| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a AgResearch, Wallaceville Animal Research Centre, Upper Hutt, New Zealand
ABSTRACT
The aim of this study was to determine 1) the time of onset and cellular localization of gene expression for steroidogenic factor-1 (SF-1), steroidogenic acute regulatory protein, 3ß-hydroxysteroid dehydrogenase/
5,4 isomerase (3ß-HSD), and the cytochrome P450 enzymes for cholesterol side-chain cleavage (P450scc), 17
-hydroxylase (P45017OH), and aromatase (P450arom) during gonadal development; and 2) the amount of progesterone, androstenedione, testosterone, and 17ß-estradiol present in the fetal sheep gonad. Fetuses were collected on Days 24, 26, 28, 30, 32, 35, 40, 55, and 75 of gestation, and gene expression was determined by in situ hybridization. The steroid content of gonads collected on Days 30, 35, 55, and 75 of gestation was determined by RIA. Developing gonads collected from both male and female fetuses were steroidogenically active around the time of morphological sexual differentiation. In the female, the steroidogenic cells were initially located at the boundary of the cortex and medulla but become increasingly restricted to the mesonephric-derived cell streams. In the male, once tubules were identifiable, steroidogenesis was restricted to the interstitial regions. Interestingly, expression of both SF-1 and 3ß-HSD was observed prior to morphological sexual differentiation. In addition, expression of both of these genes was more widespread than the other genes in both males and females.
developmental biology, gene regulation, ovary, steroid hormones, testis
INTRODUCTION
Sexual differentiation in sheep occurs around Days 30–32 of fetal life [1, 2], and in common with many mammals [3, 4] there is evidence that steroidogenesis occurs in the gonad at this time [5–7]. It is known that fetal androgen biosynthesis is required for expression of the male phenotype [4]. In contrast, estrogen synthesis does not appear to be essential for normal development of the female phenotype [4]. However, the role that steroids may play in the development of the gonads in both males and females is not well understood. In cases of androgen insensitivity, structural differences, including disorganization of the seminiferous tubules and a thickened basal lamina, have been observed in the testis [8, 9]. Estradiol-17ß has been shown to act as a germ cell survival factor in the human testis [10], and exposure of pregnant and lactating rats to secoisolariciresinol diglycoside, a naturally occurring lignan with antiestrogen properties, permanently affected ovarian function of their offspring [11].
De novo synthesis of androgens from cholesterol requires the sequential involvement of the cytochrome P450 side-chain cleavage enzyme (P450scc), 3ß-hydroxysteroid dehydrogenase/5,4 isomerase (3ß-HSD), 17-hydroxylase/C17–20 lyase (P45017OH), and 17-ketosteroid reductase. Cytochrome P450 aromatase (P450arom) can convert testosterone to 17ß-estradiol [12]. In addition, steroidogenic acute regulatory protein (StAR) facilitates transfer of cholesterol to the inner mitochondrial membrane, which is a rate-limiting step in steroidogenesis [13]. The expression of StAR, 3ß-HSD, and the steroid P450 hydroxylases is regulated by the orphan nuclear receptor, steroidogenic factor 1 (SF-1) [14].
In sheep, SF-1 mRNA was detected at Day 25 of fetal life. Thus, steroidogenic capability may develop in the gonad before sexual differentiation [7]. In fetal sheep testes, androstenedione and testosterone are present from about Day 30 of fetal life [5, 6, 15], and low levels of estrogens were detected at Day 40 [15] or after Day 70 [5]. More recently, Sweeney et al. [16], using immunocytochemistry, detected 3ß-HSD in fetal-type Leydig cells prior to and after seminiferous cord formation. With respect to the fetal sheep ovary, 17ß-estradiol secretion was detected from around the time of sexual differentiation at Day 31 until Day 62 of fetal life but not beyond [6]. Payen et al. [7], using reverse transcription-polymerase chain reaction (RT-PCR) have demonstrated aromatase gene expression in the fetal sheep ovary from Day 30 through to birth, with maximal expression between Days 35 and 49 of fetal life. Recently, Lun et al. [15] have shown the presence in sheep fetal ovaries of androstenedione and progesterone at 32 days, 17ß-estradiol at 35 days, and testosterone at 40 days of fetal life. There remains however, a paucity of data on the ontogeny and cellular localization of gene expression for the key regulatory proteins of the steroidogenic pathway in sheep gonads. Thus, the main aim of this study was to examine the spatiotemporal pattern of expression of mRNA encoding SF-1, StAR, 3ß-HSD, P450scc, P45017OH, and P450arom in male and female fetal sheep gonads before and after sexual differentiation.
MATERIALS AND METHODS
All experiments were performed with the approval of the Animal Ethics Committee at Wallaceville Animal Research Centre in accordance with the 1987 Animal Protection (Codes of Ethical Conduct) Regulations of New Zealand. Except where indicated, laboratory chemicals were obtained from BDH Chemicals New Zealand Ltd. (Palmerston North, New Zealand), Gibco BRL (Auckland, New Zealand), or Roche Diagnostics N.Z. Ltd. (Auckland, New Zealand).
Collection of Tissue Samples
Sheep fetuses were recovered at 24, 26, 28, 30, 32, 35, 40, 55, and 75 days of gestation following transfer of 4- to 5-day-old embryos to Romney ewes (gestation = 147 days). The fetuses were the results of matings of Romney ewes (with or without the Booroola gene) superovulated using eCG (Folligon; Intervet, Lane Cove, New Zealand) and Ovagen (Immuno-chemicals Products Ltd., Auckland, New Zealand) as described [17]. The superovulated ewes mated either individually or in pairs with a ram (with or without the Booroola gene). Four or 5 days after mating, embryos were recovered and transferred into estrous-synchronized recipient ewes (three embryos per ewe using a laparoscopic technique). Fetuses (with no or two copies of the Booroola gene) were recovered after a barbiturate overdose (Euthatal; South Island Chemicals, Christchurch, New Zealand) was given to their mothers. The presence or absence of the Booroola gene did not appear to affect the onset of expression of the mRNAs or gonadal steroids. For in situ hybridization, the fetuses or dissected fetal gonadal-mesonephroi complexes were fixed in 4% (w/v) phosphate-buffered paraformaldehyde and embedded in paraffin wax. For determination of steroid content, fetal gonads/mesonephroi were dissected into right and left gonad, weighed, and frozen at -70°C until RIA. The sex of fetuses recovered before Day 55 of gestation was determined by PCR using SRY gene-specific primers from tissue sections as described [18]. The sex of fetuses recovered at Days 55 and 75 was determined by examination of external genitalia.
Complementary DNA Cloning of Ovine SF-1 and P45017OH
Total cellular RNA was isolated from ovine tissues using Trizol (Gibco BRL) according to the manufacturer's instructions. Polyadenylated mRNA was isolated from total cellular RNA using the Oligotex mRNA purification system (Qiagen; Biolab, Palmerston North, New Zealand) according to the manufacturer's instructions.
For generation of ovine (o)SF-1, first-strand cDNA was produced using 4–5 µg of adult ovarian mRNA using GibcoBRL's SuperScript preamplification system for first-strand cDNA synthesis. A 1383-base pair (bp) cDNA was generated by PCR using the standard PCR buffer from Roche Diagnostics N.Z. Ltd. (1x concentration 10 mM Tris, 1.5 mM MgCl2, 50 mM KCl, pH 8.3) using 2.5 units of Taq DNA polymerase, 0.2 mM dNTPs, 0.2 µM each primer, 1/20 of the cDNA synthesis reaction mixture, and forward ATGGACTATTCGTACGACGAGGACC (corresponding to bases 120–144 of the bovine sequence) [19] and reverse AGTCTGCTTGGCCTGCAGCATCTCG (corresponding to bases 1478–1502 of the bovine sequence) [19] primers. Following an initial cycle of 3 min at 95°C, 3 min at 58°C, and 5 min at 72°C, oSF-1 cDNA was amplified using 35 cycles of denaturing at 95°C for 30 sec, annealing at 62°C for 1 min, and extension at 72°C for 2 min. For generation of oP45017OH, first-strand cDNA was produced using 1 µg of total cellular RNA isolated from a fetal ovary collected on Day 35 of gestation using a kit for first-strand cDNA synthesis (Roche Diagnostics N.Z. Ltd.). A 360-bp cDNA was generated by PCR using the standard PCR buffer from Roche and forward GCCAGCAACACGAGAACTTCTTC (corresponding to bases 186–208 of the ovine sequence; GenBank accession L40335) and reverse GATCTATGGACTGTCCATGCTGG (corresponding to bases 523–545 of the ovine sequence) primers. Following an initial cycle of 3 min at 95°C, 3 min at 65°C, and 5 min at 72°C, oP45017OH cDNA was amplified using 35 cycles of denaturing at 95°C for 30 sec, annealing at 63°C for 1 min, and extension at 72°C for 2 min. The PCR samples were then incubated at 72°C for 10 min and cooled to 4°C before analysis by gel electrophoresis. The products obtained from the ovarian cDNA were ligated into the pCR2.1 vector (oSF-1; Invitrogen; GeneWorks Pty Ltd., Adelaide, SA, Australia) or pGEMTeasy vector (oP45017OH; Promega) and their nucleotide sequences determined by automated sequence analysis (Waikato DNA Sequencing Facility; The University of Waikato, Hamilton, New Zealand). The obtained sequences were compared with known SF-1 or oP45017OH sequences to confirm identity of the amplified product [20]. The oSF-1 insert was released with XbaI and KpnI restriction enzymes for use in Northern analysis and subcloned into pGEM7Z (Promega; Dade Diagnostics PTY Ltd., Auckland, New Zealand) for use in in situ hybridization analysis. The oP45017OH insert was released by the EcoRI restriction enzyme for use in Northern analysis.
Specificity of generated cDNAs was tested with Northern analysis as previously described [21]. Samples (SF-1: 5 µg poly[A]+ RNA isolated from follicles, ovarian stroma, corpus luteum, liver, heart, adrenal, and kidney; P45017OH: 30 µg total cellular RNA from liver, heart, adrenal, ovary containing luteal tissue, fetal kidney, and fetal testis [Day 135 of gestation]) were separated on a 1.5% agarose-formaldehyde-3[N-morpholino]propanesulfonic acid gel, transferred to a nylon filter (Hybond-NX; Amersham Pharmacia Biotech New Zealand, Auckland, New Zealand) by capillary action, and cross-linked by UV light. Filters were prehybridized at 42°C overnight and hybridized at 42°C for approximately 24 h. The final wash was in 0.1x SSC (15 mM sodium chloride, 1.5 mM sodium citrate), 0.1% SDS at 65°C for 30 min. Filters were exposed to Kodak X-AR film (Eastman Kodak, Rochester, NY) for 2.5 (P45017OH) or 10 (SF-1) days.
In Situ Hybridization
Cellular localizations of mRNAs were determined using the in situ hybridization protocol described previously with minor modifications [18]. Sense and anti-sense RNA probes were generated from cDNA encoding oSF-1 (described above), oStAR (provided by Dr. G. Niswender, Animal Reproduction and Biotechnology Laboratory, Colorado State University, Ft. Collins, CO) [22], bP450scc (provided by Dr. M. Waterman, Department of Biochemistry, Vanderbilt University, Nashville, TN) [23, 24], o3ß-HSD (provided by Dr. G. Niswender) [25], oP45017OH (described above), and bovine (b)P450arom (provided by Dr. E Simpson, Prince Henry's Inst., Monash Medical Center, Clayton, Victoria, Australia) [26] with T3, T7, or SP6 RNA polymerase using the Riboprobe Gemini system (Promega). For all in situ hybridizations, 4- to 6-µm sections were incubated overnight at 50°C (P450scc and P450arom) or 55°C in hybridization solution containing 30 000–100 000 cpm/µl of 33P-labeled antisense RNA. Nonspecific hybridization of RNA was removed by RNase A digestion followed by stringent washes (2x SSC, 50% formamide, 65°C and 0.2x SSC at 37°C). Following washing, sections were dehydrated, air dried, and coated with autoradiographic emulsion (LM-1 emulsion; Amersham Pharmacia Biotech New Zealand). The emulsion-coated slides were exposed at 4°C for 2–3 wk. Slides were then developed and fixed. The sections were stained with hematoxylin and viewed and photographed using both light- and darkfield illumination on an Olympus BH-2 microscope. Nonspecific hybridization was monitored by hybridizing at least one tissue section from each sex and age with approximately equal concentrations of the sense mRNA for each gene. No specific hybridization was observed for any section hybridized with the sense mRNA for any gene.
Immunocytochemistry
Previously, we have noted that fetuses collected following embryo transfer are delayed in their development by a few days. Sweeney and coworkers have characterized expression of 3ß-HSD protein in developing testis collected from fetuses with a known single insemination date [16]. Therefore, we determined the ontogeny of 3ß-HSD protein expression in fetuses collected on Days 28–40 (n = 3–4 per age for each sex) using the same antibody (kindly supplied by Professor I.J. Mason, Department of Clinical Biochemistry, University of Edinburgh, UK) under similar conditions. The purpose of this work was to compare the developmental stage of the gonads in the present study with those in previous reports where embryos were recovered from animals that were mated naturally. Localization of 3ß-HSD protein was as previously described [18] using 8 µg IgG per ml of S683 antisera (rabbit anti-3ß-HSD) as the primary antibody. To increase sensitivity of the procedure, gonads collected from fetus on Days 28–40 and Day 75 (n = 3–4 per day for each sex) were analyzed with an additional enhancement step employing the Tyramide Signal Amplification (TSA)-Biotin System (NEN Life Science Products, Boston, MA). Briefly, immediately preceding the visualization of the immunoperoxidase activity the selected sections were incubated with biotinyl tyramide for 10 min at room temperature, washed three times (5 min) in Tris-buffered saline, and incubated for an additional 30 min at 20°C with horseradish peroxidase-labeled streptavidin. In all cases the avidin, biotin, primary antibody, secondary antibody, and avidin-biotin complex were diluted in the blocking buffer supplied with the TSA-Biotin System kit.
Determination of Gonadal Steroid Contents
Amounts of progesterone, androstenedione, testosterone and 17ß-estradiol were determined in tissue homogenates by RIA as described [15]. Due to the antisera for testosterone from Diagnostic Products Corporation (DPC, Los Angles, CA) becoming unavailable; antiserum obtained from Biogenesis (catalog no. 8680-6004; Poole, England) was utilized to determine testosterone concentrations in some samples. Validation of the antiserum included demonstration of parallelism as determined by assaying varying amounts of several standard samples and a determination of cross-reactivity. A high correlation (r2 = 0.98) was noted when assaying the standard samples with the two antisera. The major differences between the two antisera were decreased cross-reactivity with 5-dihydrostestosterone (4.65 versus 34%) and an increased cross-reactivity with 5ß-dihydrostestosterone (47.7 versus <0.01 %) for the Biogenesis versus the DPC antiserum. Sensitivities (90% of zero binding value) of the assays averaged 9, 14, 5, and 1.5 pg/sample for progesterone, androstenedione, testosterone, and 17ß-estradiol, respectively. Intra- and interassay coefficients of variations were 13% and 17%, 15% and 13%, 9% and 17%, and 7% and 15% for progesterone, androstenedione, testosterone, and 17ß-estradiol, respectively.
Statistical Analysis
All steroid values were transformed (natural logarithm) to normalize variation prior to statistical analysis, and data are presented as geometric means with 95% confidence limits. A one-sample Student t-test was performed comparing steroid content to a value 2 SD above the detection limit of the assay to determine if the steroid content of the gonads was significantly greater than the detection limit of the assay. The SD was determined by the interassay coefficient of variation of a standard sample run in each assay with approximately 80–85% binding.
RESULTS
Characterization of cDNAs Encoding oSF-1 and P45017OH
The nucleotide sequence of the cDNA encoding oSF-1 (GeneBank accession number AF299081) was 98%, 92%, and 92% identical to the bovine [19], human [27], and porcine [28] SF-1 cDNA, respectively. The deduced amino acid sequence of the cDNA was 99%, 94%, and 96% identical to the bovine, human, and porcine deduced amino acid sequence, respectively. The oSF-1 cDNA hybridized to bands at approximately 13.7, 7.1, 5.6, 4.6, and 3.0 kilobases (kb) in adrenal mRNA, and at 7.1, 5.6, 4.6, and 3.0 in ovarian follicular, stromal, and luteal mRNA (data not shown). No hybridization was observed in mRNA isolated from liver, heart, or kidney (data not shown). Previously, mRNAs of similar sizes have been shown in ovine adrenal and luteal poly(A)+ RNA following Northern analysis using a partial oSF-1 cDNA [29]. The high degree of homology between the generated oSF-1 cDNA and the characterized bovine Ad4BP/SF-1 [19], human SF-1 [27], and porcine SF-1 [28] cDNAs as well as the hybridization of this cDNA to transcripts in ovine tissues known to express SF-1 mRNA but not to mRNA isolated from tissues devoid of expression indicated that the isolated cDNA recognizes oSF-1 mRNA.
The nucleotide sequence of the generated oP45017OH cDNA was identical to that published in GeneBank (L40335). The oP45017OH cDNA hybridized to a single band of approximately 1.9 kb in RNA isolated from adrenal and fetal testis (data not shown). No hybridization was observed in RNA isolated from luteal ovary, fetal kidney, heart, or liver (data not shown). Previously, mRNA of a similar size has been shown in ovine adrenal following Northern analysis using a 45-mer oligonucleotide probe based on ovine sequence [30]. Thus, the generated partial cDNA encoding oP45017OH specifically detects mRNA encoding P45017OH.
In Situ Hybridization
SF-1: Females Expression of mRNA encoding SF-1 was not observed in three out of four animals on Day 24 of gestation (Fig. 1, a and b; Table 1). However, SF-1 mRNA was consistently observed on Day 26 of gestation (Fig. 1, c and d) and continued to be expressed through Day 75 (Fig. 1, e–h). In the female, expression of SF-1 mRNA was widespread throughout the gonad (Fig. 1, c–h), including cells of the surface epithelium as well as those of the cell streams (Figs. 1, g and h, and 6, a and b). The cell streams are composed of cells of the involuting giant glomeruli of the mesonephros that migrate into the fetal sheep ovary between Days 40 and 75 of gestation [31].
|
|
SF-1: Males The mRNA for SF-1 was not observed at Day 24 (Fig. 2, a and b; Table 2) but was observed in all other time points (Fig. 2, c–h; Table 2). In the male, SF-1 mRNA expression was widespread throughout the testis until development of the tunica albuginea (Fig. 2, c and d). Expression of SF-1 mRNA was not observed in the tunica albuginea at any age. Expression of SF-1 mRNA was most intense in the somatic cells outside the seminiferous tubules (Fig. 2, e–h) but was also observed within the seminiferous tubules (Fig. 2, g and h).
|
|
3ß-HSD: Females Expression of 3ß-HSD mRNA was first observed on Day 30 of gestation (Fig. 3, a–d; Table 1). Initially, strong widespread expression punctate in pattern was observed (Fig. 3, c–f); however, after Day 40, the intensity of the signal appeared to decrease (Fig. 3, g and h). At later ages, light widespread expression of mRNA encoding 3ß-HSD was sometimes observed (Fig. 3, g and h); often, however, the signal was most intense or limited to the mesonephric-derived cell streams (data not shown). Two animals at Days 55–75 did not express detectable amounts of 3ß-HSD mRNA (Table 1).
|
3ß-HSD: Males As in females, mRNA encoding 3ß-HSD was first observed at Day 30 of gestation in the developing testis (Fig. 4, a–d; Table 2). In the male, while 3ß-HSD mRNA was not observed in the tunica albuginea (Fig. 4, e and f), expression was widespread throughout the rest of the testis and observed both inside and outside of the developing tubules (Fig. 4, g and h). Labeling did however appear stronger outside than inside the tubules.
|
P450scc, P450arom, P45017OH, and StAR: Females
The patterns of expression for mRNAs encoding P450scc, P45017OH, P450arom, and StAR were similar. At Day 32 of gestation most animals did not express the mRNAs (e.g., P45017OH, Fig. 5, a and b; Table 1); however, expression was observed in some animals (Table 1). By Day 35, all animals expressed the mRNAs (e.g., P45017OH, Fig. 5, c and d; Table 1). Initially, expression of these mRNAs was detected in a band of cells located in the innermost regions of the ovarian cortex and outermost regions of the medulla (e.g., P45017OH, Fig. 5, c and d). Expression became increasing restricted to the mesonephric-derived cells streams by Day 75 (Figs. 5, e and f and 6, c–h), and expression of these mRNAs was not detectable in some animals at Days 55 and 75 (Table 1).
|
P450scc, P450arom, P45017OH, and StAR: Males
The patterns of expression for P450scc, P45017OH, and StAR were similar. At Day 32 of gestation most animals did not express the mRNAs (e.g., P45017OH, Fig. 7, a and b; Table 2); however, expression was observed in a limited number of the animals (Table 2). Thereafter, all animals expressed the mRNAs (e.g., P45017OH, Fig. 7, c and d; Table 2). Expression of these mRNAs was not observed in the tunica albuginea (e.g., P45017OH, Fig. 7, c and d) and was restricted to the interstitial regions once tubules formed (e.g., P45017OH, Fig. 7, e and f). The mRNA encoding P450arom was not detected in the developing testis at any age (Table 2).
|
Immunocytochemistry
With the less sensitive method, 3ß-HSD was not observed in the developing ovary at any age tested (n = 3–4 animals per age). In the male, staining for 3ß-HSD was not above background at Days 28 and 30 of gestation; thereafter, expression was observed in one of three animals at Day 32, two of three animals at Day 35, and four of four animals at Day 40.
With the more sensitive procedure that included the additional enhancement step employing the TSA-Biotin System, 3ß-HSD protein was detected in the developing ovary beginning at Day 30 of gestation (Fig. 8, a, c, and e). The pattern of protein expression was similar to that of mRNA expression, being punctate but widespread throughout the ovary (Fig. 8, c and e). Only nonspecific staining was observed when the primary antibody was replaced with the nonimmune rabbit serum (Fig. 8g). In the male, strong staining for 3ß-HSD protein was first detected on Day 30 of gestation (Fig. 8d). A faint signal was observed in one of the animals at Day 28 (data not shown); however, no signal was observed in the other two animals (Fig. 8b). Expression was punctate with intense staining observed outside the developing seminiferous tubules (Fig. 8, d, f, and h). However, weaker staining was also observed within the tubules (Fig. 8h).
|
Gonadal Steroid Contents
At Day 30, the mesonephric-gonadal complex did not contain detectable amounts of progesterone, androstenedione, testosterone, or 17ß-estradiol (Tables 3 and 4). In the female, 17ß-estradiol was detected from Day 35 onward and androstenedione and progesterone were present on Day 75 (Table 3). However, significant amounts of testosterone were not detected at any day examined (Table 3). In the male, both androstenedione and testosterone were measurable from Day 35 onward (Table 4), and all steroids examined were detectable on Days 55 and 75 of gestation (Table 4). Gonadal weights are provided to allow assessment of tissue concentrations of steroids at each age (Tables 3 and 4).
|
|
DISCUSSION
In this study the expression of a number of genes encoding regulatory proteins and enzymes involved in the steroidogenic pathway was examined in relation to the differentiation and development of sheep gonads during fetal life. In general, it was not until between Days 32 and 35 of gestation (around the time of sexual differentiation) that the expression of genes at all levels of the steroidogenic pathway was observed in the gonads of either sex. The finding that expression of mRNAs encoding proteins important for steroidogenesis is strongly correlated with steroidogenic capability in the developing gonad is supported by the detection of steroids on Day 35 and later but not at Day 30 of gestation (present data) [15]. In the female, estrogen appeared to be the predominant steroid synthesized, whereas in the male the androgens were the major steroids produced, although detectable amounts of both progesterone and 17ß-estradiol were observed on Days 55 and 75 of gestation. Earlier studies have indicated a slightly earlier onset of expression of steroidogenic enzymes [7, 16] and steroidogenesis [5] in gonadal development (i.e., Day 30). This apparent discrepancy is most likely due to technical differences. All gonads studied in the present experiment were collected from fetuses following the transfer of three embryos to each recipient ewe. This procedure appears to delay development by a few days as demonstrated by the delay of about 2 days in the onset of 3ß-HSD protein when detected by very similar immunocytochemical methods using the same antibody (present study) [16].
While expression of mRNAs encoding 3ß-HSD and SF-1 occurred prior to morphological sexual differentiation in the developing testis; expression of StAR, P450scc, and P45017OH was not observed until around the time of sexual differentiation. As the seminiferous tubules form, StAR, P450scc, and P45017OH mRNAs were all observed to be limited to the interstitial regions to putative fetal-type Leydig cells, indicating that these cells are the source of androgens in the developing testis. Similar expression patterns for P45017OH have been observed in the fetal testis of the pig [32], rat [33], and mouse [34]. However, both SF-1 and 3ß-HSD gene expression was observed inside as well as outside the tubules. Whether this was associated with germ cells and/or Sertoli cells could not be definitively determined in all cases. However, it was evident that many germ cells were not expressing either of the mRNAs or 3ß-HSD protein. P450arom mRNA was not detected in the fetal testis at any stage examined. This is in agreement with the finding of Payen et al. [7]. Very low concentrations of estrogens have been detected in the fetal testis in some studies (present study) [5, 15] while other studies did not report such a finding [6]. The reason for the apparent discrepancy among the studies is most likely due to the sensitivity of the assays because ovine fetal testis have detectable P450arom activity [7] and synthesize 17ß-estradiol in culture [15]. A very low amount of P450arom expression widely distributed throughout the testis might allow for detectible P450arom activity in the whole testis with expression of mRNA in individual cells remaining below the sensitivity of the in situ hybridization assay.
The SF-1 gene was expressed in male and female gonads well before those of StAR and the steroidogenic enzymes are first consistently observed in the developing ovary and testis at Day 26 of gestation. This is 2 days after the thickening of the coelomic epithelium is observed at the site where the gonad will develop. Similarly, using RT-PCR, Payen et al. [7] were able to detect SF-1 mRNA beginning on Day 25 of gestation in developing sheep gonads. In mice, SF-1 is expressed in the urogenital ridge beginning at an embryonic age of 9 days [35] and in transgenic mice lacking SF-1, the urogenital ridge fails to develop, leading to complete gonadal agenesis [36, 37]. In the mouse ovary, SF-1 gene expression and protein levels decline during sexual differentiation and remain low or undetectable until follicular formation. In contrast, in the developing sheep ovary, mRNA encoding SF-1 is expressed at high levels during and after sexual differentiation [7] (present study). This discrepancy may be related to the difference in the interval between sexual differentiation and first meiotic prophase between sheep and mice and the lack of steroidogenic activity in the ovary of mice during this time [38]. Intensity of signal for SF-1 mRNA was strongest in the same cell types as mRNAs encoding the steroidogenic enzymes and StAR: namely the cell streams in females and the interstitial region containing the fetal-type Leydig cells in males. However, SF-1 mRNA was also observed in other cell types within the seminiferous tubules in males and in mesenchymal cells adjacent to the newly forming follicles in females. In combination, the much earlier onset of expression of SF-1 mRNA and more widespread distribution of SF-1 mRNA is consistent with other functions for SF-1 in the developing gonad in addition to its role in the regulation of steroidogenic enzymes [14].
Interestingly, expression of 3ß-HSD mRNA and protein was not as restricted as mRNAs encoding StAR, P450scc, P45017OH, and P450arom in both males and females. This might suggest that the cells within the fetal gonad obtain the ability to convert pregnenolone to progesterone without the ability to synthesize pregnenolone. Little is known about the concentration or potential actions of pregnenolone in the developing gonad. Previously, Sweeney et al. [16] had demonstrated expression of 3ß-HSD protein beginning on Day 30 of gestation in males and hypothesized that the gonads were steroidogenically active at this time. However, they did not localize any other proteins important in steroidogenesis in that experiment. It appears that expression of 3ß-HSD precedes that of the other proteins important for steroid synthesis by 2–5 days. In addition, we detected mRNA and protein both inside and outside the seminiferous tubules. This is in contrast to the results presented by Sweeney et al. [16] and Majdic et al. [33] in which they localized 3ß-HSD protein only to the interstitial cells of sheep and rats. This discrepancy is most likely due to differences in sensitivity of the immunocytochemistry technique; however, differences between species cannot be ruled out. Because the presence of 3ß-HSD without mRNA encoding the other proteins necessary for steroid synthesis is seen in both the male and female, it seems likely that this protein is performing some as yet unknown, but potentially important, function during gonadal development.
The interval between gonadal sex differentiation and onset of female germ cell meiosis differs between species and can be described as immediate or delayed [38]. In species with a delay period, steroid synthesis occurs before the onset of meiosis [6, 39, 40]. In contrast, mouse germ cells enter the first meiotic prophase simultaneously with or shortly after sexual differentiation, and the ovary is steroidogenically inactive until follicles are formed [34, 38]. In sheep, the beginning of germ cell meiosis is delayed until about Day 55 of gestation [2], during which time the germ cells become enclosed in germ cell cords lined with a basal lamina. The mRNAs encoding SF-1, 3ß-HSD, P450scc, StAR, P45017OH, and P450arom are all expressed during this delay period. In fact, peak expression of these mRNAs occurs at Days 35–40. This corresponds to the period when germ cells and somatic cells make contact and form germ cell cords. Thus, it could be hypothesized that steroids are important in this process. At this time, expression of SF-1 and 3ß-HSD is widespread throughout the gonad but with a punctate pattern. In contrast, mRNA encoding the P450scc, P45017OH, P450arom, and StAR is more restricted, being observed only in the outer medullary-inner cortex region. By the end of the delay period at 55–60 days of gestation, mesonephric derived cell-streams can be seen invading the area around the germ cell cords [31]. Expression of mRNAs encoding SF-1, 3ß-HSD, P450scc, StAR, P45017OH, and P450arom as well as gonadal concentrations of 17ß-estradiol decreased at this time and remained low through Day 75 of gestation. Furthermore, expression of all but SF-1 and 3ß-HSD mRNAs was increasingly limited to the cell streams. In the fetal human ovary greater than 19 wk, sporadic but intense immunostaining for StAR was observed in the hilar region [41]. Interesting, as was observed in the male, mRNA encoding 3ß-HSD and SF-1 were expressed earlier and have a more widespread expression pattern than the other proteins studied. While expression of these mRNAs was strongest in the cell streams, other cell types in the developing ovary also expressed these mRNAs, suggesting an additional role for these proteins in ovarian as well as testicular development.
While the role of testicular secretion of androgens during fetal development is well characterized, little is known about what role, if any, estrogens may play in development of the ovary or the role that steroids may play in formation of the testis itself. In cases of androgen insensitivity, structural differences, including disorganization of the seminiferous tubules and a thickened basal lamina, have been observed in the testis [8, 9]. While it is known that the female phenotype is not dependent on the ovary, the transient appearance of high concentrations of estrogens around the time of morphological sexual differentiation throughout the delay period might indicate an important role of estrogens during germ cell development [39, 40, 42]. Estrogens have been shown to be a germ cell survival factor in human testis [10]. In addition, exposure of pregnant and lactating rats to secoisolariciresinol diglycoside, a naturally occurring lignan with antiestrogen properties, permanently affected ovarian function of their offspring [11]. These animals that had been exposed to this lignan during development of their ovaries had a delayed onset of puberty, decreased ovarian weight at 21 days of age, premature cycle irregularity, and acyclicity. These authors proposed that the effects seen in the lignan-exposed animals were due to loss of follicles through direct ovarian damage [11].
Mice lacking the estrogen receptor (ER) , ß, and ß as well as those lacking aromatase have been generated [43–45]. Male and female mice lacking aromatase, ER, or ERß are infertile or have reduced fertility. Often the effects observed in these mice become progressively more prominent with age. However, it is difficult to dissect out which effects are due to the lack of estrogens at the level of the ovary and which are related to changes in other hormones regulated by estrogens such as LH and FSH and/or an elevated level of androgens. As mice lacking aromatase age, a disorganization of the granulosa layer in antral follicles as well as a disorganization of the ovarian stroma with increased deposition of collagen was observed [46]. In adult mice lacking both the and ß form of the ER, a large proportion of the ovary was composed of very abnormal structures with characteristics of seminiferous tubules [43]. Thus, in the mouse, estrogen does appear important for normal ovarian morphology; however, the striking phenotypes observed in the adult mice are not observed in younger mice. Thus, these changes would not appear to be linked to ovarian formation during fetal and early neonatal life in mice. However, it is important to remember that while ovarian formation in mice is similar in many respects to that observed in sheep, two major differences in mice are the lack of estrogen secretion in the female prior to onset of follicular growth in the neonatal period and the immediate entry into meiosis following sexual differentiation. Thus, the role of estrogens in gonadal development in sheep (a single or twin ovulator) may be very different than those observed in mice, which are multiple ovulators.
In summary, fetal sheep gonads in both the male and female become steroidogenically active around the time of morphological sexual differentiation. In the male, androgens are the primary steroid synthesized; however, progesterone and 17ß-estradiol are also present in the testis by Day 55 of gestation. The steroidogenically active cells are located in the interstitial region outside the tubules. However, expression of mRNA encoding SF-1 and 3ß-HSD was observed prior to morphological sexual differentiation and was located inside the developing seminiferous tubules as well as in the interstitial region. In the female, 17ß-estradiol was the predominant steroid synthesized. At the time of morphological sexual differentiation, the steroidogenically active cells were located at the boundary of the cortex and medulla. By the end of the delay period, when meiosis begins, concentrations of steroids decreased dramatically and steroidogenically active cells were restricted to the mesonephric-derived cell streams. However, as in the male, expression of mRNA encoding 3ß-HSD and SF-1 was both more widespread and had an earlier onset than the mRNAs of other proteins important in steroidogenesis.
|
ACKNOWLEDGMENTS
The authors thank Dr. Gordon Niswender for the ovine StAR and 3ß-HSD cDNA; Dr. M. Waterman for the P450scc cDNA; Dr. E. Simpson for the P450arom cDNA; Professor I.J. Mason for the 3ß-HSD antibody; Peter Smith, Norma Hudson, Anne O'Connell, and Teeba Lundy for animal care, surgery, and tissue collection; Lee-Ann Still and Lynn O'Donovan for preparation of histological material; Lilian Morrison for statistical analysis; Alan Barkus for preparation of the figures; and Sue Swaney for secretarial assistance.
FOOTNOTES
First decision: 7 February 2001.
1 Supported by the New Zealand Foundation for Research, Science and Technology. 
2 Correspondence: Jennifer Juengel, Wallaceville Animal Research Centre, Ward Street, P.O. Box 40063, Upper Hutt, New Zealand. FAX: 64 4 922 1380; jenny.juengel{at}agresearch.co.nz
3 Current address: Infectious Diseases, Institute of Veterinary and Biomedical Sciences, Massey University, Private Bag 11222, Palmerston North, New Zealand.
Accepted: February 28, 2001.
Received: January 17, 2001.
REFERENCES
5,4 isomerase in ovine corpora lutea during the estrous cycle. Biol Reprod 1994; 51:380-384[Abstract]5,4 isomerase (3ß-HSD) and 3ß-HSD enzyme activity following treatment of ewes with prostaglandin F2. Endocrine 1998; 8:45-50[CrossRef][Medline]-hydroxylase/C17–20 lyase, and P450 aromatase in fetal mouse gonads. Endocrinology 1994; 135:262-268[Abstract] and ß. Science 1999; 286:2328-2331This article has been cited by other articles:
|
|
 |
|
  S. E. Recabarren, P. P. Rojas-Garcia, M. P. Recabarren, V. H. Alfaro, R. Smith, V. Padmanabhan, and T. Sir-Petermann Prenatal Testosterone Excess Reduces Sperm Count and Motility Endocrinology, December 1, 2008; 149(12): 6444 - 6448. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.Y. Yang and J.E. Fortune The Capacity of Primordial Follicles in Fetal Bovine Ovaries to Initiate Growth In Vitro Develops During Mid-Gestation and Is Associated with Meiotic Arrest of Oocytes Biol Reprod, June 1, 2008; 78(6): 1153 - 1161. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. J. O'Shaughnessy, P. J. Baker, A. Monteiro, S. Cassie, S. Bhattacharya, and P. A. Fowler Developmental Changes in Human Fetal Testicular Cell Numbers and Messenger Ribonucleic Acid Levels during the Second Trimester J. Clin. Endocrinol. Metab., December 1, 2007; 92(12): 4792 - 4801. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Steckler, M. Manikkam, E. K. Inskeep, and V. Padmanabhan Developmental Programming: Follicular Persistence in Prenatal Testosterone-Treated Sheep Is Not Programmed by Androgenic Actions of Testosterone Endocrinology, July 1, 2007; 148(7): 3532 - 3540. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A Forsdike, K. Hardy, L. Bull, J. Stark, L. J Webber, S. Stubbs, J. E Robinson, and S. Franks Disordered follicle development in ovaries of prenatally androgenized ewes J. Endocrinol., February 1, 2007; 192(2): 421 - 428. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Orisaka, K. Tajima, T. Mizutani, K. Miyamoto, B. K. Tsang, S. Fukuda, Y. Yoshida, and F. Kotsuji Granulosa Cells Promote Differentiation of Cortical Stromal Cells into Theca Cells in the Bovine Ovary Biol Reprod, November 1, 2006; 75(5): 734 - 740. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Pannetier, S. Fabre, F. Batista, A. Kocer, L. Renault, G. Jolivet, B. Mandon-Pepin, C. Cotinot, R. Veitia, and E. Pailhoux FOXL2 activates P450 aromatase gene transcription: towards a better characterization of the early steps of mammalian ovarian development. J. Mol. Endocrinol., June 1, 2006; 36(3): 399 - 413. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L Juengel, D. A Heath, L. D Quirke, and K. P McNatty Oestrogen receptor {alpha} and {beta}, androgen receptor and progesterone receptor mRNA and protein localisation within the developing ovary and in small growing follicles of sheep Reproduction, January 1, 2006; 131(1): 81 - 92. [Abstract] [Full Text] [PDF]
|
|
|
|
