Experimental Diets
Experimental diets with five glutathione supplementation levels (0, 100, 200, 400, and 800 mg kg-1) were prepared (Table 1). According to spectrophotometric analysis, dietary glutathione concentration were 0.25 (G1), 98.35 (G2), 201.36 (G3), 398.66 (G4), and 796.98 (G5) mg kg-1 diets, respectively. The key protein sources in the experimental diets were casein and gelatine, with fish oil as a lipid source, and dextrin and starch as carbohydrate sources. The ingredients were finely ground before mixing (< 250 µm) and then blended with vitamins and minerals. All the ingredients were thoroughly mixed for 15 min and mixed again for 10 min after the addition of the lipid source. The resultant dough was made into feed pellets with a diameter of 1.2 mm and dried in a ventilated oven at 45 °C for approximately 12 h. The pellets were then frozen at −20 °C until required. Normal procedures were used to determine the crude lipid, crude protein, moisture, ash, and gross energy of feed (AOAC 2012).
Feeding experiment
The experimental fish were adapted in a laboratory for 14 d before the experimental feed was introduced. The fish were starved for 24 h and pooled at the start of the trial. A total of 4500 fish (initial weight 5.36 ± 0.13 g) were distributed into each of 15 tanks (size: 500 L). The tanks were divided into five treatment groups with three replicates per treatment. The fish in each tank were weighed as a batch. Fish challenged by intraperitoneally injecting with 100 μL of lipopolysaccharide (3.00 mg kg-1 of body weight; Sigma, USA) diluted in sterile phosphate buffer (PBS) on day 49 (Haukenes and Barton 2004). During the 56 d feeding trial, the fish were manually fed until evident satiation, four times a day (08:00, 11:00, 14:00, and 17:00). The fish were cultured in spring water (flow rate: 1.0 L s-1) and a YSI 6600 v2-2 water quality monitoring system (YSI, Yellow Springs, OH, USA) was used for daily monitoring of water temperature (11.3–15.8 °C), pH (7.2–7.5), dissolved oxygen (7.6–9.6 mg L-1), and ammoniacal nitrogen (< 0.2 mg L-1) in the tanks. After a 24 h starvation period at the end of the feeding trial, the fish from each tank were again weighed as a batch, and the fork length and body weight of a random sample of 30 fish from each tank were determined to calculate the condition factor. The fish were handled in compliance with the Chinese Law on the Protection of Animal Health and the Instructions for the Granting of Permits for Scientific Experimental Animals (Ethics approval number: SCXK (YU) 2005-0001).
Skin mucus sampling
After starvation for 24 h, sixty specimens were randomly selected from each tank. These fish were anesthetized using a 5 mg L-1 clove oil solution before collection of their skin mucus following the method of Ross et al. (2000) The individuals were then moved to a 10 mL sodium chloride polyethylene container (50 mmol). After 2 min, the mucus was collected, placed in a sterile tube (15 mL), and centrifuged for 10 min at 1500 g and 4 ℃. The surfactant was collected and stored at −80 °C until further use.
Skin mucus protein and enzyme activity
For skin mucus from each of the five treatments, the total mucus protein was determined using the method of Lowry et al. (1951). The standard was bovine serum albumin, and the spectrophotometer absorbance was determined at 750 nm. Spectrophotometric kits were bought from the Chinese Nanjing Jiancheng Institute of Bioengineering and were used to analyze the malonaldehyde content (MDA; cat. no. A003-1), and the activity of the enzymes: lysozyme (cat. no. A059-2), alkaline phosphatase (ALP; cat. no. A059-2), glutathione peroxidase (GSH-px; cat. no. A006-2), superoxide dismutase (SOD; cat. no. A001-1), and catalase (CAT; cat. no. A007-1) in the mucus from each treatment.
Skin gene expression
RNAiso Plus Reagent (TaKaRa, Dalian, China) was used to extract total RNA from the skin of three specimens that were randomly selected from each tank according to the manufacturer's instructions. A spectrophotometer was used to examine the absorbance at 260 nm to determine the RNA concentration. Using agarose gel electrophoresis, the RNA integrity was determined, and the absorbance ratio at A260 nm/A280 nm ranged from 1.8 to 2.0. Tumor necrosis factor (TNF-α), interleukin 6 (IL-6), interleukin 8 (IL-8), NF-kappa-B inhibitor alpha-like (IκBα), mucins-2 (MUC-2), claudin-1, zonula occludens-1 (ZO-1), occludin, and β-actin expression levels were determined using quantitative real-time PCR (ABI 7500, USA) with a reaction volume of 20 µL, including 10 µL of 2 SYBR® Premix Ex Taq (TaKaRa, Dalian, China), 0.8 µL for quantitative real-time PCR, specific primers were constructed based on sequences cloned and published in the O. mykiss gene bank (Table 2). 2−ΔΔCt was used to measure the expression values.
Skin mucus antibacterial activity
Two bacterial strains (Yersini ruckeri and Aeromonas hydrophila) were used to test the antimicrobial properties of the fish skin mucus from each treatment. A traditional disc diffusion approach was used for assessing antimicrobial properties (Hellio et al. 2002). In brief, the bacteria were cultivated for 24 h at 37 °C, and then 0.1 mL of each broth culture medium (1.5 × 108 CFU mL-1) was cultured on nutrient agar. Paper discs (6 mm in diameter) were impregnated with 100 μL of the mucus samples, mounted on the medium, and incubated for 24 h at 37 °C. The diameter of the growth inhibition region was measured using a ruler. Antimicrobial activity was indicated by a clear zone around the disks. Serial dilution was used to assess the minimum inhibitory concentration (MIC) of the mucus samples, which was determined using the visual observation method (Hellio et al. 2002).
Calculations and statistical analysis
Weight gain rate (WGR, %) = 100 × (weight gain, g) / (initial weight, g); Survival rate (%) = 100 × (final amount of fish) / (initial amount of fish); Condition factor (CF) = 100 × [(body weight, g) / (total length, cm) 3]; Feed conversion ratio (FCR) = (dry dietary intake, g) / (weight gain, g).
After normality and homogeneity checking, One-way variance analysis (ANOVA) and Duncan multi-multiple-range analyses were used to examine the data. P values < 0.05 were considered significantly different. The SPSS statistical package 23.0 was used for statistical analysis (SPSS Inc., Chicago, IL, USA). The GraphPad Prism software version 9.0 is used to draw column graphs.