The Involvement of Follicle-Stimulating Hormone in Testis Differentiation in Nile Tilapia

Abstract

In Nile tilapia, one of the most important aquaculture species, males are larger than females, and an all-male monosex culture offers significant economic benefits. Although the pituitaries of genetic female (XX) and genetic male (XY) tilapia have identical expression levels of follicle-stimulating hormone (fsh), FSH receptor (fshr) expression remains relatively low in XY-undifferentiated gonads and then increases following morphological sex differentiation. The expression patterns of genes related to androgen biosynthesis in XY-undifferentiated gonads are similar to those of fshr during testis differentiation. This might imply that FSH has a potential function in testis differentiation through regulating the expression of genes related to androgen biosynthesis. To determine whether FSH signaling regulated androgen biosynthesis, we microinjected recombinant FSH (rFsh) into XY larvae during the early sex-differentiation stage. We compared the expression of various genes related to testis differentiation after injection. The genes hsd3b, cyp17a1, dmrt1, and gsdf were found to have higher expression in the rFsh treatment group. These results suggest that FSH signaling can activate androgen biosynthesis by regulating steroidogenic enzymes, including hsd3b and cyp17a1. Moreover, injected rFsh can upregulate dmrt1, which has a positive effect on the expression of gsdf. Therefore, during testis differentiation and development, FSH plays a role in both androgen synthesis and the expression of genes related to testis differentiation in Nile tilapia.

1. Introduction

The process of undifferentiated gonads developing into testes or ovaries is called gonadal sex differentiation [1]. The Nile tilapia, Oreochromis niloticus, has an XX/XY sex-determination system. The established biotechnological system of all-female (XX) or all-male (XY) populations is based on the artificial fertilization of normal females with sex-reversed males (XX) or super males (YY), respectively [2,3]. Therefore, the Nile tilapia is an ideal model for studies on sex differentiation in teleosts. Between 23 and 26 days after hatching (dah), it is possible to distinguish the ovary or testis under a microscope; this is enabled by the formation of the ovarian cavity in XX-undifferentiated gonads or the formation of the efferent duct in XY-undifferentiated gonads. Therefore, morphological sex differentiation begins 23–26 dah [3]. Moreover, the expression of genes related to sex differentiation in Nile tilapia begins at 5 or 6 dah. Thus, the period of molecular sex differentiation extends from 5 dah to the start of morphological sex differentiation [4].

In XY-undifferentiated gonads, the expression of gsdf and dmrt1 begins at 5 and 6 dah, respectively, and their expression levels increase; in contrast, in XX-undifferentiated gonads, the expression of these genes remains significantly low [4,5]. It has been reported that dmrt1 is the essential male pathway gene for testis differentiation and development [6]. In XY-undifferentiated gonads, between 5 and 25 dah, the levels of cyp11a1, hsd3b, cyp17a1, hsd17b1, and cyp19a1a remain low or zero. Beginning in the morphological sex-differentiation period, the expression of cyp11a1, hsd3b, and cyp17a1 increases [4]. In XY-undifferentiated gonads, the expression of cyp11c1 begins at 25 dah, and 11-ketotestosterone (11-KT) is synthesized for morphological sex differentiation and spermiogenesis [4]. In contrast, aromatase, which is encoded by cyp19a1a, is important for ovarian differentiation; both foxl2 and ad4bp/sf1 known as nr5a1 upregulate the transcription of cyp19a1a in Nile tilapia [4,7,8].

The hypothalamic–pituitary–gonadal axis in teleosts is the most important endocrine axis regulating reproduction [9]. Follicle-stimulating hormone (FSH) and luteinizing hormone (LH) are secreted by the pituitary and belong to the glycoprotein hormone family [10]. In humans, combined treatment with human chorionic gonadotropin and FSH was found to be effective; this treatment is able to increase the spermatogenesis rate in hypogonadotropic hypogonadism patients [11,12]. FSH plays a key role not only in Sertoli cell proliferation and differentiation but also in gametogenesis, whereas LH is mainly responsible for the final maturation of gametes [10,13,14]. The functions of FSH signaling were also investigated in teleosts. Single fshr knockout in XY medaka resulted in normal testis with fertility [15], whereas, in male Atlantic salmon, the loss of fshr resulted in abnormal testis development with low 11-KT [16]. In the pituitary of Nile tilapia, fsh initiates expression at 3 dah without a sexual dimorphic expression pattern between XX and XY during sex differentiation. The expression of fshr in XY gonads remains lower than that in XX gonads from 5 to 25 dah and increases after 25 dah [17]. A Nile tilapia-specific recombinant FSH (rFsh) with biological activity has been generated in HEK293F cells [18]. Considering the similar expression patterns of fshr and genes related to steroid biosynthesis, it is necessary to determine the function of FSH signaling in testis differentiation and development in Nile tilapia.

Here, we characterized the larvae between 5 and 10 dah as the initial stage of sex differentiation at the molecular level. The aim of this study was to elucidate the role of FSH in testis differentiation through the expression of genes related to androgen biosynthesis and testis differentiation. Accordingly, we microinjected rFsh into the larvae specifically at 5 to 6 dah, 7 to 8 dah, and 9 to 10 dah.

2. Materials and Methods

2.1. Experimental Animals

Adult Nile tilapia were reared in freshwater (26 °C) under a 14L:10D photoperiod in tanks and fed commercial trout pellets ad libitum (Marubeni Nisshin Feed Co., Ltd., Tokyo, Japan). Under these conditions, females can spawn repeatedly about every 14 to 18 days. Eggs were stripped on the day of spawning and fertilized by the usual drying method. All-female (XX) or all-male (XY) populations were obtained by the artificial fertilization of eggs from a normal female (XX) with sperm from a sex-reversed male (XX) or a super male (YY), respectively [3]. Fertilized eggs were cultured in a rotating glass tube at 26 °C. Fertilized eggs were hatched 4 days after fertilization. At 4 dah, all-female (XX) and all-male (XY) larvae were transferred into 60 L tanks.

2.2. Preparation of Micropipettes and Microinjection Plates

Micropipettes were fabricated by heating and pulling glass capillaries with filament (Narishige Scientific Instrument Lab, Tokyo, Japan) in a micropipette puller device (Narishige Scientific Instrument Lab, Tokyo, Japan). Melted 4% agarose prepared in distilled water was poured into a Petri dish. After the agarose solution had completely solidified, several wedge-shaped wells were made, and the agarose bed was immersed in fresh water, which was used to hold the tilapia larvae during the injection.

2.3. Preparation of Nile Tilapia rFsh and 1×PBS

rFsh synthesized by HEK293F cells [15] was stored at −80 °C until ready for use. On the day of microinjection, it was kept on ice for melting. Phosphate-buffered saline (PBS) was prepared with a composition of 10 mM sodium phosphate and 0.15 M sodium chloride, at a pH of 7.5.

2.4. Microinjection of rFsh into Genetic XX and XY Larvae

The rFsh solution was prepared freshly before experiments: rFsh was dissolved into PBS at a final concentration of 0.226 mg/mL. The freshly prepared rFsh solution was incubated on ice during microinjection. XX female fish and XY male fish were arbitrarily separated into three groups, respectively. The PBS-injection group and non-injection group were the control and blank groups; the rFsh treatment group received rFsh injections once a day at 5 to 6, 7 to 8, and 9 to 10 dah. The micropipette was attached to 5 mL syringe-fitted tubing, and the distal tip of the micropipette was slightly submerged in the working solution. The working solution was siphoned through the micropipette’s distal end by pulling the syringe’s plunger. Before injection, the tilapia larvae were anesthetized with ice and then transferred to the Petri dish with 4% agarose. Each tilapia larva was manipulated with forceps to make its body cavity visible under the dissecting microscope. For the PBS- or rFsh-injected groups, all the XX female fish and all XY male fish were placed on the plate containing 4% agarose and injected with 30 nL of PBS buffer or 30 nL of rFsh. The larva was allowed to recover in a plate at room temperature in freshwater after injection. The injection experiments and sampling at 6 dah, 8 dah, and 10 dah were repeated at least 3 times.

2.5. Gonad Isolation and Quantitative PCR

After 8 h of injection, the gonads were isolated from the XX female and XY male rFsh- or PBS-injected group. The gonads from the blank groups were isolated independently at 6, 8, and 10 dah. We isolated 40 to 50 gonads from the larvae at 6 dah, 30 to 40 gonads from the larvae at 8 dah, and 20 to 30 gonads from the larvae at 10 dah to ensure the quality of RNA extraction. After sampling, all the gonads were stored at −80 °C immediately until used for RNA extraction.

Total RNA was extracted using an RNA extraction kit (NucleoSpin RNA XS; Macherey-Nagel, Duren, Germany). The RNA extraction procedures were carried out 3 times independently as biological triplicates. cDNA synthesis was carried out using the ReverTra Ace kit (Toyobo Co., Ltd., Osaka, Japan) with random primers (Invitrogen, Carlsbad, CA, USA).

Quantitative PCR (qPCR) was performed with the PowerUp SYBR Green Master Mix (Thermo Fisher Scientific Inc.). Each qPCR mixture contained 2 μL of cDNA, 2.4 μL of double-distilled water, 0.3 μL of 10 mM forward primer, 0.3 μL of 10 mM reverse primer, and 5 μL of PowerUp SYBR Green Master Mix. The qPCR cycling conditions were as follows: 50 °C for a 2 min hold, 95 °C for a 2 min hold, and 40 cycles of amplification (95 °C for 3 s, 60 °C for 30 s). A melting curve analysis was performed for each reaction to confirm a single amplification. The qPCR primer pairs are listed in

Table 1. Primer sequences for qPCR.

Gene	Forward Primer	Reverse Primer	EFF%
cyp11a1 [4]	AAGTCTGGGCTTCGGCTTTG	CTCGAGAATGTGGATAAGAAAGAGTTG	93
hsd3b [4]	GGTGAATGTCAAAGGAACGCA	TTCTCCTGAATACATGCCTCCA	101
cyp17a1 [4]	GCTACGTGGAAGTTCCACAGGAA	GCCTCTGCACAAATGGTCTTCTC	90
hsd17b1 [4]	AAACATTCAAAGTGTATGCAACAATG	CAGGCCTTTCATGCTCTCTAAAA	93
cyp19a1a [4]	GCATAGGCACAGCCAGCAAC	GTGCACTGCTGAAGATCTGCTTAGTA	92
ad4bp/sf1 [4]	TCTCCAGTCTGGTCCAAAGAGGT	CCAGCAATTTTACATTTGGGTTGA	99
foxl2 [4]	AAGAGGAGCCGGTTCAGGACAA	GCTCTCCCGGATAGCCATGG	91
dmrt1 [4]	CGGCCCAGGTTGCTCTGAG	CCAACTTCATTCTTGACCATCA	95
gsdf [5]	ACCCGAAGCTGCCGTCTT	GACTGCTGGGGTTGCAGTATG	95
fshr [17]	CGGGCTGAGGATTTTTCCA	TGTTGTCCTGAAGATCCAGCAG	98
lhr [17]	CAGTGCAGAATATCAACAGCCTGA	TGTTAGAGATGCTCAAATATTCCAGCTT	92
β-actin [17]	CCCCAGGCATCAGGGTGT	TTGCTCTGGGCCTCATCAC	95

. qPCR was carried out using an Applied Biosystems StepOne plus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). A 10-fold dilution series of plasmids expressing target genes was used to construct a standard curve. The expression levels of the target genes were normalized to β-actin.

2.6. Statistical Analysis

The results are reported as the mean ± standard error (SE) (n ≥ 3). All data were tested for normality and homoscedasticity. A one-way ANOVA, followed by a post hoc Tukey–Kramer test for mean comparisons, was used to assess the significant differences in the data. The statistical analyses were performed using Excel Statistical Analysis Ver. 7.0 (ESUMI Co. Ltd., Tokyo, Japan) to determine the significant differences. Significance for the differences was set at p < 0.05.

3. Results

3.1. mRNA Levels of Genes Related with Steroid Synthesis at 6, 8, and 10 Dah

In XY Nile tilapia, the expression levels of cyp11a1 for the rFsh-injected group and controls were similar at 6, 8, and 10 dah (Figure 1A). The expression level of hsd3b was significantly higher in the rFsh-injected group compared with blank and PBS groups at 6 dah (Figure 1B), followed by an upward trend at 8 and 10 dah, without statistical significance thereafter (Figure 1B). Moreover, in XY Nile tilapia, the expression of hsd3b was similar between the control groups (Figure 1B).

In the XY Nile tilapia, the cyp17a1 levels tended to be higher in the rFsh-injected groups compared with controls at 6, 8, and 10 dah (Figure 1C). In XY gonads, the transcripts of hsd17b1 and cyp19a1a at 6, 8, and 10 dah were low and there was no significant difference among the three groups (Figure 1D,E).

3.2. mRNA Levels of foxl2 and ad4bp/sf1 at 6, 8, and 10 dah

In XY Nile tilapia, there was no significant difference in the mRNA levels of foxl2 among the treatments and controls at 6, 8, and 10 dah (Figure 2A). The ad4bp/sf1 levels in the rFsh treatments were similar to those in the blank groups and slightly higher than those in the PBS groups at 6 and 8 dah (Figure 2B). The mRNA levels of ad4bp/sf1 were also similar among the rFsh-injected group and the control groups at 10 dah (Figure 2B).

3.3. mRNA Levels of fshr and lhr at 6, 8 and 10 dah

In XY Nile tilapia, the expression levels of fshr showed nonsignificant differences (Figure 3A). The lhr level showed a tendency to be lower in the rFsh group compared with controls at 6 dah (Figure 3B), whereas the rFsh group at 8 dah showed a mild increase in lhr expression compared with the control (Figure 3B). At 10 dah, the lhr levels were similar among the rFsh-injected group and controls (Figure 3B).

3.4. mRNA Levels of dmrt1 at 6, 8, and 10 dah

In XY Nile tilapia, the dmrt1 level showed a minor increase in the rFsh groups compared with controls at 6 and 8 dah (Figure 4A). At 10 dah, the dmrt1 level was slightly elevated in the rFsh group compared with the PBS group but was similar to that in the blank (Figure 4A). Moreover, the expression of dmrt1 in the XY gonads tended to be higher from 6 to 10 dah, whereas the expression of dmrt1 in XX gonads remained at a low level from 6 to 10 dah (Figure 4).

In XX Nile tilapia, at 6 dah, the dmrt1 level showed a clear increase in the rFsh group (Figure 4B) and exerted a sustained effect during sex differentiation at the molecular level (Figure 4B). In the undifferentiated gonads of genetic XX tilapia, the dmrt1 expression was comparable between the controls at 6, 8, and 10 dah, respectively (Figure 4B).

3.5. mRNA Levels of gsdf at 6, 8, and 10 dah

Previous studies have reported that gsdf can promote testis differentiation in XY individuals [5,19]. We analyzed the relative expression of gsdf in the rFsh-injected group and controls to determine the effect of rFsh on testis differentiation via gsdf.

In XY Nile tilapia, gsdf had higher expression in the rFsh group than in the controls at 6 dah (Figure 5A); the gsdf levels in the rFsh group tended to be higher than those in the PBS group and were slightly higher than those in the no-injection group at 8 and 10 dah (Figure 5A). Moreover, the expression levels of gsdf in XX and XY gonads were similar at 6 dah, whereas gsdf levels were higher in XY gonads than in XX gonads at 8 and 10 dah (Figure 5).

In XX Nile tilapia, gsdf levels were significantly higher in the rFsh group compared with the controls at 6 and 8 dah (Figure 5B). At 10 dah, the expression of gsdf was slightly higher in the rFsh group than in the controls (Figure 5B). In the undifferentiated XX gonads, the gsdf levels were similar between the controls at 6, 8, and 10 dah, respectively (Figure 5B).

4. Discussion

To investigate the regulation of FSH signaling regarding its role in testis differentiation, qPCR analysis was performed to examine steroidogenic enzymes and testis-differentiation-related genes following rFsh injection. We first report the effects of rFsh on the expression of steroidogenic enzymes and testis-differentiation-related genes in XY-undifferentiated gonads during the early stage of sex differentiation at the molecular level.

In XY Nile tilapia, the expression of cyp11a1 and fshr begins at 5 dah, although levels remain low until 25 dah and then increase [4,17]. In the testis of Japanese eel, both cyp11a1 and fshr are expressed in Leydig cells [20,21]. The high concentration of Japanese eel-specific rFsh produced in Drosophila S2 cells can induce the expression of cyp11a1 in the testis in vitro [21]. In this study, we found that Nile tilapia rFsh was unable to induce cyp11a1 expression in the undifferentiated gonads of genetic XY fish during this stage at the molecular level; this might be due to the development stage (undifferentiated gonads). The positive effect of rFsh may be stronger during testis development.

In this study, the hsd3b levels in the undifferentiated gonads of XY Nile tilapia were higher in the rFsh-injected group. In XY Nile tilapia, hsd3b and fshr expression levels have been shown to increase from 25 dah onwards [4,17]. When the testicular fragments of male Japanese eels with 400–600 g body weight were cultured with Japanese eel-specific rFsh produced by HEK293F cells, high concentrations of rFsh induced the expression of hsd3b [22]. Recombinant human FSH21, which was hypo-glycosylated, induced hsd3b expression in porcine granulosa cells [23]. Therefore, in Nile tilapia, FSH signaling may also involve hsd3b expression during molecular and morphological sex differentiation to contribute to 11-KT biosynthesis during testis development.

In XY Nile tilapia, both cyp17a1 and fshr expression levels increase when morphological sex differentiation starts at 25 dah [4,17]. High concentrations of Japanese eel-specific rFsh produced by either HEK293F cells or Drosophila S2 cells can induce the expression of cyp17a1 when Japanese eel testes are cultured in vitro [21,22]. In this study, although we detected a weakly positive effect of rFsh on cyp17a1 expression, FSH signaling may involve cyp17a1 expression during molecular and morphological sex differentiation to contribute to androgen biosynthesis for testis differentiation and development.

It has been demonstrated that foxl2 and dmrt1 play antagonistic roles in the sex differentiation of Nile tilapia [4,7]. In this study, although rFsh was injected, the expression of foxl2 in the undifferentiated gonads of genetic XY tilapia remained low due to the high expression of dmrt1. In XY gonads, cyp19a1a was still low during the sex-differentiation stage at the molecular level due to the low foxl2 level, even though rFsh was injected.

In this study, the fshr level in genetic XY-undifferentiated gonads after rFsh injection was similar to that in the control groups. In XY Nile tilapia, almost no 11-KT was synthesized before 25 dah [4]. Although rFsh was injected into XY Nile tilapia before 25 dah, it did not have to express more fshr for 11-KT biosynthesis during molecular sex differentiation. When the testicular fragments of male Japanese eels with 400–600 g body weight were cultured with Japanese eel-specific rFsh produced by HEK293F cells, only the highest concentrations of Japanese eel-specific rFsh led to the significant induction of fshr expression [22]. Therefore, the amount of rFsh appears to influence whether fshr expression is induced in vivo or in vitro. The expression of lhr was not changed by rFsh injection in XY larvae, which indicates that FSH signaling is unable to induce lhr expression during this developmental stage. A similar result has been observed in XX larvae during this period, when genetic XX Nile tilapia was injected with rFsh [18].

In the undifferentiated gonads of XY Nile tilapia, the dmrt1 level increased during the molecular sex-differentiation process in the rFsh-injected group. To confirm the relationship between FSH signaling and dmrt1, rFsh was also injected into genetic XX female Nile tilapia during the same period, as dmrt1 remained low in the undifferentiated gonads of genetic XX females. In this study, the expression of dmrt1 was significantly increased by the injection of rFsh. However, dmrt1 might still be repressed by E2 in XX Nile tilapia. Therefore, the expression of dmrt1 remained at a low level and did not reach the level of that in undifferentiated gonads in XY male Nile tilapia at 10 dah. Recently, it has been reported that dmrt1 was important in sex reversal from female to male when Nile tilapia were treated with androgen during sex differentiation [24]. Moreover, after 25 dah in the gonads of XY Nile tilapia, as fshr expression increases, the expression of dmrt1 also increases during morphological differentiation [4,17]. In postnatal rat Sertoli cells, FSH signaling directly activates dmrt1 transcription [25]. In the orange-spotted grouper, a hermaphroditic fish, porcine FSH can induce female-to-male sex reversal, affecting sex fate, by increasing the expression of dmrt1 [26]. Moreover, dmrt1 promotes testis differentiation and development not only in Nile tilapia but also in Chinese tongue sole [27]. However, during molecular sex differentiation in Nile tilapia, the expression of dmrt1 began at 6 dah and then increased, whereas the expression of fshr remained low until 25 dah. Therefore, in Nile tilapia, although FSH signaling may be involved in dmrt1 expression, it may not be the most important factor.

Although the mRNA levels of gsdf were upregulated in the rFsh-injected groups at 6, 8, and 10 dah in XX and XY Nile tilapia, the relationship between FSH and gsdf remains unknown. It has been shown that dmrt1 and ad4bp/sf1 can increase the expression level of gsdf according to promoter assays in Nile tilapia [19], spotted scat [28], and gibel carp [29]. Moreover, gsdf expression was detected in dmrt1 homozygous knockout XY Nile tilapia, whereas dmrt1 expression was absent in gsdf homozygous knockout XY Nile tilapia. The gsdf gene might be downstream of dmrt1 [19]. Therefore, increased expression of gsdf might be stimulated by increased expression of dmrt1.

In summary, in XY Nile tilapia, although FSH signaling has a positive effect on the expression of hsd3b and cyp17a1 (and is thus involved in androgen biosynthesis), FSH signaling is not important during molecular sex differentiation; the expression levels of cyp11a1, hsd3b, and cyp17a1 were low during this period. After 25 dah, the expression of fshr, cyp11a1, hsd3b, and cyp17a1 increased, suggesting that FSH signaling has roles in activating steroidogenesis to synthesize 11-KT during morphological sex differentiation in XY Nile tilapia. Moreover, based on the expression patterns of dmrt1 in both XX and XY Nile tilapia after rFsh injection, FSH signaling may be involved in dmrt1 expression. Therefore, FSH signaling plays a role in testis differentiation and testis development through modulating androgen synthesis and dmrt1 expression.