Animals
Nile tilapia (O. niloticus) used in this study was first introduced by Prof. Nagahama (Laboratory of Reproductive Biology, National Institute for Basic Biology, Okazaki, Japan) and kept in aerated recirculating freshwater tanks at 26 °C under a natural photoperiod in our lab. Animal experiments were conducted in accordance with the regulations of the Guide for Care and Use of Laboratory Animals and were approved by the Committee of Laboratory Animal Experimentation at Southwest University.
Establishment of cyp11a1 mutant line
The sequence of tilapia cyp11a1 (Gene ID: 100692956) was obtained from NCBI. The target site with a Sau96I restriction site (underlined) was designed in the 4th exon of cyp11a1 using the online software ZiFit (http://zifit.partners.org/ZiFiT/). The guide RNA and Cas9 mRNA were prepared as previously reported32. Embryos at the one-cell stage were co-injected with gRNA and Cas9 mRNA with a final concentration of 500 and 1000 ng/µl, respectively. After mutation screening by polyacrylamide gel electrophoresis (PAGE) and Sanger sequencing, F0 chimeric adult XY males were crossed with wild-type (WT) XX females to produce F1 progeny. Siblings in F1 carrying a 22 bp deletion at the same locus were incrossed to produce F2 progeny. Sau96I digestion and PAGE was used for mutation screening in the F2 population. The genetic sex of each fish was identified by a sex-linked marker as described previously31. Primers used for gRNA synthesis, mutation screening, and genetic sex identification are listed in Supplementary Table 3. After mutation screening, the proportion of cyp11a1 homozygous mutant (cyp11a1−/−) fish in the F2 population at 5, 15, 25, 35, and 90 dah was calculated (n = 4 populations at each time point, more than 48 fish were randomly selected in each population).
Measurement of suffocation point
After genotyping, the 90-dah-old WT XY and cyp11a1−/− XY fish (n = 6/genotype) with similar body size (body length: WT XY 6.63 ± 0.34 cm, cyp11a1−/− XY 6.60 ± 0.30 cm; body weight: WT XY 10.37 ± 0.90 g; cyp11a1−/− XY 10.35 ± 0.78 g) were kept in a tank containing 20 L of water at 26 °C. Oxygen in the water was deprived by continuous filling of nitrogen. The suffocation point, which refers to the minimal oxygen concentration at which the fish loses balance, was determined by using the average value of six fish after measurement of the dissolved oxygen concentration when each fish loses balance. The dissolved oxygen concentration was measured by HQ30D portable dissolved oxygen meter (HACH, Colorado, USA).
Measurement of steroid hormones
Enzyme immunoassay (EIA) kit was used to measure the serum P5 (Spbio, Wuhan, China) and cortisol level of WT XY and cyp11a1−/− XY fish, and serum estradiol (E2), 17,20 β-dihydroxy-4-pregnen-3-one (DHP), testosterone (T) and 11-ketotestosterone (11-KT) (Cayman, Michigan, USA) level of WT XX, WT XY, cyp11a1−/− XX, and cyp11a1−/− XY fish at 90 dah. After anesthesia with MS-222 (250 mg/L, Sigma-Aldrich, Missouri, USA), blood was collected from the caudal vasculature of fish from each genotype (serum from three fish as one sample, n ≥ 3/genotype) and placed at 4 °C overnight to allow clot. The serum was separated by blood centrifugation (1000×g, 10 min, 4 °C) and stored at −80 °C. EIA was performed according to the manufacturer’s instructions. All samples were run in triplicate on one 96-well plate to ensure the comparability of each assay.
Quantitative real-time PCR
Twelve tissues (brain, pituitary, gill, heart, spleen, liver, intestine, ovary, testis, kidney, head kidney, and muscle) of WT XX and WT XY fish at 180 dah (n = 4/sex) were collected for expression analysis of cyp11a1 in different tissues. Gonads of WT XX, WT XY, cyp11a1−/− XX, and cyp11a1−/− XY fish at 90 dah were collected for gene expression analysis of foxl2, dmrt1, cyp19a1a, and cyp11c1 (n = 6/genotype). Total RNA was isolated using RNAiso Plus (Takara, Tokyo, Japan). RNA quality and concentration were determined using NanoDrop 2000, and ≥500 ng total RNA was used for cDNA synthesis with PrimeScript RT Master Mix Perfect Real Time kit (Takara, Tokyo, Japan). Quantitative real-time PCR (qPCR) was performed using a TB Green Premix Ex Taq II kit (Takara, Tokyo, Japan). All experiments were performed according to the manufacturer’s instructions. The value was detected on the StepOne Plus Real-time PCR system (Thermo Fisher Scientific, Massachusetts, USA). Gene expression was normalized to β-actin using the 2−ΔΔCt method71. Primers used for qPCR are listed in Supplementary Table 3.
In situ hybridization
In situ hybridization was performed to investigate the cellular location of cyp11a1 in gonads and head kidneys. Ovaries, testes, and head kidneys of WT fish at 240 dah were dissected and fixed in 4% PFA (4 °C, overnight). After fixation, the gonads were embedded in paraffin and sliced into 5 µm sections (cross-section) for use. The digoxigenin (DIG)-labeled sense and antisense probes of cyp11a1 were prepared as follows: a fragment (604 bp) of cyp11a1 were amplified from cDNA by PCR and recovered. After subjected to TA clone, the plasmid containing T7 promoter and insertion or inverted insertion of cyp11a1 fragment were extracted and amplified by PCR. The sense and antisense probes were synthesized by in vitro transcription using the recovered PCR products as templates and a kit containing T7 RNA polymerase and DIG-labeled rNTP mix according to the manufacturer’s instructions (Roche, Basel, Switzerland). The in situ hybridization were performed as follows: Briefly, after deparaffinization in xylene, the slides were placed in 100, 90, 80, and 70% ethanol and 1× PBS for rehydration. Then, the slides were fixed in 4% PFA again, and rinsed in 2 mg/mL glycine, 0.1 M TEA, 0.25% acetic anhydride/0.1 M TEA, successively. Next, the slides were placed in pre-hybridization buffer (66% formamide; 10% 20× SSC) at 65 °C for 2 h and in hybridization buffer (60% formamide; 7.5% dextran sulfate; 0.3 M NaCl, 0.02 M Tris-HCL PH 8.0; 0.0025 M EDTA; 1X Denhardt solution) containing tRNA (20 μg/mL) and antisense probe (500 ng/mL) for hybridization (65 °C, 16 h). After hybridization, the slides were washed stringently with 50% formamide/ 2× SSC, 1× SSC, 0.2× SSC, DIG buffer (0.1 M Maleic acid; 0.15 M NaCl; PH 7.4) and blocked in DIG buffer containing 5% BSA at 37 °C for 1 h. After these steps, the slides were incubated with anti-DIG-AP (Roche, Basel, Switzerland) at 4 °C overnight. Finally, the slides were rinsed in a DIG buffer five times and hatched in a detection buffer (1 M Tris-HCL PH 9.5; 5 M NaCl; 1 M MgCL2) containing NBT/BCIP (Roche, Basel, Switzerland) for signal detection. Images were captured under a BX53 microscope (Olympus, Tokyo, Japan). Primers used for probe preparation are listed in Supplementary Table 3.
Hematoxylin and eosin staining
Gonads from WT XX, WT XY, cyp11a1−/− XX, and cyp11a1−/− XY fish at 90 dah and WT XY, cyp11a1−/− XX, and cyp11a1−/− XY fish at 180 dah were dissected and fixed in Bouin’s solution (24 h, room temperature). After fixation, gonads were embedded in paraffin and sliced into 5 µm sections (middle cross-section) for use. For hematoxylin and eosin (H&E) staining, tissue sections were firstly deparaffinized in xylene and rehydrated in decreasing concentration of ethanol. Then, tissue sections were stained with hematoxylin and eosin successively. Finally, sections were dehydrated in increasing concentration of ethanol and mounted. Images were captured under a BX51 optical microscope (Olympus, Tokyo, Japan). The cross-sectional area of testes of WT XY, cyp11a1−/− XX, and cyp11a1−/− XY fish at 90 and 180 dah was quantified by Image J Pro 1.51 software using default parameters (n = 6 fish/genotype). The gonadosomatic index (GSI) of WT XY, cyp11a1−/− XX, and cyp11a1−/− XY fish at 180 dah was calculated as (gonad weight/body weight) × 100% (n = 6 fish/genotype).
Papanicolaou staining, mobility test, and fertilization assay
Since no sperm could be squeezed out from the genital pore of cyp11a1 mutants at 180 dah, the testes of WT XY, cyp11a1−/− XX, and cyp11a1−/− XY fish were dissected and minced with scissors for further Papanicolaou (Pap) staining, mobility test and in vitro fertilization (IVF) assay to check whether fertile sperm was produced in the mutants (n = 6/genotype). For Pap staining, the homogenates were firstly fixed in 4% PFA for 10 min then stained with Papanicolaou solution EA50 (Solarbio, Beijing, China) for 3 min. Images were captured under a BX51 optical microscope (Olympus, Tokyo, Japan). For the mobility test, the homogenates were mixed with 20 times the volume of water and analyzed under a Sperm Quality Analyzer (Zoneking Software, Beijing, China). For the IVF assay, the homogenates of WT and mutant fish was used to inseminate eggs from WT XX fish. The embryos were checked at 16 h post-fertilization under Leica M205 FA Stereomicroscope (Leica, Wetzlar, Germany) to assess whether fertilization is successful. The fertilization rate was calculated as (the number of eggs with normal blastoderm at 16 hpf/the total number of eggs) × 100%.
Immunofluorescence
Immunofluorescence (IF) was performed to evaluate the expression of Cyp19a1a, Cyp11c1, and Vasa in the gonads of the WT XX, WT XY, cyp11a1−/− XX, and cyp11a1−/− XY fish at 90 dah (n = 4/genotype). Gonads were dissected and fixed in Bouin’s solution (24 h, room temperature). After fixation, gonads were embedded in paraffin and sliced into 5 µm sections (cross-section) for use. After deparaffinization and hydration, sections were subjected to antigen retrieval and blocked in donkey serum (37 °C, 1 h). Sections were then incubated with primary antibody (37 °C, 1 h). The rabbit polyclonal antibodies against Cyp19a1a (2 mg/mL, 1:2000) and Cyp11c1 (2 mg/mL, 1:500) were produced by our lab, and their specificity has been verified previously24,65. Alexa Fluor 488-conjugated donkey anti-rabbit secondary antibody (1:500, Thermo Fisher Scientific, Massachusetts, USA) was used to detect the primary antibodies (37 °C, 40 min). The nuclei were stained with DAPI (1:1000, Sigma-Aldrich, Missouri, USA). Images were captured under the FV3000 confocal laser scanning microscope (Olympus, Tokyo, Japan).
Whole-mount fluorescence in situ hybridization and immunofluorescence
Whole-mount fluorescence in situ hybridization (FISH) and IF were combined to track the expression of gsdf and Cyp19a1a in the gonads of the WT XX, WT XY, cyp11a1−/− XX, and cyp11a1−/− XY fish at 5, 10, 15, 25 dah (n = 5/genotype) and the expression of foxl3 and Dmrt1 in the gonads of the WT XX, WT XY, and cyp11a1−/− XX fish at 15 dah (n = 6/genotype). The DIG-labeled antisense probes of foxl3 and gsdf were prepared as described in the in situ hybridization section of this study. Primers used are listed in Supplementary Table 3. The rabbit antibodies against tilapia Cyp19a1a (2 mg/mL, 1:2000), Dmrt1 (2 mg/mL, 1:500), and Vasa (2 mg/mL, 1:1000) were produced by our lab, and their specificity has been verified previously24,35. After removing the viscera, fish with gonads were fixed in 4% PFA at 4 °C overnight, and placed in 100% methanol at −20 °C for at least 2 h. Then, the fish were placed in 75, 50, 25% methanol, and 1× PBS successively for rehydration and in a 3% H2O2/0.5% KOH medium under light to remove pigment (room temperature, 10 min). After permeabilization in precooled acetone (−20 °C, 30 min), the fish were rinsed four times in 1× PBS and were placed in pre-hybridization buffer (50% formamide; 25% 20× SSC; 0.01 M citric acid; 0.1% Tween-20) at 65 °C for 2 h. Subsequently, the fish were placed in a pre-hybridization buffer containing tRNA (500 μg/mL), heparin (50 μg/mL), and an antisense probe (50–500 ng/mL) for hybridization (65 °C, 16 h). After hybridization, the fish were washed stringently with pre-hybridization buffer, 1× SSC, 0.2× SSC, TN buffer (0.1 M tris-HCL; 0.15 M NaCl; PH 7.5) successively and blocked in TN buffer containing blocking regent (Lot 46925300, Roche, Basel, Switzerland) and 5% donkey serum (37 °C, 1 h). Next, the fish were incubated with anti-DIG-POD (Roche, Basel, Switzerland) at 4 °C overnight. For signal detection, tyramide signal amplification (TSATMR) was performed according to the manufacturer’s instructions (Akoya Biosciences, Marlborough, USA). After TSA amplification, the fish were rinsed four times in 1× PBS and incubated with primary antibody (37 °C, 1 h). Alexa Fluor 488-conjugated donkey anti-rabbit secondary antibody (1:500, Thermo Fisher Scientific, Massachusetts, USA) was used to detect the primary antibody (37 °C, 40 min). The nuclei were stained with DAPI (1:1000, Sigma-Aldrich, Missouri, USA). Images were captured under the FV3000 confocal laser scanning microscope (Olympus, Tokyo, Japan). The positive signal of gsdf and Cyp19a1a was quantified by Image J Pro 1.51 software using default parameters. The zones selected for quantification were in the middle of the gonads. The number of foxl3-positive germ cells was quantified manually.
TUNEL staining
Terminal deoxynucleotidyl transferase (TdT) dUTP Nick-End Labeling (TUNEL) kit (Roche, Basel, Switzerland) was used to assess the apoptosis in the gonads of the WT XX, WT XY, and cyp11a1−/− XX fish at 15 dah (n = 4/genotype). Rabbit antibody against tilapia Vasa (2 mg/mL, 1:500) was used to mark the germ cells. The TUNEL staining was conducted as follows: after removing the viscera, fish with gonads were fixed in 4% PFA at 4 °C overnight, and placed in 100% methanol at −20 °C for at least 2 h. Then, the fish were placed in 75, 50, 25% methanol, and 1× PBS successively for rehydration. After permeabilization in 0.1% triton X-100 (37 °C, 10 min), fish were rinsed four times in 1× PBS and were incubated with TUNEL reaction mixture containing 5 μL enzyme solution and 45 μL label solution (37 °C, 1 h). The positive control was treated with 3 U/mL DNase I (37 °C, 10 min) before reaction mixture incubation. The negative control was incubated only with 50 μL label solution. After TUNEL incubation, the fish were rinsed four times in 1× PBS and blocked in 5% donkey serum (37 °C, 1 h). Fish were next incubated with Vasa antibody (37 °C, 1 h). Alexa Fluor 488-conjugated donkey anti-rabbit secondary antibody (1:500, Thermo Fisher Scientific, Massachusetts, USA) was used to detect the primary antibody (37 °C, 40 min). The nuclei were stained with DAPI (1:1000, Sigma-Aldrich, Missouri, USA). Images were captured under the FV3000 confocal laser scanning microscope (Olympus, Tokyo, Japan). The positive signal of TUNEL in the gonads was quantified using Image J Pro 1.51 software.
Transcriptome analyses
The expression of cyp11a1 in the gonads of tilapia at 5, 7, 20, 30, 40, 90, and 180 dah was determined by analyzing the gonadal transcriptome in our lab29,72. To evaluate the global gene expression in the gonads of the cyp11a1 mutants, the gonads of the WT XX, WT XY, and cyp11a1−/− XX fish at 15 dah (pooled sample, 30 fish/genotype) and the gonads of the WT XX, WT XY, cyp11a1−/− XX, and cyp11a1−/− XY fish at 90 dah (pooled sample, 4 fish/genotype) were collected. After total RNA was extracted, mRNA was enriched by Oligo(dT) beads. Then the enriched mRNA was reversely transcribed into cDNA and disrupted into short fragments for cDNA library construction. The resulting cDNA library was sequenced on Illumina Novaseq 6000 by Gene Denovo Biotechnology Co. (Guangzhou, China). Raw reads were filtered by fastp to remove adapters and low-quality reads. Reads were submitted to the NCBI SRA database (Accession number: PRJNA1074672). Clean reads from each library were aligned to the reference genome (https://www.ncbi.nlm.nih.gov/datasets /genome/ GCF_001858045.2/) using HISAT2 with default parameters73. The reads per kb per million mapped reads (RPKM) method was used to quantify the gene expression level. Pearson correlation analysis of gene expression in the gonads of WT XX, WT XY, cyp11a1−/− XX, and cyp11a1−/− XY fish at 90 dah was performed to evaluate the sample relationship. It was calculated by sample relationship analysis tools on the Omicshare web server (https://www.omicshare.com) based on the global gene expression of each sample. Differential expression analysis was performed by edgeR software74. The p values was calculated according to the suggested BCV (square-root-dispersion) value (BCF = 0.1). The log2 (FC, fold change) ≥2, P < 0.01 (90 dah), or P < 0.05 (15 dah) were set as the threshold for significantly differential expression. GO enrichment analysis and clustering gene expression analysis were also finished on the Omicshare web server (https://www.omicshare.com).
Rescue experiment and androgen treatment
The direct product of Cyp11a1 (P5) and two main female hormones (E2, DHP) were used to rescue the sex reversal of cyp11a1−/− XX fish. After randomized grouping, the fish with mixed genotypes were firstly raised in aerated water containing 100 µg/L P5, E2, or DHP or the same volume of ethanol (ctrl group) from 5 to 25 dah. The rescue effect of P5, E2, and DHP was first evaluated by FISH at 25 dah using gsdf antisense probe (n = 5/genotype). After evaluation, the successfully rescued groups were subjected to treatment with the same hormone until 90 dah. The treatment of fish from 25 to 90 dah was conducted with a diet containing P5 or E2 at a dosage of 50 µg/g. The gonad phenotype of rescued cyp11a1−/− XX fish was evaluated by histological analysis (n = 5/genotype). Androgen treatment of cyp11a1−/− XX fish was conducted from 75 to 90 dah with a diet containing 11-KT at a dosage of 50 μg/g. The expression of Cyp19a1a was analyzed by IF after treatment (n = 4/genotype). The positive signal of Cyp19a1a was quantified by Image J Pro 1.51 software using default parameters. The zones selected for quantification were the whole gonads.
Dual-luciferase reporter assay
The promoter of cyp19a1a and the open reading frame of sf1, dmrt1 were cloned into pGL3 and pcDNA3.1 vector, respectively, in our previous study75,76. The open reading frame of ar1 and ar2 was inserted into Hind III and Xho I restriction sites of the pcDNA3.1 vector in this study. The Luciferase assay was conducted as follows: HEK293 cells were plated on a 24-well plate at a density of 1 × 106 cells per well. After growing to 70–80% confluence, cells were transfected with pGL3-cyp19a1a (250 ng/well), pcDNA3.1-sf1, –dmrt1, –ar1, –ar2 (100 ng/well), and pRL-TK (internal control, 50 ng/well) constructs using Lipofectamine 2000 (Invitrogen, California, USA). At 12 h post-transfection, 200 ng 11-KT was added to each well to activate the ARs. At 48 h post-transfection, the cells were harvested and lysed. The firefly luciferase enzyme activity (pGL3-) was measured and normalized to the Renilla luciferase enzyme activity (pRL-TK) according to the manufacturer’s instruction (Promega, Wisconsin, USA). Primers used for plasmid construction are listed in Supplementary Table 3.
Statistics and reproducibility
Values are presented as mean ± SD. A two-tailed independent Student’s t-test was used to determine the differences between the two groups. One-way ANOVA, followed by Tukey multiple comparison, was used to determine the significance of differences in more than two groups. All analyses were performed using SPSS 22.0 (IBM, New York, USA). P < 0.05 was used as a threshold for statistically significant differences. Experiments except transcriptome sequencing in this study were performed twice to ensure the reproducibility.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.