FoxH1 Represses the Promoter Activity of cyp19a1a in the Ricefield Eel (Monopterus albus)

Forkhead box H1 (FoxH1) is a sexually dimorphic gene in Oreochromis niloticus, Oplegnathus fasciatus, and Acanthopagrus latus, indicating that it is essential for gonadal development. In the present study, the molecular characteristics and potential function of FoxH1 and the activation of the cyp19a1a promoter in vitro were evaluated in Monopterus albus. The levels of foxh1 in the ovaries were three times higher than those in the testes and were regulated by gonadotropins (Follicle-Stimulating Hormone and Human Chorionic Gonadotropin). FoxH1 colocalized with Cyp19a1a in the oocytes and granulosa cells of middle and late vitellogenic follicles. In addition, three FoxH1 binding sites were identified in the proximal promoter of cyp19a1a, namely, FH1 (−871/−860), FH2 (−535/−524), and FH3 (−218/−207). FoxH1 overexpression significantly attenuated the activity of the cyp19a1a promoter in CHO cells, and FH1/2 mutation increased promoter activity. Taken together, these results suggest that FoxH1 may act as an important regulator in the ovarian development of M. albus by repressing cyp19a1a promoter activity, which provides a foundation for the study of FoxH1 function in bony fish reproductive processes.


Introduction
Ovarian follicular cell proliferation and differentiation are critical processes in follicle development that are regulated by multiple endocrine, paracrine, and autocrine factors.These factors simultaneously enhance the crosstalk between oocytes and follicular cells.In teleost ovaries, interstitial cells, granulosa cells (GCs), and thecal cells (TCs) mediate steroidogenesis due to the abundance of intracellular steroidogenic enzymes, such as CYP11, CYP17, and CYP19 [1,2].Previous studies have shown that fox transcription factors regulate the promoter activity of cyp19a1 through a conserved N-terminal Forkhead domain that binds to DNA [3,4].In teleosts, FoxL2 [5][6][7], FoxO1 [8], and FoxO3 [9] bind to the cyp19a1a promoter via the Forkhead domain and significantly upregulate its transcriptional activity.In addition, the conserved potential binding sites of FoxO4, FoxF1/2, and FoxL1 in the cyp19a1a promoter have been predicted [10,11].Some fox genes are sexually dimorphic, including foxj2, foxj1a, and foxh1, and even exhibit male-specific expression [12].These results indicate that fox genes are widely involved in sex determination and folliculogenesis.
Forkhead box H1 (FoxH1) is a critical Smad2/3 cofactor [13,14] involved in regulating gene expression downstream of Activin/Nodal/TGF-β signals during early embryo development [15,16].FoxH1 interacts with Smad2/3/4 via the C-terminal Smad interaction domain (SID) to form the Smad/FoxH1 complex, which binds to DNA via a conserved N-terminal Forkhead domain and regulates target gene transcription [17].The presence of FoxH1 mRNA in the unfertilized eggs of Xenopus [18], zebrafish [19], and mice [20] indicates that this gene is maternally expressed.In mouse ovaries, FoxH1 is expressed in oocytes and TCs, but FoxH1 is expressed specifically in TCs during ovarian follicle development, ovulation, and luteinization [21].In Oplegnathus fasciatus [22], Acanthopagrus latus [23], and Oreochromis niloticus [12], foxh1 is a sexually dimorphic gene, and its expression level in the ovaries is higher than that in the testes.These findings reveal that foxh1 might play essential roles during ovarian development.
The ricefield eel (Monopterus albus) is a protogynous hermaphrodite fish with few oocytes and low fecundity that is increasingly being used as a model vertebrate for the study of gonad development and sex change [24,25].Our previous RNA-seq data showed that foxh1 expression was highest in the ovaries, followed by the intersex gonads, and was very low in the testes [26].In addition, foxh1 −/− XX tilapia oogenesis was arrested, and Cyp19a1a expression was markedly decreased.However, the mechanism by which FoxH1 regulates the expression of cyp19a1a in M. albus remains unclear.To explore the function of FoxH1 in follicle development and the transcriptional regulation of cyp19a1a in M. albus, the sequences and expression patterns of FoxH1 were determined, and ovarian tissue fragments were incubated with gonadotropin in vitro.Then, the localization of FoxH1 and Cyp19a1a in the ovaries was analyzed.Finally, the mechanism whereby FoxH1 transcriptionally regulates cyp19a1a was analyzed.We found that FoxH1 plays a critical role in follicular cell differentiation through possible effects on cyp19a1a transcription and estrogen synthesis in ricefield eel.

Sequence Analysis of Ricefield Eel foxh1
The coding sequence of foxh1 was 1533 base pairs (bp) in length and consisted of three exons (encoding a putative protein of 510 amino acids) (Figure 1).A phylogenetic tree was constructed based on the full protein sequences of FoxH1 orthologs.FoxH1 of the ricefield eel clustered with those of other teleost species and clustered closely with the sequence of A. latus (Figure 2A).Consistent with the FoxH1 orthologs of other vertebrates, FoxH1 contains three typical domains: a Forkhead domain, a FoxH1 domain (FM1/2), and an SID.In the ricefield eel, the FoxH1 domain was the most conserved, followed by the Forkhead domain and the SID.In mammals, amphibians, and teleosts, the FM1 domain was fairly conserved, but the FM2 domain was not (Figure 2B).

Expression Patterns of foxh1 in M. albus
foxh1 mRNA was widely expressed in all examined tissues, including the brain, heart, liver, kidney, intestine, spleen, blood, ovary, and testis.The highest levels of foxh1 transcripts were detected in the ovaries, where they were approximately three times higher than those in the testes (p < 0.01) (Figure 3A).There were no significant differences in foxh1 expression levels during the five stages of ovarian development (p > 0.05) (Figure 3B).

Expression Patterns of foxh1 in M. albus
foxh1 mRNA was widely expressed in all examined tissues, including the brain, heart, liver, kidney, intestine, spleen, blood, ovary, and testis.The highest levels of foxh1 transcripts were detected in the ovaries, where they were approximately three times higher than those in the testes (p < 0.01) (Figure 3A).There were no significant differences in foxh1 expression levels during the five stages of ovarian development (p > 0.05) (Figure 3B).However, foxh1 expression levels increased from the primary growth (PG) to early vitellogenic (EV) stages and decreased from the EV to the mature ovary (OM) stages.They were highest in the EV stage, followed by the mid-to-late vitellogenic (MLV) stage.These results suggest that foxh1 is a sexually dimorphic gene that may play a crucial role in female sexual cycle maintenance and ovarian vitellogenesis.
Int. J. Mol.Sci.2023, 24, x FOR PEER REVIEW 5 of 15 However, foxh1 expression levels increased from the primary growth (PG) to early vitellogenic (EV) stages and decreased from the EV to the mature ovary (OM) stages.They were highest in the EV stage, followed by the mid-to-late vitellogenic (MLV) stage.These results suggest that foxh1 is a sexually dimorphic gene that may play a crucial role in female sexual cycle maintenance and ovarian vitellogenesis.

Expression of foxh1 after FSH and hCG Incubation In Vitro
Overall, FSH and hCG had different stimulatory effects on foxh1 expression levels.In the 0.05 ng/L and 1 ng/L FSH groups (Figure 4A), the expression levels of foxh1 had a similar pattern of change, in which foxh1 levels were significantly elevated at 1 h (p < 0.05) and then slowly decreased to the control level.The highest concentration of FSH (5 ng/L) significantly increased the foxh1 expression at 2 h (p < 0.05).Furthermore, in the 10 IU/mL hCG group, the foxh1 expression levels decreased significantly at 4 h (p < 0.05) (Figure 4B).In the 50 IU/mL hCG group, the foxh1 level increased significantly at 2 h compared to other time points (p < 0.05).In the 100 IU/mL hCG group, foxh1 expression levels increased at 1 h but were not significant.However, foxh1 expression levels at 2 h, 4 h, and 10 h were lower than those of the control.

Expression of foxh1 after FSH and hCG Incubation In Vitro
Overall, FSH and hCG had different stimulatory effects on foxh1 expression levels.In the 0.05 ng/L and 1 ng/L FSH groups (Figure 4A), the expression levels of foxh1 had a similar pattern of change, in which foxh1 levels were significantly elevated at 1 h (p < 0.05) and then slowly decreased to the control level.The highest concentration of FSH (5 ng/L) significantly increased the foxh1 expression at 2 h (p < 0.05).Furthermore, in the 10 IU/mL hCG group, the foxh1 expression levels decreased significantly at 4 h (p < 0.05) (Figure 4B).In the 50 IU/mL hCG group, the foxh1 level increased significantly at 2 h compared to other time points (p < 0.05).In the 100 IU/mL hCG group, foxh1 expression levels increased at 1 h but were not significant.However, foxh1 expression levels at 2 h, 4 h, and 10 h were lower than those of the control.However, foxh1 expression levels increased from the primary growth (PG) to early vitellogenic (EV) stages and decreased from the EV to the mature ovary (OM) stages.They were highest in the EV stage, followed by the mid-to-late vitellogenic (MLV) stage.These results suggest that foxh1 is a sexually dimorphic gene that may play a crucial role in female sexual cycle maintenance and ovarian vitellogenesis.

Expression of foxh1 after FSH and hCG Incubation In Vitro
Overall, FSH and hCG had different stimulatory effects on foxh1 expression levels.In the 0.05 ng/L and 1 ng/L FSH groups (Figure 4A), the expression levels of foxh1 had a similar pattern of change, in which foxh1 levels were significantly elevated at 1 h (p < 0.05) and then slowly decreased to the control level.The highest concentration of FSH (5 ng/L) significantly increased the foxh1 expression at 2 h (p < 0.05).Furthermore, in the 10 IU/mL hCG group, the foxh1 expression levels decreased significantly at 4 h (p < 0.05) (Figure 4B).In the 50 IU/mL hCG group, the foxh1 level increased significantly at 2 h compared to other time points (p < 0.05).In the 100 IU/mL hCG group, foxh1 expression levels increased at 1 h but were not significant.However, foxh1 expression levels at 2 h, 4 h, and 10 h were lower than those of the control.

Colocalization of FoxH1/Cyp19a1a in EV-Stage M. albus Ovaries
The subcellular colocalization of FoxH1 and Cyp19a1a in EV ovaries was determined by immunofluorescence (Figure 5).FoxH1 was mainly localized in the nuclei and cytoplasm of primary growth oocytes (PGOs) (Figure 5B and Figure S1).However, FoxH1 was present only in the follicular cells of EV follicles (Figure 5B).Cyp19a1a was localized in the nuclei and cytoplasm of PGOs, cortical alveoli stage oocytes (CAOs), and early vitellogenic-stage oocytes (EVOs) (Figure 5A,C and Figure S2).It was also localized in the follicular cell nuclei of EV follicles (Figure 5C), and the fluorescence signal of Cyp19a1a was stronger than that of FoxH1 in follicular cells.In general, FoxH1 and Cyp19a1a colocalized in the cytoplasm and nuclei of PGOs, the cytoplasm of EVOs, and the follicular cells of EV follicles (Figure 5A).

Colocalization of FoxH1/Cyp19a1a in EV-Stage M. albus Ovaries
The subcellular colocalization of FoxH1 and Cyp19a1a in EV ovaries was deter by immunofluorescence (Figure 5).FoxH1 was mainly localized in the nuclei and plasm of primary growth oocytes (PGOs) (Figures 5B and S1).However, FoxH1 wa sent only in the follicular cells of EV follicles (Figure 5B).Cyp19a1a was localized nuclei and cytoplasm of PGOs, cortical alveoli stage oocytes (CAOs), and early v genic-stage oocytes (EVOs) (Figures 5A,C and S2).It was also localized in the fol cell nuclei of EV follicles (Figure 5C), and the fluorescence signal of Cyp19a1 stronger than that of FoxH1 in follicular cells.In general, FoxH1 and Cyp19a1a coloc in the cytoplasm and nuclei of PGOs, the cytoplasm of EVOs, and the follicular cells follicles (Figure 5A).

Colocalization of FoxH1 and Cyp19a1a in MLV-Stage M. albus Ovaries
The subcellular colocalization of FoxH1 and Cyp19a1a in MLV ovaries was examined based on immunofluorescence (Figure 6).In MLV follicles, follicular cells proliferate and differentiate to form inner GCs and outer TCs (Figure 6D).FoxH1 was localized in the nuclei of oocytes and GCs.Cyp19a1a was localized in the nuclei of oocytes, GCs, and TCs (Figure 6A,C).FoxH1 and Cyp19a1a colocalized in the nuclei of oocytes and GCs, whereas no fluorescent signal was observed in the cytoplasm of oocytes (Figure 6A).Moreover, the specific signal of Cyp19a1a was stronger than that of FoxH1 in TCs (Figure 6C).

Colocalization of FoxH1 and Cyp19a1a in MLV-Stage M. albus Ovaries
The subcellular colocalization of FoxH1 and Cyp19a1a in MLV ovaries was examined based on immunofluorescence (Figure 6).In MLV follicles, follicular cells proliferate and differentiate to form inner GCs and outer TCs (Figure 6D).FoxH1 was localized in the nuclei of oocytes and GCs.Cyp19a1a was localized in the nuclei of oocytes, GCs, and TCs (Figure 6A,C).FoxH1 and Cyp19a1a colocalized in the nuclei of oocytes and GCs, whereas no fluorescent signal was observed in the cytoplasm of oocytes (Figure 6A).Moreover, the specific signal of Cyp19a1a was stronger than that of FoxH1 in TCs (Figure 6C).

Colocalization of FoxH1 and Cyp19a1a in MLV-Stage M. albus Ovaries
The subcellular colocalization of FoxH1 and Cyp19a1a in MLV ovaries was based on immunofluorescence (Figure 6).In MLV follicles, follicular cells prolif differentiate to form inner GCs and outer TCs (Figure 6D).FoxH1 was localiz nuclei of oocytes and GCs.Cyp19a1a was localized in the nuclei of oocytes, GCs (Figure 6A,C).FoxH1 and Cyp19a1a colocalized in the nuclei of oocytes and GCs no fluorescent signal was observed in the cytoplasm of oocytes (Figure 6A).Mor specific signal of Cyp19a1a was stronger than that of FoxH1 in TCs (Figure 6C).

Activation of the cyp19a1a Promoter by Foxh1 via the Forkhead Binding Site In Vi
As shown in Figure 7A,B, three FoxH1 binding sites were identified in the promoter by JASPAR online software (2022, the 9th release of the open-access da transcription factor binding profiles), namely, FH1 (−871/−860), FH2 (−535/−524) (−218/−207).To investigate whether FoxH1 was a transcription factor of the cyp19 wild-type and mutated luciferase reporter vectors of cyp19a1a were constructe transfected with pcDNA3.1-FoxH1into CHO cells.Luciferase activity analysis that FoxH1 overexpression significantly decreased the promoter activity of the gene (p < 0.01) (Figure 7C).Moreover, the luciferase activity of the cyp19a1a-mut1 and cyp19a1a-mut2 (p < 0.05) vectors were significantly increased in FoxH1-ove ing cells compared to control cells, whereas the promoter activity of cyp19 showed no change (Figure 7D), suggesting that FoxH1 regulated the transcriptio ity of cyp19a1a through the FH1 and FH2 motifs.

Activation of the cyp19a1a Promoter by Foxh1 via the Forkhead Binding Site In Vitro
As shown in Figure 7A,B, three FoxH1 binding sites were identified in the cyp19a1a promoter by JASPAR online software (2022, the 9th release of the open-access database of transcription factor binding profiles), namely, FH1 (−871/−860), FH2 (−535/−524), and FH3 (−218/−207).To investigate whether FoxH1 was a transcription factor of the cyp19a1a gene, wild-type and mutated luciferase reporter vectors of cyp19a1a were constructed and cotransfected with pcDNA3.1-FoxH1into CHO cells.Luciferase activity analysis revealed that FoxH1 overexpression significantly decreased the promoter activity of the cyp19a1a gene (p < 0.01) (Figure 7C).Moreover, the luciferase activity of the cyp19a1a-mut1 (p < 0.01) and cyp19a1a-mut2 (p < 0.05) vectors were significantly increased in FoxH1-overexpressing cells compared to control cells, whereas the promoter activity of cyp19a1a-mut3 showed no change (Figure 7D), suggesting that FoxH1 regulated the transcriptional activity of cyp19a1a through the FH1 and FH2 motifs.

Discussion
FoxH1 family members have been described in many vertebrate groups, including mammals [27], amphibians [18], and teleosts [28].These proteins showed high homology in the Forkhead DNA binding domain and SID but very little conservation outside those domains.In the present study, we cloned 1533 bp cDNA sequences and characterized them.The results showed that M. albus FoxH1 is highly conserved between bony fish and amphibians, including species such as A. latus, Oryzias latipes, and Xenopus laevis, especially in the Forkhead domain and FoxH1 domain (FM1/2).These results imply that FoxH1 may have a conserved function in bony fish.
Previous studies have demonstrated that foxh1 is a maternal and zygotic gene that plays an important role in early embryonic development.Similar to the results of studies on zebrafish [28], Xenopus [18,29], and mice, M. albus foxh1 is expressed maternally.Gonad RNA-seq data of nile tilapia [12], rock bream [22], and yellowfin seabream [23] have shown that foxh1 expression levels are significantly higher in the ovary than in the testis.Notably, yellowfin seabream is a protandrous hermaphroditic fish, and foxh1 expression levels in the ovary and ovo-testis are approximately 20 times higher than those in the testis [23].As determined in the present study, M. albus foxh1 is a sexually dimorphic gene, and its expression levels in the ovary are three times higher than those in the testis.Moreover, foxh1 expression levels increase from PG to EV and decrease from EV to OM during ovarian development.This finding was consistent with the TGF-β family bmpr2 [30], smad2 [24] and smad3 (unpublished data from our lab) expression patterns in M. albus ovaries.These results reveal that foxh1 may be involved in early folliculogenesis and previtellogenesis as a Smad2/3 transcriptional partner mediating TGF-β signaling.
Gonadotropins participate in ovarian development by regulating numerous gene networks, such as those related to steroid synthesis, cell proliferation, and differentiation [31].The expression levels of FoxH1 remained stable after pregnant mare serum gonadotropin (PMSG) and hCG treatment, and FoxH1 was localized in the newly formed corpus luteum, but its expression decreased as the corpus luteum degenerated [21].In the present study, neither time nor dose dependency was observed in either FSH or hCG treatments.In the FSH group, FSH stimulated foxh1 expression, but the stimulation weakened with time.However, the effect of hCG on foxh1 expression was not regular.FSH in teleosts is primarily involved in the control of oocyte growth [2,32].Fshb immunoreactive signals and fshb mRNA in the M. albus pituitary increased at the onset of secondary oocyte growth, indicating that FSH is key to the onset of first puberty and vitellogenesis [33].This is consistent with foxh1 expression patterns during ovarian development.Taken together, these results suggest that foxh1 is an FSH-responsive gene and might play important roles in previtellogenesis, follicular cell layer formation, and FSH-modulated ovary development.
During zebrafish oogenesis, foxh1 mRNA changed its localization from the vegetal Balbiani body to the animal pole between stage I and II oocytes [34].FoxH1 was strongly expressed in oocytes and TCs throughout folliculogenesis in mouse ovaries [21].In XX tilapia, foxh1 signals were present in the cytoplasm of stage I and II oocytes in the ovary by in situ hybridization [12].Moreover, foxh1 −/− XX tilapia oocytes failed to transition from phase II to phase III, and follicle cells were blocked from transitioning from one to two layers [35].In the present study, FoxH1 was localized mainly in oocytes and GCs.Notably, the transfer of FoxH1 signals from oocytes to GCs followed subsequent development.
FoxH1 was not observed in the cytoplasm of oocytes during vitellogenesis, whereas it was observed in GCs and TCs.Additionally, FoxH1 and Cyp19a1a colocalized in GCs.Considering these results together, the cell-specific expression pattern of FoxH1 in M. albus raises the possibility that FoxH1 may promote oocyte growth, GC proliferation, and steroid hormone synthesis.
FoxH1 acts as an activator or repressor, alone or in concert with other transcription factors, to regulate target genes.For example, FoxH1 activated lim gene transcription alone or in conjunction with Smad2/4 [36].FoxH1 also interacted with Smad2/3 to activate downstream genes of Nodal signaling [37,38].Furthermore, FoxH1 bound to a corepressor to repress xrn1 gene transcription [39,40].FoxH1 repressed androgen receptor (AR) transcriptional activity and colocalized with AR [41].FoxH1 repressed the liganddependent and ligand-independent transcriptional activity of the estrogen receptor through the estrogen response element [42].In the present study, the levels of foxh1 in ovaries peaked at the EV stage.However, cyp19a1a mRNA levels were highest in the MLV stage [43].FoxH1 and Cyp19a1a colocalized in GCs.In addition, three FoxH1-binding sites were predicted in the cyp19a1a promoter, namely, FH1 (−871/−860), FH2 (−535/−524), and FH3 (−218/−207).pcDNA3.1-FoxH1and cyp19a1a-Luc were cotransfected into CHO cells, and FoxH1 significantly suppressed transcription of the cyp19a1a gene.However, cyp19a1a transcriptional activity was significantly increased when FH1/2 and FH1/3 were mutated.In addition, the promoter activity of cyp19a1a mut-1 (FH1/2 mutant type) was higher than that of cyp19a1a mut-2 (FH1/3 mutant type).The transcriptional activity of cyp19a1a mut-3 (FH2/3 mutant type) did not change.These results suggest that FoxH1 may act as a repressor to regulate the promoter activity of cyp19a1a via FH1 and FH2.In contrast, Cyp19a1a expression levels were significantly reduced in foxh1 −/− XX tilapia, possibly because the transition from stage II to III and follicular cells from one to two layers were blocked [4].It is well known that teleost aromatase is synthesized predominantly at the follicular cell layer [5].In addition, ovarian transcriptomics of foxh1 −/− XX tilapia have shown decreased foxl2 expression and increased dmrt1 expression [4].FoxL2 was a transcriptional activator of cyp19a1a in tilapia [6] and M. albus [7].Dmrt1 inhibited the transcription of cyp19a1a in tilapia [8].Therefore, the decrease in cyp19a1a in foxh1 −/− tilapia may be caused by the FoxH1 regulated gene network.In addition, no study has reported the regulatory relationship between FoxH1 and cyp19a1a.However, the present study determined that there is a regulatory relationship between them, and in vivo studies are needed to reveal the regulatory mechanisms and the effects on the downstream pathways and gonadal development, which will be the focus of future studies.
To date, most studies on FoxH1 functions have focused on the role of FoxH1 during embryogenesis, while little is known about the roles of FoxH1 reproduction.To date, most studies on FoxH1 functions have focused on the role of FoxH1 during embryogenesis, while little is known about the roles of FoxH1 reproduction.The ricefield eel is a protogynous hermaphrodite fish for which it is difficult to phenotypically differentiate sex in the nonspawning season; sex discrimination requires histological observation or detection of the expression of sex-specific genes, such as cyp19a1a (a female-specific gene) and dmrt1 (a malespecific gene).We found evidence that FoxH1 is a sexually dimorphic gene that is highly expressed in the ovaries and expressed at low levels in the testes; thus, foxh1 can be used as a female marker gene.Furthermore, FoxH1 acts as a suppressor of cyp19a1a transcription and maintains its expression in sexually mature females, potentially prolonging the spawning cycle of females.These findings provide new insights for artificial sex control and for the improvement of spawning quality.Overall, this study demonstrates that FoxH1 regulates cyp19a1a transcription in vitro, but further in vivo studies are needed to understand the role of FoxH1 in bony fish ovarian development.

Sample Collection and Preparation
The wild M. albus (n = 100, body length = 34.93 ± 4.52 cm and body weight = 37.99 ± 21.88 g) used in the present study were obtained from a local market in Chengdu, Sichuan.These fish were kept under the natural temperature and photoperiod in the laboratory.All experimental procedures involving animal research were subject to approval and performed in accordance with the guidelines of the ethics committee (Approval No. 20190031).
Fish were anesthetized with 0.02% tricaine buffer (80 µg/L) (Sigma, West Hollywood, LA, USA) for 10 min after a 24 h fast, and the tissues, including half of the gonads, pituitary, eyes, heart, kidneys, intestines, spleen, muscles, and blood, were collected and immediately stored in liquid nitrogen.A portion of the fresh gonads was immediately fixed in Bouin's solution for 24 h and embedded in paraffin.Sections were serially cut at a thickness of 5 µm using a slicer (Leica, Nussloch, Germany) and stained with hematoxylin/eosin.The histological classification of the gonad, including the PG, previtellogenic stage (PV), EV, MLV, and OM, has been described previously [44].

RNA Extraction and cDNA Synthesis
Total RNA was extracted using TRIzol (Invitrogen, Chicago, IL, USA) following the manufacturer's instructions.A RevertAid First-strand cDNA Synthesis Kit (Thermo Scientific, Waltham, MA, USA) was used to generate cDNA according to the manufacturer's instructions.cDNA quality was verified by the successful amplification of ef1α and rpl17 [26].

Cloning of M. albus foxh1 cDNA
Specific primers (foxh1-F1, foxh1-R1, foxh1-F2, and foxh1-R2 (Table 1)) were designed to clone foxh1 based on the coding sequence from the M. albus genome (Accession No: 109965524).After an initial 0.5 min denaturation at 95 • C, PCR was conducted for all of the above genes with the following cycling conditions: 35 cycles of 94 • C for 0.5 min, 55 • C for 0.5 min, and 72 • C for 0.5 min, with a final extension at 72 • C for 30 min.All target products were ligated into pMD19-T (TaKaRa, Dalian, China) and sequenced by TsingKE Biological Technology Company Limited (Chengdu, Sichuan, China).Multiple alignments of the amino acid sequences were conducted with ClustalX 1.83.Based on the deduced amino acid sequences, a phylogenetic tree was constructed via the neighbor-joining method with bootstrap values calculated from 1000 replicates in the MEGA 11 software package.

Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR)
cDNA was obtained from gonads at the five developmental stages and from other tissues.qRT-PCR was performed with a Bio-Rad CFX Connect system (Bio-Rad, Chicago, IL, USA) to determine the expression levels of foxh1.The sequences of the primers are listed in Table 1.The cycling parameters were as follows: 95 • C for 5 min followed by 40 amplification cycles of 95 • C for 10 s, 59 • C for 15 s, and 72 • C for 20 s.To minimize variation due to differences in cDNA loading, the geometric mean expression levels of rpl17 and ef1α were utilized to normalize the expression levels of the target genes.Target gene expression was calculated with the equation C target gene / C ef1a × C rpl17 .

Expression Patterns of foxh1 after Incubation of Ovaries with hCG and FSH In Vitro
Gonads of female M. albus at the MLV stage were washed and dissected in Leibovitz L-15 medium (Gibco, Carlsbad, CA, USA) on ice.Ovarian tissues (50-100 mg) were placed in 24-well tissue culture dishes in 1 mL of Leibovitz L-15 medium (0.1 U/mL penicillin and 0.1 mg/mL streptomycin) with FSH (0.05, 1.0, and 5.0 ng/mL), hCG (10, 50, and 100 IU/mL), or saline solution (control group), and then incubated at 28 • C for 1, 2, 4, and 10 h.

Plasmid Construction and Dual-Luciferase Reporter Assays
To generate luciferase reporters, fragments of the cyp19a1a promoter (1866 bp) were amplified, cloned, and inserted into a pGL3-Basic reporter vector between the NheI and XhoI restriction sites.For FoxH1 expression vector construction, the foxh1 full-length coding sequence (1533 bp) of M. albus was amplified, double-digested with NheI and EcoRI, and then cloned and inserted into the pcDNA3.1 vector.To further evaluate the effects of FoxH1 on the transcriptional activity of M. albus cyp19a1a, its promoter containing wild-type FoxH1 binding sites (FHs) was amplified, cloned, and inserted into the pGL3-Basic reporter vector between the KpnI and XhoI restriction sites.Additionally, FH mutant-type vectors were constructed by using a TaKaRa MutanBEST Kit (#R401, TaKaRa, Beijing, China) according to the manufacturer's instructions with the wild-type plasmids as templates.The sequences of wild-type and mutant FH binding sites are shown in Table 2.All recombinant plasmids were constructed by Bioengineering (Shanghai, China) Co., Ltd. and verified by Sanger sequencing.The primers for plasmid construction are listed in Table 1.For luciferase activity detection, after transfection for 48 h, the cells were harvested, and their lysates were collected for dual-luciferase analysis with a Dual-Luciferase Reporter Assay System (#E1910, Promega, Madison, WI, USA) following the kit's manual.The GloMax detection system (Promega) was used to measure firefly and Renilla luciferase activity in cell lysates.

Statistical Analysis
Statistical analyses were performed by using GraphPad Prism v8.0 software (Graph-Pad, California, CA, USA) and SPSS v20.0 (IBM, Armonk, NY, USA).All data are presented as the mean ± SEM from three independent experiments.Comparisons among three or more different groups were conducted by using one-way ANOVA followed by Duncan's multiple comparisons test.* p < 0.05 was considered to indicate statistical significance, and the significance levels are stated in the corresponding figure legends.

Conclusions
In summary, foxh1 is a sexually dimorphic gene.The foxh1 levels in the ovaries were three times those in the testes, and they were regulated by the gonadotropins FSH and hCG in vitro.FoxH1 colocalized with Cyp19a1a in the oocytes and GCs of middle and late vitellogenic follicles.Furthermore, three FoxH1 binding sites were identified in the proximal promoter of cyp19a1a, and FoxH1 overexpression significantly attenuated the activity of the cyp19a1a promoter in CHO cells.This study demonstrates that FoxH1 regulates cyp19a1a transcription in vitro, but further in vivo studies are needed to understand the role of FoxH1 in bony fish ovarian development.

15 Figure 1 .
Figure 1.Nucleotide and amino acid sequences of the coding region of FoxH1 in the ricefield eel, Monopterus albus.The untranslated regions and translated regions are indicated by lowercase letters and uppercase letters, respectively.The initiation codon (ATG) and stop codon (TGA) are marked in red.Asterisks (*) indicate the translation stop codon.

Figure 1 .
Figure 1.Nucleotide and amino acid sequences of the coding region of FoxH1 in the ricefield eel, Monopterus albus.The untranslated regions and translated regions are indicated by lowercase letters and uppercase letters, respectively.The initiation codon (ATG) and stop codon (TGA) are marked in red.Asterisks (*) indicate the translation stop codon.

Figure 4 .
Figure 4. Regulation of foxh1 expression in the ovary by FSH and hCG in vitro.(A) Expression levels of foxh1 after FSH incubation.(B) Expression levels of foxh1 after hCG incubation.The results are presented as the means ± SEMs (n = 5).FSH, follicle stimulating hormone; hCG, human chorionic gonadotropin; * p < 0.05.

Figure 4 .
Figure 4. Regulation of foxh1 expression in the ovary by FSH and hCG in vitro.(A) Expression of foxh1 after FSH incubation.(B) Expression levels of foxh1 after hCG incubation.The resu presented as the means ± SEMs (n = 5).FSH, follicle stimulating hormone; hCG, human ch gonadotropin; * p < 0.05.

Figure 7 .
Figure 7. FoxH1 acts as a transcription factor and represses cyp19a1a promoter transcriptional activity.(A) Marker sequence of the FoxH1 binding site based on the JASPAR database and FoxH1

Figure 7 .
Figure 7. FoxH1 acts as a transcription factor and represses cyp19a1a promoter transcriptional activity.(A) Marker sequence of the FoxH1 binding site based on the JASPAR database and FoxH1 binding

Table 1 .
Primers used for cloning and mRNA expression analysis.