| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Embryo |
Centre de recherche en reproduction animale,3 Faculté de médecin vétérinaire, Université de Montréal, St.-Hyacinthe, Québec, Canada J2S 7C6
Murdoch Children's Research Institute and Department of Pediatrics,4 University of Melbourne, Royal Children's Hospital, Flemington Road, Melbourne, VIC 3052, Australia
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
developmental biology, gene regulation, Sertoli cells
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Most of the genes currently known to play a role in the process of sex determination and differentiation have been identified from human and mouse models. Studying human patients showing sex reversal or dysgenesis of the gonads has revealed roles for genes such as Sry itself, and Sox9, Dmrt1, Dax1, Wt1 [1, 12–15]. Phenotypic anomalies in sex determination observed in knockout mouse models have further implicated genes such as M33, Lhx9, and Sf1 [16– 18]. One gene, Dmrt1, was identified via a combination of homology with Drosophila and Caenorhabditis elegans genes [19] as well as human dysgenesis of the gonads [20]. More recently, to identify additional candidate genes as well as signaling pathways implicated in these processes, differential expression-based screening approaches were undertaken by several labs [21–25]. Although procedures varied somewhat, all of the screens reported to date have in common the tissue level comparison of developing testis with developing ovaries at a given developmental date.
To characterize in more detail the expression profile of pre-Sertoli cells at the beginning of sex differentiation as well as to identify new candidate genes involved in sex determination and differentiation, we used a transgenic mouse model expressing green fluorescent protein (GFP) under the control of 4.5 (kilobases) kb of the pig SRY 5' flanking region (pSRYp-GFP) [7] to isolate and purify pre-Sertoli cells via fluorescence. RNA from 12.0 to 12.5 dpc pre-Sertoli cells was subtracted against nonfluorescent cells from the male genital ridge using suppression subtraction hybridization (SSH). Gene sequencing followed by whole-mount in situ hybridization (WISH) on selected clones allowed us to identify new candidate genes involved in mammalian sex determination and differentiation and to confirm the utility of using transgenically labeled cell populations for SSH applications in development.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The cloning and characterization of 4.5 kb of pig SRY 5' flanking sequences and the generation of pSRYp-GFP transgenic mice have been described previously [7, 26]. Mice were housed and handled according to guidelines established by the Guide for Care and Use of Laboratory Animals (copyright 1996, National Academy of Science) and according to the standards of the institution, and protocols were approved by the institutional ethics committee on use of animals in research. For staging of embryonic development, noon on the day of vaginal plug detection was designated as Day 0.5 postcoitum (0.5 dpc). To isolate male genital ridge cells from 12.0–12.5 dpc embryos, timed matings were performed. Genital ridges were dissected, male ridges were identified via fluorescence, and mesonephroi removed. The male genital ridges were then pooled and digested for 30 min at 37°C in M2 media (Sigma, St. Louis, MO) supplemented with collagenase (50U ml–1) (Gibco, Canadian Life Technologies, Burlington, ON, Canada) and dispase (2.4U ml–1) (Gibco). Fluorescent cells (pre-Sertoli) and nonfluorescent cells (remainder of genital ridge) were separated using fluorescent-activated cell sorting (FACS) isolation on a FACSTAR-Plus machine (Becton Dickinson, Mountain View, CA) equipped with an EGFP filter. In total 21 400 fluorescent cells and 22 000 nonfluorescent cells were recovered. Total RNA was extracted using RNeasy total RNA extraction kit (Qiagen, Mississauga, ON, Canada). The concentration of total RNA was quantified by measurement of optical density at 260 nm and was deemed sufficient to continue with the procedure (data not shown).
Suppression Subtractive Hybridization (SSH)
The SSH procedure was performed essentially as previously described [27], with changes appropriate to the mouse model system as described herein. SSH was used to compare gene expression between fluorescent and nonfluorescent cells of 12.0–12.5 dpc pSRYp-GFP transgenic mouse testis. To generate sufficient amounts of double-stranded cDNA for SSH, fluorescent and nonfluorescent cDNAs were amplified separately using the SMART polymerase chain reaction (PCR) cDNA synthesis kit according to the manufacturers' instructions (BD Biosciences Clontech, Mississauga, ON, Canada). To generate the first strand cDNA, 0.3 µg of total RNA from fluorescent and nonfluorescent cells was reverse-transcribed with an oligo-dT30 primer (CDS: 5'-AAGCAGTGGTAACAACGCAGAGTACT(30)(A/C/G/T)(A/G/C)-3') using PowerScript reverse transcriptase (BD Biosciences Clontech). Second cDNA strands were produced with the SMART II 5'-anchored oligo, followed by PCR amplification for 21 cycles using Advantage 2 DNA polymerase (BD Biosciences Clontech). The final PCR products consisted of pools of cDNA fragments in the range of 300 base pairs (bp) to 4 kb. The fluorescent cell cDNAs (F) were subtracted against the nonfluorescent cell cDNAs (NF) using PCR-Select cDNA subtraction technology according to the provided protocol (user manual: PT1117-1; BD Biosciences Clontech), to give the (F-NF) subtraction cDNA pool.
For the (F-NF) cDNA pool, the efficiency of normalization and subtraction was assessed by comparing the abundance of given cDNAs before and after subtraction. This was accomplished via PCR amplification, using mouse specific primers for the constitutively expressed gene glyceraldehyde 3-phosphatase dehydrogenase gene (Gapdh sense: 5'-TGTTCCAGTATGATTCCACCCACG-3;' Gapdh antisense: 5'-CTGTTGAAGTCGCAGGAGACAACC-3'), for Sox9, a gene known to be expressed and upregulated in pre-Sertoli cells at the time of sex differentiation (Sox9 sense: 5'-GCAAGCTGGCAAAGTTGA-3;' Sox9 antisense: 5'-TGCTGCTTCGACATCCACACGTG-3') and for GFP (GFP sense: 5'-AGCAAGGGCGAGGAGCTGTTC-3;' GFP antisense: 5'-CCAGGATGTTGCCGTCCTCCT-3'). PCR amplification was performed using Advantage 2 DNA polymerase (BD Biosciences Clontech), and 5-µl aliquots were removed at various numbers of PCR cycles for analysis.
Cloning of Subtracted cDNAs
Sequences from the (F-NF) genital ridge cell subtracted cDNA pool were cloned into the pT-Adv plasmid (BD Biosciences Clontech), to construct the fluorescent minus nonfluorescent (F-NF) subtracted cDNA library. Ligated DNA was used to transform competent TOP10F' Escherichia coli, which was plated onto X-Gal/LB agar containing kanamycin (40 µg/ml). Individual colonies were transferred into 96-well plates, grown in LB freezing media (8.8% glycerol, 55 mM K2HPO4, 1 mM MgSO4, 15 mM NH4(SO4), 26 mM KH2PO4) for 14 h at 37°C and then frozen at –70°C.
Differential Hybridization Screening
Differential hybridization screening was performed following previously described methodologies [27]. The (F-NF) genital ridge cell cDNA library was used to established macroarrays for differential screening. The insert of each cDNA clone was amplified in 96-well plates by PCR (25 cycles) using the PCR-nested primers 1 and 2R as supplied by the manufacturer (Clontech) and AmpliTaq DNA polymerase (Roche Molecular Systems Inc, Laval, PQ, Canada). To establish the cDNA macroarrays, an aliquot of each amplification product was denatured in 0.3 M NaOH with 5% bromophemol blue, and 10 µl was vacuum transferred onto a nylon membrane (Hybond-N+, Amersham Pharmacia Biotech, Baie d'Urfé, PQ, Canada) using a 96-well dot-blot apparatus. Membranes were then exposed to 150 mJ ultraviolet light to perform DNA cross-linking (Gs Gene Linker, Bio-Rad, Mississauga, ON, Canada). Control cDNAs (Gapdh, GFP) were transferred onto the microarrays. For each 96-well plate, three identical cDNA microarray replicate membranes were generated.
The (F-NF), (F), and (NF) cDNA pools were used to generate complex hybridization probes for differential screening of microarrays of the (F-NF) genital ridge cell cDNA library. The cDNA pools were first amplified by performing the secondary nested PCR, and products were purified using Qiaquick PCR purification kit (Qiagen). To prevent nonspecific interaction of the probes to cDNAs on microarrays during hybridization, the adaptors were removed by three successive digestions with AfaI, SmaI, and EagI. The cDNA mixtures were again purified (Qiaquick PCR purification kit; Qiagen) and 75–100 ng was labeled with 32P-dCTP via random priming (Megaprime DNA Labeling System; Amersham Pharmacia Biotech). The radioactive probes were purified (QIAquick Nucleotide Removal kit; Qiagen) and quantified using a beta-counter.
The cDNA microarray membranes were incubated for 4 h at 71°C in 20 ml of prehybridization solution (600 mM NaCl, 120 mM Tris, pH 7.4, 4 mM EDTA, 0.1% Na4P2O7, 0.2% SDS, and 500µg/ml heparin), to which heat denatured (5 min, 100°C) oligos were added to prevent nonspecific binding. These oligos included 1 mM each of PCR-Nested 1, PCR-Nested 2R, PCR-Nested 1 INV (5'-ACCTGCCCGGGCGGCCGCTCGA-3'), PCR-Nested 2R-INV (5'-ACCTCGGCCGCGACCACGCT-3'), as well as 2 mM poly dAdT (Amersham Pharmacia Biotech). Equal counts-per-minute amounts of each of the three heat-denatured cDNA probes were added to separate 15-ml aliquots of hybridization solution and used to individually hybridize the replicated (F-NF) cell microarray membranes. The hybridization solution was identical to the prehybridization solution with the addition of 15 µg salmon sperm DNA and 10% dextran sulfate. The membranes were hybridized for 16 h at 71°C, then were washed for 15 min in 2x saline sodium citrate (SSC) and 0.1% SDS followed by two 1-h washes in 0.1x SSC and 0.1% SDS at 71°C. Membranes were exposed to a phosphor screen for 4 h and the images were digitized (Storm 840; Amersham Pharmacia Biotech). The differentially hybridizing cDNA clones were further characterized by DNA sequencing and WISH.
DNA Sequencing and Sequence Analysis
The cDNA clones identified as differentially expressed by the (F-NF) cell cDNA probe were amplified by PCR for 15 cycles with the PCR-nested 1 and PCR-nested 2 oligos from the PCR product generated initially for the microarray analysis. The PCR product was purified (QIAquick; Qiagen) and verified by agarose gel analysis for the presence of a single cDNA band before proceeding with sequencing. Sequencing reactions were performed on cDNA clones via the dideoxy sequencing method (Big dye terminator 3.0; ABI prism; Applied Biosystems, PE, Branchburg, NJ) using the oligo PCR-Nested 1 (1 mM). Sequencing reactions were analyzed on an ABI Prism 310 sequencer (Applied Biosystems). Nucleic acid sequences were analyzed by BLAST (Basic Local Alignment Search Tool) against GenBank data banks (NR, EST, and mouse genome). A cDNA sequence was considered homologous to a GenBank sequence when at least 100 bp matched with an E probability value of less than e–10.
Whole-Mount In Situ Hybridizations
The cDNA clones of the known genes that were differentially expressed in the SSH experiment were further characterized via WISH. To provide a standardized method for generating linear template DNA, clones were amplified by PCR using SP6-PCR-Nested1 (5'-TAATACGACTCACTATAGGGAGCGTGGTCGCGGCCGAGGT-3') and T7-PCR-Nested2 (5'-ATTTAGGTGACACTATAGAATCGAGCGGCCGCCCGGGCAGGT-3'). The PCR products were used as templates to generate digoxigenin-labeled antisense riboprobes for WISH as previously described [28]. The PCR products were transcribed using Sp6 or T7 RNA polymerase as appropriate. Expression was analyzed in 11.5, 12.5, 13.5, and 14.5 dpc gonads, using two pairs of embryos of each sex per gene.
Depletion of Germ Cells by Busulfan
Two cDNA sequences were selected for an expression analysis in embryos lacking germ cells. These embryos were generated by injecting pregnant females with 40 mg/kg busulfan at 9.5 dpc and then harvesting embryos at 12.5–13.5 dpc [29]. As a control for this experiment, expression of the Chk1 gene, known to be expressed in germ cells but not in pre-Sertoli cells, was verified in germ cell-depleted embryos.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The pSRYp-GFP-positive embryos from timed matings were dissected at 12.0–12.5 dpc and fluorescent (pre-Sertoli) cells and nonfluorescent cells from the genital ridge were efficiently separated via FACS purification, with an efficiency of 98%. Reverse transcription-polymerase chain reaction (RT-PCR) was performed on RNA from fluorescent cells to reveal the presence of endogenous Sry transcripts and the absence of Oct4 transcripts (data not shown), consistent with the fluorescent cells being of pre-Sertoli cell lineage and not of germ cell lineage as previously shown [7].
Production of Subtracted Libraries and Assessment of Normalization and Subtraction Efficiencies
To verify the efficiency of the SSH procedure, several control experiments were performed. To ensure that normalization of cDNAs had occurred, the quantity of the housekeeping gene Gapdh was assessed in subtracted (F-NF) and unsubtracted (F) cDNA pools. As expected, Gapdh abundance was decreased following subtraction (Fig. 1A). To assess the efficiency of enrichment of differentially expressed genes, the abundance of transcripts for GFP and Sox9 (pre-Sertoli cell expressed gene) was examined. Both GFP and Sox9 transcripts were enriched in the pre-Sertoli cell subtracted (F-NF) cDNA pools (Fig. 1, B and C). Interestingly, GFP but not Sox9 transcripts were still visible in PCR cycles taken from the unsubtracted cDNA pool. This can be explained by the PCR cycling conditions chosen and possibly also reflects relative differences in concentrations of message at the start of the procedure.
|
Once the subtraction efficiency was shown to be satisfactory, a differential hybridization screening procedure was performed on 658 bacterial colonies randomly selected from the (F-NF) subtracted cDNA library. The (F-NF) subtracted cDNA pool as well as (F) and (NF) nonsubtracted cDNA pools were used to generate three different hybridization probes for the differential screening of the selected clones spotted on three identical sets of microarrays. Representative differential screening results are illustrated in Figure 2. The differential screening procedure identified 154 cDNA clones as true positives, i.e., showing enhanced expression via the (F-NF) probe compared with the (F) probe. One hundred ten of these clones were analyzed by sequencing and 90 generated adequate sequencing results. Comparison of the cDNA sequences obtained with GenBank sequence databases revealed that 38.8% (35/90) corresponded to 32 known genes (taking into account redundancy of clones), 16.6% (15/90) corresponded to GFP, and 44.4% (40/90) of the sequenced clones corresponded to novel genes (analysis to be presented elsewhere). Only six sequences were identified more than once, except for GFP, which was identified 15 times. Among the known expressed genes identified was Sf1, a gene expressed in the genital ridge and implicated in sex determination/differentiation. In addition, Vanin-1, Vcp1, Sparc, and Aldh3a1 were identified, representing genes that were previously identified in screens of fetal testes versus ovaries [21, 22, 25].
|
Analysis of mRNA Expression
To verify that genes identified by SSH are differentially expressed between pre-Sertoli cells and the Leydig/endothelial and germ cell compartments and to test whether these genes are also differentially expressed between male and female embryos, WISH was performed on male and female embryos. Initial screening was carried out using 12.5-dpc embryos of both sexes for the 32 known genes. Most of these genes (23) were expressed in both sexes; 5 genes were overexpressed in testis; 1 gene was overexpressed in ovary; and no expression was detected for 3 genes (Table 1). Representative examples of WISH results are shown in Figure 3 for the genes Catna, Sparc, Fdxr, Hmgcs2, and Rgs2. For the clones showing sex-specific or overexpression patterns in gonads of one sex, additional in situ experiments were performed at 11.5, 12.5, 13.5, and 14.5 dpc. We present here the expression patterns for Bromodomain-containing 3 (Brd3) (Fig. 4, A–D) and Palmitoyl-protein thioesterase 1 (Ppt1) (Fig. 4, E–H). At the detection level of the WISH procedure, Brd3 is ubiquitously expressed but is more strongly expressed in the testis compared with the ovary (see Fig. 4B) and to other tissues (data not shown). In contrast, Ppt1 was only detected within the developing testis. To further confirm that these genes are expressed in the pre-Sertoli cell population and not within germ cells, WISH was performed on embryos lacking germ cells after treating pregnant females with busulfan (Fig. 5). Expression of Brd3 and Ppt1 was not diminished within busulfan-treated compared with nontreated testes.
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Screening the (F-NF) subtracted cDNA library by membrane hybridization allowed the identification of 32 known gene candidates, the analysis of which is presented herein. These genes included Sf1, known for its role in sex determination/differentiation, as well as Vanin-1, Sparc, Vcp1, and Aldh3a1, four genes previously identified in published screens for genes overexpressed in developing testis compared with ovary [21, 22, 25]. Furthermore, we confirmed by RT-PCR that Sry is expressed in fluorescent cells and that Sox9 is overexpressed within the (F-NF) subtracted cDNA pool compared with the (F) nonsubtracted cDNA pool. The identification of these genes provides an important confirmation of the biological validity of our current model. Other genes known to be implicated in sex determination, such as Wt1, Dax1, Lhx9, or Gata4 were not identified in our current hybridization screen, and although Sry and Sox9 were subsequently identified by RT-PCR, they were not identified within the hybridization screen. This is possibly due to the relatively small sample size of clones screened as well as the potentially large complexity and low redundancy of the cDNA population of the pre-Sertoli cell. These observations are consistent with the findings of other reported genital ridge differential screens [21, 22, 25] but are in contrast with expression profiling experiments reported for other tissues such as granulosa cells of the developing bovine follicle [27].
The known genes identified by SSH were further characterized via WISH to determine differential expression patterns within the developing testis and also between developing testis and ovary. A high proportion of the genes tested by WISH (21/32) were expressed in both developing testis and ovary. This is not surprising due to the common origin of Sertoli and granulosa cells [6] and the fact that our subtraction was not between testicular and ovarian cells. This included genes known to be implicated with Sertoli cells, such as Cftr, implicated in nutrition of spermatozoids [30] and Gatm involved in creatine biosynthesis. In addition, genes involved with energy metabolism were identified, including Ptpn1, which acts as a negative regulator of insulin signaling. Recently, it was demonstrated that insulin receptor family members are needed for testes determination [31]. The fact that, for three genes, there was no WISH signal detected could be explained by the moderate sensitivity of the WISH method or could simply be due to technical failure.
For the six genes showing a differential expression between testes and ovaries, Sf1, Vanin-1, and Ppt1 were only expressed in pre-Sertoli cells, Brd3 and Sel1L were more strongly expressed in pre-Sertoli cells compared with interstitial cells, while Catna1 and Sparc were expressed equally in pre-Sertoli cells and interstitial cells. Two genes, Fdxr and Hmgcs2, were expressed in both testicular and ovarian genital ridges. One gene, Rgs2, was found to be overexpressed in ovaries compared with testes at 12.5 dpc but on closer examination was observed to be downregulated in ovaries at 13.5 dpc, underscoring the fact that sex differentiation occurs later in ovaries than in testis.
Ppt1 and Brd3 were selected for more detailed characterization of genital ridge expression. Ppt1 encodes a soluble lysosomal enzyme that hydrolyzes the thioester bond that attaches long-chain fatty acids, mostly palmitate, to the cysteine residue of S-acylated proteins [32, 33]. Palmitoylation has been shown to be important in different cellular processes, including neural cell differentiation (review: [34]) and possibly in insulin-regulated cellular functions [35]. Recently, nonlysosomal roles for Ppt1 were described in synaptosomes [36] and during apoptosis [37]. It was previously reported that Ppt1 is ubiquitously expressed in adult tissue, with increased expression within the testis [32, 38], while in Drosophila, Ppt1 shows testicular expression levels five time that of other tissues [39]. Here, we report that, within the mouse genital ridge, Ppt1 was only expressed in pre-Sertoli cells of the developing testis via SSH and WISH but was not seen in developing ovary via WISH. These observations suggest that Ppt1 may be involved in mammalian sex determination and differentiation.
Brd3 is a potential transcription factor containing two bromodomains. Members of this family include Brd2, a gene that is strongly expressed in ovary and testis (both Sertoli and germ cells) [40] and BrdT [41], a gene only expressed in the germ cells of the testis. Significantly, this family of genes is related to the female sterile homeotic protein gene in Drosophila, a gene required maternally for proper expression of other homeotic genes, such as Ubx, which is involved in pattern formation. Ubiquitous expression of UBX protein within the mesoderm in Drosophila results in the formation of ectopic gonad tissue anterior to the normal location of the gonads [42]. We now provide evidence that Brd3 may play a role in testis formation in mammals. Identification of Brd3 expression within the male genital ridge, with increased expression within pre-Sertoli cells compared with other cell types, provides evidence that Brd3 may play a role in testis formation in mammals and may help to define the developmental tissue code required for testis formation. Further studies are now required to either confirm or refute these ideas.
In summary, we have combined fluorescent marking of cell populations in transgenic models with differential expression screening to study gene expression in pre-Sertoli cells of the mammalian genital ridge. Using these methods, we have identified new candidate genes, including Ppt1 and Brd3, the genital ridge expression patterns of which suggest they may be involved in the processes of sex determination and differentiation.
|
FOOTNOTES |
|---|
2 Correspondence: David W. Silversides, Faculté de médecine vétérinaire, Université de Montréal, PO Box 5000, St.-Hyacinthe, Québec, Canada J2S 7C6. FAX: 1 450 778 8103; david.w.silversides{at}umontreal.ca
Received: 7 March 2004.
First decision: 29 March 2004.
Accepted: 27 April 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C.-H. Lee, C.-C. Wu, Y.-N. Wu, and H.-S. Chiang Gene copy number variations in Asian patients with congenital bilateral absence of the vas deferens Hum. Reprod., December 17, 2008; (2008) den413v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Beverdam and P. Koopman Expression profiling of purified mouse gonadal somatic cells during the critical time window of sex determination reveals novel candidate genes for human sexual dysgenesis syndromes Hum. Mol. Genet., February 1, 2006; 15(3): 417 - 431. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Ottolenghi, S. Omari, J. E. Garcia-Ortiz, M. Uda, L. Crisponi, A. Forabosco, G. Pilia, and D. Schlessinger Foxl2 is required for commitment to ovary differentiation Hum. Mol. Genet., July 15, 2005; 14(14): 2053 - 2062. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
