2. Materials and Methods

2.1. Animal Ethics Statement

All experimental procedures involving live fish were carried out in full accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health (NIH Publications, Number 8023, revised in 1978). In addition, all animal experimentation was reviewed and approved by the Shantou University Animal Care and Use Committee (Number 132, revised on 6 July 2022).

2.2. Diets

A set of five diets was formulated to provide ~42% crude protein with fishmeal and soy protein concentrate (SPC) as the predominant sources of protein and around 16% lipid using mixed vegetable oil (palm oil/rapeseed oil/perilla oil, 3:1:1 by vol.) and soybean lecithin as the major added fat sources. On this basis, diets were designed with graded levels of OLE of 0.0%, 0.5%, 1.0%, 1.5%, and 2.0%. The commercial OLE product contained 40.3% oleuropein, 9.3% hydroxytyrosol, 4.2% apigenin, 3.5% luteolin, 1.6% oleanolic acid, 1.1% maslinic acid, 0.5% chlorogenic acid, 0.6% caffeic acid, and around 3.0% ash and 35.0% fibrous organic substances insoluble in water and alcohol and was obtained from Yaan Huijie Biotechnology Co., Ltd. (Yaan city, China). The dietary ingredients and analyzed proximate compositions are shown in Table 1.

Ingredient (g/100 g diet) | Dietary OLEa level (%)
0.0 | 0.5 | 1.0 | 1.5 | 2.0
Fishmeal | 30.0 | 30.0 | 30.0 | 30.0 | 30.0
Soy protein concentrate | 18.0 | 18.0 | 18.0 | 18.0 | 18.0
Corn gluten | 10.0 | 10.0 | 10.0 | 10.0 | 10.0
Fermented soybean meal | 10.0 | 10.0 | 10.0 | 10.0 | 10.0
Cassava starch | 6.0 | 6.0 | 6.0 | 6.0 | 6.0
α‐Starch (corn starch) | 5.0 | 5.0 | 5.0 | 5.0 | 5.0
Mixed vegetable oilb | 10.0 | 10.0 | 10.0 | 10.0 | 10.0
Soybean lecithin | 2.0 | 2.0 | 2.0 | 2.0 | 2.0
Carboxymethyl cellulose | 3.0 | 2.5 | 2.0 | 1.5 | 1.0
Vitamin premixc | 2.0 | 2.0 | 2.0 | 2.0 | 2.0
Mineral premixd | 2.0 | 2.0 | 2.0 | 2.0 | 2.0
Lutein | 0.2 | 0.2 | 0.2 | 0.2 | 0.2
Betaine | 0.5 | 0.5 | 0.5 | 0.5 | 0.5
Choline chloride | 0.5 | 0.5 | 0.5 | 0.5 | 0.5
Calcium dihydrogen phosphate | 0.8 | 0.8 | 0.8 | 0.8 | 0.8
Olive leaf extract (OLE) | 0.0 | 0.5 | 1.0 | 1.5 | 2.0
Total | 100.0 | 100.0 | 100.0 | 100.0 | 100.0
Proximate composition (% dry weight)
Dry matter | 89.87 | 90.21 | 90.04 | 90.02 | 90.27
Organic matter | 79.46 | 79.55 | 79.38 | 80.18 | 80.14
Crude protein | 41.79 | 41.81 | 41.86 | 41.88 | 41.75
Crude fat | 15.34 | 15.33 | 15.93 | 15.49 | 15.62
Crude fiber | 3.07 | 3.18 | 3.06 | 2.93 | 2.81

For the preparation of each diet, the dry ingredients were ground and thoroughly mixed before oil and water were incorporated to form a paste, which was extruded into appropriately sized pellets utilizing a laboratory pelleting machine (SLC‐45, Fishery Machinery and Instrument Research Institute, Shanghai, China). The pellets were air‐dried, sealed in plastic bags, and stored at −20°C until utilization.

2.3. Experimental Fish and Feeding Trial Design

The nutritional trial was conducted from the beginning of September to the end of November 2022 in floating sea cages ~150 m offshore of Nan Ao Marine Biology Station (NAMBS) of Shantou University. Juvenile T. ovatus, sourced from a fish hatchery in Zhangzhou, Fujian Province, were placed in a cuboid floating net cage (6.0 m × 6.0 m × 3.0 m) for 2 weeks prior to the commencement of the feeding trial. During this period, the fish were fed a diet containing a mixture of equal amounts of the five experimental feeds. By 1 week prior to the commencement of the feeding trial, the T. ovatus were acclimated to the experimental conditions such that there were no mortalities, and the fish were eating normally and were in good condition. The T. ovatus were starved for 24 h and anesthetized with 0.01% phenoxyethanol, and 300 healthy fish (average weight, 11.57 ± 0.10 g) were distributed randomly into 15 net sea cages (1.0 m × 1.0 m × 1.5 m) at a stocking density of 20 fish per cage. The experimental diets were assigned at random to three cages, and the fish were fed by hand to satiation twice daily (6:00 and 16:00) for 8 weeks, with the quantity of feed administered to each cage recorded each day. During the trial, the temperature of the seawater varied naturally with season between 20.0 and 29.6°C, while salinity levels ranged from 30 to 32‰, dissolved oxygen levels were in excess of 5.0 mg/L, and ammonia nitrogen levels were maintained below 0.05 mg/L.

2.4. Sample Collection

At the termination of the trial, all fish in each cage were starved for 24 h, anesthetized with 0.01% phenoxyethanol, counted, and weighed individually. Three fish were selected randomly from each cage and frozen at −20°C prior to the analysis of whole‐body composition. Samples of blood from three additional fish per cage were drawn from the caudal vein, and serum was obtained by centrifugation at 1500 × g for 10 min at 4°C for subsequent analysis of biochemical indicators. The viscera collected from the same three fish were weighted for the determination of viscerosomatic indices (VSIs). Then the liver was removed for these viscera and weighted to calculate the hepatosomatic indices (HSIs). Two portions of liver, intestine, and muscle were collected from each tissue sample for biochemical and enzyme activity analyses. Additionally, another sample from each liver and intestine was collected into cryotubes for molecular (gene expression) analyses. All the above tissue samples were frozen immediately in liquid nitrogen and stored at −80°C prior to the analyses. The samples of tissue from the three fish per cage were individually analyzed and then averaged to provide data per cage (n = 3). The liver and anterior intestine samples for histological analysis were collected from another three fish per cage and placed in a 4% solution of paraformaldehyde. In addition, on the day prior to the main sampling and before feed was withdrawn, intestinal contents were collected from four or six fish per cage and pooled in two pools of two or three per cage (n = 6) for microbiota analysis.

2.5. Growth Performance Indices

Growth indices were calculated as follows:

Weight gainWG,%=100×final body weight−initial body weight×initial body weight−1;

Specific growth rateSGR,%/d=100×lnfinal body weight−lninitial body weight×days of feeding trial−1;

Feed conversion ratioFCR=total dry weight of feed fed×final body weight−initial body weight−1;

Survival rateSR,%=100×final fish number×initial fish number−1;

Hepatosomatic indexHIS,%=100×liver weight×fish body weight−1;

Viscerosomatic indexVSI,%=100×visceral weight×fish body weight−1;

Condition factorCF,gcm−3=100×fish body weight,g×fish body length,cm−3.

2.6. Proximate Composition Analysis

Proximate compositions (moisture, protein, lipid, fiber, and ash) of the experimental diets and whole fish body (moisture, protein, lipid, fiber, and ash) were determined following standard methods. Specifically, moisture was assessed by drying the samples at 105°C until constant weight (AOAC, 2006). Protein contents were measured by determining nitrogen (6.25) using a Kjeldahl System (Kjeltec8400; FOSS) according to the Chinese national standard method (GB/T 6432‐2006). Lipid contents were determined using an automatic Soxhlet extraction apparatus (SZF‐06A; Xinjia, Shanghai, China) with petroleum (boiling range 40–60°C) extraction for 4 h at 60–65°C according to the Chinese national standard method (GB/T 6433‐2006). Ash contents were obtained by incinerating the samples in a muffle furnace (CWF1100; Carbolite) at 550°C for 6 h based on the Chinese national standard method (GB/T 6438‐2007). Fiber content was analyzed using a fully automatic fiber analyzer (F2000; Hanon, China) according to the Chinese national standard method (GB/T6434‐2006). Total organic matter contents of feeds were obtained by calculation, subtracting ash contents from dry matter contents.

2.7. Total Lipid and Fatty Acid Composition Analysis

Liver fatty acid compositions were determined using a Shimadzu GC‐17A gas chromatograph (Kyoto, Japan). In brief, total lipids were isolated from the liver tissue using chloroform/methanol (2:1, by volume). Subsequent methylation was conducted with boron trifluoride etherate (concentration, 48%; Acros Organics, NJ, USA) to generate fatty acid methyl esters, following the analytical procedures detailed in a previous study [22]. Likewise, the specific gas chromatography settings and experimental protocols followed those outlined previously [23]. Fatty acids were identified and measured using established commercial reference standards (Sigma–Aldrich, St. Louis, MO, USA) and the Shimadzu CLASS‐GC10 gas chromatography workstation.

2.8. Serum Biochemical Parameters and Enzyme Activities Analysis

Serum triglyceride (TG, A110‐1‐1; 500 nm), total cholesterol (TC, A111‐1‐1; 500 nm), high‐density lipoprotein cholesterol (HDLC, A112‐1‐1; 550 nm), and low‐density lipoprotein cholesterol (LDLC, A113‐1‐1; 550 nm), along with serum antioxidant status parameters including malondialdehyde (MDA, A003‐1‐1; 532 nm), aspartate aminotransferase (AST, C010‐2‐1; 505 nm), alanine aminotransferase (ALT, C009‐2‐1; 505 nm), alkaline phosphatase (AKP, A059‐2; 520 nm), superoxide dismutase (SOD, A001‐3‐2; 450 nm), and total antioxidant capacity (TAOC, A015‐2‐1; 405 nm), were all measured using commercial assay kits (Jiancheng Biotech Co., China) that employed spectrophotometric assays (absorbance wavelengths as indicated). All assays were conducted in detail according to the manufacturer’s instructions, as described previously [20, 24].

2.9. Liver and Intestine Histological Analysis

Samples of both liver and intestine were preserved in a 4% solution of paraformaldehyde for fixation, and the acquisition and scanning of slices were carried out commercially (Wuhan Servicebio Technology Co., Ltd., Wuhan, China). Briefly, liver specimens were frozen and sectioned into 5 μm slices and subsequently stained with oil red O (ORO) following standard procedures as detailed previously [25]. In contrast, intestinal samples underwent dehydration and embedding in paraffin wax, and sections of 5 μm thickness were produced by a paraffin slicing machine (PM‐24, Servicebio, China) before being stained with hematoxylin and eosin (H&E). Histological images obtained were captured and analyzed utilizing CaseViewer software (SWE‐CX63, Servicebio). Lipid droplets stained by ORO were quantified using Image‐Pro Plus Version 6.0 (Media Cybernetics, Rockville, MD, USA).

2.10. Intestinal Enzyme Activities Analysis

The activities of amylase, trypsin, and lipase in the intestine were measured using commercial assay kits according to the manufacturer’s instructions (Nanjing Jiancheng Bioengineering Institute). Briefly, intestinal samples (three fish per replicate cage) were individually weighed and mechanically homogenized at a ratio of 0.1 g tissue to 0.9 mL in phosphate‐buffered saline (PBS) at 4°C, and the homogenates were centrifuged at 3500 × g for 10 min at 4°C to obtain supernatants for the assay. Samples of the supernatant were combined with starch and an iodine solution, and a blue complex was produced after hydrolysis of starch and then measured at 660 nm to determine amylase activity. A red product was generated after reacting supernatants with 6‐methyl resorufin‐labeled lipid and then measured at 580 nm to determine lipase activity. The activity of trypsin was calculated by detecting the change in absorbance at 253 nm after reacting supernatants with the arginine ethyl ester substrate. Protein contents of the supernatants were determined by a commercial kit (BCA Protein Assay Kit, G3522, GBCBIO Technologies Inc.). All enzyme activities were presented as units per milligram of protein (U/mg prot).

2.11. Intestinal Microbiome Analysis

DNA was extracted from the intestinal samples utilizing the MagicPure Stool and Soil Genomic DNA Kit (Tiangen, Beijing, China), following the manufacturer’s prescribed protocol. The quantity and quality of the extracted DNA were assessed using the Qubit dsDNA HS Assay Kit and Qubit 3.0 Fluorometer (Invitrogen, Thermo Fisher Scientific, Oregon, USA). Subsequent to standardization at 1 ng/μL, equal volumes of DNA samples from three fish in each replicate were pooled across treatments, resulting in six composite samples per dietary group. These pooled samples underwent microbial community profiling via high‐throughput sequencing, outsourced to Biomarker Technologies Co., Ltd. (Beijing, China). Once sequences were optimized, they underwent operational taxonomic unit (OTU) clustering analysis and species classification annotation. Analysis of the diversity index was performed based on OTU clustering, and in‐depth examination of species structure and differential species analysis was conducted using the available taxonomic information [26]. All the above bioinformatic analyses were carried out using BMK Cloud (Biomarker Technologies Co., Ltd.).

2.12. RNA Extraction, cDNA Synthesis, and Quantitative Real‐Time PCR

Total RNA from liver and intestine samples was extracted using the Trizol reagent (Vazyme, China) following the manufacturer’s instructions, and the concentration and purity of the extracted RNA were quantified using a NanoDrop One Spectrophotometer (Thermo Fisher Scientific, USA). Only samples exhibiting an A260/A280 ratio between 1.8 and 2.2, with a total yield exceeding 5 μg, were utilized. The integrity of RNA was further validated via 1.0% agarose gel electrophoresis. Subsequently, cDNA synthesis from high‐quality RNA was performed using HiScript II Q Select RT SuperMix for gPCR kits (Vazyme, China), following the manufacturer’s instructions. Transcript levels of genes of lipid metabolism and inflammatory responses were determined by SYBR Green‐based real‐time quantitative PCR (qPCR) using primers, as shown in Table S1. All qPCRs were carried out in a LightCycle 480 Real‐Time PCR System (Roche, Switzerland) using the ChamQ Universal SYBR qPCR Master Mix kit (Vazyme, China). The reaction systems were filled with 3 μL of nuclease‐free water, 5 μL of SYBR Green Supermix, 0.5 μL each of forward and reverse primers, and 1 μL of cDNA. The thermal cycling conditions were initial denaturation at 94°C for 5 min; 45 cycles of denaturation at 95°C for 10 s, annealing at 58°C for 20 s, and extension at 72°C for 20 s, followed by a final dissociation step at 95°C for 5 s, 65°C for 1 min, and cooling at 40°C for 10 s. Detailed operational steps and experimental conditions have been published previously [20]. Relative mRNA levels of target genes were normalized with β-actin expression and calculated by the 2−ΔΔCT method, as described in detail previously [27].

2.13. Calculation and Statistical Analysis

All data were analyzed statistically using the general linear mixed model in SPSS Statistics Version 20.0 software (SPSS Inc., Chicago, IL, USA) as follows: where Y is the observed dependent variable; X is the design matrix for fixed effects (constructed from user‐specified predictors like dietary OLE level); β is a vector of fixed‐effect coefficients (parameters estimated, representing the average effects of predictors); Z is the design matrix for random effects (links observations to the groups/individuals from which random effects are drawn); γ is a vector of random‐effect coefficients; and ε is a vector of residual errors.

Y=Xβ+Zγ+ε,

Assessments of the linear and quadratic effects of dietary OLE levels were carried out by orthogonal polynomial analysis. Significant differences among groups were determined by one‐way analysis of variance (ANOVA) using Tukey’s honest significant difference (HSD) test. Differences between means with a statistical p value of <0.05 were considered significant.