Green tea extract protects against nonalcoholic steatohepatitis in ob/ob mice by decreasing oxidative and nitrative stress responses induced by proinflammatory enzymes

Abstract

Oxidative and nitrative stress responses resulting from inflammation exacerbate liver injury associated with nonalcoholic steatohepatitis (NASH) by inducing lipid peroxidation and protein nitration. The objective of this study was to investigate whether the anti-inflammatory properties of green tea extract (GTE) would protect against NASH by suppressing oxidative and nitrative damage mediated by proinflammatory enzymes. Obese mice (ob/ob) and their 5-week-old C57BL6 lean littermates were fed 0%, 0.5% or 1% GTE for 6 weeks (n=12-13 mice/group). In obese mice, hepatic lipid accumulation, inflammatory infiltrates and serum alanine aminotransferase activity were markedly increased, whereas these markers of hepatic steatosis, inflammation and injury were significantly reduced among obese mice fed GTE. GTE also normalized hepatic 4-hydroxynonenal and 3-nitro-tyrosine (N-Tyr) concentrations to those observed in lean controls. These oxidative and nitrative damage markers were correlated with alanine aminotransferase (P<.05; r=0.410-0.471). Improvements in oxidative and nitrative damage by GTE were also associated with lower hepatic nicotinamide adenine dinucleotide phosphate oxidase activity. Likewise, GTE reduced protein expression levels of hepatic myeloperoxidase and inducible nitric oxide synthase and decreased the concentrations of nitric oxide metabolites. Correlative relationships between nicotinamide adenine dinucleotide phosphate oxidase and hepatic 4-hydroxynonenal (r=0.364) as well as nitric oxide metabolites and N-Tyr (r=0.598) suggest that GTE mitigates lipid peroxidation and protein nitration by suppressing the generation of reactive oxygen and nitrogen species. Further study is warranted to determine whether GTE can be recommended as an effective dietary strategy to reduce the risk of obesity-triggered NASH.

Fig. 1

Representative hematoxylin and eosin-stained liver sections from a lean mouse (A) and ob/ob mice fed no GTE (B), 0.5% GTE (C), or 1% GTE (D) for 6 weeks. Hepatic steatosis and inflammatory cell infiltration were apparent in obese mice fed no GTE compared with lean mice. Obese mice fed GTE at 1% markedly reduced obesity-induced steatosis and inflammation (original magnification 200×).

Fig. 2

(A) Hepatic 4-HNE concentrations in lean and obese (ob/ob) mice fed 0%, 0.5% or 1% GTE for 6 weeks (n=12–13 mice/group; means±S.E.). Liver homogenates were prepared in RIPA buffer, and HNE was measured by ELISA. (B) Hepatic NADPH oxidase activity in lean and obese (ob/ob) mice fed GTE at 0%, 0.5% and 1% for 6 weeks (n=12–13 mice/group; means±S.E.). NADPH oxidase from liver homogenates was evaluated by lucigenin-enhanced chemiluminescence following incubation with lucigenin and NADPH. Means not sharing a common superscript are significantly different, P<.05.

Fig. 3

(A) Hepatic N-Tyr in lean and obese (ob/ob) mice fed 0%, 0.5% and 1% GTE for 6 weeks (n=12–13 mice/group; means±S.E.). N-Tyr was measured by ELISA from liver homogenates prepared in PBS. (B) Hepatic total nitrate/nitrite (NO_x) in lean and obese (ob/ob) mice fed GTE at 0%, 0.5% and 1% for 6 weeks (n=12–13 mice/group; means ±S.E.). NO_x was measured by the Griess reaction from liver homogenates prepared in PBS. Means not sharing a common superscript are significantly different, P<.05.

Fig. 4

Protein expression of hepatic MPO in lean and obese (ob/ob) mice fed GTE at 0%, 0.5% and 1% for 6 weeks (n=12–13 mice/group; means±S.E.). MPO was measured by Western blot analysis from liver homogenates prepared in RIPA buffer. Proteins (50 μg/lane) was electrophoresed through an 8% polyacrylamide gel, transferred onto PVDF membranes and probed with antibody against MPO (1:1000, top) or β-actin (1:5000, bottom). Data are relative density normalized to β-actin expression. Means not sharing a common superscript are significantly different, P<.05.

Fig. 5

Protein expression and localization of hepatic iNOS in lean and obese (ob/ob) mice fed GTE at 0%, 0.5% and 1% for 6 weeks (n=12–13 mice/group; means±S.E.). (A) iNOS was measured by Western blot analysis from liver homogenates prepared in RIPA buffer. Proteins (20 μg/lane) were electrophoresed through a 10% polyacrylamide gel, transferred onto a PVDF membrane and probed with antibodies against iNOS (1:2000, top) or β-actin (1:5000, bottom). Data are relative densities normalized to β-actin expression. Means not sharing a common superscript are significantly different, P<.05. (B). Localization of iNOS was determined by immunohistochemistry using formalin-fixed and paraffin-embedded sections. Liver sections were incubated with antibody against iNOS, followed by HRP-conjugated antirabbit IgG, and visualized with DAB substrate after counterstaining with hematoxylin. Representative images are shown at 400× magnification.