Expression of Stimulated by Retinoic Acid Gene 8 (Stra8) in Spermatogenic Cells Induced by Retinoic Acid: An In Vivo Study in Vitamin A-Sufficient Postnatal Murine Testes

Abstract

Vitamin A is required for male fertility and normal spermatogenesis. Retinoic acid (RA), an active metabolite of vitamin A, is necessary for spermatogonial maturation and proper entry of germ cells into meiotic prophase in the postnatal testes. The expression of Stra8, which is essential for successful meiosis in both male and female gonads and normal spermatogenesis, is directly related to the availability of RA. This study examined the developmental expression pattern of Stra8 transcript in both male and female gonads, provided specific cellular localization of STRA8 protein in the postnatal and adult testis, and investigated RA actions in adult germ cells in a vitamin A-sufficient condition. The peak of Stra8 mRNA expression coincided with the onset of meiosis in postnatal testes. STRA8 protein was detected in gonocytes as early as 5 days postpartum. The expression of STRA8 protein in the neonatal testes was not uniform among spermatogonia, perhaps heralding the asynchronous beginning of spermatogenesis. In adult testes, the highest level of Stra8 mRNA and protein was found in seminiferous epithelial stages VI–VIII. STRA8 protein was localized to some type A and B spermatogonia, preleptotene spermatocytes, and early leptotene spermatocytes. In the vitamin A-sufficient adult testes, RA but not retinol acetate stimulated Stra8 mRNA expression. STRA8 protein expression in adult spermatogonia was induced by RA stimulation, suggesting its role in spermatogonial differentiation. Retinoic acid also increased the number of preleptotene spermatocytes exhibiting 5-bromo-2-deoxyuridine incorporation, indicating a more synchronized premeiotic DNA replication.

INTRODUCTION

Vitamin A plays a pivotal role in maintaining many important physiological functions within the body [1]. Retinoic acid (RA), an active metabolite of vitamin A, is a vital signaling molecule for normal fetal development, pattern formation, cell proliferation and differentiation, and apoptosis [1]. The actions of RA are mediated through six distinct ligand-dependent transcription factors, including three retinoic acid receptors (RARA, RARB, and RARG) and three retinoid X receptors (RXRA, RXRB, and RXRG). It has been known since 1925 that vitamin A is required for male fertility and normal spermatogenesis. Spermatogenesis ceases and most germ cells degenerate in vitamin A-deficient (VAD) animals [2–6]. Only type A spermatogonia remain in VAD mouse testes, whereas in VAD rats, some preleptotene spermatocytes are also present. Supplementing vitamin A (retinol) or RA to VAD animals reinitiates spermatogenesis in a synchronous fashion throughout the testis, and the synchrony of spermatogenesis may last for a few months [2, 3].

Retinoic acid has been demonstrated to regulate sex-specific timing of meiotic initiation via a gene named Stra8 [7]. Stra8 was first identified as a gene stimulated by RA in cultured embryonic carcinoma cells and embryonic stem cells [8]. Stra8 expression was later found in embryonic ovarian germ cells over a period of 4 days, beginning at 12.5 days postcoitum (dpc), just 1 day before onset of meiosis. This is in contrast to the testis, where it is absent in embryonic tissue but expressed in postnatal male premeiotic cells [9]; however, the subset of premeiotic spermatogenic cells that express Stra8 has yet to be identified. Koubova et al. [7] demonstrated that RA is required for Stra8 mRNA expression in embryonic ovaries and that RA was sufficient to induce Stra8 mRNA expression in embryonic and VAD adult testes in vivo. Recent studies used cultured gonocytes, undifferentiated spermatogonia, and neonatal testes in vitro to demonstrate the direct RA induction of Stra8 mRNA and STRA8 protein in the absence of somatic cells [10]. Therefore, the presence of Stra8 mRNA and protein is considered to be a reliable indicator of RA action in both male and female premeiotic germ cells.

Both Stra8-deficient male and female mice are infertile [7, 11]. In embryonic ovarian germ cells lacking Stra8, mitotic division is normal, but these cells fail to undergo premeiotic DNA replication and all subsequent meiotic events [11]. Collectively, these data suggest that the decision to enter meiosis is determined primarily, if not exclusively, by RA induction of Stra8 preceding premeiotic DNA replication [11]. It is less clear whether this is also the case in postnatal testes, even though a severely reduced number of spermatogenic cells are apparent in Stra8-deficient testes, and nearly all germ cells appear to be premeiotic [11].

In this study, several in vivo experiments were performed to achieve three major objectives: 1) identify the premeiotic germ cells that express Stra8; 2) determine whether RA regulates STRA8 expression in adult spermatogonia; and 3) investigate whether RA induction of Stra8 regulates the initiation of premeiotic DNA replication in postnatal spermatogenic cells. To address these objectives, we generated a developmental expression profile of Stra8 mRNA in embryonic male and female gonads and postnatal testes and also examined STRA8 protein in postnatal testes. We also examined STRA8 expression in adult spermatogonia with or without RA treatment in a vitamin A-sufficient condition and investigated potential temporal sequences of premeiotic DNA replication in preleptotene spermatocytes by BrdU incorporation with or without RA stimulation.

MATERIALS AND METHODS

Animal Care and Treatments

All animal experiments were approved by Washington State University Animal Care and Use committees and were conducted in accordance with the Guiding Principles for the Care and Use of Research Animals of the National Institutes of Health. A BL/6–129 mouse colony was maintained in a temperature- and humidity-controlled room with food and water provided ad libitum. To test the effects of exogenous RA, normal adult male mice were injected subcutaneously with all-trans RA (90% oil and 10% ethanol) at a dose of 750 µg per mouse or were injected with vehicle as control. To investigate the induction of STRA8 protein expression by RA in neonatal testes, 8 days postpartum (dpp) male mice were injected subcutaneously with all-trans RA at a dose of 350 µg per mouse or were injected with vehicle as control. For 5-bromo-2-deoxyuridine (BrdU) labeling experiments, mice were given BrdU solution via intraperitoneal injections (dissolved in sterile saline at a level of 100 mg/kg body weight) 4 h before they were killed. Twenty-four hours after RA injection, mice were killed and testes fixed in neutral-buffered formalin (NBF). Tissues used for microarray profiling were collected as previously described [12, 13]. In brief, for embryonic testes and ovaries, gonad tissues from 11.5, 12.5, 14.5, 16.5, and 18.5 dpc were collected. For microarray profiling of postnatal testes, tissues were obtained from postnatal males at ages of 0, 3, 6, 8, 10, 14, 18, 20, 30, 35, and 56 dpp. In both cases, replicate samples were obtained and applied to microarrays, except for 12.5 dpc ovaries.

Microarray Processing

A total of 10 µg (from postnatal testes) or 2–4 µg (from embryonic testes and ovaries) of total RNA from each sample was used to create the microarray target. The microarray processing was performed as previously described [13]. Mouse Genome 430 2.0 arrays (Affymetrix, Santa Clara, CA) were used in accordance with the manufacturer’s standard protocol. Two independent samples were analyzed for each tissue. The resulting data were viewed and analyzed using GeneSpring software (Agilent Technologies, Santa Clara, CA). All reactions and microarray hybridization procedures were performed in the Laboratory for Biotechnology and Bioanalysis I at Washington State University.

Dissection of Stage-Specific Tubules from Adult Testes

Stage-specific seminiferous tubule clusters were collected using the transillumination-assisted microdissection technique described previously [14]. Harvested tubules were divided into four stage clusters, including stages XII–I (weak spot area), II–V (strong spot area), VI–VIII (dark zone), and IX–XI (pale zone). Three independent dissection repeats were performed. Total RNA was prepared and used for real-time RT-PCR.

Immunohistochemistry of STRA8 Protein in Postnatal Mouse Testes

Immunohistochemistry using the Zymed LAB-SA System per the manufacturer’s instructions (Zymed Laboratory Inc., San Francisco, CA) was performed as previously reported [10] with minor modifications. Briefly, tissue sections were incubated with 10% goat serum to block nonspecific binding, followed by incubation with STRA8 antiserum (1:2000) or preimmune serum (1:2000) at 4°C overnight. The slides were incubated with biotinylated secondary antibody at room temperature for 45 min, followed by application of streptavidin-horseradish peroxidase for 15 min at room temperature. The diaminobenzidine solution was applied until brown color developed (in about 1–2 min). Sections were analyzed with a Nikon Microphoto-FX (Meridian Instrument Company Inc., Kent, WA) microscope. Photomicrographs were taken with an Olympus OLY-200 digital camera (Olympus America Inc., Center Valley, PA) using Olympus MagnaFire Camera Imaging and Control version 1.0 (Olympus America).

Immunohistochemical Double Staining of BrdU and STRA8 Protein

Normal mouse adult testes were fixed in 4% (w/v) paraformaldehyde (PFA) solution for 24 h and then transferred to 70% ethanol, embedded in paraffin, and sectioned. Mouse BrdU antibody (1:200 dilution; Vision BioSystems, Norwell, MA) was applied to testis sections and incubated at 4°C overnight. After three washes in PBS, tissues were incubated with secondary antibody (TRITC-conjugated donkey anti-mouse immunoglobulin G; 1:100; Jackson ImunoResearch Laboratory, West Grove, PA) at 37°C for 1 h. Tissues were blocked again with 10% donkey serum for 20 min. After further PBS washes, tissues were incubated with rabbit STRA8 antibody (1:200 dilution) at 37°C for 2 h, and then secondary antibody (1:100 dilution; FITC-conjugated donkey anti-rabbit IgG; Jackson ImunoResearch Laboratory) was applied at 37°C for 1 h. Sections were mounted with medium for fluorescence with 4′-6-diamidino-2-phenylindole (DAPI). Sections incubated in PBS without primary antibody were used as negative controls. Images were captured with ProgRes digital camera (Olympus America) using Olympus MagnaFire Camera Imaging and Control (version 1.0; Olympus America) and compiled using Adobe PhotoShop 7.0 (Adobe Systems, San Jose, CA).

RA Treatment of Normal Adult Mice and Tissue Collection

Adult (75–90 days old) male mice (BL/6–129; The Jackson Laboratory) fed a regular (vitamin A-sufficient) diet were used in this experiment. One testis was removed from each animal and placed in TRIzol (Invitrogen) for RNA extraction. This served as a preinjection control for RT-PCR analysis. Incisions were sealed, and the animals were allowed to recover for 24 h. Eight mice were injected subcutaneously with 100 µl of 15 mg/ml all-trans retinol acetate (ROL; Sigma-Aldrich, St. Louis, MO) in 10% ethanol and sesame oil. Nine mice were injected subcutaneously with 100 µl of 7.5 mg/ml all-trans RA (Sigma-Aldrich) in 10% ethanol and sesame oil. Three (control) animals were injected with 100 µl of 10% ethanol and sesame oil. The remaining testes from the experimental and control animals were harvested 24 h after the injection and placed in TRIzol for RNA extraction.

Real-Time RT-PCR

A two-step real-time RT-PCR was used to measure the expression of candidate genes as previously described [15]. For each experiment, at least three independent biological samples were collected. One cDNA was synthesized for each independent biological sample, and triplicate reactions were run in real-time PCR for each cDNA sample for each gene tested. Stra8 primers amplify a 151-bp product (primers: 5′-GTTTCCTGCGTGTTCCACAAG-3′ and 5′-CACCCGAGGCTCAAGCTTC-3′); Cathepsin L (Ctsl) primers amplify a 101-bp product (primers: 5′-CTGTTGCTATGGACGCAAGC-3′ and 5′-CAGAACCCCATGGTCGAGG-3′); control Ribosomal protein S2 (Rps2) primers amplify a 112-bp product (primers: 5′-CTGACTCCCGACCTCTGGAAA-3′ and 5′-GAGCCTGGGTCCTCTGAACA-3′). Expression of Stra8 and Ctsl was normalized to Rps2 expression. All data sets were first tested for normal distribution in SigmaStat for Windows Version 3.01 (SPSS, Chicago, IL). The pairwise Student t-test (P < 0.05) was used to analyze the normally distributed datasets, and a signed rank test was used to analyze the skewed dataset using SigmaStat. Data represent mean ± SEM.

Analysis of STRA8-Immunopositive Spermatogonia in Stages Other Than VI–VIII Seminiferous Tubules of Adult Testes after Treatment of Retinoic Acid

Adult (75–90 days old) male mice (BL/6–129; The Jackson Laboratory) were maintained on a regular vitamin A-sufficient diet. Animals (four mice for vehicle control and three mice for RA) were injected subcutaneously with 0.75 mg all-trans RA per mouse. The duration of RA treatment was 24 h. Testes were fixed in 4% of PFA and embedded in paraffin. For staging of the mouse seminiferous epithelium, testis sections were stained with hematoxylin (Sigma). The following numbers were counted: total number of seminiferous tubules at stages other than VI–VIII per testis cross section; total number of seminiferous tubules at stages other than VI–VIII containing the STRA8-immunopositive spermatogonia per testis cross section; the number of STRA8-immunopositive spermatogonia in seminiferous tubules at stages other than VI–VIII per testis cross section. All data sets were first tested for normal distribution in SigmaStat for Windows Version 3.01 (SPSS, Chicago, IL). Student t-test was used for the statistical analysis (P < 0.05).

Analysis of BrdU Incorporation in Normal Adult Mouse Testes

Adult (75–90 days old) male mice (BL/6–129; The Jackson Laboratory) were maintained on a regular vitamin A-sufficient diet. Animals (six mice for vehicle control and six mice for RA) were injected subcutaneously with 0.75 mg all-trans RA per mouse or the same volume of vehicle oil. The duration of RA treatment was 24 h. Four hours before harvesting the testes, mice were injected with BrdU in sterile saline at a concentration of 100 µg/g body weight. Testes were fixed in NBF and embedded in paraffin. Slides were stained with BrdU antibody according to the procedures described above. The following numbers were counted: total number of seminiferous tubules per testis cross section; the number of seminiferous tubules at stages VII–VIII per testis cross section; the total number of BrdU-positive germ cells per testis cross section; and the number of BrdU-positive germ cells in the stage VII–VIII tubules per testis cross section. Student t-test was used for the statistical analysis (P < 0.05).

RESULTS

Developmental Expression of Stra8 Transcript in Testes and Ovaries

The gene expression profile of the Stra8 transcripts was obtained using microarray analysis (Fig. 1A). Stra8 was absent from ovaries at 11.5 dpc and 12.5 dpc and from 11.5- to 18.5-dpc testes. A dramatic peak of Stra8 expression appeared in the ovary at 14.5 dpc. In the testis, high levels of the Stra8 transcripts first appeared at 6 dpp, peaked at 10 dpp, and then significantly decreased in 14-dpp and 18-dpp testes. The expression of the Stra8 transcripts in adult testes remained at low levels, similar to those observed at 3 dpp, when only premeiotic germ cells are present (Fig. 1B).

FIG. 1.

Stra8 expression in developing ovaries and testes (GeneChip MOE 430.2.0). Values represent raw signal of Stra8 ± SEM. A) Stra8 mRNA expression in embryonic ovaries and testes. Solid line represents ovaries, and dotted line represents testes. Raw signal values detected by microarray analysis are shown on the y-axis; days postcoitum are shown on the x-axis. B) Stra8 mRNA expression in postnatal testes. Raw signal values detected by microarray analysis are shown on the y-axis; days postpartum are shown on the x-axis.

Differential Expression of Stra8 Transcript in Stage Clusters of Seminiferous Tubules

As the adult mouse testes appeared to contain only low levels of Stra8, it was of interest to investigate whether the expression of Stra8 mRNA was consistent and present in all stages of the seminiferous tubules or if it was stage specific. Four stage clusters of seminiferous tubules were manually dissected from adult mouse testes, including stages XII–I (weak spot), stages II–V (strong spot), stages VI–VIII (dark zone), and stages IX–XI (transparent zone) [14]. Using real-time RT-PCR, the dissection accuracy was validated by monitoring the expression level of Ctsl, known to be specifically expressed in stages VI–VIII (Fig. 2A) [16]. The level of Stra8 transcript was significantly higher in stage VI–VIII tubules than in the other stages tested (Fig. 2B).

FIG. 2.

Relative levels of Stra8 mRNA expression in manually dissected stage clusters of adult seminiferous tubules. Values represent relative level of Stra8 ± SEM. A) Relative levels of Ctsl mRNA expression in four stage clusters of adult seminiferous epithelium using stages IX–XI as the baseline. B) Relative levels of Stra8 mRNA expression in four stage clusters of adult seminiferous epithelium using XII–I as the baseline.

Developmental Expression of STRA8 Protein in Postnatal Testes

Immunolocalization of STRA8 protein in the testes of mice aged 1, 5, 10, 15, 20, 35, and 70–90 (adult) dpp was evaluated to determine the cellular location of STRA8 during postnatal testis development. In 1-day-old animals, no STRA8 protein was detected (data not shown), but at 5 dpp, moderate to strong immunostaining of STRA8 was observed in a small number of germ cells (Fig. 3A). Most STRA8-positive germ cells were immediately adjacent to the basal lamina (Fig. 3G), with very few in the center of the tubule (Fig. 3G, arrow). Some of these STRA8-positive germ cells exhibited the morphological features of gonocytes, with large, round nuclei (Fig. 3G), whereas some possessed the oval-shaped spermatogonialike nuclei (Fig. 3G, arrowhead). Although quantitative analysis was not performed, the number of STRA8-positive germ cells at 10 dpp (Fig. 3B) appeared to increase in comparison to 5 dpp, whereas most tubules were still STRA8 negative. Moreover, the identity of the STRA8-positive germ cells became identifiable at 10 dpp. Three types of STRA8-positive tubules could be observed: 1) tubules contained mostly STRA8-positive spermatogonia (Fig. 3H); 2) tubules contained mostly STRA8-positive preleptotene spermatocytes (Fig. 3I); 3) tubules contained both STRA8-positive spermatogonia and preleptotene spermatocytes (Fig. 3J). At 15 and 20 dpp, some seminiferous tubules contained STRA8-positive preleptotene/leptotene spermatocytes and some spermatogonia (Fig. 3, C, D, and K–N). The intensity of the positive immunostaining increased in line with the increased number of STRA8-positive germ cells in all subsequent ages. The uneven distribution of STRA8 expression across seminiferous tubules appeared as early as 10–15 dpp and continued through all subsequent ages (Fig. 3, E and F). The majority of STRA8-immunopositive cells in adult testes were preleptotene/early leptotene spermatocytes at stage VIII (Fig. 3, F, S, and U). STRA8-positive preleptotene spermatocytes were also observed at stage VI–VII; however, they were not as prevalent, and not all preleptotene spermatocytes at stage VIII were STRA8 positive. STRA8 protein also localized to preleptotene-leptotene transitional cells (Fig. 3T) and early leptotene spermatocytes in stage VIII, but it was not present in most of the leptotene spermatocytes in stages IX and X (Fig. 3U). Very weak immunostaining was detected in only a few leptotene spermatocytes in stage IX (data not shown). None of the leptotene spermatocytes in stage X were positive for STRA8. In stages other than VII–VIII, STRA8 was detected in some but not all spermatogonia, including type A (Fig. 3O), intermediate (Fig. 3P), and type B (Fig. 3Q). The distribution of STRA8-positive spermatogonia displayed no stage specificity, and no immunostaining of STRA8 protein was observed in Sertoli cells or interstitial cells (Fig. 3F).

FIG. 3.

Immunostaining of STRA8 protein in the postnatal testes of mice aged 5 days (A), 10 days (B), 15 days (C), 20 days (D), 35 days (E), and 70–90 days (F). STRA8-positive germ cells in 5-dpp (G) and 10-dpp (H, I, and J) testes have different morphological features and locations. STRA8-positive cells in 20-dpp testes include type A spermatogonia (A Sp) in K, A₃ spermatogonia (A₃ Sp) in L, B spermatogonia (B Sp) in M, and preleptotene spermatocytes (Prelep) in N. STRA8-positive cells in adult testes include type A spermatogonia in O, intermediate spermatogonia (In Sp) in P, B spermatogonia in Q, dividing spermatogonia in R (arrow), preleptotene spermatocytes (Prelep) in S, preleptotene-leptotene transitional cells (Transitional) in T, and leptotene spermatocytes (Lept) in U. V shows preimmune negative control (adults). Original magnifications ×125 (A–F and V); ×400 (G–J); ×750 (K–P and R); ×1000 (Q and S–U).

Immunohistochemical Double Staining of BrdU with STRA8

To investigate the potential temporal sequence of STRA8 expression and premeiotic DNA replication in preleptotene spermatocytes, adult mice were treated with BrdU, and testis sections of these mice were examined for the colocalization of STRA8 and BrdU. Both STRA8 (green) and BrdU (red) incorporation was detected in some preleptotene/early leptotene spermatocytes at stages VII–VIII (Fig. 4, A–C), indicated by the yellow stain visible in these cells. The overlay view of some stage VII–VIII tubules indicates both the presence of STRA8 without BrdU incorporation in preleptotene spermatocytes (Fig. 4, D–F) and BrdU-positive cells not expressing STRA8 in spermatogonia in stages other than VII–VIII (Fig. 4, G–I).

FIG. 4.

Immunohistochemical double staining of BrdU (red color in A, D, and G) and STRA8 (green color in B, E, and H) in adult mouse testes. Tissues were counter-stained with DAPI (purple color in C, F, and I after the images merged). Arrows indicate the BrdU- and/or STRA8-positive spermatogonia and preleptotene spermatocytes. Interstitial cells presented autofluorescence in both preimmune control and antibody-treated conditions, and colors in those cells were not considered positive. Original magnification ×400.

Stra8 mRNA Expression in Adult Testes after Treatment of RA

To investigate whether exogenous RA could induce Stra8 mRNA expression in vitamin A-sufficient adult testes, real-time RT-PCR was performed to examine the level of Stra8 transcript 24 h after RA or ROL treatment. Retinoic acid increased Stra8 transcript levels 3-fold (P < 0.01). Neither vehicle oil alone nor ROL changed the level of Stra8 transcripts (Fig. 5).

FIG. 5.

Quantitative RT-PCR analysis of Stra8 expression in control (oil-injected), ROL-injected, or RA-injected adult vitamin A-sufficient testes compared with preinjection (Pre-inj.) contralateral testes. Values represent relative fold change of Stra8 ± SEM. Relative level of Stra8 expression using preinjection contralateral testes as the baseline is shown on the y-axis. Asterisk indicates a statistically significant difference (P < 0.05) between the labeled sample and preinjection sample.

Number of STRA8-Immunopositive Spermatogonia in Stages Other Than VI–VIII Seminiferous Tubules of Adult Testes after Treatment of RA

To investigate whether exogenous RA had an effect on STRA8 protein levels in adult spermatogonial cells, the number of STRA8-immunopositive spermatogonia in stages other than VI–VIII were counted, and a comparison was made between the vehicle- and RA-treated testes. The duration of the treatment was 24 h. Because of the technical difficulty of distinguishing A_al/A₁ spermatogonia and preleptotene spermatocytes in a large-scale quantification study under current experimental conditions, spermatogonia in stages VI–VIII were excluded from this experiment. Retinoic acid increased the percentage of STRA8-positive tubules in stages other than VI–VIII from 14% to 40% (P < 0.01), and the number of STRA8-positive spermatogonia in each of these tubules increased from three to five (P < 0.01; Fig. 6).

FIG. 6.

Regulation of STRA8 protein expression in spermatogonia in stages other than VI–VIII of adult vitamin A-sufficient testes after 24 h of RA treatment. A) The percentage of tubules at stages other than VI–VIII containing STRA8-positive spermatogonia. Values represent the average percentage ± SEM. B) The number of STRA8-positive spermatogonia in each of the tubules at stages other than VI–VIII containing STRA8-positive spermatogonia. Values represent average number of STRA8-positive spermatogonia per tubule ± SEM. Asterisks (*) indicate a statistically significant difference (P < 0.05) between the labeled sample and control sample (C).

BrdU Incorporation in Adult Mouse Testes after RA Treatment

To investigate the effect of exogenous RA on DNA replication in adult preleptotene spermatocytes, a BrdU incorporation study in adult mouse testes treated with either RA or vehicle control was performed. Almost all of the BrdU-positive cells in the adult testes were germ cells, with the majority (55%–85%) being preleptotene spermatocytes in stages VII–VIII (Supplemental Fig. 1, A and B, available online at www.biolreprod.org). There were moderate numbers of BrdU-positive germ cells in stages V–VI, but they likely represent a mixture of B spermatogonia and preleptotene spermatocytes (Supplemental Fig. 1C, available online at www.biolreprod.org). BrdU-positive germ cells in stages other than VI–VIII were mitotic spermatogonia (Supplemental Fig. 1C). Although the number of BrdU-positive germ cells per tubule after RA stimulation tended to increase, the level was not statistically significant (Fig. 7). The percentage of stage VII–VIII tubules in each testis did not change after RA treatment, nor did the number of BrdU-positive spermatogonial cells in each tubule other than stages VII–VIII (Fig. 7). However, there was a 44% increase in BrdU-positive preleptotene spermatocytes in each VII–VIII tubule (from 27 to 39, P < 0.01) 24 h after RA injection (Fig. 7).

FIG. 7.

BrdU incorporation in the germ cells of adult vitamin A-sufficient testes after RA treatment for 24 h. Number of BrdU-positive germ cells in each tubule in a cross section (a); number of BrdU-positive germ cells in each of the stage VII–VIII tubules (b); number of BrdU-positive germ cells in each of the stages other than VII–VIII tubules (c); and percentage of stage VII–VIII tubules in each cross section (d). Error bars represent SEM. Asterisk indicates a statistically significant difference (P < 0.05) between the labeled sample and control sample (C).

RA Induction of STRA8 Protein Expression in 8-Dpp Vitamin A-Sufficient Testes

STRA8 protein expression was examined by immunohistochemistry in the spermatogonia and preleptotene spermatocytes of 8-dpp testes of mice treated by oil vehicle alone (for 24 h). Only a small number of the tubules contained STRA8-positive germ cells (Fig. 8A). However, after 24 h, the RA-treated, 8-dpp testes showed STRA8 protein was induced in most of spermatogonia and preleptotene spermatocytes, and the majority of tubules contained STRA8-positive germ cells (Fig. 8B).

FIG. 8.

Immunohistochemical detection of STRA8 protein induced by RA in 8-dpp neonatal testes. A) Immunostaining of STRA8 after 24 h of oil treatment (8-dpp mouse testis). B) Immunostaining of STRA8 after 24 h of RA treatment (8-dpp mouse testis). Original magnification ×100.

DISCUSSION

The current study examined the developmental expression pattern of Stra8 transcripts in both male and female gonads, provided specific cellular localization of STRA8 protein in the testes during postnatal development and in adult mouse, and investigated RA actions in adult germ cells under vitamin A-sufficient conditions. The findings include: 1) the maximal levels of Stra8 mRNA expression coincided with the onset of meiosis in both male and female gonads; 2) the highest level of Stra8 mRNA expression in adult mouse testis was in seminiferous tubule stages VI–VIII; 3) STRA8 protein expression in germ cells started as early as 5 dpp, with heterogeneous expression of STRA8 protein observed in different seminiferous tubules; 4) in the adult testes, STRA8 protein localized to type A, intermediate, and B spermatogonia, and preleptotene and early leptotene spermatocytes; 5) RA but not ROL increased Stra8 mRNA expression in normal adult testes; 6) RA increased the number of STRA8-positive seminiferous tubules in stages other than VI–VIII and also increased the number of STRA8-positive spermatogonia per tubule in vitamin A-sufficient adult mouse testes; 7) RA increased the number of BrdU-positive preleptotene spermatocytes in stages VII–VIII in vitamin A-sufficient adult mouse testes; and 8) STRA8 protein expression became ubiquitous in 8-dpp testes treated by RA.

Stra8 was initially considered to be a spermatogonial marker [8, 17, 18]. Oulad-Abdelghani et al. [8] reported the expression of STRA8 protein in spermatogonia from two lines of evidence: 1) the localization of the STRA8 protein was restricted to the peripheral layer of the tubule, and 2) immunopositive staining was detected in the cytoplasm of spermatogonia-like germ cells by electron microscopy immunocytochemistry. A possible expression in preleptotene spermatocytes was proposed based on the peripheral expression pattern. Subsequent studies showed Stra8 expression in embryonic ovaries and upregulation of Stra8 mRNA immediately prior to the entry of embryonic female germ cells into meiosis [9]. The presence of Stra8 mRNA was also accepted as a premeiotic germ cell marker [7, 19, 20]. In our previous study, we found that Stra8 mRNA and protein could be artificially induced by RA in gonocytes of 2-dpp mouse testes [10]. In the current study, STRA8 was detected in some of the neonatal gonocytes as early as 5 dpp. During the first wave of spermatogenesis and also in the adult testis, STRA8 localized to several different types of male germ cells, including type A and type B spermatogonia, preleptotene spermatocytes, and early leptotene spermatocytes. Analysis of Stra8-deficient males demonstrated that spermatogenesis ceased at preleptotene/leptotene spermatocytes, with these cells undergoing apoptosis [11]. The identification of STRA8 protein in preleptotene and leptotene spermatocytes provides the foundation for the future functional study of Stra8 in these male germ cells.

The developmental pattern of Stra8 mRNA expression in postnatal testes revealed by our microarray study showed a peak of expression around 10–14 dpp. In male rodents, this time point coincides with the onset of meiosis, as well as a large wave of spermatogonial differentiation in prepubertal testes. The appearance of preleptotene spermatocytes can be observed as early as 8–10 dpp [11]. It has been proposed that some gonocytes give rise to spermatogonial stem cells (SSCs) at the beginning of spermatogenesis, and others directly differentiate to A₁ or A₂ spermatogonia [21, 22]. On the other hand, McLean et al. [23] reported the development of SSCs capable of initiating donor-derived spermatogenesis after 3–4 dpp in neonatal mouse testes [23]. The differentiating spermatogonia derived from this newly formed SSC population would also appear in the neonatal testes around 10–14 dpp [24]. Therefore, it seems that the overlapping arrival of two developing germ cell populations, preleptotene/leptotene spermatocytes derived from gonocytes directly and differentiating spermatogonia derived from newly formed SSCs (both with Stra8 expression), created an enhanced peak expression of Stra8 in 10- to 14-dpp mouse testes.

In an earlier study, we reported an increased BrdU incorporation and Stra8 (both mRNA and protein) expression in 2-dpp neonatal gonocytes treated with RA for 24 h. It was possible that RA was promoting self-renewal or differentiation-related mitosis in these neonatal gonocytes. However, RA treatment in either 2-dpp mouse testes or isolated THY1-positive gonocytes/spermatogonia increased expression of Kit, a marker for spermatogonial differentiation. Moreover, the increase of Stra8 was observed to precede the increase of Kit in these experiments. In the current study, RA treatment of vitamin A-sufficient testes transformed STRA8-negative spermatogonia into STRA8-positive spermatogonia. Retinoic acid treatment of adults for 24 h was sufficient to induce both Stra8 mRNA and protein expression in spermatogonia, but it was not sufficient to statistically increase BrdU incorporation in these cells. These data suggest that induction of Stra8 mRNA and protein by RA in adults could be an indicator of spermatogonial differentiation, and this induction might be an upstream event of mitotic DNA replication. Whether a causative relationship can be established between RA induction of Stra8 and mitotic DNA replication associated with spermatogonial differentiation requires further studies.

The stage-specific expression of Stra8 mRNA (in stage VI–VIII tubules) and protein (in preleptotene/early leptotene spermatocytes in stages VII–VIII) suggests that there is a narrow window of entry by male germ cells into meiosis that is strictly controlled by RA. By treating the vitamin A-sufficient adult male testes with RA, BrdU incorporation in preleptotene spermatocytes in stage VII and VIII was increased significantly. The data not only demonstrate that RA regulates preleptotene spermatocytes and STRA8 is a mediator of this RA action, but also suggest that exogenous RA treatment under vitamin A-sufficient conditions made this premeiotic DNA replication and entry more synchronous, rather than occurring over three different stages (VI–VIII). It is also in good agreement with the observations obtained from the analysis of Stra8-deficient ovaries [11]. Baltus et al. [11] reported that female germ cells of Stra8-deficient mice failed to undergo premeiotic DNA replication and all subsequent meiotic events, and they concluded that in female germ cells, the decision to enter meiosis preceded premeiotic DNA replication. Our results showed a high correlation between RA induction of STRA8 and premeiotic DNA replication in postnatal male germ cells. In addition, it is worth noting that RA also regulates at least two other major steps of germ cell development in seminiferous epithelium stages VII–VIII: conversion of undifferentiated spermatogonia to differentiating spermatogonia [25, 26] and spermiation [27]. Therefore, it seems there is a very intensive, if not exclusively confined, RA action on germ cell development in stages VII–VIII, indicating a strictly regulated, stage-specific synthesis and delivery of retinoids across the seminiferous epithelium stages.

Retinoic acid is transported at a low level via serum, and it was thought that the oxidation of retinol to RA generally takes place in target tissues [2]. The testis was demonstrated to be a tissue very resistant to exogenous RA delivery [28]. A strong degradation mechanism is also present in the interstitial and Sertoli cells [7, 28, 29], and one of the major RA-metabolizing p450 enzymes, CYP26B1, was demonstrated to be critically important for RA metabolism in embryonic and adult testes [29]. These two pieces of evidence suggest that in the testis, a strong somatic metabolic barrier to exogenous RA exists in order to deliver RA to germ cells in a temporally and spatially controlled manner. This, in turn, would allow control of spermatogonial differentiation and, ultimately, entry into meiosis. Therefore, it is not unexpected that we observed that a high dose of exogenous RA administered to vitamin A-sufficient adult males only generated a very limited response in the germ cells based on the measure of Stra8 induction. Meanwhile, liver and Sertoli cells appear to function similarly in the uptake of retinol and the maintenance of large pools of retinol esters [1, 2]. In the current study, a high dose of retinol acetate treatment did not generate significant changes in Stra8 expression in vitamin A-sufficient adult males. This is in agreement with the idea that Sertoli cells may serve not only as a metabolic barrier but also as a storage pool of vitamin A to maintain homeostasis in the face of ever-changing dietary supplies [1]. It is noteworthy that Sertoli cells are also the primary sites for RA synthesis in the testis. Whether Sertoli cells are the suppliers of RA or retinoids to germ cells is unclear. Although quantitative analysis was not performed, it is apparent that the induction of STRA8 by RA treatment in 8-dpp testes was more extensive and more dramatic than in the adult testes. This is a strong indication that the storage pool of vitamin A in Sertoli cells and a metabolic barrier to RA is well established, at least after puberty, and the developmental program of male germ cells regulated by RA can be perturbed more easily in neonates than in adults.