SEARCH

SEARCH BY CITATION

Keywords:

  • mouse;
  • sex determination;
  • XY sex reversal;
  • SRY protein;
  • MIS;
  • Sertoli cell

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Sry, a single-copy gene on the Y-chromosome, acts dominantly to trigger differentiation of a testis from a gonadal primordium that otherwise develops into an ovary in mammals. Sry encodes a protein with a DNA-binding domain and probably acts as a transcription factor. However, the mode of SRY action in testis determination remains largely unknown. In the present study, we detected the endogenous SRY protein in normal XY fetal mouse gonads by Western blotting and immunohistochemistry. The tissue-specificity and ontogeny of the detected protein were consistent with those of Sry transcripts. Immunofluorescent double labeling revealed that the SRY protein was detected in the Sertoli cell lineage and was swiftly down-regulated concurrently with testis cord organization. Surprisingly, however, the SRY protein was detected in the entire gonad from the onset of its expression, not in parallel to the spatiotemporal pattern of testis cord organization. The SRY protein was also detected in the entire region of all B6.YTIR fetal gonads, which were anticipated to undergo either partial or complete sex reversal. SRY down-regulation was considerably delayed, compared with control B6.XY gonads and was not associated with testis cord organization in B6.YTIR gonads. We conclude that the testis-determining pathway is impaired at the site of SRY action in the B6.YTIR gonad. Developmental Dynamics 233:612–622, 2005. © 2005 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

In normal mammalian development, the presence or absence of a Y-chromosome dictates the differentiation of a gonadal primordium into a testis or an ovary. Consequently, the hormones produced by a testis induce development of male reproductive organs as well as sexual behaviors. On the other hand, the absence of testicular hormones results in female development as a default pathway (Jost et al., 1973). SRY/Sry, a single-copy gene on the Y-chromosome, has been identified to play the critical role in initiating testicular differentiation during gonadal development (Berta et al., 1990; Gubbay et al., 1990; Sinclair et al., 1990; Koopman et al., 1991). SRY/Sry encodes a protein with an HMG-box type of DNA-binding domain (Nasrin et al., 1991). It has been demonstrated that in vitro-translated SRY proteins bind to a specific DNA sequence, resulting in a sharp bend of its flanking DNA (Ferrari et al., 1992; Giese et al., 1994). This structural change in the target DNA may be important for its regulation of downstream genes (Pontiggia et al., 1994; Ukiyama et al., 2001). On the other hand, a role for the SRY protein in pre-mRNA splicing has been suggested recently (Ohe et al., 2002). Despite the progress in our understanding of the chemical nature of Sry products in vitro, the mode of SRY action during gonadal differentiation in vivo remains unknown. Expression of the SRY protein, the ultimate mediator of Sry functions during gonadogenesis, needs to be demonstrated in well-defined developmental systems such as laboratory mice.

Sex reversal models have been providing powerful tools for studying the mechanism of sex determination in mammals. Examples are B6.YPOS and B6.YTIR mouse strains, in which the Y-chromosomes of certain variations of Mus musculus domesticus (DOM) have been placed on a B6 genetic background by repeated backcrosses (Eicher et al., 1982; Nagamine et al., 1987; Biddle and Nishioka, 1988). After eight backcross generations, all XY progeny fail to develop normal testes and instead develop ovaries or ovotestes in fetal life. Because the Y-chromosome of either the B6.YPOS or B6.YTIR mouse is fully functional on a genetic background other than B6, which belongs to Mus musculus molossinus (MOL), this sex reversal suggests the importance of coordination between Y-encoded and autosomal genes for testis determination (Eicher et al., 1995, 1996). Previous studies have suggested the impairment of Sry posttranslational events, because Sry transcript levels do not correlate with the extent of sex reversal (Lee and Taketo, 2001; Albrecht et al., 2003). Although polymorphisms have been identified within the Sry open reading frame between DOM and MOL, it remains to be demonstrated whether structural differences in Sry products are responsible for sex reversal (Coward et al., 1994; Carlisle et al., 1996).

To clarify the molecular mechanism of testis determination, it is essential to detect and characterize the endogenous SRY protein in fetal gonads at the time of sex determination. Detection of the SRY protein to date has been reported only in human tissues (Rossi et al., 1993; Poulat et al., 1995; Salas-Cortes et al., 2001). Since SRY is expressed in both Sertoli and germ cells in human testes through all developmental stages, these studies have provided only a limited view on the mechanism of SRY action. In the mouse, most studies on Sry expression have been limited to analyses of its mRNA (Koopman et al., 1990; Capel et al., 1993; Lee and Taketo, 1994; Hacker et al., 1995; Jeske et al., 1995; Bullejos and Koopman, 2001). Some of these studies have demonstrated the existence of several types of Sry transcripts, including a circular form and linear forms with multiple transcription initiation and polyadenylation sites (Capel et al., 1993; Hacker et al., 1995; Jeske et al., 1995). Therefore, analyses of Sry transcripts alone are not sufficient to clarify the mechanism or site of SRY action. In the present study, we analyzed the expression of the endogenous SRY protein in fetal mouse gonads during sexual differentiation by both Western blotting and immunohistochemistry. Our results demonstrated that SRY expression was tightly regulated in the normal XY gonad before testis cord organization. On the other hand, SRY expression was initiated over the genital ridge but its down-regulation was delayed and did not coordinate with testis cord organization in the B6.YTIR gonad, which was anticipated to undergo partial or complete sex reversal.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Detection of the SRY Protein in CD1 Fetal Gonads by Western Blotting

Starting with 1 mg of total protein extract from CD1.XY gonads at 11.5 days post coitum (dpc) for immunoprecipitation, followed by Western blotting, an approximately 38-kDa protein band was detected by anti-SRY monoclonal antibody (Mab) #15 (Fig. 1a). The immunoprecipitation step was necessary to concentrate the SRY protein sufficient for the immunodetection. No such protein band was detected in XY gonads at 12.5 or 13.5 dpc, XX gonads at 11.5–13.5 dpc, mesonephroi of either sex at 11.5–13.5 dpc, or adult testes. When protein loading amounts were increased threefold, the 38-kDa protein was also detected in XY gonads at 12.5 dpc but not in any other fetal tissues (Fig. 1b) or adult testes (not shown). When the gel after electrotransfer was stained with silver, protein loading was seen comparable in all samples (Fig. 1c).

thumbnail image

Figure 1. Western blotting with anti-SRY antibody in tissues of the CD1 mouse strain. A: One milligram of total protein extract each was used for immunoprecipitation. The arrow indicates a band at 38 kDa. Molecular weight (MW) standards, 78.0 and 39.5 kDa, are indicated at the left. T, testis (or testicular primordium at 11.5 days post coitum, dpc); O, ovary (or ovarian primordium at 11.5 dpc); M, mesonephros. The content of tubulins in each supernatant, after immunoprecipitation with an anti-SRY antibody, was immunoprecipitated and Western blotted, both with the rabbit anti–Y-tubulin antibody. B: Three milligrams of total protein extract each from various tissues at 12.5 dpc was used for immunoprecipitation. The 38-kDa protein (arrow) is discernible only in XY testes. The major band seen in all samples corresponds to the heavy chain of immunoglobulins. C: The sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel after electrotransfer onto the membrane shown in Figure 1b was stained with silver. Protein loading was comparable in all samples.

Download figure to PowerPoint

Localization of the SRY Protein in CD1 Fetal Gonads by Immunohistochemistry

To determine the localization of the endogenous SRY protein, XY gonads of the CD1 strain from 10.5 through 13.5 dpc were processed for immunohistochemical staining with anti-SRY Mab#15. SRY staining was absent from all three gonads examined at the 10 tail somite (ts)-stage and two of five gonads examined at the 11 ts-stage (Fig. 2a). Faint SRY staining was seen in a small number of cells scattered over the genital ridge in two gonads at the 11 ts-stage, but the staining intensity was not convincingly higher than the background (not shown). On the other hand, faint but distinct SRY staining was seen in many cells at both poles of one gonad at the 11 ts-stage (Fig. 2b). SRY-positive cells were scattered over the entire gonads examined at the 12-ts stage (Fig. 2c). In these gonads at the 11 and 12 ts-stages, SRY staining was diffuse over the cells and only partially concentrated in the cell nuclei (Fig. 3a). Furthermore, a limited number of cells composing the genital ridge were positive. SRY staining became intense in the entire genital ridge by 14 ts-stage and remained so up to 23 ts-stage (Fig. 2d–f). At these later developmental stages, SRY staining was intense in the entire cell nuclei within the genital ridge, whereas it was absent from germ cells, the surface epithelium, and the adjacent mesonephros (Fig. 3b–d). Double immunofluorescent labeling demonstrated that germ cells, identified by cytoplasmic staining of mouse vasa homolog (MVH), were negative for SRY and were surrounded by SRY-positive somatic cells (Fig. 4c). At the 24–25 ts-stages, SRY-positive cells were seen in clusters only at the posterior pole in three of five gonads examined (Fig. 2h), whereas SRY-positive cells with diminished staining intensity were seen at both poles of one gonad (Fig. 2g) and scattered within the testis cords of another gonad (Figs. 2i, 3e). No SRY staining was seen at later stages (not shown).

thumbnail image

Figure 2. Immunohistochemical staining of the SRY protein in CD1 fetal gonads. a–j: Sections are oriented with the genital ridge at the right, the mesonephros at the left, the anterior pole at the top, and the posterior pole at the bottom. a: XY gonad at 11 tail somite (ts)-stage. No SRY staining is seen. b: XY gonad at 11 ts-stage. Cells with faint SRY-staining are scattered over the both poles. c: XY gonad at 12 ts-stage. Cells with faint SRY staining are scattered over the genital ridge. df: XY gonads at 14, 17, and 23 ts-stages. SRY staining has intensified, and the SRY-positive cells occupy the entire genital ridge but are absent from the adjacent mesonephros. g: XY gonad at 25 ts-stage. SRY-positive cells with reduced staining intensity are scattered at the anterior pole as well as the posterior pole (not shown). h: XY gonad at 25 ts-stage. SRY-positive cells are seen only at the posterior pole. i: XY gonad at 25 ts-stage. SRY-positive cells are scattered within the testis cords. j: XX gonad at 21 ts-stage. No SRY staining is seen. k: Sagittal section of an XY fetus at 17 ts-stage. SRY staining is seen only in the genital ridge (asterisked area is shown at a higher magnification in inset). DA, dorsal aorta; L, liver; S, somite; SC, spinal cord; T, tongue; V, ventricle. l: XY gonad at 19 ts-stage. The staining solution containing anti-SRY Mab#15 alone was preincubated for 20 min before its application to sections. Intense SRY staining is seen over the genital ridge (top left). m: Adjacent section of the XY gonad shown in Figure 2l. The staining solution containing anti-SRY Mab#15 was preincubated with a recombinant SRY protein for 20 min before its application to sections. No SRY staining is seen in the genital ridge (top left). Scale bars = 80 μm in j (applies to a–j,l–m), 640 μm in k, 100 μm in the inset of k.

Download figure to PowerPoint

thumbnail image

Figure 3. Immunohistochemical staining of SRY proteins in CD1.XY fetal gonads, followed by counterstaining with methyl green. a: Posterior pole of the gonad at 11 tail somite (ts)-stage shown in Figure 2b. Faint SRY staining is seen over the genital ridge but not in the mesonephros or surface epithelium. Inset: diffuse SRY staining is seen over the nuclei of the cells surrounding two germ cells (double asterisks), whereas no SRY-staining is seen in the cells surrounding another germ cell (single asterisks), shown at 2.5 times higher magnification. b: Anterior–medial area of gonad at 14 ts-stage. SRY-positive cells cluster in the genital ridge, whereas the cells in the mesonephros and the surface epithelium are negative. c: Central area of gonad at 23 ts-stage. Most somatic cells deep in the genital ridge are stained for SRY, whereas the cells in the mesonephros and the surface epithelium are negative. A thick layer of somatic cells (tunica albuginia) under the surface epithelium is also negative. SRY-negative large and round cells surrounded by SRY-positive cells are likely germ cells. d: Posterior pole of gonad shown in c. SRY-positive cells cluster in the genital ridge where germ cells are scarce. e: Anterior–medial area of gonad at 25 ts-stage. Faint SRY staining is seen in some cells within the testis cords (arrowheads). The intense staining in the interstitium indicates nonspecific binding of the secondary antibody to the blood cells. Scale bar = 80 μm in e (applies to a–e).

Download figure to PowerPoint

thumbnail image

Figure 4. Double immunofluorescent labeling in CD1.XY fetal gonads. a: XY gonad at 22 tail somite (ts)-stage. SRY staining (green) is seen in numerous cells in the entire genital ridge, whereas MIS staining (red) is predominantly seen in the central region near the mesonephros. b: An area (asterisks) of a at a higher magnification. Intense SRY staining (green) is seen in the nuclei, whereas MIS staining (red) is seen in the cytoplasm of the same cells (arrowheads). c: XY gonad at 21 ts-stage. MVH staining (red) is localized in the cytoplasm of germ cells, whereas SRY staining (green) is seen in the nuclei of somatic cells surrounding the germ cells. d: Medial (M)–posterior (P) region of XY gonad at 25 ts-stage. Intense SRY staining (green) is seen in the nuclei of abundant cells in the posterior pole, whereas much reduced staining is seen in the Sertoli cells within the testis cords near the medial region. On the other hand, intense MIS staining (red) is seen in the cytoplasm of Sertoli cells within the testis cords near the medial region, although it is less frequently seen near the posterior pole. Both staining colors are seen in the same cells in an area indicated with asterisks and also in a few cells within the testis cords. Yellow staining indicates nonspecific binding of both secondary antibodies to the blood cells. The dark green staining over the tissue is also nonspecific binding of the secondary antibody. e: An area (asterisk) of d at a higher magnification. Intense SRY staining is seen in the nuclei, and MIS staining is seen in the cytoplasm of the same cells (arrowheads). f: The area marked with three asterisks in d is shown here at a higher magnification. Most cells have intense SRY staining in their nuclei. A few cells have both SRY staining in their nuclei and MIS staining in their cytoplasm (arrowheads). Scale bars = 80 μm a–d, 32 μm in e (applies to e,f).

Download figure to PowerPoint

In whole fetuses at the 17–18 ts-stages, SRY staining was seen in the genital ridge, but not in other organs (Fig. 2k). When a recombinant SRY protein was introduced into the staining solution containing anti-SRY Mab#15, the immunostaining in the genetic ridge was completely eliminated (Fig. 2l,m). Double immunofluorescent labeling demonstrated the expression of SRY and MIS, a marker of Sertoli cells, in the same cells within the genital ridge at the 22 ts-stage (Fig. 4a,b). In XY gonads in which testis cords had developed in the central but not the polar region, SRY-positive cells were scarce in the central region and abundant in the polar region (Fig. 4d). By contrast, MIS staining was more distinct within testis cords in the central region and less frequently seen in the polar region. Selected cells were positive for both SRY and MIS (Fig. 4d–f). SRY staining was never detected in XX gonads at comparable stages (Fig. 2j).

Detection of SRY Proteins in B6.YTIR and B6.XY Gonads by Western Blotting

We compared the levels of endogenous SRY proteins between B6.YTIR and control B6.XY gonads at 11.5 dpc (15–18 ts-stages) by combining immunoprecipitation and Western blotting. Starting with 500 μg of total protein extract each, a 38-kDa band was discernible in the B6.YTIR lane but not in the B6.XY lane (Fig. 5a). After longer exposure of chemiluminescent signals, a 73-kDa band appeared in the B6.XY lane but not in the B6.YTIR lane (Fig. 5b). The contents of protein loading were comparable between the samples from B6.YTIR and B6.XY gonads (Fig. 5c). Many bands from the rabbit antibody added for immunoprecipitation were detected by the secondary antibody in the original Western blotting. An extra blocking step using both goat anti-rabbit F(ab)1 and goat anti-mouse F(ab)1 fragments diminished these bands, whereas the bands corresponding to SRY proteins were enhanced as shown in Figure 5a,b.

thumbnail image

Figure 5. Comparison of SRY protein levels between B6.YTIR and B6.XY gonads. a,b: A total of 500 μg of total protein extract each from B6.YTIR and B6.XY gonads at 11.5 days postcoitum (dpc) were used for immunoprecipitation. a,b: The film was exposed for 30 sec (a) and 3 min (b); the arrows indicate 38- (a) and 73- (b) kDa bands unique to B6.YTIR and B6.XY gonads, respectively. The major bands seen in both lanes correspond to the heavy chain of immunoglobulins. c: Each supernatant, after immunoprecipitation with an anti-SRY antibody, was divided into three aliquots and processed for Western blotting with the anti–Y-tubulin antibody. The contents of Y-tubulin proteins were comparable between the proteins collected from B6.YTIR and B6.XY gonads.

Download figure to PowerPoint

Localization of SRY Proteins in B6.YTIR and B6.XY Fetal Gonads by Immunohistochemistry

We also compared the localization of SRY proteins between B6.YTIR and control B6.XY fetal gonads by immunohistochemistry. Faint SRY staining was seen in a small number of cells scattered over the genital ridge at the 12 and 13 ts-stages and the staining intensity increased by the 14 ts-stage in B6.XY gonads (Fig. 6a,b). By contrast, SRY staining was not seen until the 15 ts-stage in B6.YTIR gonads (Fig. 6f,g). SRY-positive cells were concentrated at both poles in this gonad (Fig. 6g) but were distributed over the entire genital ridge in other gonads of both strains examined at 11.5 dpc (Fig. 6a–c,h). Germ cells, identified by cytoplasmic staining of MVH, were negative for SRY in both strains (Fig. 7a,b). SRY staining in B6.XY gonads considerably diminished by the 24 ts-stage and was almost absent at the 25 ts-stage (Fig. 6d,e). Testis cords had developed in the entire gonad by this stage (Fig. 7c,d). By contrast, SRY staining in B6.YTIR gonads was intense up to the 27 ts-stage and less intense but clearly visible at the 28 ts-stage (Fig. 6i–e). Because a minority of B6.YTIR gonads was anticipated to develop testis cords in the central region (Taketo et al., 1991), we examined a large number of gonads in this mouse strain. SRY staining was seen in the entire region of all 17 sexually undifferentiated gonads examined at the 15–27 ts-stages (Fig. 6g–j). Of the seven gonads examined at the 28 ts-stage, one had testis cords in its central region (occupying approximately 20% of the entire gonad); SRY staining was seen in the entire region of all seven gonads (Fig. 6k,l). At the 30 ts-stage and later, no SRY staining was seen in any B6.YTIR gonads examined, including four ovaries and six ovotestes (data not shown). Double immunofluorescent labeling revealed intense MIS signals within the testis cords in B6.XY gonads at the 25 ts-stage or later (Fig. 7c,d). By contrast, MIS signals were detected in B6.YTIR ovotestes at 13.5 dpc or later, but their testis cords were not well organized until 14.5 dpc (Fig. 7e–h).

thumbnail image

Figure 6. Immunohistochemical detection of SRY proteins in B6.XY and B6.YTIR fetal gonads. All sections are oriented with the genital ridge at the right, the mesonephros at the left, the anterior pole at the top, and the posterior pole at the bottom. a: B6.XY gonad at 12 tail somite (ts)-stage. A small number of cells with faint SRY-staining are scattered over the genital ridge (arrowheads). b,c: B6.XY gonads at 14 and 18 ts-stages. SRY-positive cells are seen in the entire genital ridge but absent from the adjacent mesonephros. d: B6.XY gonad at 24 ts-stage. SRY-positive cells are distributed in the entire gonad, but the staining intensity has diminished. e: B6.XY gonad at 25 ts-stage. Faint SRY staining is seen within the testis cords at the posterior pole. f: B6.YTIR gonad at 14 ts-stage. No SRY staining is seen. g: B6.YTIR gonad at 15 ts-stage. SRY-positive cells are scattered over the genital ridge but are more abundant at both poles. hj: B6.YTIR gonads at 17, 25 and 27 ts-stages. SRY-positive cells are seen in the entire genital ridge. k: B6.YTIR gonad at 28 ts-stage. SRY-positive cells with reduced staining intensity are seen in the entire gonad. (No testis cords were seen in this gonad with transmitting light at the time of dissection.) l: B6.YTIR ovotestis at 28 ts-stage. SRY-positive cells are seen in the entire gonad. (Testis cords were seen in the central 20% of this gonad with transmitting light at the time of dissection.) Scale bar = 80 μm in a (applies to a–l).

Download figure to PowerPoint

thumbnail image

Figure 7. Double immunofluorescent labeling in B6.XY and B6.YTIR fetal gonads. a: Medial area of B6.XY gonad at 18 tail somite (ts)-stage. MVH staining is localized in the cytoplasm of germ cells, whereas SRY staining is seen in the nuclei of somatic cells around the germ cells. b: Medial area of B6.YTIR gonad at 26 ts-stage. Localization of SRY and MVH staining is similar to that in the B6.XY gonad (a). c: B6.XY gonad at 25 ts-stage. SRY-staining is hardly visible, whereas MIS staining is seen in the entire gonad. d: Central area of the section shown in c, at a higher magnification. Intense MIS staining is seen in the cytoplasm of most cells, whereas SRY staining is seen in the nuclei of a few cells within the testis cords. e: B6.YTIR ovotestis at 13.4 days post coitum (dpc). (Testis cords were seen in the central 60% of this gonad with transmitting light at the time of dissection.) MIS staining is seen in the central region. f: A part of the section shown in e, at a higher magnification. SRY staining is hardly visible while MIS staining is seen in cell clusters in the central region (arrowheads). Testis cord organization is not distinct. g: B6.YTIR ovotestis at 14.4 dpc. MIS staining (red) is seen in clusters in the central region. h: A part of the section shown in g, at a higher magnification. MIS staining is intense within well-organized testis cords, whereas it is less intense in the testis cords that are only partially organized (asterisks). Scale bars = 80 μm a (applies to a,b,d,f,h), in c (applies to c,e,g).

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Detection of the Endogenous SRY Protein

The present study detected a 38-kDa protein specific to XY fetal gonads of the CD1 strain by combining immunoprecipitation and Western blotting. We believe that this protein is the endogenous Sry protein for several lines of evidence. First, Sry is the only gene known to be expressed in the XY and not in the XX gonad at 11.5 dpc. Many other genes are initially expressed in both types of gonads and undergo up- or down-regulation as a consequence of sexual differentiation. Second, the ontogeny of the 38-kD protein agrees with that of Sry transcripts (Koopman et al., 1990; Hacker et al., 1995; Bullejos and Koopman, 2001). Third, the mobility of the 38-kD protein in sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) is consistent with that of the protein transcribed and translated in vitro from the DOM-type Sry genomic DNA (Coward et al., 1994).

We also detected endogenous SRY proteins in both B6.YTIR and B6.XY fetal gonads. The mobility of the endogenous SRY protein in SDS-PAGE corresponded to that of the protein transcribed and translated in vitro from the respective-type Sry genomic DNA. Although the molecular weights of SRY proteins predicted from the amino acid sequence are 28 and 49 kDa for DOM and MOL strains, in vitro-translated proteins from respective Sry genomic DNAs migrate with apparent molecular weights of 38 and 73 kDa (Coward et al., 1994). These abnormal electrophoretic migrations can be attributed to the highly basic nature of the polyglutamine/histidine-rich domain in the carboxy termini of the SRY protein. Our results further confirm that the predicted early stop codon in the DOM-type Sry genomic DNA is used in vivo during translation of the SRY protein in the fetal gonad at the time of testis determination. Endogenous SRY protein levels in B6.YTIR fetal gonads were not less than those in normal B6.XY fetal gonads, in agreement with the findings of Sry transcripts (Lee and Taketo, 2001; Albrecht et al., 2003). These results support the hypothesis that Sry expression levels alone cannot explain sex reversal in the B6.YTIR mouse.

Expression of SRY Proteins During Gonadal Sex Differentiation

Immunohistochemical detection of SRY proteins was also consistent with the ontogeny of Sry transcripts in normal XY gonads of both CD1 and B6 strains. The SRY protein was detectable in gonads as early as the 11 ts-stage (10.5 dpc), its staining levels reached a peak before testis cord organization at the 21–23 ts-stages (11.5 dpc), and then SRY expression was down-regulated concurrently with testis cord organization at the 24–26 ts-stages (12.5 dpc). The variable patterns of SRY staining observed at the onset of gonadal differentiation can be attributed to the variation of developmental stages. Although we used the ts-stage as a landmark, tail somite development is progressive and one ts-stage covers a range of developmental events. We also observed variable patterns of SRY staining at the onset of testis cord organization. We have seen localization of intense SRY staining only at the posterior pole of 60% gonads, in agreement with the findings by whole-mount in situ hybridization (Bullejos and Koopman, 2001). However, we have also seen faint SRY staining in scattered cells at the anterior pole of one gonad and within the testis cords of another gonad. We speculate that, although Sry transcription was down-regulated, SRY proteins persisted, in some cells forming the testis cords.

Our study has provided the first direct evidence that the endogenous SRY protein is expressed in the Sertoli-cell lineage, although such a hypothesis has been postulated for many years (Palmar and Burgoyne, 1991). In sexually undifferentiated CD1 gonads, the SRY protein was detectable in the nuclei of almost all somatic cells in the genital ridge, likely including Sertoli-precursor cells, and not in germ cells or cells in the mesonephros and surface epithelium. SRY expression in the Sertoli-cell lineage was further substantiated by colocalization of MIS, a Sertoli-cell marker, in the same cells at late stages of SRY expression. As soon as testis cord organization became evident, SRY expression was down-regulated, while MIS staining became intense within the testis cords.

Testis cord organization begins in the central region and extends to both poles in normal XY gonads (Taketo et al., 1991). It has been suggested that this spatiotemporal pattern of testicular organization coincides with the pattern of Sry expression, detected by whole-mount in situ hybridization or GFP expression under the Sry promoter (Bullejos and Koopman, 2001; Albrecht and Eicher, 2001). Our results did not show the central localization of SRY staining from the earliest stage of detection. It is conceivable that the detection of Sry transcripts may not correspond to proper translation or processing of SRY proteins. Nuclear localization of SRY proteins in our results agrees with the predicted function of SRY as a transcription factor. Expression of Sry reporter genes involves translation and nuclear translocation and has been detected in the correct cell lineage (Albrecht and Eicher, 2001; Sekido et al., 2004). However, the timing of their down-regulation is not precisely the same as that of the endogenous SRY protein (current studies). For example, the reporter gene by Albrecht and Eicher (2001) remains expressed long after the onset of testis cord organization, whereas that by Sekido et al. (2004) is down-regulated at the onset of SOX9 expression, which precedes testis cord organization. These discrepancies can be explained by different expression levels of proteins or detection sensitivities. Our results suggest that expression of the endogenous SRY protein does not directly influence the spatiotemporal pattern of testis cord organization. There may be a mechanism that coordinates with the SRY protein and facilitates testis cord organization in the central-to-pole direction.

Throughout our current studies, we observed weaker SRY staining in B6.XY and B6.YTIR gonads compared with CD1.XY gonads even at their peak developmental stages. Although almost all somatic cells in the genital ridge were positive for SRY in CD1 gonads, a considerably fewer cells were positive for SRY in B6.XY and B6.YTIR gonads. It is unclear whether a limited number of somatic cells expressed SRY or lower expression levels reflected in the number of cells stained with SRY in the latter gonads. Nonetheless, it is intriguing that the lower level of SRY expression is sufficient for normal testicular organization in the B6.XY gonad. By Western blotting, the SRY protein in CD1 gonads was much easier to be detected than those in B6.XY and B6.YTIR gonads. Although these quantitative comparisons are not accurate, they are in agreement with the levels of Sry transcripts in different mouse strains (Lee and Taketo, 2001; Albrecht et al., 2003). The genetic background appears to influence the levels of Sry transcription and translation.

Mechanism of Sex Reversal in the B6.YTIR Gonad

Our results have revealed two possible mechanisms of sex reversal in the B6.YTIR gonad. First, the earliest stage at which the SRY protein became detectable in the B6.YTIR gonad was considerably later than that in the normal B6.XY or CD1.XY gonad. This observation leaves the possibility that the SRY protein acts for a short period to successfully initiate testicular differentiation. Further studies are needed to support this hypothesis. Second, the SRY protein was detectable in the entire B6.YTIR gonad in which testicular structures develop, if any, only in the central region. We have reported previously that only 7% of B6.YTIR gonads develop MIS-positive testis cords at 12.5 dpc and the maximum 35% at 14.5 dpc (Taketo et al., 1991). Because we detected the SRY protein in the entire region of all 17 gonads examined at the 17–27 ts-stages (11.5 dpc), we conclude that the SRY protein was expressed in the B6.YTIR gonad or region that failed to develop testicular structures. Furthermore, the SRY protein was detectable in all B6.YTIR gonads until later developmental stages and down-regulated simultaneously, unlike normal XY gonads in which SRY down-regulation was tightly associated with testis cord organization. All these results suggest that SRY is expressed, but its action is impaired in the entire region of the B6.YTIR gonad. We further speculate that testis cords develop only when other B6 factors that favor testicular organization overcome the inefficient action of the TIR-type SRY protein.

The SRY protein is anticipated to interact with cofactors to form a functional transcriptome in regulation of downstream target genes (Dubin et al., 1995; Poulat et al., 1997; Lau and Zhang, 1998; Oh et al., 2005). SRYTIR and SRYB6 proteins are not only different in sizes but also polymorphic at multiple sites (Coward et al., 1994). Therefore, differences in their biochemical properties or three-dimensional structures may be sufficient to impair the ability of the SRYTIR protein to interact with cofactors or target genes of the B6 strain. One candidate target of the SRY protein is the Sox9 gene (Morais da Silva et al., 1996; Bergstrom et al., 2000; Vidal et al, 2001; Chaboissier et al., 2004; Sekido et al., 2004). It has been reported recently that the SOX9 protein is up-regulated only in the testis cords, similarly to MIS, in the B6.YTIR ovotestis (Moreno-Mendoza et al., 2004). It is conceivable that impairment of SRY action results in failure of Sox9 up-regulation, which in turn abolishes testicular organization. However, regulation of Sox9 by SRY appears to be indirect, because SOX9 expression is limited to the central region, while SRY is expressed in the entire region of the B6.YTIR gonad. Further studies are needed to address the exact relation between Sry and Sox9 during the course of testis determination. Availability of specific antibodies against the mouse SRY protein provides an opportunity for identifying the endogenous SRY interactive proteins and evaluating the roles of such proteins in SRY functions during testis determination.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Primary Antibodies

A rabbit anti-SRY antibody was raised against a GST-SRY fusion protein consisting of the bridge region (corresponding to amino acid 82-144), which was flanked by the HMG box and polyQ/H repeats. A total of 1.5 mg of purified protein was used to immunize a rabbit at multiple intervals at Caltag Laboratory (Healdsburg, CA). Mouse anti-SRY Mab were produced against the SRY recombinant protein without the HMG-box motif (corresponding to amino acids 110-395) using a hybridoma technique as described previously (Kuo et al., 1989). The specificities of all antibodies were tested with Western blotting, with the respective antigens used for immunization. A rabbit antibody against MVH was a kind gift from Dr. T. Noce (Mitsubishi Kagaku Institute of Life Sciences, Tokyo, Japan). A goat antibody against Mullerian Inhibiting Substance (MIS) and a rabbit antibody against Y-tubulin were purchased from Santa-Cruz Biotechnology (Santa Cruz, CA) and Sigma-Aldrich Canada Ltd. (Oakville, ON, Canada), respectively.

Animals

All animal procedures were performed in accordance with the Canadian Council on Animal Care and approved by the McGill University Animal Care Committee. CD1 and B6 mice were purchased from Charles River Laboratory (St. Constance, QC, Canada) and the Jackson Laboratory (Bar Harbor, ME), respectively. B6.YTIR male mice (N30-45 backcross generations) carrying a B6 genetic background and the Y chromosome originally from a DOM house mouse (caught in Tirano, Italy; Nagamine et al., 1987) were maintained in our mouse colony. Females were caged with males up to 3 days, and separated from the males on the morning when copulation plugs were identified. The gestation age was defined as dpc, assuming that the copulation occurred at 1:00 AM.

Staging of Fetuses

Throughout the present studies, we counted the number of tail somites (ts) from the base of the genital tubercle (Nagamine et al., 1999; Lee and Taketo, 2001). However, many other reports used a different method, counting the number of ts from the posterior to the hind limb bud (Hacker et al., 1995; Bullejos and Koopman, 2001; Albrecht et al., 2003). We found the latter method more difficult and less reliable, particularly at advanced developmental stages, in our laboratory settings. Nonetheless, we compared the two methods in approximately 30 litters covering 10.4–12.5 dpc. The ts-number counted by our method was smaller than that counted by the second method by two in average. In the CD1 strain, fetuses at the 10–12 ts-stages were collected at 10.4–10.8 dpc. Similarly, fetuses at the 14–23 ts-stages and those at the 24–28 ts-stages were collected at 11.3–11.8 and 12.3–12.5 dpc, respectively. B6.XY fetuses at the 12–18 and 24–26 ts-stages were collected at 11.3–11.8 and 12.3–12.5 dpc, respectively. B6.YTIR fetuses at the 13–25 and 26–32 ts-stages were collected at 11.3–11.8 and 12.3–12.5 dpc, respectively. The representative stages, 10.5 dpc, 11.5 dpc, and so on, were used when the result of a group of gonads was discussed.

Immunoprecipitation

Fetal gonads (genital ridges only) and mesonephroi were collected at 11.5–13.5 dpc, snap-frozen in liquid nitrogen, and stored at −80°C. The number of ts was counted for accurate staging at 11.5 dpc. A part of the remaining fetal body was used to identify the chromosomal sex by polymerase chain reaction (PCR) amplification of Zfy (Amleh and Taketo, 1998). Pooled tissues were homogenized in an ice-cold immunoprecipitation (IP) buffer (50 mM Hepes, pH 7.5, 150 mM NaCl, 10% glycerol, 1% Triton X-100, 0.5% deoxycholate, 1 mM MgCl2, 4 mM NaF, 100 μM Na3VO4, 5 mM Na2MO4, 2 mM benzamidine, 5 mM iodoacetamide, 5 mM pNpp) containing a protease inhibitor cocktail (Roche Diagnostics, Indianapolis, IL), incubated on ice for 30 min, and centrifuged at 12,000 rpm for 10 min. An aliquot of the supernatant was used to determine the protein content by the Bradford method (Bio-Rad Laboratories, Ltd., Mississauga, ON, Canada) with bovine serum albumin (BSA) as a standard. IP buffer was added to the remaining supernatant to a total volume of 500 μl, which was incubated with 5 μl of normal rabbit serum for 1 hr, and then mixed with 75 μl of 20% slurry of protein A–Sepharose (Sigma-Aldrich Canada, Ltd.) in an HBS buffer (50 mM Hepes, pH 7.5; 150 mM NaCl; 10% glycerol) for 1 hr, followed by centrifugation at 12,000 rpm for 30 sec. The supernatant was then incubated with 5 μl of rabbit anti-SRY antibody at 4°C overnight, and immunoprecipitated as described above. The precipitate was boiled in a sample buffer (100 mM Tris-HCl, pH 6.8; 200 mM dithiothreitol; 4% SDS; 0.2% BPB; 5% Ficoll) for 10 min, snap frozen in liquid nitrogen, and stored at −20°C.

Western Blotting

Half of the immunoprecipitated proteins were applied to 12.5% SDS-PAGE followed by electrotransfer to either nitrocellulose or polyvinylidene difluoride membranes in a Towbin buffer. The membranes were rinsed, blocked with 1% PVP/2% BSA, and incubated with anti-SRY Mab#15 at 1:10,000 dilution. For the detection of proteins from B6.XY and B6.YTIR gonads, an extra blocking step using both goat anti-rabbit F(ab)1 and goat anti-mouse F(ab)1 fragments (Jackson ImmunoResearch, Mississauga, ON, Canada) was included. Binding of the anti-SRY antibody was detected by using a sheep anti-mouse Ig antibody conjugated with horseradish peroxidase (Amersham-Pharmacia Biotechnology, Baie d'Urfe, QC, Canada), followed by Lumi Light Plus electrochemiluminescence reagents (Roche Diagnostics) and exposure to ECL hyperfilm (Amersham-Pharmacia) according to the manufacturer's protocols. The supernatant, after immunoprecipitation with the rabbit anti-SRY antibody, was immunoprecipitated with the rabbit anti–Y-tubulin antibody and processed for Western blotting as described above using the same anti–Y-tubulin antibody. To determine protein molecular weights, an aliquot of Kaleidoscope Prestained Standards (Bio-Rad Laboratories, Hercules, CA) was applied to each SDS-PAGE gel and electrotransferred to the membrane together with protein samples. Individual bands in the membrane were remarked with a fluorescent pen before exposure to an ECC hyperfilm. For accuracy, the molecular weights of Kaleidoscope bands were compared with the bands of biotinylated SDS-PAGE standards (Bio-Rad Laboratories).

Immunohistochemistry

Fetal gonads with adjacent mesonephroi were isolated at 10.5–14.5 dpc, staged according to the number of ts up to 12.5 dpc, fixed with 2% formaldehyde (Electron Microscopy Sciences, Hartford, PA) in a microtubule-stabilizing buffer (Messinger and Albertini, 1991) for 1 hr at room temperature, and stored in 70% ethanol at 4°C until being embedded in paraffin. Serial paraffin sections of 5 μm thickness were placed on Plus-coated histology slides (Fisher Scientific, Fair Lawn, NJ) and stored in a box containing silica gel at 4°C. Whole fetuses at 11.5 dpc were similarly processed and sectioned at 6–10 μm thickness. Selected slides were deparaffinized, treated with 50 mM Tris-HCl (pH 10) at 95°C for 30 min, and processed for inactivation of the endogenous peroxidase followed by blocking with an anti-mouse IgG reagent as recommended in the protocol for the MOM-ABC kit (Vector Laboratory, Burlington, ON, Canada). The slides were then washed twice in Holding Buffer (HB, 1% goat serum, 0.3% BSA, 0.005% Triton X-100 in PBS) and incubated with anti-SRY Mab#15 at 1:500 dilution in HB at 4°C overnight. (Of the panel of SRY-specific antibodies, only Mab#4 and #15 detected the endogenous SRY protein in histological sections of mouse fetal gonads.) The slides were incubated with a secondary antibody against mouse IgG, conjugated with biotin (MOM, Vector) for 30 min, and processed for ABC staining according to the manufacturer's protocol (Vector). All slides were dehydrated and mounted in Histoclad (Clay Adams, Franklin Lakes, NJ). To confirm the specificity of SRY staining, a full-length recombinant protein made from the MOL-type Sry genomic DNA at the final concentration of 10 μg/ml was mixed with anti-SRY Mab#15 at 1:500 dilution in PBS for 20 min by rotation before staining of sections. The protein concentrate (MOM, Vector) was added to the staining solution as recommended in the instruction. Histochemical staining was observed under a transmission microscope (Zeiss Axiophot, Germany). All images were captured with a digital camera (Retiga 1300, QImaging, Burnaby, BC, Canada) and processed with Northern Eclipse digital imaging software, version 6.0 (Empix Imaging, Mississauga, ON, Canada).

Immunofluorescent Double Labeling

All slides were deparaffinized and treated for antigen retrieval as described above. The slides were then incubated with mouse IgG blocking reagent for 60 min, followed by anti-SRY Mab#15 at 1:500 dilution and either the goat anti-MIS or rabbit anti-MVH antibody at 1:1,000 dilution in PBS containing the protein concentrate (MOM, vector) at 4°C overnight. All washings were performed in PBS. On the next day, all slides were washed and incubated with the anti-mouse IgG antibody conjugated with biotin (MOM, Vector) and either a donkey anti-goat IgG antibody conjugated with RRX (Jackson ImmunoResearch) or a goat anti-rabbit IgG antibody conjugated with rhodamine (Pierce Biotechnology, Rockford, IL). All slides were washed, incubated with avidin conjugated with fluorescein isothiocyanate (Pierce Biotechnology), and mounted in the Prolong Antifade mounting medium (Molecular Probe, Eugene, OR). Fluorescent signals were examined under an epifluorescence microscope (Zeiss, Axiophot) and recorded as described above.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We thank Kevin Kemball for his advice on antigen retrieval. Y.-F.C.L is a Research Career Scientist in the Department of Veterans Affairs and was funded by the National Institutes of Health. T.T. was funded by the Canadian Institutes of Health Research.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES