Eight-week-old wildtype mice were kept under standard diet (SD) for 4 weeks, or fed Western

Transcription

1 Supplemental information M. Itoh et al. 0 Supplemental Methods Inducible NASH model using wildtype mice Eight-week-old wildtype mice were kept under standard diet (SD) for weeks, or fed Western diet (WD) for weeks, and were received a single injection of CCl (WAKO, Osaka, Japan) intraperitoneally at a dose of 0. ml/kg diluted :0 in olive oil. Then, the mice were kept on SD or WD and were sacrificed at each time point. Induction of hepatocyte death with anti-fas antibody After -week WD feeding, MCR-KO mice received a single injection of anti-fas antibody (Jo, BD Bioscience, San Jose, CA) intravenously at a dose of 0 µg/kg. Then, the mice were kept on WD and were sacrificed at each time point.

2 Supplemental information M. Itoh et al. Supplemental Table. Body and tissue weights in MCR-KO mice reconstituted with bone marrow cells from CCR-KO mice. WT/SD MCR/WD 0 WT-BM WT-BM CCR KO-BM Body weight (g). ± 0.. ± 0.**.0 ± 0. Liver (g). ± 0.0. ± 0.**.0 ± 0. Epididymal fat (g) 0. ± 0.0. ± 0.**. ± 0.0 Blood glucose (mg/dl, ad lib). ± ±.**. ±.0 WT, wildtype mice; MCR, melanocortin receptor-deficient mice; CCR-KO, C-C chemokine receptor -deficient mice; SD, standard diet; WD, Western diet; BM, bone marrow. Data represent mean ± SEM. **P < 0.0 versus WT-BM WT/SD (Tukey-Kramer test). n = -.

3 Supplemental information M. Itoh et al. Supplemental Table. Primers used in this study 0 0 Genes Ccr Clecf Cola Emr Itgam Itgax Lyc Siglec Tgfb Timp Tnfa S Primers Fw: ACAAATCAAAGGAAATGGAAGACAAT Rv: TGCCGTGGATGAACTGAGG Fw: GATGGGACACCATTCAACAATG Rv: CTCTCCGTTCCTATGTCTCCAGTT Fw: CCTCAGGGTATTGCTGGACAAC Rv: ACCACTTGATCCAGAAGGACCTT Fw: CTTTGGCTATGGGCTTCCAGT Rv: GCAAGGAGGACAGAGTTTATCGTG Fw: TTACCTGGGTTATGCTTCTGCAG Rv: AAGCTTTGGACACGGTTCCTC Fw: GCCATTGAGGGCACAGAGA Rv: GAAGCCCTCCTGGGACATCT Fw: GCAGTGCTACGAGTGCTATGG Rv: ACTGACGGGTCTTTAGTTTCCTT Fw: GCAGCCTCTTTCAATGCTAAGG Rv: TGTATTTGACGGTGTGATGACCA Fw: CCTGAGTGGCTGTCTTTTGACG Rv: AGTGAGCGCTGAATCGAAAGC Fw: CATCACGGGCCGCCTA Rv: AAGCTGCAGGCACTGATGTG Fw: ACCCTCACACTCAGATCATCTTC Rv: TGGTGGTTTGCTACGACGT Fw: GTAACCCGTTGAACCCCATT

4 Supplemental information M. Itoh et al. B Rv: CCATCCAATCGGTAGTAGCG Fw: GGCCCTGCACTCTCGCTTTC Rv: TGCCAGGACGCGCTTGT

5 Supplemental information M. Itoh et al. 0 Supplemental Figure. Effect of CCR deficiency on inflammatory changes in the adipose tissue. (A) F/0 immunostaining in the epididymal fat from wildtype mice on standard diet (SD) and MCR-KO mice reconstituted with CCR-KO or wildtype mice-derived bone marrow cells on Western diet (WD) for 0 weeks. Scale bars, 0 µm. (B) The number of F/0-positive cells and crown-like structure (CLS). (C) The insulin tolerance test at weeks of WD feeding. MCR-KO mice were challenged with U/kg insulin after hour fasting, and blood glucose was measured at indicated time point. Data represent mean ± SEM. ** P < 0.0 (Tukey-Kramer test). n = -.

6 Supplemental information M. Itoh et al. Supplemental Figure. Histological analysis and lipid content of the livers from bone marrow-specific CCR-deficient MCR-KO mice. Histological scores for hepatic steatosis, lobular inflammation, ballooning degeneration, NAFLD activity score (NAS) (A), and fibrosis stage (B). (C) Hepatic content of triglyceride and total cholesterol. (D) Representative images of immunostaining for αsma. CV, central veins. Scale bars, 0 µm. Data represent mean ± SEM. ** P < 0.0 (Tukey-Kramer test); n.s., not significant. n = -. 0

7 Supplemental information M. Itoh et al. Supplemental Figure. Dose-dependent effect of CCl on hcls formation and liver fibrosis. (A) Serum alanin aminotransferase (ALT) concentrations day after CCl injection. (B) hcls number evaluated by F/0 immunostaining. (C) Fibrosis area evaluated by Sirius red staining days after CCl injection. Data represent mean ± SEM. * P < 0.0, ** P < 0.0 vs. veh (day ) (-tailed unpaired Student s t test). n =. 0

8 Supplemental information M. Itoh et al. 0 Supplemental Figure. Effect of low-dose CCl on wildtype mice fed standard diet. (A) Experimental protocol of CCl injection to wildtype mice fed SD. Eight-week-old wildtype mice were kept under SD for weeks, and received a single injection of CCl at a dose of 0.ml/kg or olive oil as vehicle (Veh) intraperitoneally. (B) Time course of serum ALT concentrations after CCl administration. Data represent mean ± SEM. ** P < 0.0 vs. veh (day ); ## P < 0.0 (Tukey-Kramer test). Representative images of F/0 immunostaining (C) and Sirius red staining (D). Arrows, swollen hapatocytes. CV, central veins; PV, portal veins. Scale bars, 0 µm. n = -.

9 Supplemental information M. Itoh et al. 0 Supplemental Figure. Effect of low-dose CCl on wildtype mice fed Western diet. (A) Experimental protocol of CCl injection to wildtype mice fed WD. Eight-week-old wildtype mice were fed WD for weeks, and received a single injection of CCl at a dose of 0.ml/kg or olive oil as vehicle (Veh) intraperitoneally. (B) F/0 immunostaining, and (C) Sirius red staining days after CCl injection. Arrows, hcls. CV, central veins. Scale bars, 0 µm. Data represent mean ± SEM. * P < 0.0, ** P < 0.0 (-tailed unpaired Student s t test). n = -.

10 Supplemental information M. Itoh et al. 0 0 Supplemental Figure. Anti-Fas antibody-induced hepatocyte death and hcls formation. (A) Experimental protocol for induction of hepatocyte death by anti-fas antibody. Eight-week-old MCR-KO mice were fed WD for weeks, and received injection of anti-fas antibody at a dose of 0 µg /kg (Fas) or normal saline (Veh) intravenously. n=. (B) Changes in serum ALT levels days after antibody injection. (C) TUNEL staining, (D) F/0 immunostaining, and (E) Sirius red staining days after antibody injection. Arrowheads, TUNEL-positive cells; Arrows, hcls. Scale bars, 0 µm. (F) Hepatic mrna expression of genes related to inflammation (Emr (F/0), Tnfa, and Itgax (CDc)) and fibrogenesis (Tgfb and Cola). Data represent mean ± SEM. * P < 0.0, ** P < 0.0 (-tailed unpaired Student s t test).

11 Supplemental information M. Itoh et al. 0 Supplemental Figure. Role of CDc-positive macrophages in hcls formation in wildtype mice fed WD. Representative images of immunofluorescent staining for Clecf and CD (A), and F/0 and CDc (B) of the livers from wildtype fed WD for months. (C) Experimental protocol for depletion of CDc-positive cells in CDc-DTR bone marrow-chimeric wildtype (CDc-DTR-BM WT) mice fed WD for 0 weeks. (D) Representative images of F/0 immunostaining and the number of hcls. WT/SD, SD-fed WT-BM wildtype mice with DT treatment; Dep (-), WT-BM wildtype mice fed WD for 0 weeks with DT treatment; Dep (+), CDc-DTR-BM wildtype mice fed WD for 0 weeks with DT treatment. Arrows, hcls. scale bars, 0 µm. N.D., not detected. ** P < 0.0 (Tukey-Kramer test). n = -.

12 Supplemental information M. Itoh et al. 0 Supplemental Figure. Expression of CDc in the restored macrophages after diphtheria toxin treatment in the conventional NASH model using MCR-KO mice. Representative images of immunofluorescent staining for F/0 and CDc of the livers from CDc-DTR-BM MCR-KO mice with or without macrophage depletion ( and 0 days after DT treatment). Dep (-), WT-BM MCR-KO mice fed WD for 0 weeks with DT treatment; Dep (+)_day or 0, CDc-DTR-BM MCR-KO mice fed WD for 0 weeks at each time point after DT treatment. scale bars, 0 µm.

13 Supplemental information M. Itoh et al. 0 Supplemental Figure. Potential role of CDc-positive resident macrophages in hepatocyte death-triggered liver fibrosis in NASH. During the development of NASH, resident (CD + F/0 hi ) macrophages aggregate arround dying or dead hepatocytes to constitute hcls, where CDc-positive resident macrophages promote liver fibrosis. Our data suggests that these macrophages in hcls become CDc-positive in response to hepatocyte death, with unique polarization profiles. Therefore, CDc-positive resident macrophages in hcls would be a novel macrophage subset that drives hepatocyte death-induced liver fibrosis.