Liver kinase B1/AMP‐activated protein kinase‐mediated regulation by gentiopicroside ameliorates P2X7 receptor‐dependent alcoholic hepatosteatosis

Figure S1 Schematic hypothesis for suppression of alcoholic hepatosteatosis by gentiopicroside. Hypothesis 1: gentiopicroside P2X7 receptor‐dependetly regulates lipid accumulation in the livers of acute or chronic‐plus‐binge alcohol fed mice and alcoholic‐induced steatotic hepatocytes. Hypothesis 2: gentiopicroside P2X7 receptor‐dependently inhibits Inflammasome‐mediated release of IL‐1β from macrophages. Hypothesis 3: IL‐1β from activated macrophages promotes lipid accumulation of hepatocytes during alcoholic hepatosteatosis.

Figure S2 Gentiopicroside down‐regulates mRNA expression of P2X7 receptor and Nlrp3 in acute and chronic plus binge alcohol fed‐mice steatotic liver. mRNA expression of P2X7 receptor and Nlrp3 from livers of mice treated as described in Animal experimental procedures (n = 6 in each group). #, P < 0.05, significantly different when compared with normal group; * P < 0.05, significantly different when compared with ethanol‐treated group; one‐way ANOVA followed by Tukey's test. All histograms represent the mean ± SD of five independent assays.

Figure S3 Gentiopicroside regulates lipid accumulation in ethanol‐induced steatotic mouse primary hepatocytes. Mouse primary hepatocytes were incubated with gentiopicroside in the absence or presence of ethanol (50 mM) except untreated cells for 24 h. (A) Lipid deposition in mouse primary hepatocytes was assessed with Oil Red O staining. Images are taken by light microscopy at a magnification of 400×. (B) Immunofluorescence staining of SREBP1. Original images were taken at 200 × magnification. SREBP1 staining (red) and nuclei with DAPI (blue) are shown. Representative images were shown. (F) Relative fluorescence intensity of SREBP1 was analysed with Image Pro‐Plus 6.0. (C) Protein levels of intracellular total‐ and Phosphor‐AMPKα and extracellular procasepae‐1, cleaved caspase‐1, mature IL‐1β were determined using Western Blot analysis. (D, E) Each immunoreactive band was normalized against GAPDH, and relative to normal to control for unwanted sources of variation. (G) IL‐1β protein released to culture medium was determined by elisa assay. The in vitro data are expressed as the mean ± SD of five independent assays. #P < 0.05, significantly different when compared with untreated cells; * P < 0.05, significantly different when compared with ethanol‐treated cells; one‐way ANOVA followed by Tukey's test. Representative images were shown. All histograms represent the mean ± SD.

Figure S4 P2X7 receptor‐NLRP3 inflammasome activation is involved in ethanol‐induced steatosis. HepG2 cells were incubated with A438079 or caspase‐1 inhibitor VI in the absence or presence of ethanol (50 mM) for 24 h. (A) Immunofluorescence staining of SREBP1. Original images were taken at 200 × magnification. SREBP1 staining (red) and nuclei with DAPI (blue) are shown. Representative images were shown. (B) Relative fluorescence intensity of SREBP1 was analysed with Image Pro‐Plus 6.0. (C) Total‐ and Phosphor‐AMPKα and LKB1 protein levels were determined using Western Blot analysis with the whole cell lysates. (D) Each immunoreactive band was normalized against GAPDH, and relative to normal to control for unwanted sources of variation. # P < 0.05, significantly different when compared with untreated cells; * P < 0.05, significantly different when compared with ethanol‐treated cells; one‐way ANOVA followed by Tukey's test. Representative images were shown. All histograms represent the mean ± SD.

Figure S5 IL‐1β from activated macrophages accelerates lipid accumulation in hepatocytes. (A) PMA‐differentiated THP1 macrophages was primed with LPS (1 μg/mL) for 4 h then continuously stimulated ATP (3 mM) for additional 30 min. Intracellular and extracellular pro‐IL‐1β and mature IL‐1β were detected by Western blotting. (B) HepG2 cells were incubated in normal medium containing LPS/ATP or conditioned medium from LPS/ATP‐activated THP‐1 macrophages for 24 h. Lipid deposition in HepG2 cells was assessed with Oil Red O staining (400 × magnification). (C) Total‐ and Phospho‐AMPKα and LKB1 protein levels were determined using Western Blot analysis with the whole HepG2 cells lysates. (D) Each immunoreactive band was normalized against GAPDH, and relative to normal to control for unwanted sources of variation. # P < 0.05, significantly different when compared with untreated normal cells; a P < 0.05, significantly different when compared with cells treated with normal medium containing LPS/ATP; one‐way ANOVA followed by Tukey's test. (E) Immunofluorescence staining of SREBP1. Original images were taken at 200 × magnification. SREBP1 staining (red) and nuclei with DAPI (blue) are shown. Representative images were shown. (F) Relative fluorescence intensity of SREBP1 was analysed with Image Pro‐Plus 6.0. One‐way ANOVA followed by Tukey's test. # P < 0.05, significantly different when compared with untreated normal cells; * P < 0.05, significantly different when compared with cells treated with conditioned medium from LPS/ATP‐activated THP‐1 macrophages. (G) HepG2 cells were incubated with conditioned medium from LPS/ATP‐activated THP‐1 macrophages presence or absence of IL‐1Ra (100 μM) for 24 h. Total‐ and Phospho‐AMPKα and LKB1 protein levels were determined using Western Blot analysis with the whole HepG2 cells lysates. (H) Each immunoreactive band was normalized against GAPDH, and relative to normal to control for unwanted sources of variation. # P < 0.05, significantly different when compared with conditioned medium from untreated THP‐1 macrophages; * P < 0.05, significantly different when compared with cells treated with conditioned medium from LPS/ATP‐activated THP‐1 macrophages; one‐way ANOVA followed by Tukey's test. (I) HepG2 cells were incubated with conditioned medium from LPS/ATP‐activated THP‐1 macrophages containing increasing amounts of neutralizing anti‐human IL‐1β IgG or isotype‐matched IgG for 24 h. (J) Each immunoreactive band was normalized against GAPDH, and relative to normal to control for unwanted sources of variation. #P < 0.05, significantly different when compared with conditioned medium from untreated THP‐1 macrophages; *P < 0.05, significantly different when compared with cells treated with conditioned medium from LPS/ATP‐activated THP‐1 macrophages; one‐way ANOVA followed by Tukey's test. Data are representative of five independent experiments, and all histograms represent the mean ± SD.

Figure S6 Gentiopicroside attenuates alcoholic hepatosteatosis even in absence of macrophages. C57BL/6 mice were pretreated with clodronate liposomes to deplete kupffer cells, then 48 h after injection mice were intragastrically treated with ethanol (5 g kg^{, body weight) or gentiopicroside (40 mg kg, body weight) every 12 h for a total of 3 doses. (A) Depletion efficiency was verified by immunofluorescence staining of F4/80 (original magnification 200×). Red arrows mean F4/80 positive stain cells. Representative images were shown. Serum ALT (B), serum and hepatic TG content (C and D). P values were shown in histogram. Statistics was performed by one‐way ANOVA followed by Tukey's test. #}P < 0.05, significantly different when compared with PBS/liposomes‐treated mice; *P < 0.05, significantly different when compared clodronate/liposomes‐treated mice; a P < 0.05, significantly different when compared with clodronate/liposomes plus gentiopicroside‐treated mice; b P < 0.05, significantly different when compared with PBS/liposomes plus ethanol‐treated mice. Data are representative of five independent experiments and all histograms represent the mean ± SD (n = 6 per group).

Figure S7 Kupffer cell depletion alleviates alcoholic hepatic steatosis. C57BL/6 mice were pretreated with clodronate liposomes to deplete Kupffer cells, then 48 h after injection mice were intragastrically treated with ethanol (5 g kg^{, body weight) or gentiopicroside (40 mg kg, body weight) every 12 h for a total of 3 doses. (A) HE staining (original magnification 200×). (B) Nile red staining (original magnification 200×). (C) Total‐ and phospho‐AMPKα and LKB1 protein levels were determined using Western Blot analysis. (D) Each immunoreactive band was normalized against GAPDH, and relative to normal to control for unwanted sources of variation. P values were shown in histogram. Statistics was performed by one‐way ANOVA followed by Tukey's test. #}P < 0.05, significantly different when compared with PBS/liposomes‐treated mice; b P < 0.05, significantly different when compared with PBS/liposomes plus ethanol‐treated mice; c P < 0.05, significantly different when compared with clodronate/liposomes plus ethanol‐treated mice Data are representative of five independent experiments and all histograms represent the mean ± SD (n = 6 per group).

Table S1 Primers sequences used in Real Time PCR.