| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Reproductive Technology; |
Institute of Bioscience and Biotechnology3 and Department of Aquaculture,4 National Taiwan Ocean University, Keelung 202, Taiwan
ABSTRACT
Protandrous black porgy fish, Acanthopagrus schlegeli, have a striking life cycle, with a mono-male sex differentiation at the juvenile stage and male-to-female sex change at 3 yr of age. We report for the first time integrative molecular data on these interesting phenomena. Sex differentiation occurred between 4 and 5 mo of age. Testicular nr5a4 transcripts increased to high levels during sex differentiation (5 mo old), whereas nr0b1 (Dax-1) did not increase until the age of 8 mo. High nr5a4 and nr0b1 expression in testicular tissue, in contrast to low nr5a4 and high nr0b1 expression in ovarian tissue, were found in the male phase of 0+- to 2-yr-old fish (before sex change). Increased nr5a4, decreased nr0b1, and increased cyp19a1a were found in the ovarian tissues undergoing development from primary oocytes to vitellogenic oocytes during the natural sex change in 2+-yr-old fish. Removal of testicular tissue in 1+-yr-old fish resulted in both increased ovarian nr5a4 and genes in the steroidogenic pathway and decreased nr0b1 together with the appearance of vitellogenic oocytes. Ovary developed into the active stage with the increased expression of star and steroidogenic enzymes, including aromatase, in concordance with the decreased expression of nr0b1 in the testis-excised fish. Long-term estradiol (E2) administration resulted in early sex change, but the ovaries were mainly with primary oocytes. Low nr5a4, high nr0b1, and low steroidogenic enzymes, including cyp19a1a expression, were also observed in these E2-fed ovarian tissues. Thus, nr5a4 but not nr0b1 was associated with male sex differentiation. Testicular development required cooperative functions of both nr5a4 and nr0b1. The present study suggests that nr5a4 and nr0b1 have an antagonistic interaction for the oocyte development. Testicular tissue exerted inhibitory effects on ovarian development. It is probable that nr0b1 regulates the timing of vitellogenic development and sex change in black porgy.
early development, gene regulation, oocyte development, ovary, testis
Black porgy, Acanthopagrus schlegeli Bleeker, a marine protandrous hermaphrodite fish, have a striking life cycle. The fish are functional males for the first two years of life but begin to sexually change to females during the third year [1, 2]. However, only about 40% of black porgies change to females, whereas the rest remain functional males during the third spawning season. Bisexual gonad with testicular and ovarian tissue separated by connective tissues was found in black porgy before sex change. The changes occurring in gonad tissue during the process of natural sex change are from a "bisexual gonad" to "full ovarian gonad," with primary oocytes and regressed testicular tissue, and then finally to an "advanced (vitellogenic) ovary" [3]. Vitellogenic oocytes did not appear in the ovary until the complete regression of testicular tissue. The development of vitellogenic oocytes is an important and critical event for the success of natural sex change.
High levels of plasma estradiol-17β (E2) during the prespawning and spawning season were correlated with the natural sex change in 3-yr-old black porgy [1, 2]. Application of exogenous E2 to sex-differentiated black porgy of 0+ and 1+ yr old for the induction of sex change resulted in a sex change from male to female, whereas in 0+-yr-old fish, the ovaries remained at the primary oocyte stage, followed by a reversible sex change (from an E2-induced female to male) after withdrawal of E2 administration [4]. In 1+-yr-old black porgy, only a few vitellogenic oocytes were observed following a treatment with E2 for at least 5–6 mo [5, 6]. Thus, the development of vitellogenic oocytes is important for the success of E2-induced sex change in 1+-yr-old fish. Furthermore, cyp19a1a (P450arom, aromatase), the enzyme responsible for conversion of testosterone (T) to E2, was necessary for the occurrence of natural sex change in black porgy, as demonstrated by the ability of aromatase inhibitor to block natural sex change [7]. It was concluded that E2 may be important for both controlled and natural (from 2+ to 3 yr of age) sex change in protandrous black porgy [5, 7, 8]. The molecular mechanism of the development of vitellogenic oocytes during sex change is still unknown.
Nr5a1 (steroidogenic factor 1, Sf-1; or adrenal 4 binding proten, Ad4BP) has been implied as an important factor for steroidogenesis and sex differentiation in animals (in favor of male: rat [9], pig [10], and turtle [11]; in favor of female: chicken [12], American alligator [13], and American bullfrog [14]). Nr0b1 (dosage-sensitive sex reversal, adrenal hypoplasia congenital critical region on the X-chromosome, gene 1; Dax-1) is an orphan nuclear receptor that regulates the expression of multiple steroidogenic enzymes in mice [15–17]. Nr0b1 also was reported to downregulate nr5a4-mediated cyp19a1a expression in ovarian follicles in medaka [18]. Nr0b1 also is reported to be important for either the female or male sex differentiation in various species (in favor of male: mouse [19] and frog [20]; in favor of female: pig [10] and chicken [21]; no sex dimorphism: American alligator [13], sea turtle [22], and teleost-tilapia [23]). In addition to sex differentiation, the involvement of nr5a4 and nr0b1 in steroidogenesis further implies the possible roles of these sex-related genes in the late gonadal development, which has not yet been well studied. Therefore, the temporal expression and possible roles of nr5a4 and nr0b1 during male sex differentiation and gonadal development in protandrous black porgy is of great interest.
In the present study, black porgy was chosen to study the gene profiles in 1) male sex differentiation (i.e., from an undifferentiated stage to a male in juvenile [0+-yr-old fish]), because of the mono-male sex pattern observed in this species during sex differentiation; 2) the natural sex change from a male to female in 2+- to 3-yr-old fish; 3) the controlled sex change to a female ovary induced by long-term E2 administration in 0+- or 1+-yr-old fish; 4) reversible sex change from an E2-induced female to a male after E2 withdrawal in 0+-yr-old fish; and 5) the interaction of testicular and ovarian tissue in the bisexual gonad by in vivo manipulation of testicular tissue and ovarian tissue. We propose that, in addition to the possible roles in sex differentiation, nr5a4 and nr0b1 may also interact with each other and regulate steroidogenesis, and thus play important roles in the gonadal development, especially in relation to sex change in the protandrous black porgy.
Four batches of black porgy, A. schlegeli (from a private farm in Chiayi county, Taiwan), were obtained for the experiments. Total body weight and gonad weight were measured for the calculation of gonadosomatic index (GSI = gonad weight/body weight x 100%). Fingerlings 2 mo old (0+-yr-old fish; n = 270; body weight [BW]: 0.48 ± 0.02 g; standard body length [BL]: 3.25 ± 0.05 cm) were obtained from a pond culture in April to study sex differentiation and for the experiment on E2 administration. Black porgy 1+ yr old (n = 70; BW = 394.39 ± 11.65 g; GSI = 0.14 ± 0.02) were obtained in August for the E2 administration experiment, and another batch (n = 40; BW = 82.0 ± 2.69 g; BL = 16.6 ± 0.21 cm) was obtained for the experiment on testis excision. Black porgy 2+ yr old (n = 40; BW = 363.93 ± 18.19 g; GSI = 1.04 ± 0.19) were obtained in April 2002 to study natural sex change. Sex change was identified by the presence of vitellogenic oocytes during the spawning period (January to March 2003).
The experimental fish were acclimated to the pond condition at the university culture station in seawater and natural light system. The water temperatures ranged from 19°C to 26°C. The fish were fed with commercial feed (Fwa Sou Feed Co., Taichung, Taiwan) ad libitum. All procedures and investigations were approved by the National Taiwan Ocean University institutional animal care and use committee and were performed in accordance with standard guiding principles.
Experiment 1: morphological changes and gene profiles of male sex differentiation and E2-induced female sex differentiation in 0+-yr-old fish. Sexually undifferentiated juveniles were obtained to examine the gene profiles and morphological changes in gonads during sex differentiation (Fig. 1). Fish (3 mo old) were divided equally into two groups: control and E2-administered groups (6 mg E2 per kg of feed) in May 2002. Estradiol treatment by oral administration was conducted for 3 mo, from July to October. Fish gonads (n = 8 fish) were collected monthly from May (3 mo old, undifferentiated gonad) to October (8 mo old, differentiated gonad) for gene analyses (n = 8 fish each group) and histology (n = 6 fish per group).
|
Experiment 2: gene profiles of natural sex change in 2+- to 3-yr-old fish. In order to examine gene profiles in gonads during natural sex change (Fig. 1), 2+-yr-old black porgy were collected (n = 8 fish) at 2-mo intervals from July 2002 (before sex change) to January 2003 (after sex change). Gonad samples were collected to examine the status of sex change (male or female) and also for gene analyses.
Experiment 3: gene profiles of reversible sex change in 0+-yr-old fish. In order to further clarify the gene profiles during testicular differentiation, reversible sex change experiments were conducted (Fig. 1). Estradiol administration was withdrawn from the E2-administered fish (from experiment 1) in October 2002, and E2-terminated (n = 56) and control fish (n = 52) were fed with a control diet. Reversible sex change (from an E2-induced female to a male) occurred during the period of E2 termination (October 2002 to January 2003). Gonad (testicular and ovarian tissues) from E2-terminated fish (n = 8 fish per month) was collected monthly from October 2002 to January 2003 for gene analyses (n = 8 fish per group) and histological studies (n = 6 fish per group).
Experiment 4: gene profiles of long-term E2 administration in 1+-yr-old fish. In order to examine the long-term effects of E2 (Fig. 1), 1+-yr-old black porgy were divided into two groups (n = 16 per group)—control and E2-administered (6 mg/kg feed) groups—in September 2001. Gonads were collected after 1.5 and 4.5 mo of E2 administration for gene analyses.
Experiment 5: the effects of testis excision in the ovarian development in bisexual gonad of 1+-yr-old fish. In order to investigate the interaction between testicular and ovarian tissue in the bisexual gonad (Fig. 1), 1+-yr-old fish were divided into testis-excised and sham groups in June 2005. Testicular tissue was carefully and completely removed from the bisexual tissue (keeping the ovarian tissue intact) in the testis-excised group by in vivo surgical methods. Sham group received the surgery, but the testis was not removed (intact bisexual gonad). Gonad and plasma were collected in October 2005 (n = 10 fish per group) and January 2006 (n = 10 fish per group) to examine the gene expression, aromatase activity, and gonad histology. Full gonad was collected to calculate GSI. Plasma E2 concentrations were measured by E2 enzyme immunoassay.
Parameters Related to Gonad Development
Steroidogenesis-related factors or enzymes such as star (steroid acute regulatory protein, StAR), cyp11a1 (cytochrome P450 side-chain cleavage enzyme, P450scc), cyp17a1 (17
-hydroxylase/C-17–C-20 lyase, P450c17), hsd3b1 (3β-hydroxysteroid dehydrogenase/
5,4-isomerase, 3βHSD), cyp11b2 (11β-hydroxylase, P45011β), and cyp19a1a (P450arom) were measured to study the changes in their expressions with respect to nr5a4 and nr0b1 expressions. The important steroidogenic enzyme, cyp11b2, the enzyme responsible for the conversion of 11β-hydroxyandrostenedione to 11-ketotestosterone mainly in the testicular tissue [24], and cyp19a1a mainly in ovarian tissue [25] were also monitored.
Total RNA (1 µg) extracted from gonad of the representative fish was reverse transcribed to the first-strand cDNA using Superscript II (Invitrogen; Carlsbad, CA) with the oligo (dT)12–18 primers. This first-strand cDNA was used for the PCR and quantitative real-time PCR analyses. Genes of nr5a4 (GenBank accession no. AY491379), nr0b1 (EF 423617), cyp19a1a (AY 273211), cyp11b2 (EF 423618), and gapdh (glyceraldehyde-3-phosphate dehydrogenase; DQ 399798) were cloned from black porgy, and specific primers were designed for PCR (Table 1). As an internal control, gapdh was used. Polymerase chain reaction conditions were as follows: 94°C for 1 min, 52.5°C for 1 min, and 72°C for 1.5 min for 30 cycles. This number of PCR cycles was preliminarily tested and was in the range of the linear curve for the relationship between the number of cycles and the amount of PCR product.
|
Quantification of nr5a4, nr0b1, cyp19a1a, and cyp11b2 Transcripts by Quantitative Real-Time PCR Analysis
Quantitative real-time PCR analysis for gene transcripts of nr5a4, nr0b1, star (GenBank accession no. AY870248), cyp19a1a (AY870246), cyp17a1 (AY870249), hsd3b1 (AY674377), cyp19a1a (AY 273211), cyp11b2, and gapdh were conducted according to the methods described in previous studies [26, 27]. Specific primers were designed for the quantitative real-time PCR (Table 2). Gene quantification of standards, samples, and controls were conducted simultaneously by a real-time PCR (GeneAmp 5700 Sequence Detection System; Applied Biosystems, Foster City, CA) with SYBR Green I as a double-stranded DNA minor-groove binding dye. The correlation of the respective standard curve of log (transcript concentrations) versus CT (the calculated fractional cycle number at which the PCR fluorescence product is detectable above a threshold) was –0.999. The values detected from different amount RNA (10 times of series dilution) of the representative samples were parallel with the respective standard curve. The transcript values of each gene were calibrated with internal control gapdh and then normalized (x100%).
|
Gonadal Sex Differentiation and Sex Change: Histology and dmrt1 Levels
Gonadal tissue was fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned at 5 µm. Transverse sections were stained with hematoxylin-eosin or immunohistochemical staining. Gonad development in relation to the stages of male and female germ cells was determined. Gonadal sex differentiation was further confirmed by the quantification of dmrt1 transcripts. DM-related transcription factor 1, a (double-sex/Mab-3)-domain gene (dmrt1) is an important sex differentiation gene in animals included black porgy (GenBank accession no. AY323953) [28]. Specific primers were designed for the quantitative real-time PCR for dmrt1 (Table 2).
Measurement of Aromatase Activity and E2 Enzyme Immunoassay
Gonad tissue was homogenized with a potassium phosphate buffer (100 mM KCl, 10 mM KH2PO4, 1 mM EDTA, and 10 mM dithiothreitol; pH 7.4) and centrifuged at 1000 x g for 10 min at 4°C. Aromatase (Cyp19a1a) activity was measured in the supernatant with 1β-[3H] androstenedione as a substrate according to previous studies in black porgy [7]. Aromatase activity was expressed as fentomoles of 3H2O per hour per milligram of protein (fentomole/h.mg protein). Plasma E2 was extracted with ethyl ether and then measured with an enzyme immunoassay (Cayman Chemical Co., Ann Arbor, MI).
All data are expressed as mean ± SEM. The values were subjected to analysis by one-way ANOVA, followed by a Student-Newman-Keuls multiple test with P < 0.05 as significant difference. Student t-test was also conducted to compare the significant difference (P < 0.05) between treatments.
Morphologic Changes and Gene Profiles of Male Sex Differentiation and E2-Induced Female Sex Differentiation in 0+-Yr-Old Fish (Experiment 1)
Gonad morphology during male sex differentiation and E2-induced female sex differentiation. In control fish, undifferentiated gonad tissue with early germ cell (oval shape with 12.5 µm and 6.7 µm in average cell diameter) was observed in 3- and 4-mo-old fish (May and June 2002; Fig. 2, a and b). Spermatogonia (oval-sphere shape with 8.6 µm and 7.1 µm in average cell diameter) started to appear in July (5 mo old; Fig. 2c). Lobular testicular tissue with late spermatogonia and spermatocytes appeared in August to September (6–7 mo old; Fig. 3, a and b). Active spermatogenesis was observed from October 2002 to January 2003 (8- to 11-mo-old; Fig. 3, c, g, and h), and mature sperm were found in the control gonad in January 2003 (spawning period, 11 mo old; Fig. 3h). Ovarian tissue formed a very small portion in the bisexual gonad in 0+- to 1-yr-old fish only with few primary oocytes in central cavity of the bisexual gonad (data not shown).
|
|
Testicular differentiation to male germ cells (spermatogonia) in the 5-mo-old gonad was further confirmed by the specific expression of a testis-specific gene: dmrt1. The dmrt1 was differentially expressed in the gonad tissue at 4 mo and 5 mo of age. Gonadal dmrt1 transcripts increased at 5 mo (differentiated testis, with 5-fold increase) compared with 4-mo-old gonad (undifferentiated, very low and constant levels).
After 1 mo of E2 administration, development of ovarian tissue was detected at 6 mo of age in August 2002 (Fig. 3d). Many primary oocytes were found in these ovarian tissues at 6 and 7 mo of age (August and September; Fig. 3, d and e). Full ovary with primary oocytes together with the complete regression of testicular tissue was detected in the E2-treated fish at 8 mo of age (E2 administration for 3 mo; Fig. 3f).
Gonadal nr5a4, nr0b1, and cyp19a1a during male sex differentiation. The levels of nr5a4 expression were low in May to June 2002 (3–4 mo old, before sex differentiation), followed by a significant increase to high levels (3.4-fold) in gonad (mainly testicular tissue) from July (sex differentiated) to September, which reached the highest levels (11-fold) in October 2002 (Fig. 4a). The expression levels of nr0b1 did not change significantly during the period from May to September (3–7 mo old), but increased significantly (2.3-fold) in the control fish in October 2002 (Fig. 4b). Transcripts of cyp19a1a remained at the same levels during male sex differentiation from May (3 mo old) to August (6 mo old), and then significantly decreased (3.7-fold) in September and October 2002 (Fig. 4c).
|
Gonadal nr5a4, nr0b1, cyp11b2, and cyp19a1a Gene Profiles in 1+- to 3-Yr-Old Fish During Natural Sex Change (Experiment 2)
Testicular tissue. According to RT-PCR data (Fig. 5A), nr5a4 and nr0b1 were expressed at high levels in 1+- and 2+-yr-old testicular tissue (male phase, Fig. 5A). Transcripts of cyp11b2 were also high in testicular tissue in 1+- and 2+-yr-old fish (Fig. 5A). Expressions of nr5a4, nr0b1, and cyp11b2 were at low levels in the regressed testicular tissue of 3-yr-old female (lane 3 in Fig. 5A). Quantitative real-time PCR (Fig. 5, a–c) further demonstrated that testicular nr5a4 remained at constant high levels from July 2002 to January 2003 in 2+- to 3-yr-old males (Fig. 5a). However, testicular nr0b1 and cyp11b2 transcripts decreased by 4.7-fold and 2.7-fold, respectively, and reached the lowest levels in January 2003 in 3-yr-old males compared with the data in July 2002 (Fig. 5, b and c).
|
Ovarian tissue. RT-PCR analyses (Fig. 5B) revealed that ovarian nr5a4 transcripts were low in 1+-yr-old fish (in male phase, with inactive primary oocytes in ovarian tissue) and 2+-yr-old fish (in male phase before sex change, with active primary oocytes in ovarian tissue), and became high in the vitellogenic ovary of 3-yr-old sex-changed female in January 2003 (Fig. 5B). In contrast, nr0b1 transcripts were high in ovarian tissue in 1+-yr-old male (lane 1 in Fig. 5B) and then dramatically decreased to low levels in ovarian tissue of both 2+-yr-old males (before sex change) and 3-yr-old females (lanes 2 and 3 in Fig. 5B). Ovarian cyp19a1a transcripts were low in both 1+- and 2+-yr-old males and then increased in 3-yr-old females with vitellogenic ovary (Fig. 5B). During the postspawning season (with regressed ovarian tissue), profiles with low nr5a4, high nr0b1, and low cyp19a1a were found in the ovarian tissue in 3+-yr-old females (lane 4 in Fig. 5B).
Quantitative real-time PCR data (Fig. 5, d–f) showed that nr5a4 transcripts remained low in the ovarian tissue in 1+- and 2+-yr-old fish from July to November 2002 and increased (3.2-fold increase in January compared with the tissue in July in 2+-yr-old fish) in the female ovary in January 2003 (Fig. 5d). Ovarian nr0b1 transcripts decreased by 2.7-fold in 2+-yr-old fish (July 2002) compared with 1+-yr-old fish (Fig. 5e), followed by a further decrease (3-fold) during the process of sex change from bisexual gonad (July 2002) to female ovary (January 2003; Fig. 5e). In contrast, ovarian cyp19a1a transcripts gradually increased (2-fold increase in November 2002 and 13.9-fold increase in January 2003) from bisexual gonad (July 2002) to female ovary (January 2003) in 2+- to 3-yr-old fish (Fig. 5f).
Gonadal nr5a4, nr0b1, and cyp19a1a in the E2-induced female sex differentiation in 0+-yr-old fish. Sex change, together with the development of ovarian tissue, was observed in 0+-yr-old fish in August after E2 administration (from July to October 2002; Fig. 3, d–f). Transcripts of nr5a4 decreased significantly (3.8-fold) in the gonad in August 2002 (after 1 mo of E2 administration) and remained at low levels in the induced ovary during E2 treatment (August to October 2002; Fig. 4a). Transcripts of nr0b1 decreased significantly (2.5-fold; P < 0.05) in August and September and then increased (P < 0.05) in October 2002 (2.5-fold compared with September 2002; Fig. 4b). In contrast, cyp19a1a transcripts were significantly high (1.1-fold, P < 0.05) in E2-administered group compared with the control in August 2002 (Fig. 4c). But cyp19a1a transcripts were significantly decreased (1-fold, P < 0.05) from August to October 2002 (Fig. 4c).
Gonadal nr5a4, nr0b1, and cyp19a1a During the Reversible Sex Change in 0+-Yr-Old Fish (Experiment 3)
Estradiol administration induced an early sex change, and fish became E2-induced females in October in 0+-yr-old fish. Estradiol administration was withdrawn (E2 termination), and fish were fed with a control diet from October 2002 to January 2003. Reversible sex change from an E2-induced female to male occurred in these E2-terminated fish. No clear testicular tissue could be collected in October and November 2002 in the E2-terminated group. After E2 termination (November 2002 to January 2003), testicular tissue regenerated in concomitance with the regression of ovarian tissue, and the gonad became a bisexual gonad (Fig. 3, i and j). After 1 mo (November 2002) and 3 mo (January 2003) of E2 termination, the development of spermatids and sperm was observed in testicular tissue of the previously E2-treated group (Fig. 3, i and j), which was similar to the control (Fig. 3, g and h). The transcripts of nr5a4, nr0b1, and cyp19a1a showed no change in ovarian tissue after E2 termination (November 2002 to January 2003) compared with the levels in the ovary in October 2002 (Fig. 6). Transcripts of nr5a4 were significantly higher (3.1-fold in December 2002, 8.9-fold in January 2003) in testicular tissue than in the ovarian tissue in December 2002 to January 2003 (Fig. 6a), and reached the levels similar to the testis of the control male in October 2002, as shown in Figure 4a. Transcript levels of nr0b1 were high and similar to the levels in the regenerating testicular tissue and regressing ovarian tissue in December 2002 to January 2003 in the E2-terminated fish (Fig. 6b). In contrast, cyp19a1a transcripts were significantly lower in testicular tissue in E2-terminated fish in December 2002 (1.9-fold) and January 2003 (16.5-fold) than the respective ovarian tissue (Fig. 6c).
|
Gene Profiles of Long-Term E2 Administration in 1+-Yr-Old Fish (Experiment 4)
Long-term E2 administration stimulated nr0b1 but suppressed nr5a4 expression and steroidogenesis in the ovarian tissue. In the long-term experiment, E2 administration for 4.5 mo caused the regression of testicular tissue and the development of ovarian tissue, and fish became females with the ovaries containing mainly primary oocytes. Quantitative real-time PCR analyses revealed that the transcripts of nr5a4 (4-fold), star (1.5-fold), cyp11a1 (0.7-fold), and hsd3b1 (3-fold) significantly decreased, whereas nr0b1 transcripts significantly increased (1.3-fold) in ovarian tissue after 1.5 mo of E2 administration compared with the control fish (Fig. 7). Expressions of cyp19a1a and cyp17a1 were not different in ovarian tissue between control and 1.5 mo of E2-fed fish (October; Fig. 7).
|
The differential expression of nr5a4, nr0b1, and cyp19a1a in the inactive and active ovarian tissue. We define ovarian tissue with primary oocytes in the bisexual gonad of 1+-yr-old fish as the inactive stage and vitellogenic ovary (after natural sex change) as the active tissue. Low nr5a4, high nr0b1, but low cyp19a1a expression levels were detected in the regressing ovarian tissue during the reversible sex change from an E2-induced female (full ovary with primary oocytes) to male (a bisexual gonad, maturing testis together with regressing ovarian tissue) after 3 mo of termination of E2 feeding (lane 1 in Fig. 8). Similarly, faint nr5a4, high nr0b1, and low cyp19a1a expression profiles were also found in the ovarian tissue in 1+-yr-old fish (inactive ovarian tissue; lane 2 in Fig. 8). In contrast, high nr5a4, low nr0b1, and high cyp19a1a expression profiles were found in the vitellogenic ovary (active stage; lane 3 in Fig. 8).
|
The Effects of Testis Excision in the Ovarian Development in Bisexual Gonad of 1+-Yr-Old Fish (Experiment 5)
The development of vitellogenic oocytes after removal of testicular tissue in the bisexual gonad. Three months after surgery, GSI in the control and testis-excised groups of 1+-yr-old fish in October 2005 (prespawning season) was significantly different (P < 0.05) and was 0.16 ± 0.04 versus 0.27 ± 0.01, respectively. GSI increased with the approaching spawning season and reached the highest value in both control and testis-excised groups in January 2006 (4.60 ± 0.91 vs. 5.76 ± 0.81; P > 0.05), respectively.
Primary oocytes were observed in the ovary in October 2005, with the oocyte diameter being significantly different (P < 0.05) in control and testis-excised groups (33.2 ± 1.2 µm vs. 48.8 ± 2.9 µm, respectively). Vitellogenic oocytes (356.7 ± 12.9 µm) and mature oocytes (770.5 ± 5.3 µm) appeared in the ovaries in the testis-excised group in January 2006 (no testicular tissue in this group). In contrast, ovarian tissue formed a very small portion of bisexual gonad (testicular tissue as the main tissue) and was mainly with small primary oocytes (21.2 ± 1.5 µm) in control group (as males) in January 2006.
Decrease of nr0b1 and increase of nr5a4 and steroidogenesis in the ovarian tissue after testis-removal. Quantitative real-time PCR showed that transcripts of nr5a4 and genes in steroidogenic pathway, such as star (34-fold), cyp11a1 (2.8-fold), cyp17a1 (2.1-fold), hsd3b1 (4-fold), and cyp19a1a (14-fold), significantly increased (P < 0.05) in female ovary in the testis-excised group in January 2006 compared with ovarian tissue in the control group (Fig. 9a). In contrast, nr0b1 transcripts decreased (2.4-fold; P < 0.05) in the female ovary compared with the control fish in January 2006 (Fig. 9a). Furthermore, ovarian aromatase activity (25-fold; Fig. 9b) and plasma E2 concentrations (1.6-fold; Fig. 9c) increased significantly in the testis-excised group in January 2006 compared with the control group (Fig. 9, b and c).
|
We report for the first time molecular data on sex differentiation and sex change together with the in vivo integrative interaction of nr5a4, nr0b1, and steroidogenic enzymes in a nonmammalian model animal, protandrous black porgy fish. Despite increasing concern on sex determination and differentiation, only a few studies, using mammalian animals, have been performed to date to investigate the possible roles of nr0b1 in the development of advanced oocytes.
Histological analysis showed that gonad sex differentiation occurred between 4 and 5 mo of age in black porgy. The characteristics of mono-male sex differentiation let us consider that spermatogonia instead of oogonia appeared in the male gonad. Gonad dmrt1 transcripts, a differentiated testis-specific gene in some teleosts, including black porgy [28], was further quantified in this study to confirm that a gonad at 5 mo of age differentiated to male testis with spermatogonia compared with the undifferentiated gonad at 4 mo of age.
Expression of nr5a4 increased significantly during testicular sex differentiation. The experiment on reversible sex change in 0+-yr-old fish showed that nr5a4 was also highly expressed in the regenerating testicular tissue during the reversible sex change in December. The importance of nr5a4 expression in testicular differentiation and development in black porgy is in agreement with previous reports in other species. Expression of nr5a1 was higher in testicular tissue than in ovarian tissue during sex differentiation in rat [9], pig [10], and turtle [11]. Disruption of nr5a1 in male mice resulted in the testis regression and persistence of mullerian structure [29]. Expression of nr5a1 was mainly in Leydig cell and Sertoli cell in mouse [30] and rat [9]. In contrast, nr5a4 expression was high in the differentiating ovary compared with testicular gonad in chicken [12], American alligator [13], and American bullfrog [14]. It is considered that nr5a1 is required for steroidogenesis and several P450 steroid hydroxylases in the active gonad [30, 31].
There was not significant nr0b1 expression during early male sex differentiation in black porgy, but expression increased in the differentiated gonad at 8 mo of age, when the fish underwent intensive spermatogenesis. Therefore, nr0b1 is probably more important in the testicular spermatogenesis and development than sex differentiation in this species. The importance of nr0b1 in sex differentiation and development in either male or female gonad had been reported in several other species. The nr0b1 was expressed at higher levels in male than in female gonads of frog, and these changes in nr0b1 expression were related to female-to-male sex reversal in frog [20]. But no sex difference in the expression of nr0b1 during gonadal differentiation was found in the teleost-tilapia [21], American alligator [13], and sea turtle [22]. In chicken, higher nr0b1 transcripts were found in female than in male during late gonadal development [21].
Expression of nr5a4 was much higher in the testicular tissue than in ovarian tissue in the bisexual gonad from 0+- to 2-yr-old fish (male phase). Expression of nr5a1 increased in the active gonad of either male or female phase in gobiid fish [32]. The nr0b1, in addition to nr5a4, was also actively expressed in the testicular tissue of black porgy in the male phase from 10-mo-old to 2+-yr-old fish. High expression levels of nr5a4 and nr0b1 genes in the developing testicular tissue were highly correlated, indicating a possible interaction between them. The expression of cyp11b2, the key steroidogenic enzyme for the production of 11-ketotestosterone, was also positively correlated to the expression of nr0b1, when nr5a4 expression remained at high levels in active testicular tissue. In contrast, nr5a4, nr0b1, and cyp11b2 expressions were low in the regressed testis when the fish underwent sex change to female. It is therefore suggested that nr5a4 and nr0b1 may cooperatively and positively regulate cyp11b2 enzyme expression in the testis of black porgy.
It has been shown that nr5a1 plays a central role in the differentiation and development of steroidogenic tissues, such as gonad and adrenal [33]. The nr5a1 regulates the promoter activities and controls the expression of all steroidogenic enzymes and some genes related to sex determination and differentiation, such as sry, amh (anti-mullerian hormone), and nr0b1 in mammals [33]. Knockdown of nr5a1 resulted in gonadal agenesis or sex reversal in mice [33, 34]. The nr5a4 may also act at the cyp19a1a promoter and enhances cyp19a1a expression in fish [35]. Increased nr5a1 is in parallel with the occurrence of female phase in a sex-changing gobiid fish [32]. Our data also demonstrated the close correlation of nr5a4 expression with male sex differentiation in juvenile and the natural sex change to female in adult. It is suggested that nr0b1 plays a crucial role in testis differentiation by regulating the formation of intact testis cords in mice [19]. The nr0b1 is not an ovarian-determining gene, but rather has a critical role in spermatogenesis according to studies in nr0b1-deficient mice [36, 37]. Sex reversal from a male to female occurred in the mutant nr0b1 XY mice [38]. The nr0b1 could regulate the expression of multiple steroidogenic enzymes in mice [15–17] and also suppress nr5a4-mediated cyp19a1a expression in medaka [18].
In the long-term steroid administration experiment, E2 suppressed nr5a4 and stimulated nr0b1 expression in the ovarian tissue of 0+- and 1+-yr-old black porgy. E2 administration caused the expression profiles of nr0b1 and cyp19a1a in a reciprocal manner and further decreased gene expression in steroidogenic pathway. Estradiol administration also resulted in a decreased expression of testicular nr5a2 in Arctic char [39]. Methyl testosterone administration decreased ovarian nr5a1 expression in the ovary of grouper [40]. No published information is available on the regulation of nr0b1 by sex steroids. Therefore, we suggest that E2 administration induces ovarian development, which is still at the inactive stage. In the inactive ovarian tissue, the general expression profiles were low nr5a4, high nr0b1, and low cyp19a1a. The profiles with low nr5a4 and high nr0b1 are not in favor of the development of vitellogenic oocytes, as indicated by low expression of cyp19a1a. Our data on the negative correlation between the high expression of nr0b1 and low expression of steroidogenic enzymes were consistent with previous data in mammalian cells. Nr0b1 inhibited the promoter activity of star, cyp11a1, and hsd3b1 [15] and suppressed star gene expression, leading to a dramatic decrease in steroid production [17]. Nr0b1 has also been reported to repress the Nr5a1-mediated transactivation of the cyp19a1a promoter, but it did not inhibit other proteins (e.g., star and cyp17a1) in mice [16] and medaka (nr5a4) [18].
We report for the first time that cyp19a1a was expressed in very high levels during the period of male sex differentiation and early development of gonad (3–6 mo old), and then decreased with the advancement in spermatogenesis (7 and 8 mo old). After E2 termination in 0+-yr-old fish, reversible sex change occurred with the E2-induced ovary reverting to a bisexual gonad, consisting of regenerating testicular tissue and regressing ovarian tissue. Higher cyp19a1a expression was also detected in the developing testicular tissue (December) than in the mature testis (January). From these data on male sex differentiation and reversible sex change, it is suggested that aromatase enzyme is important for the early sex differentiation of male gonad in the protandrous fish. Our previous studies also revealed that low dose of exogenous E2 (1.0 or 0.25 mg/kg feed) given to 8-mo-old sexually differentiated male black porgy stimulated testicular development and spermiation [41]. In contrast, high E2 administration (4–6 mg/kg feed) induced early sex change to a female ovary in 0+-yr-old black porgy [4–6].
In adult fish, cyp19a1a expression increased during the course of natural sex change from July 2002 to January 2003 in 2+- to 3-yr-old fish. Therefore, cyp19a1a was highly correlated to the ovarian development during natural sex change. The cyp19a1a may be important for the development of vitellogenic oocytes [42, 43]. The critical role of Cyp19a1a in the natural sex change (3-yr-old fish) was previously demonstrated by applying fadrozole (an aromatase inhibitor) to 2+-yr-old black porgy [7]. We report for the first time that the increase of ovarian cyp19a1a expression was correlated with the increase of nr5a4 and decrease of nr0b1 in the ovarian tissue during the natural sex change. In the present study, decreased nr0b1 transcripts are considered to relieve the suppression on genes in the steroidogenic pathway, including cyp19a1a. In mouse, several possibilities have been considered to explain the relief for the suppression by Nr0b1: by binding to the DNA hairpin structure, recruiting the corepressor, or affecting at the transcriptional/posttranscriptional levels [15–17, 44, 45]. In contrast, nr5a1 has been suggested to activate nr0b1 expression in both male and female mice [46, 47].
Testis excision experiment conducted on 1+-yr-old fish resulted in an enhanced expression of nr5a4 and genes in steroidogenic pathway and a decreased expression of nr0b1 in the ovarian tissue in the testis-excised fish compared with the control group. During natural sex change in 2+-yr-old fish, the decrease of ovarian nr0b1 occurred prior to the increase of ovarian nr5a4 expression. Thus, similar profiles of high nr5a4 and low nr0b1 in the ovarian tissue (vitellogenic ovary) were found both in the testis-excised and natural sex-changed black porgy. High cyp19a1a expression and aromatase activity in ovarian tissue and high E2 concentrations in plasma were also detected in fish with active ovary, where the ovary had high nr5a4 and low nr0b1 expression in both the testis-excised and natural sex-changed fish.
Removal of testicular tissue resulted in the ovarian tissue becoming the active tissue in September 2005 (3 mo after removal of testis), as revealed by increased oocyte diameter and further sex change. Thus, we induced by a nonchemical approach an early sex change in the 1+-yr-old black porgy. The removal of testicular tissue in 1+-yr-old, testis-excised fish mimicked natural sex change (regressed testis and development of ovarian tissue), which occurs only in 2+-yr-old fish. Our findings suggest for the first time that testicular tissue exerts inhibitory effects on the development of ovarian tissue and vitellogenic oocytes. Further experiments on the effect of the removal of ovarian tissue (keeping the testicular tissue intact) are also being carried out currently on 2+-yr-old fish with the intention of investigating whether the ovariectomized fish will remain as male (with testicular gonad) or change to a female (with ovarian gonad) during the period of natural sex change at 3 yr of age. Results of the testis-excised experiment in the present study, together with the results from the ongoing ovariectomized experiment, may provide valuable information about the interaction between testicular and ovarian tissue in the bisexual gonad.
Estradiol administration was not able to maintain the appropriate levels of nr5a4 and nr0b1 similar to the microenvironment in the testis-excised ovary, and the induction of a vitellogenic ovary by E2 administration was not easy to obtain. We suggest that high nr5a4 and low nr0b1 expression in the ovarian tissue is important for the further gonad development, such as vitellogenic oocytes, and the success of sex change in the protandrous black porgy. Some factors from testicular tissue may play important roles on the neighboring tissue (ovarian tissue) by regulating the expression of nr5a4 and nr0b1 in the ovarian tissue. Studies on the direct role of nr0b1 in controlling sex change are being conducted by knockdown approach.
In conclusion, the present study demonstrates that protandrous black porgy is an unique model fish to study male sex differentiation and development. The data revealed that nr5a4 but not nr0b1 expression was important for the early male sex differentiation in black porgy. High nr5a4 and nr0b1 favored the maintenance of testicular development in male phase. In contrast, low nr5a4 and high nr0b1 expression was found in ovarian tissue before sex change. Therefore, our data suggest that nr5a4 and nr0b1 have a cooperative function for the testicular development and an antagonistic interaction for oocyte development. The expression levels of nr0b1 probably regulate the timing of oocyte development and vitellogenesis for sex change in the protandrous black porgy. The successful induction of an early sex change in the testis-excised fish further suggests that testicular tissue may exert some inhibitory effects on the advanced ovarian (vitellogenic) development in bisexual gonad. Our results provide new insight into the development of advanced oocytes during the natural sex change in black porgy.
ACKNOWLEDGMENTS
We are grateful to Dr. Sylvie Dufour (USM 0401, UMR 5178 CNRS/MNHN/UPMC, Biologie des Organismes Marins et Ecosystèmes, Département des Milieux et Peuplements Aquatiques, Muséum National d'Histoire Naturelle, Paris, France) for the suggestion and English editing.
FOOTNOTES
1Supported by the National Science Council (Taiwan). 
Correspondence: 2FAX: 886 2 24621579; e-mail: B0044{at}mail.ntou.edu.tw
Received: 4 May 2007.
First decision: 7 June 2007.
Accepted: 19 October 2007.
REFERENCES
-methyltestosterone-induced precocious sex change Mol Reprod Dev 2007 74665–673[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
  G.-C. Wu, S. Tomy, M. Nakamura, and C.-F. Chang Dual Roles of cyp19a1a in Gonadal Sex Differentiation and Development in the Protandrous Black Porgy, Acanthopagrus schlegeli Biol Reprod, December 1, 2008; 79(6): 1111 - 1120. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
