2.1. Experimental Diets

Three isonitrogenous and isocaloric experimental diets (crude protein 51.5% and crude lipid 10.5%) were formulated, containing different levels of DHA-enriched oil (52% DHA, in the form of triglycerides): 0.10%, 0.95%, and 2.65% (Table 1). The DHA content in the three experimental diets accounted for 5.16%, 8.93%, and 16.47% of the total fatty acids (TFA, Table 2), respectively, and were designated as control (C), low DHA group (DHA-L), and high DHA group (DHA-H), respectively. All ingredients were sieved through a 60-mesh sieve, accurately weighed, and mixed thoroughly. Subsequently, the pellets were made using a pelleting machine and dried at 56°C. The dried diets were preserved at 4°C before the feeding trial [24].

2.2. Experimental Fish and Feeding Management

The same batch of experimental juvenile Chinese tongue sole (5 months old, average initial body weight, 10.46 ± 0.15 g) with no dark color on the abdominal side were selected for the experiment. The juveniles were in the gonadal development stage of II. The fish were purchased from Chengtai Aquaculture Co., Ltd. (Rizhao, China). The experimental fish were transported to Haiyang Aquaculture Co., Ltd. (Haiyang, China), where the feeding experiment was conducted. First, fish were fed a commercial feed for 30 days to acclimate to the experiment conditions. Subsequently, the experimental fish were fasted for 24 h before randomly distributed into nine polyethylene tanks (72 cm × 72 cm × 56 cm), with 15 fish per tank and three replicates per group. The fish were fed to satiation twice daily (7:30 and 19:00). Residual feed on the tank bottom was cleaned daily, and the tank walls were cleaned every 5 days. The experimental water was sourced from both seawater and shallow groundwater and mixed in a reservoir before use. The water temperature ranged from 19 to 28°C; salinity, 28–30; and dissolved oxygen greater than 6.08 mg/L during the experimental period.

After 58 days of feeding, the experimental fish were fasted for 24 h before sampling. The number and weight of fish in each tank were recorded, and 12 fish were selected from each tank. After anesthetized with MS-222, the weight of each fish was measured, and the lateral fin samples were collected. Blood was collected from the caudal vein, which was allowed to clot for 2 h at room temperature and then for another 6 h at 4°C. Following centrifugation at 4000 × g for 10 min, the serum was collected and stored at −80°C. Subsequently, the brain, liver, gonad, muscle, and skin on the eyeless side (both skin samples with and without dark pigment) of each fish were harvested and placed in cryopreservation tubes, which were then stored in liquid nitrogen and subsequently transferred to a −80°C freezer for future use. All handling and sampling practices in this study were reviewed and approved by the Animal Care and Use Committee of Yellow Sea Fisheries Research Institute (protocol code ACUC202208111518).

2.3. Proximate and Fatty Acid Compositions of Fish Tissues

The proximate composition in the experimental diets was determined according to the AOAC methods. The moisture content in the liver and muscle was measured by freeze-drying for 24 h with a freeze dryer (FDU-1100, Tokyo Rikakikai, Co. Ltd., Tokyo, Japan). The crude lipid content was determined via Soxhlet extraction (FOSS Soxtec 2050, Hillerod, Denmark) using petroleum ether as the solvent. The crude protein content was measured using the Kjeldahl method with a Kjeldahl analyzer (FOSS 2300, Hillerod, Denmark). The ash content was assayed with incineration at 550°C for 8 h.

The fatty acid compositions of diets, liver, and muscle were assayed by the gas chromatography method (GC-2010 pro, Shimadzu, Tokyo, Japan). The fatty acids in lyophilized samples were esterified with KOH-methanol and BF₃ successively in a 75°C water bath. Fatty acid methyl esters were extracted with hexane and then subjected to the gas chromatography, which was equipped with a flame ionization detector and a fuzed silica capillary column (SH-RT-2560, 100 m × 0.25 mm × 0.20 μm). The column temperature increase was programed: from 100°C up to 190°C at a rate of 10°C/min; then from 190°C–200°C at a rate of 0.3°C/min; and then from 200 to 230°C at a rate of 4°C/min. Both injector and detector temperatures were 230°C. The fatty acid contents were expressed as %TFA. The detailed program was similar to Liu et al. [25].

2.4. Sex Identification of Experimental Fish

Briefly, the DNA was extracted from the fin using a Marine Animal Tissue Genomic DNA Extraction Kit (TIANGEN BIOTECH, Beijing, China). After extraction, the DNA was subjected to PCR amplification with sex-specific markers (F: CCTAAATGATGGATGTAGATTCTGTC, R: GATCCAGAGAAAATAAACCCAGG). The PCR products were then electrophoresed on a 2% agarose gel at 180 V for 30 min. At the end of the electrophoresis, the bands were observed under UV conditions. A single band (169 bp) indicated genetic male fish, while two bands (169 bp/134 bp) indicated genetic female fish. The genetic female Chinese tongue sole may include both female and pseudomale fish, but all Chinese tongue sole with one band can be considered male fish. In this study, male fish were selected for subsequent experiments. More details on sex identification of Chinese tongue sole can be found in a previous publication of Liu et al. [25].

2.5. The Analysis of Androgen in Serum

The measurement of T and 11-KT in serum was performed using commercial ELISA kits (T: #H090-1-1; 11-KT: #H088) provided by Nanjing Jiancheng Bioengineer Institute (Nanjing, China). Briefly, the standard curve was plotted following the procedures stipulated in the instruction manual using the standards provided in the kit. Subsequently, the serum was processed in accordance with the kit’s instructions. After that, the absorbance is measured with a microplate reader (TECAN Infinite M200, Switzerland), and the obtained values are put into the standard curve to calculate the final concentration.

2.6. Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR)

The total RNA in the gonad was extracted using RNA iso Plus (TaKaRa Biotechnology Co., Ltd., Dalian, China). Subsequently, reverse transcription was carried out with the Evo M-MLVRT Mix kit with gDNA Clean for qPCR (provided by Accurate Biotechnology Co., Ltd., China) in accordance with the manufacturer’s instructions.

Based on the sequences from GenBank, specific primers for the target gene and reference genes were designed and synthesized by Qingke Biotechnology Co., Ltd. (Qingdao, China) (Table 3). In this experiment, β-actin and β2M of Chinese tongue sole were used as the reference genes. The primer validation steps, reaction system, and PCR program were according to Xiong et al. [26]. The mRNA expression level of the target gene was calculated using the 2^−ΔΔCt method.

2.7. Western Blot (WB) Analysis

Briefly, the protein was extracted from the testis tissue of the Chinese tongue sole using RIPA lysis buffer, and the protein concentration of the samples was precisely quantified using a BCA kit (#P0010S, Beyotime, China). SDS–PAGE gel electrophoresis was then employed to separate the protein in the samples, and the protein was transferred onto a polyvinylidene fluoride (PVDF) membrane. After blocking, washing, and incubation with the target protein antibody, as well as incubation with the corresponding secondary antibody (Table 4), visualization was achieved using BeyoECL Star (super-sensitive ECL chemiluminescent kit, #P0018AS, Beyotime, China) from Beyotime. Exposure and photography were carried out using the Tanon 5200 series automatic chemiluminescence image analyzer (Tianneng Life Science Co., Ltd. Shanghai, China), and quantitation was performed using Image J.

2.8. The In Vitro DHA Incubation Study

The Chinese tongue sole testis cell was provided by the laboratory of Songlin Chen in Yellow Sea Fisheries Research Institute, Chinese Academy of Fishery Sciences. The complete medium consisted of 20% fetal bovine serum (#A5669701, Gibco, China), 2% triple antibiotics (penicillin-streptomycin-amphotericin B, #C0223, Beyotime, China), 5 ng/mL growth factors, including bFGF, EGF, and LIF (#P5453, #P6114, #P5354, Beyotime, China), and 27.5 μmoL/L mercaptoethanol (#GUSA-R001, Haixing Biotechnology Co., Ltd., China). L-15 (#11415064, Gibco, China) was used as the basic medium. Once the cells grew to cover the bottom of the T25 flask (#FFLK075, Beyotime, China), the medium was discarded, and the flask was washed twice with DPBS. Then, 3 mL of 0.25% trypsin (#T1300, Solarbio, China) was added for digestion for about 10 s. Two milliliters of trypsin were aspirated, and the digestion was allowed to continue for ~5 min. After the cells were detached, the complete medium containing serum was added to neutralize the trypsin. The cell suspension was evenly distributed into a 6-well plate, and the wells were filled with the complete medium to 2 mL each. DHA (#D2534, Sigma-Aldrich, USA) was added to the complete medium using 2% bovine serum albumin (BSA) (#A8020, Solarbio, China) as the solvent. The final DHA concentrations were 10, 50, and 100 umol/L, respectively. The C group received the same volume of 2% BSA. After incubation for 24 h, the medium was discarded, and the cells were washed twice with PBS. RNA lysis buffer was then added, and the cells were left to sit on ice for 5 min. Subsequently, the total RNA of cells was extracted for subsequent q-PCR experiments.

2.9. Statistical Methods

All data were analyzed using one-way ANOVA in SPSS 16.0 (SPSS Inc. Chicago, USA). After passing the test of homogeneity of variances, multiple comparisons were conducted using Tukey’s test. The calculated results in the tables and figures are presented as mean ± standard error. Significant differences were considered when p < 0.05.

3.1. Genetic Sex Determination

The total number of genetic female fish in group C, DHA-L, and DHA-H was 10, 10, and 7, respectively, and that of male fish in the three groups was 25, 28, and 25, respectively (Figure 1; Table 5).

3.2. Growth Performance and Proximate Composition of Fish Tissue

All experimental fish initially weighed about 10 g. At the end of the feeding experiment, the weight difference between the female and male fish in the same treatment group had begun, and the weight of genetically female fish was generally larger than the male fish (Figure 2). Dietary DHA content had no significant effects on the body weight of the male and female fish (p > 0.05) and had no significant effects on total lipid content in muscle and liver of male fish (p > 0.05).

3.3. Fatty Acid Composition in Tissues of Male Fish

The liver DHA content in the DHA-H group was significantly (p < 0.05) higher than that in the C and DHA-L groups, while no significant (p > 0.05) difference was observed between the latter two groups (Table 6). In contrast, the liver 18:1n-9 content in the DHA-H group was significantly (p < 0.05) lower than that in the C and DHA-L groups. Very similar results were observed in the muscle.

3.4. Concentrations of T and 11-KT in Serum

The serum T content exhibited an increasing trend with increasing dietary DHA levels, but no significant (p > 0.05) difference was observed among groups (Figure 3). The content of 11-KT increased significantly (p < 0.05) with the increasing DHA contents in the diet, and the DHA-H group showed a significantly higher level than the C group.

3.5. Expression of Steroid Hormone Biosynthetic Genes in the Testis

The expression levels of p450c17, hsd17b1, hsd3b1, hsd11b2, p45011b, and aromatase were significantly (p < 0.05) upregulated by increasing DHA levels in the diet (Figure 4). The expression of hsd17b3 showed a similar changing trend, but no significant difference was observed among groups (p > 0.05).

3.6. Gene Expression of Gonadotropin-Releasing Hormone (GnRH) and Its Receptor (GnRHR), As well As Those Associated With the PKC/MAPK Pathway in the Brain of Male Fish

The expression of gnrh2 showed decreasing trends with increasing DHA levels in the diet, but no significant (p > 0.05) differences were observed among groups (Figure 5). There were also no significant differences in the expression of other genes studied (p > 0.05).

3.7. The Protein Expressions of P450c17 and Hsd17b1

The protein expression of P450c17 showed an increasing trend with increasing DHA levels in the diet, although no significant difference (p > 0.05) was observed among groups (Figure 6). The DHA-H group showed significantly (p < 0.05) higher protein expression of Hsd17b1 than the other two groups.

3.8. Expression of Steroid Hormone Biosynthetic Genes in the In Vitro Study

The expression level of p450c17 increased with the increase of DHA concentration up to 50 μmol/L but decreased thereafter (Figure 7). The p450c17 expression level in the group with 50 μmol/L DHA was significantly (p < 0.05) higher compared to the C group. The expression of hsd17b1 in the group with 50 μmol/L DHA was significantly (p < 0.05) lower compared to the groups with 10 or 100 μmol/L DHA. There were no significant differences (p > 0.05) in the gene expression of star and aromaste among experimental groups.