Objective To investigate the preventive effects of emilia sonchifolia on experimental hepatic steatosis in rats and its molecular mechanism.Methods Seventy Sprague-Dawley (SD) rats were randomly divided into five groups: normal control, model, high dose emilia sonchifolia, low dose emilia sonchifolia groups and high dose emilia sonchifolia + phosphorylated extracellular signal regulated protein kinase 1/2 (pERK1/2) inhibitor (PD98059) group (PD group). In normal control group, the rats were fed with normal diet, and in the other four groups, the rats were fed with high fat and low protein diet combined with 30% carbon tetrachloride (CCl4) peanut oil 2 mL/kg subcutaneous injection, once every 3 days for consecutive 3 weeks to establish animal models with hepatic steatosis. In emilia sonchifolia high and low dose groups, 5.0 g/kg and 2.5 g/kg doses of emilia sonchifolia were given respectively by gavage, once a day. In PD group, after administration of emilia sonchifolia high dose by gavage once a day, additionally PD98059 0.3 mg/kg was injected through a tail vein, once a week. After 3 weeks, all rats were switched to normal diet and treatment continued as before. At the end of the 5th week, liver tissues were taken for pathological analyses. The serum levels of alanine transaminase (ALT), aspartate transaminase (AST), total cholesterol (TC), and triglyceride (TG) were determinated by automatic biochenical analyzer. The positive cell count and protein expressions of sterol-regulatory element binding protein 1 (SREBP-1), pERK1/2, toll like receptor 4 (TLR4) and high mobility group box-1 protein (HMGB1) were tested by immunohistochemistry, Western Blot and flow cytometry. The levels of superoxide dismutase (SOD) and malonaldehyde (MDA) in liver cell homogenate were detected by hydroxylamine and TBA method.Results Compared with the model group, the lobular inflammation in high and low dose emilia sonchifolia groups and PD group was attenuated (1.50±0.53, 1.80±0.43, 1.20±0.42 vs. 2.30±0.48), and ALT, AST, TC, TG, SREBP-1, and MDA were significantly decreased, the decrease in high dose emilia sonchifolia group being the most significant [ALT (U/L): 51.91±6.95 vs. 66.50±12.15, AST (U/L): 125.70±5.62 vs. 147.10±10.52, TC (mmol/L): 1.79±1.04 vs. 2.81±1.08, TG (mmol/L): 0.87±0.55 vs. 1.17±0.67, SREBP-1: (30.60±5.56)% vs. (53.10±5.02)%, MDA (nmol/mg): 5.20±0.87 vs. 10.61±5.45,P < 0.05 orP < 0.01]; the relative expression levels of pERK1/2, TLR4, and HMGB1 showed no statistically significant differences between each treated group and the model group [pERK1/2: (43.77±4.93)% vs. (46.83±5.27)%, TLR4 (rmfi): 69.12±24.64 vs. 69.08±24.32, HMGB1 (rmfi): 22.93±14.88 vs. 33.17±13.29, allP > 0.05]. While the above index values in PD group were close to those in high dose emilia sonchifolia group, showing that PD98059 had no impact on emilia sonchifolia's action.Conclusions Emilia sonchifolia can alleviate hepatic injury and attenuate lobular inflammation in rat experimental hepatic steatosis. Its mechanism is possibly related to the reduction of oxidative stress reaction, and SREBP-1 may be as a mediator involved in the action.