The degree of liver injury due to alcohol treatment was assessed by measuring the levels of AST, ALT and γ-GT. Silymarin (SML) is a substance extracted from Silybum marianum that contains flavonolignans, which are capable of preventing liver injury caused by chemicals and toxic substances such as carbon tetrachloride and alcohol [26,27]. One hundred mg/kg body weight or higher doses of SML have been widely used as a standard hepatoprotective agent [27,28]. Hence, SML served as a positive control in the present study. As shown in Figure 1A, the ASL and ALT levels were significantly increased in the alcohol-treated group compared with the control group (p < 0.05). SML treatment significantly reduced the AST and ALT levels by 35.6% and 51.6%, respectively. LS treatment lowered the AST and ALT levels by 47.5% and 55.4%, respectively (p < 0.05), whereas LJ treatment lowered the AST and ALT levels by 42.2% and 55.6%, respectively (p < 0.05). γ-GT, which is secreted by the liver and gallbladder, is a common biochemical marker for alcoholic liver injury, as the levels of γ-GT become elevated as the result of such injury. SML treatment lowered γ-GT levels by 61.3%, while LS and LJ treatments also significantly reduced γ-GT levels by 52.7% and 59.8%, respectively (Figure 1B).

Thiobarbituric acid-reactive substances (TBARS) analysis mainly measures the levels of secondary products of lipid peroxidation-malondialdehyde (MDA), which is a common indicator for oxidative stress. MDA was used in the study to investigate the degree of oxidative stress caused by alcohol treatment. As expected, MDA concentrations were significantly higher in the alcohol group as compared with the control group (Figure 2). SML treatment decreased MDA concentrations by 31.4%, while the LS and LJ groups also showed notable reductions in MDA levels of 22.0% and 28.1%, respectively (p < 0.05).

To confirm the above in vitro findings, we investigated the effects of LS and LJ treatments on γ-GT in a rat model of acute alcohol exposure. After 10 days of feeding with Lactobacillus strains, the animals had an average body weight of 274.2 g. No significant difference was found among the groups, which indicated that the heat-killed LS and LJ had no effect on the normal growth of the rats. Moreover, no difference in liver and kidney weights was found among the groups (Table S1). Hence, no obvious toxicities were observed based on the above findings. As shown in Figure 3A, the alcohol group had significantly higher levels of AST and ALT. SML treatment resulted in lower AST and ALT levels. Unlike in the in vitro studies, only LS treatment significantly lowered the AST levels, which it did by 26.2%. Meanwhile, no significant effects on the ALT levels were exerted by either the LS or LJ treatment. On the other hand, the LS and LJ treatments more effectively lowered γ-GT levels (by 36.6% and 40.5%, respectively) than the SML treatment (27.3%; Figure 3B).

MDA levels in the alcohol-treated group were elevated as compared with the control group. The SML, LS and LJ treatments all significantly lowered MDA concentrations (by 36.1%, 44.4% and 41.7%, respectively (Figure 4; p < 0.05).

Excess consumption of alcohol may lead to hypertriglyceridemia and an elevated level of very low density lipoprotein [29]. After feeding with heat-killed LS and LJ for ten consecutive days, followed by alcohol induction (6 g/kg body weight), the SML treatment did not result in reduced TG levels, while the LS treatment significantly lowered TG levels (p < 0.05; Table 1). No difference was observed for high density lipoprotein (HDL)-cholesterol, low density lipoprotein (LDL)-cholesterol or total cholesterol levels among the groups.

Bovine serum albumin (BSA) and SML were purchased from Sigma-Aldrich (St. Louis, MO, USA). MRS broth was purchased from Difco Laboratories (Detroit, MI, USA). Dulbecco’s Modified Eagle Medium (DMEM), trypsin, streptomycin, sodium pyruvate and nonessential amino acids were purchased from GIBCO/BRL (Rockville, MD, USA). Fetal bovine serum (FBS) was purchased from Hyclone (Logan, UT, USA). The above reagents were of reagent grade I or molecular biology grade.

LS and LJ were provided by New Bellus Enterprises Co., Ltd., Tainan, Taiwan. These two strains were incubated for 24 h at 37 °C in 3 mL MRS broth medium containing 0.5% cysteine. The revived culture was then added to 5 mL MRS broth medium and incubated for 24 h at 37 °C. The culture was then enumerated and the suspension adjusted, such that it contained 2 × 10⁹ CFU/mL. One milliliter of the suspension of LS or LJ was autoclaved for 20 min at 121 °C, and samples prepared in this manner served as the experimental samples used in the present study.

HepG2 cell line (human hepatocellular liver carcinoma cell line) was purchased from the Bioresource Collection and Research Center (Hsinchu, Taiwan). HepG2 cells were cultured in DMEM medium containing (10% FBS, 0.37% sodium hydrogen carbonate, 100 U/mL penicillin and 100 U/mL streptomycin) at a density of 1 × 10⁵ cells/mL at 37 °C and 5% CO₂. When the cells grew to 80% confluence, the medium was changed to DMEM medium (FBS free). Fifty microliters heat-killed suspension, SML (30 µg/mL) and 100 mM alcohol were also added. After 24 h, the cells were washed twice with phosphate-buffered saline. The medium and the cells were then used for analysis.

Six-week old male Spraque-Dawley (140–170 g, n = 45) rats were purchased from BioLASCO Taiwan Co., Ltd. (Taipei, Taiwan). The animals were housed individually in hanging wire mesh cages in a temperature (25 ± 2 °C) and humidity (65% ± 5%) controlled room with an alternating 12-h light:dark cycle. Upon arrival, the animals were acclimated for 1 week, during which they were fed a standard rodent diet (Lab 5001; Purina Mills, St. Louis, MO, USA) and water ad libitum. After one week of acclimation, the animals were randomly assigned to five groups of nine rats each. Each group was administered a dose of one of five treatments at respective time points. The groups were: (i) the control group; (ii) the alcohol group; (iii) the SML group; (iv) the LS group; and (v) the LJ group. The animal study was designed according to the methods described by Seth et al. [3]. Alcohol, SML, LS and LJ were mixed with normal saline. During the study, animals in control and alcohol group were orally gavaged with normal saline. The SML, LS and LJ groups were administered SML (200 mg/kg body weight), heat-killed LS (1 × 10¹¹ CFU/kg body weight) and heat-killed LJ (1 × 10¹¹ CFU/kg body weight), respectively. On Day 10, the control group was given an equal volume of normal saline while the other groups were orally gavaged with a single dose of alcohol (1 mL of 6 g/kg body weight as 30% saline solution). The animals were anesthetized with isoflurane. After sacrifice, blood and liver tissue were collected, snap frozen with liquid nitrogen and stored at −80 °C until analysis. The study protocol was approved by the Animal Research Committee at the Hungkuang University (Taichung, Taiwan, approval #98014).

The extent of alcohol-induced damage was assessed by measuring ALT, AST and γ-GT activity. The activities of ALT, AST and γ-GT were measured by respective commercial assay kits according to the manufacturer’s protocol (Randox laboratories Ltd., Antrim, UK; kits cat. No. AL1268, AS1267 and GT523, respectively).

The measurements of lipid peroxidation were performed according to Buege and Aust [42] and Imamoglu et al. [43]. Dibutyl hydroxy-toluene was added to 1 mL of cell supernatant or liver homogenate to the final concentration of 0.5 mM. One milliliter of 2.5% trichloroacetic acid and 1 mL 0.7% thiobarbituric acid were also added and the mixture was incubated for 10 min in a 100 °C water bath. After cooling, 3 mL of butanol was added and the solution was centrifuged at 1000× g for 5 min. One milliliter of solution was analyzed by the spectrophotometer (Excitation: 515 nm; Emission: 555 nm). MDA from the reaction of 1,1,3,3-tetramethoxypropane and 1 N H₂SO₄ was used as standards. The amount of TBARS was expressed as nmol MDA/mg protein.

Serum total cholesterol, HDL-cholesterol and LDL-cholesterol levels were measured with commercial kits according to the manufacturer’s protocol (Randox laboratories Ltd., Antrim, UK; kits cat. No. TR 213 and CH201; Fortress diagnostics limited, Antrim, UK; kits cat. No. BXC0421E and BXC0431E, respectively).

The SOD, catalase and GPx activities were each assayed with a commercial assay kit according to the manufacturer’s protocol (Cayman Chemical Co., Ann Arbor, MI, USA; Item No. 706002, 707002 and 703102, respectively).

The data were expressed as mean ± SD. Experimental data were analyzed using the t-test (Student’s t-test) procedure of SPSS 18. A comparison was made between each group and the alcohol group. In addition, we compared group means by using the one-way factorial analysis of variance (ANOVA) followed by Duncan’s multiple-range test. Differences were considered significant if p < 0.05.