatocytes, resulting in the suppression of lipogenesis.

Supplementary Materials

  • www.sciencesignaling.org/cgi/content/full/10/491/eaal3336/DC1

    Fig. S1. Trif−/− mice have similar body weight and fat mass as wild-type mice but increased fasting blood glucose after HFD feeding.

    Fig. S2. Hepatic inflammation, serum insulin concentrations, and serum lipids were similar between HFD-fed wild-type and Trif−/− mice.

    Fig. S3. HFD did not increase CHOP abundance or apoptosis in wild-type and Trif−/− mice.

    Fig. S4. Lipogenic gene expression did not increase in wild-type and Trif−/− mice fed an NCD.

    Fig. S5. Poly(I:C)–stimulated phosphorylation of IRF3 and suppressed palmitic acid–induced Scd1 expression in hepatocytes in a TLR3-dependent manner.

    Fig. S6. RNA was detected in the serum of HFD-fed mice and taken up by hepatocytes.

    Fig. S7. Lv-SCD1-siRNA decreased SCD1 abundance.

    Fig. S8. Activation of TRIF by poly(I:C) decreased fasting blood glucose concentrations in HFD-fed wild-type mice without significantly affecting body weight, fat mass, or hepatic inflammation.

    Table S1. IRF3 binding motifs in the Scd1 promoter in different species.

    Table S2. Primer sequences for specific genes.

  • Supplementary Materials for:

    TRIF-dependent Toll-like receptor signaling suppresses Scd1 transcription in hepatocytes and prevents diet-induced hepatic steatosis

    Jing Chen, Jin Li, Jensen H. C. Yiu, Jenny K. W. Lam, Chi-Ming Wong, Bernhard Dorweiler, Aimin Xu,* Connie W. Woo*

    *Corresponding author. Email: cwhwoo{at}hku.hk (C.W.W.); amxu{at}hku.hk (A.X.)

    This PDF file includes:

    • Fig. S1. Trif−/− mice have similar body weight and fat mass as wild-type mice but increased fasting blood glucose after HFD feeding.
    • Fig. S2. Hepatic inflammation, serum insulin concentrations, and serum lipids were similar between HFD-fed wild-type and Trif−/− mice.
    • Fig. S3. HFD did not increase CHOP abundance or apoptosis in wild-type and Trif−/− mice.
    • Fig. S4. Lipogenic gene expression did not increase in wild-type and Trif−/− mice fed an NCD.
    • Fig. S5. Poly(I:C)–stimulated phosphorylation of IRF3 and suppressed palmitic acid–induced Scd1 expression in hepatocytes in a TLR3-dependent manner.
    • Fig. S6. RNA was detected in the serum of HFD-fed mice and taken up by hepatocytes.
    • Fig. S7. Lv-SCD1-siRNA decreased SCD1 abundance.
    • Fig. S8. Activation of TRIF by poly(I:C) decreased fasting blood glucose concentrations in HFD-fed wild-type mice without significantly affecting body weight, fat mass, or hepatic inflammation.
    • Table S1. IRF3 binding motifs in the Scd1 promoter in different species.
    • Table S2. Primer sequences for specific genes.

    [Download PDF]

    Technical Details

    Format: Adobe Acrobat PDF

    Size: 926 KB


    Citation: J. Chen, J. Li, J. H. C. Yiu, J. K. W. Lam, C.-M. Wong, B. Dorweiler, A. Xu, C. W. Woo, TRIF-dependent Toll-like receptor signaling suppresses Scd1 transcription in hepatocytes and prevents diet-induced hepatic steatosis. Sci. Signal. 10, eaal3336 (2017).

    © 2017 American Association for the Advancement of Science

Stay Connected to Science Signaling

Navigate This Article