To investigate the protective effects of melatonin on carbon tetrachloride (CCl4)-induced hepatic fibrosis in experimental rats. All rats were randomly divided into normal control group, model control group treated with CCl4 for 12 wk, CCl4 + NAC group treated with CCl4 + NAC (100 mg/kg, i.p.) for 12 wk, CCl4 + MEL-1 group treated with CCl4 + melatonin (2.5 mg/kg) for 12 wk, CCl4 + MEL-2 group treated with CCl4 + melatonin (5.0 mg/kg) for 12 wk, and CCl4 + MEL-3 group treated with CCl4 + melatonin (10 mg/kg). Rats in the treatment groups were injected subcutaneously with sterile CCl4 (3 mL/kg, body weight) in a ratio of 2:3 with olive oil twice a week. Rats in normal control group received hypodermic injection of olive oil at the same dose and frequency as those in treatment groups. At the end of experiment, rats in each group were anesthetized and sacrificed. Hematoxylin and eosin (HE) staining and Van Gieson staining were used to examine changes in liver pathology. Serum activities of alanine aminotransferase (ALT), aspartate aminotransferase (AST) and protein concentration were measured with routine laboratory methods using an autoanalyzer. Hydroxyproline (HYP) content in liver and malondialdehyde (MDA) and glutathione peroxidase (GPx) levels in liver homogenates were assayed by spectrophotometry. Serum hyaluronic acid (HA), laminin (LN), and procollagen III N-terminal peptide (PIIINP) were determined by radioimmunoassay. Pathologic grading showed that the fibrogenesis was much less severe in CCl4 + MEL3 group than in model control group (u = 2.172, P < 0.05), indicating that melatonin (10 mg/kg) can significantly ameliorate CCl4-induced hepatic fibrotic changes. The serum levels of ALT and AST were markedly lower in CCl4 + MEL treatment groups (5, 10 mg/kg) than in model control group (ALT: 286.23 +/- 121.91 U/L vs 201.15 +/- 101.16 U/L and 178.67 +/- 103.14 U/L, P = 0.028, P = 0.007; AST: 431.00 +/- 166.35 U/L vs 321.23 +/- 162.48 U/L and 292.42 +/- 126.23 U/L, P = 0.043, P = 0.013). Similarly, the serum laminin (LN) and hyaluronic acid (HA) levels and hydroxyproline (HYP) contents in liver were significantly lower in CCl4 + MEL-3 group (10 mg/kg) than in model control group (LN: 45.89 +/- 11.71 microg/L vs 55.26 +/- 12.30 microg/L, P = 0.012; HA: 135.71 +/- 76.03 microg/L vs 201.10 +/- 68.46 microg/L, P = 0.020; HYP: 0.42 +/- 0.08 mg/g tissue vs 0.51 +/- 0.07 mg/g tissue, P = 0.012). Moreover, treatment with melatonin (5, 10 mg/kg) significantly reduced the MDA content and increased the GPx activity in liver homogenates compared with model control group (MDA: 7.89 +/- 1.49 noml/mg prot vs 6.29 +/- 1.42 noml/mg prot and 6.25 +/- 2.27 noml/mg prot, respectively, P = 0.015, P = 0.015; GPx: 49.13 +/- 8.72 U/mg prot vs 57.38 +/- 7.65 U/mg prot and 61.39 +/- 13.15 U/mg prot, respectively, P = 0.035, P = 0.003). Melatonin can ameliorate CCl4-induced hepatic fibrosis in rats. The protective effect of melatonin on hepatic fibrosis may be related to its antioxidant activities.