A 2-year-old wild and healthy giant wrasse was purchased from the Hainan Qionghai Fishing Port Terminal and Guangzhou Huangsha Fishery Market. The fish was anesthetized in ice, and samples from the brain and gonads were taken and stored in liquid nitrogen for future use. All animal handling procedures were conducted under the Institutional Animal Care and Use Committee of Hainan University.

RNA was extracted using TRIzol reagent, according to the manufacturer’s instructions. The Prime Script™ RT Reagent Kit with a gDNA Eraser (Perfect Real Time) was used to synthesize cDNA templates. The reverse transcription products were stored at −20°C for later use. cDNA fragments were cloned into PGEM-Teasy plasmids using the SF-1 F1, SF-1 R1, FOXL2 F1, and FOXL2 R1 primers to amplify the target fragments. For this study, SF-1 open reading frame (ORF) F1, SF-1 ORF R1, FOXL2 ORF F1, and FOXL2 ORF R1 were used to amplify HindIII, XhoI, and XhoI and EcoRI restriction sites. According to the restriction endonuclease site existing in pCDNA3.1, a suitable enzyme that was the same as the fragment multi-band restriction site was selected to double-digest the pCDNA3.1 vector. The SF-1 and FOXL2 ORF sequences were ligated into the pCDNA3.1 plasmid, and then the recombinant plasmid was transferred into the DH5α competent cell. Finally, the recombinant plasmid containing the target fragment was screened and identified, and the promoter sequence was obtained using the GenomeWalker method. The cloning method was performed according to the Universal GenomeWalker™ 2.0 User Manual. Table 1 lists all the primers used for cloning and vector construction.

Pomoter regions of the cyp19a1a and cyp19a1b were isolated by genome walking. The online software ALGGEN-PROMO (http://alggen.lsi.upc.es/cgi-bin/promo_v3/promo/promoinit.cgi?dirDB=TF_8.3) was used to predict the cyp19a1a and cyp19a1b transcription binding sites of SF-1 and FOXL2. The giant wrasse’s NR5A1 and FOXL2 gene sequences were spliced using DNAMAN software. The ORF was identified using the ORF Finder (https://www.ncbi.nlm.nih.gov/orffinder/). The National Center for Biotechnology Information (NCBI) database (https://www.ncbi.nlm.nih.gov/) was used to identify the coding regions and amino acid sequences. The online software Signal P4.1 was used to predict the signal peptides of NR5A1 and FOXL2 genes in the giant wrasse. In addition, an adjacency (neighbor joining (NJ)) phylogenetic tree was constructed based on amino acid sequence alignment using MEGA 5.0. GeneDoc software was used for multiple sequence alignments.

In this study, the cyp19a1a and cyp19a1b promoters were analyzed using online software, and the SF-1 and FOXL2 binding positions in their sequences were predicted. The promoters were deleted from the analysis. Primers were designed with restriction sites AAGCTT and GGTACC of restriction endonuclease (HindIII and KpnI). Progressive deletion fragments were constructed using a pGL-4.10 fluorescent expression vector. Based on the effect of transcription factors on the progressive deletion of the promoter, the scope of key transcription binding sites was narrowed, and specific primers were designed to clone the full-length sequence of site-directed mutations by fusion PCR. The ORFs of NR5A1 and FOXL2 were ligated using gene-specific primers, restriction endonucleases, and the pCDNA3.1 vector to construct a site-mutant expression vector.

HEK 293T and COS7 cell lines were grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum, 2 mM of l-glutamine, 100 U/ml of penicillin, and 0.1 mg/ml of streptomycin in 5% CO₂ and cultured at 37°C. The confluent cells were seeded in a 24-well plate. The transfection rate was maintained at ~ 90% at the time of transfection. The specific experimental components were as follows: 1) 250 ng of complete 5′ upstream or deletion constructs (normal and mutated) of cyp19a1a and cyp19a1b promoters cloned into the pGL-4.10 (firefly luciferase) vector and 2) 200 ng of pcDNA3.1 expression plasmid (Invitrogen, Carlsbad, CA, USA) containing ORF cDNAs encoding SF-1 and FOXL2. A 5 ng/well of Renilla luciferase (CMV) vector (Promega, Madison, WI, USA) was used as an internal control. Luciferase assays were performed 48 h after transfection using the Dual-Luciferase Reporter Assay System (Promega). Renilla luciferase activity was used to normalize firefly luciferase activity. Luminescence was measured using a GloMax^®96 microplate luminometer (Promega). The plasmids used in the transfection experiments were prepared using an Endo Free Plasmid Isolation Kit.

The experimental cell data were set up in four parallels per group. A one-way ANOVA was performed, and the results are expressed as the mean ± SD. Statistical significance was set at p < 0.05. * represents a significant difference between the data. SPSS and DPS software was used to perform the statistical analyses and constructed graphs using GraphPad Prism version 5.

The SF-1 and FOXL2 cDNA sequences were cloned and sequenced. The gene encoding SF-1 had a 1,467-bp ORF encoding 488 amino acid residues ( Figure 1A ). The predicted protein (SF-1) had a molecular weight of 55 kDa and a theoretical isoelectric point (pI) of 7.06. FOXL2 had a 921-bp ORF encoding 306 amino acid residues ( Figure 1B ). The predicted FOXL2 protein had a molecular weight of 34.59 kDa and a theoretical pI of 9.21.

Multiple sequence alignment was conducted to compare the amino acid sequences of SF-1 with those of other vertebrates. It showed that SF-1 was conserved in different aquatic animals ( Figure 2A ). Conserved domain analysis indicated that the SF-1 amino acid sequence contained two domains: a conserved DNA-binding domain (DBD) and a C-terminal LBD. FOXL2 consisted of an evolutionarily conserved FH domain ( Figure 2B ). In addition, a phylogenetic tree was constructed based on the SF-1 amino acid sequences from giant wrasse and 13 other species. As Figure 3A shows, SF-1 was most closely related to Labrus bergylta SF-1, clustered with Larimichthys crocea, and formed a subgroup. SF-1 also showed high similarity to other teleosts and had a relatively long homology with mammalian, avian, and amphibian species. The FOXL2 homology comparison results were similar to those of SF-1. FOXL2 showed the closest relationship with L. bergylta FOXL2. FOXL2 also showed relatively distant homology with mammalian, avian, and amphibian species ( Figure 3B ). In summary, the molecular evolution of SF-1 and FOXL2 was consistent with species evolution.

We constructed a luciferase reporter system for cyp19a1a and cyp19a1b promoters and co-transfected the eukaryotic expression vectors of SF-1 and FOXL2 into HEK293T and COS7 cell lines to understand how SF-1 and FOXL2 influence the transcriptional activities of promoters. We used different processing times for different cell lines because the expression time of dual luciferase was unknown after transfection.

In the HEK293T cell line, we found that SF-1 significantly upregulated the transcriptional activity of cyp19a1 promoters 24 and 48 h after transfection, and FOXL2 significantly increased cyp19a1b promoter transcriptional activity only after 48 h of transfection ( Figure 4 , p < 0.05). At the same time, after SF-1 was transfected into COS-7 cells for 48 h, we observed that the transcriptional activities of cyp19a1a and cyp19a1b promoters were significantly increased. However, FOXL2 did not increase the transcriptional activity of the cyp19a1 promoters. Conversely, SF-1 and FOXL2 co-transfection significantly upregulated cyp19a1 promoter activity ( Figure 5 , p < 0.05). The results of the two cell lines showed that co-transfection of the two transcription factors significantly increased cyp19a1 promoter activity.

We found that there may be three binding sites for the SF-1 transcription factor and no binding sites for FOXL2 using online software to predict the region of the gonadal aromatase cyp19a1a promoter ( Figure 6 ). We identified five SF-1 and three FOXL2 transcription binding sites in the brain aromatase cyp19a1b promoter ( Figure 7 ).

We adopted the technology of constructing a progressive deletion of the promoter, fusion PCR, and site-directed mutagenesis to determine the transcription binding sites of the two transcription factors in the aromatase promoters and identify the key elements in transcription regulation. Finally, we identified the key SF-1 and FOXL2 transcription binding sites in the aromatase cyp19a1 promoters.

Evidence supports that NR5A members (FTZ-F1, Ad4BP/SF-1, or LRH-1) and FOXL2 play a role in the brain. Previous studies have potentiated that NR5A members and FOXL2 are involved in cyp19a1a regulation in some fish and mammals (23–26). Furthermore, the combination of FOXL2 and SF-1 can significantly enhance aromatase cyp19a1 transcription (27). Similarly, our results showed that co-transfection of SF-1 and FOXL2 can enhance the transcriptional activity of the cyp19a1 promoters, which indicated a synergistic effect between SF-1 and FOXL2. The synergistic effects of SF-1 and FOXL2 have also been verified in other teleosts. For example, in catfish, FTZ-F1/SF-1 and FOXL2 bind to the promoter and upregulate transcription of the aromatase gene in the brain (28). In tilapia, FOXL2 binds to the promoter and interacts with Ad4BP/SF-1 to upregulate aromatase gene transcription in a female-specific manner (15). Similar results were also obtained in ricefield eels, where NR5A1A (one of the genes encoding NR5A homologs) cooperated with FOXL2 to upregulate cyp19a1a promoter activity (29). However, the synergistic mechanism by which SF-1 and FOXL2 act remains unclear. Studies have shown that SF-1-induced anti-Mullerian hormone (AMH) regulation via functional FOXL2 occurs through protein–protein interactions between FOXL2 and SF-1 in human granulosa cells (30). In addition, Nagahama et al. reported that FOXL2 could interact with the LBD of NR5A1 through the FH domain to form a heterodimer in tilapia (15). Researchers in Quebec, Canada, provided a more detailed explanation regarding this synergistic mechanism and found that the zinc finger region of GATA binding protein 4 (GATA-4) mediated GATA-4/SF-1 synergy. This synergy is the result of direct protein–protein interactions (31). These results collectively indicate that FOXL2 and SF-1 may affect cyp19a1 transcriptional regulation through a specific connection or interaction.

In many bony fish, SF-1 binding sites exist in the promoter region of the gonadal aromatase cyp19a1 (32–35). In Oncorhynchus mykiss, it was found that only −150 to −142 bp and −118 to −110 bp were the core transcription binding sites in the cyp19a1a promoter (36). In the orange-spotted grouper, deletion of the region (−246 to −112 bp) containing two conserved FTZ-F1 sites significantly reduced the transcriptional activity of the cyp19a1a promoter (37). We hypothesized that −120 to −112 bp were the key transcriptional binding sites of SF-1 in the cyp19a1a promoter. Collectively, these results indicated that SF-1 family proteins are common transcriptional regulators of gonad-type cyp19a1a genes in fish and provided evidence that they regulate cyp19a1 expression in vertebrate ovaries. Interestingly, we did not find the core regulatory site of SF-1 in the cyp19a1b promoter. From fish to mammals, SF-1 appears to be the common transcription factor of gonad-type cyp19a1a but not brain-type cyp19a1b. Other studies on aromatase promoters have found that in some fish, including ricefield eel (38), Epinephelus akaara (37), zebrafish (34), and Sparus macrocephalus (39), only the cyp19a1a promoter contains the SF-1 binding site, while the brain-type cyp19a1b promoter does not. Therefore, we speculated that Cheilinus undulatus may be the same as these fish. Thus, this explanation is reasonable. For example, the SF-1 binding site was found in the mouse brain-type cyp19a promoter (40). A previous study found that NR5A1 knockout mice still contained aromatase in some of their brain cells (41), suggesting that SF-1 may not be necessary for cyp19a1b expression.

Previous studies have shown that FOXL2 is involved in regulating cyp19a1 promoter transcription in vertebrates. FOXL2 and SF-1 co-regulate aromatase transcription (15, 27, 28). However, some researchers believe that FOXL2 directly binds to some DNA sites through its FH. Other studies have shown that FOXL2 binds to the half-site sequence of nuclear receptors, namely, the SF-1 response element (SFRE). The SFRE acts on the aromatase promoter (42). Our experiment found that deleting or mutating −890 to −872 bp significantly reduced cyp19a1b promoter activity. Therefore, we speculated that −890 to −872 bp is the core element for FOXL2 to regulate cyp19a1b transcriptional activity. Our results are consistent with those reported for tilapia. A mutation in the FOXL2 binding site (−545 to −538 bp) in the cyp19a1b promoter of tilapia significantly reduced the activity of the promoter (15). Similarly, when a FOXL2 binding site in the promoter of catfish cyp19a1b was mutated from −977 to −954 bp, cyp19a1b transcriptional activity was significantly downregulated, indicating that this region was the core element of FOXL2 to regulate cyp19a1b transcriptional activity (28). These studies all point to the possible involvement of FOXL2 in regulating cyp19a1b promoter transcription.

In conclusion, this study found that SF-1 and FOXL2 can regulate cyp19a1a and cyp19a1b promoters, respectively. SF-1 and FOXL2 contain key binding elements in the cyp19a1a and cyp19a1b promoters, respectively. The combination of SF-1 and FOXL2 enhanced aromatase promoter activity in giant wrasse. This study lays a theoretical foundation for identifying the aromatase gene transcriptional regulatory pathway in giant wrasse.