Search tips
Search criteria 


Logo of plosonePLoS OneView this ArticleSubmit to PLoSGet E-mail AlertsContact UsPublic Library of Science (PLoS)
PLoS One. 2013; 8(6): e66062.
Published online Jun 7, 2013. doi:  10.1371/journal.pone.0066062
PMCID: PMC3676325
The Ligand Binding Domain of GCNF Is Not Required for Repression of Pluripotency Genes in Mouse Fetal Ovarian Germ Cells
Leah M. Okumura,1,2,3 Bluma J. Lesch,1,2,3* and David C. Page1,2,3*
1Howard Hughes Medical Institute, Chevy Chase, Maryland, United States of America
2Whitehead Institute, Cambridge, Massachusetts, United States of America
3Department of Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts, United States of America
Gabriel Livera, Editor
University Paris Diderot/University Paris 7, France
* E-mail: leschb/at/ (BJL); dcpage/at/ (DCP)
Competing Interests: The authors have declared that no competing interests exist.
Conceived and designed the experiments: LMO DCP. Performed the experiments: LMO BJL. Analyzed the data: LMO BJL. Wrote the paper: LMO BJL DCP.
Received February 9, 2013; Accepted May 4, 2013.
In mice, successful development and reproduction require that all cells, including germ cells, transition from a pluripotent to a differentiated state. This transition is associated with silencing of the pluripotency genes Oct4 and Nanog. Interestingly, these genes are repressed at different developmental timepoints in germ and somatic cells. Ovarian germ cells maintain their expression until about embryonic day (E) 14.5, whereas somatic cells silence them much earlier, at about E8.0. In both somatic cells and embryonic stem cells, silencing of Oct4 and Nanog requires the nuclear receptor GCNF. However, expression of the Gcnf gene has not been investigated in fetal ovarian germ cells, and whether it is required for silencing Oct4 and Nanog in that context is not known. Here we demonstrate that Gcnf is expressed in fetal ovarian germ cells, peaking at E14.5, when Oct4 and Nanog are silenced. However, conditional ablation of the ligand-binding domain of Gcnf using a ubiquitous, tamoxifen-inducible Cre indicates that Gcnf is not required for the down-regulation of pluripotency genes in fetal ovarian germ cells, nor is it required for initiation of meiosis and oogenesis. These results suggest that the silencing of Oct4 and Nanog in germ cells occurs via a different mechanism from that operating in somatic cells during gastrulation.
In mice, several factors are essential for maintaining pluripotency in the early embryo and in embryonic stem (ES) cells, including the POU family transcription factor OCT4 (also called POU5F1) and the homeobox protein NANOG. Oct4 is expressed in the inner cell mass (ICM) of the blastocyst. In its absence, ICM cells differentiate into trophectoderm-like cells and embryos are inviable [1]. Nanog is expressed in the epiblast when it differentiates from the hypoblast, and functions to maintain pluripotency in the early embryo. Nanog-deficient embryos fail to form an epiblast and die shortly after implantation [2].
Both Oct4 and Nanog are down-regulated in somatic cells during gastrulation, but they continue to be expressed in primordial germ cells (PGCs), along with several other pluripotency genes, including Sox2 and Dppa3 (also known as Stella) [3], [4]. Pluripotency gene expression is maintained in germ cells until after they migrate to the developing gonads, which they enter by E11.5 [5], [6]. In particular, Oct4 and Nanog are both required for survival of PGCs. Conditional ablation of Oct4 or Nanog specifically in PGCs results in apoptosis [7], [8].
In XX (female) embryos, germ cell expression of pluripotency genes is maintained until about E14.5, when the germ cells enter meiosis [5], [8]. However, it is not known whether this repression of pluripotency genes in fetal ovarian germ cells is required for meiotic initiation. In addition, the mechanism of repression of pluripotency genes in fetal germ cells is largely unknown. Recent studies in C57BL/6 mice demonstrated that silencing of pluripotency genes in fetal germ cells depends on the gene Dazl (Deleted in azoospermia-like), which encodes an RNA-binding protein expressed specifically in germ cells; Dazl-deficient germ cells display extended expression of pluripotency markers [9]. However, as Dazl regulates a broad transition in fetal germ cell state, that of PGCs into gametogenesis-competent cells (GCCs) [9], its role in down-regulating pluripotency genes may be indirect.
We set out to determine what factors might directly mediate the effects of Dazl in repressing the pluripotency genes Oct4 and Nanog in fetal ovarian germ cells. GCNF (germ cell nuclear factor, Nr6a1) was a likely candidate: it is an orphan nuclear receptor known to repress Oct4 and Nanog in the soma during gastrulation [10], [11]. The GCNF protein binds directly to DR0 sites in the promoters of both the Oct4 and Nanog genes and is thought to repress them via interaction with the nuclear corepressors SMRT and NCoR [10], [11], [12]. We hypothesized that GCNF might also be required for repression of Oct4 and Nanog in fetal ovarian germ cells.
Using high-throughput mRNA sequencing and single molecule fluorescence in situ hybridization, we found that, in fetal ovarian germ cells, Gcnf expression peaks at the time when Oct4 and Nanog are repressed. To explore the function of Gcnf in fetal ovarian germ cells, we generated whole-body and conditional-knockout mice in which the ligand-binding domain (LBD) of GCNF was disrupted. We found that, although disruption of the LBD of GCNF prevented Oct4 repression in the soma during gastrulation, it did not affect Oct4 or Nanog repression in fetal ovarian germ cells. In Gcnf-mutant embryos, all aspects of germ cell development examined were similar to those found in wild-type embryos. We conclude that Gcnf is not required for fetal ovarian germ cell development, and that repression of pluripotency genes must be differentially regulated in somatic and germ cells.
Expression of Gcnf in fetal ovarian germ cells
We first determined if Gcnf expression coincided with the down-regulation of Nanog and Oct4 in fetal ovarian germ cells. Two different approaches were used: analysis of high-throughput mRNA sequencing data and single molecule fluorescence in situ hybridization (smFISH). Both NANOG and OCT4 proteins are detectable at E13.5, are down-regulated at E14.5, and are undetectable by E16.5 [5], [13]. If Gcnf is required for this down-regulation, it should be expressed in germ cells at the time when Nanog and Oct4 repression occurs.
First, we analyzed high-throughput mRNA sequencing data from wild-type gonads as well as gonads lacking germ cells (derived from KitW/KitW-v mutant mice), at E12.5, E14.5, and E16.5 (data kindly provided by Jacob Mueller and Mark Gill). We found that Gcnf transcripts are present in wild-type fetal ovaries as early as E12.5, with peak expression at E14.5, and a precipitous drop in expression at E16.5 (Fig. 1A). Gcnf transcript levels are very low at all three time points in wild-type fetal testes as well as in germ-cell-deficient W/Wv gonads of both sexes (Fig. 1A).
Figure 1
Figure 1
Gcnf expression in fetal ovarian germ cells.
Second, we used smFISH to examine Gcnf transcripts in wild-type mesentery, genital ridges, or gonads at E9.5, E10.5, and E14.5 (Fig. 1B). Germ cells were identified by staining for the germ-cell-specific cell-surface antigen SSEA1, as well as by their distinctive nuclear morphology, visualized by DAPI staining. We found that, at E9.5, consistent with published data [14], Gcnf transcripts were widespread and easily detectable in both somatic and germ cells. By contrast, at E10.5, most cells did not express Gcnf; its mRNA could be detected in only scattered somatic and germ cells. By E14.5, mRNA expression of Gcnf had increased dramatically in germ cells, but no expression was detected in somatic cells. Taken together, these results indicate that at the time when pluripotency markers are repressed in the fetal ovarian germ cells, Gcnf is expressed specifically in these cells.
Generation and phenotypic characterization of mutant mice lacking the ligand-binding domain of Gcnf
To determine whether Gcnf is required for the silencing of Oct4 and Nanog in fetal ovarian germ cells, we generated Gcnf-mutant mice from targeted ES cells in which the genomic region encoding the LBD of GCNF is disrupted. We obtained targeted ES cells from the EUCOMM (European Conditional Mouse Mutagenesis) Project. This targeted allele, referred to here as Gcnfgt, contains a gene trap cassette and loxP sites flanking exon 7 (Fig. 2A,B). Exon 7 encodes the N-terminal portion of the LBD, which has been shown to be important for interaction of GCNF with the corepressors NCoR and SMRT [11], [12].
Figure 2
Figure 2
Targeted disruption of Gcnf.
We first tested whether disruption of the LBD of GCNF resulted in defects similar to those observed when the DNA binding domain (DBD) is disrupted. Previously, disruption of the DBD has been shown to abrogate GCNF function, resulting in severe morphological defects by E9.5, including failure to turn from the lordotic to the fetal position, failure of the neural tube to close completely, a truncation in somite development, and the failure to downregulate Oct4 expression throughout the embryo. DBD-mutant embryos eventually die, surviving only until about E10.5 [11], [15], [16].
We tested LBD-mutants for prolonged expression of Oct4 and morphological defects at E9.5. Mice carrying the Gcnfgt allele were first bred to mice carrying a ubiquitously expressed FLP recombinase, ACTB:FLPe [17], in order to excise the gene trap and NeoR cassette. This left behind a single FRT site and LoxP sites flanking exon 7, enabling conditional deletion of exon 7 (Fig. 2C). This allele is referred to as Gcnffl. We then generated mice carrying both the Gcnffl allele and the ubiquitously-expressed tamoxifen-inducible Ubc-Cre-ERT2 [18], and injected adult Gcnffl/+; Ubc-Cre-ERT2 males with tamoxifen to obtain adult male mice heterozygous for a full-body deletion of Gcnf exon 7. Deletion of exon 7 should cause a frameshift and a premature stop codon near the 5′ end of exon 8, thereby disrupting translation of the LBD (Figure 2C,D). We refer to this allele as GcnfΔ. We crossed the GcnfΔ/+ founder males obtained in this manner with wild-type females to propagate the GcnfΔ allele.
We examined mutant embryos homozygous for the full-body Gcnf deletion allele (GcnfΔ/Δ) as well as wild-type (Gcnf+/+) and heterozygous (GcnfΔ/+) littermate controls at E9.5. As expected, quantitative RT-PCR on whole embryos at E9.5 showed that wild-type Gcnf+/+ controls expressed low levels of Oct4 transcript at this time point. In comparison, GcnfΔ/Δ homozygotes had greatly increased levels of Oct4 expression, suggesting that the LBD of GCNF is required for the repression of Oct4 during gastrulation (Fig. 3A). Additionally, at E9.5, GcnfΔ/Δ homozygotes lacking the LBD displayed severe morphological defects similar to those previously reported in mutants lacking the DBD [11], [15]. GcnfΔ/Δ homozygotes were smaller than their wild-type littermates, had truncated somitogenesis, and failed to turn from the lordotic to the fetal position (Fig. 3B).
Figure 3
Figure 3
GcnfΔ/Δ mutants fail to down-regulate Oct4 and display morphological defects.
Because homozygotes lacking the LBD of GCNF recapitulate the phenotype of Gcnf mutants lacking the DBD, we conclude that the LBD is required for GCNF function during early embryogenesis.
Conditional deletion of Gcnf ligand-binding domain
Ovarian germ cells down-regulate expression of pluripotency markers at about E14.5 [5], [13]. We next wished to determine whether Gcnf is required for the repression of Oct4 and Nanog in ovarian germ cells at this time. Because Gcnf is required for embryonic development, and full-body mutants are inviable beyond E10.5 [15], [16], we generated conditional knockout mice.
To generate embryos conditionally lacking exon 7, we mated heterozygous Gcnffl/+ females to heterozygous Gcnffl/+ males carrying the ubiquitously expressed tamoxifen-inducible Ubc-Cre-ERT2. Pregnant females were injected with tamoxifen twice, at E10.5 and E11.5, to induce Cre-mediated recombination throughout the embryo. This time interval was chosen for Cre induction because it is after Gcnf is down-regulated throughout the embryonic soma [14] but before Gcnf is up-regulated in ovarian germ cells at E12.5.
To determine the efficiency of recombination in the fetal ovary, we collected ovaries from tamoxifen-treated mutant embryos (Gcnffl/fl; Ubc-Cre-ERT2TAM) and tamoxifen-treated wild-type littermate controls (Gcnf+/+TAM) at E15.5 and E16.5 and used quantitative RT-PCR (qRT-PCR) to measure levels of Gcnf exon 7-containing transcript. In all mutants examined, recombination was nearly complete, with an average of less than 1.5% of exon 7 transcript detectable compared to controls (Fig. 4A).
Figure 4
Figure 4
Exon 7 of Gcnf is efficiently deleted in mutants.
To ensure that the inability to detect exon 7-containing transcript was not due to an absence of germ cells in mutant embryos, we performed immunostaining with MVH (mouse vasa homolog), a known germ-cell-specific marker [19]. MVH staining was readily apparent in ovaries from both wild-type controls and Gcnf-mutant embryos (Fig. 4B). In Gcnf-mutant ovaries, germ cells appeared to be of normal number and morphology as compared to wild-type ovaries.
Gcnf is not required for repression of Oct4 and Nanog in fetal ovarian germ cells
Because Gcnf is required for the repression of Oct4 and Nanog during gastrulation [10], [11], we hypothesized that it might also be required for their repression in fetal ovarian germ cells. If this were the case, Gcnf-deficient germ cells would continue to express OCT4 and NANOG at E15.5 and E16.5, when these proteins are undetectable in wild-type ovarian germ cells [5], [13]. We used immunostaining to examine OCT4 and NANOG protein in ovaries from Gcnf-mutant (Gcnffl/fl; Ubc-Cre-ERT2TAM) and wild-type control (Gcnf+/+TAM) littermates. For each embryo, we examined one ovary by immunostaining and extracted RNA from the other to verify by qRT-PCR whether exon 7 of Gcnf was efficiently deleted. As expected, both OCT4 and NANOG proteins were readily detectable in Gcnf+/+ wild-type ovarian germ cells at E13.5 (Fig. 5A,D), but neither was detectable at E16.5 in Gcnf+/+TAM wild-type ovaries (Fig. 5B,E). Gcnf-mutant ovaries were indistinguishable from those of wild-type littermates, with neither OCT4 nor NANOG protein detectable at E16.5 (Fig. 5C,F), indicating that both genes are down-regulated appropriately in the absence of GCNF (Fig. 5). Similar results were observed at E15.5 (data not shown). We conclude that Gcnf is not required for the proper down-regulation of OCT4 and NANOG in fetal ovarian germ cells.
Figure 5
Figure 5
Gcnf-mutant germ cells down-regulate OCT4 and NANOG similarly to wild type.
Gcnf is not required for meiotic initiation or early oogenesis
In light of this negative result, we next considered whether other aspects of germ cell development were affected in Gcnf-mutant ovaries, namely meiotic initiation and oogenesis. We began by examining expression of meiotic markers. SYCP3 (synaptonemal complex protein 3) is part of the lateral element of the synaptonemal complex, which forms between homologous chromosome pairs during meiosis. Phosphorylated histone H2AX (γH2AX) marks sites of DNA double-strand breaks that form during meiotic prophase. Both SYCP3 and γH2AX are readily detectable in wild-type ovarian germ cells at E16.5 [20]. We used immunostaining to examine wild-type control (Gcnf+/+TAM) and Gcnf-mutant (Gcnffl/fl; Ubc-Cre-ERT2TAM) ovaries for SYCP3 and γH2AX at E16.5, using the germ-cell marker GCNA (germ cell nuclear antigen) to identify germ cells (Fig. 6). In Gcnf-mutant as well as wild-type control ovaries, SYCP3 was expressed in the nuclei of all germ cells (Fig. 6A–D). Higher magnification revealed the threadlike structures of condensed chromosomes (Fig. 6B,D). γH2AX was also found in all germ cell nuclei of Gcnf-mutant as well as wild-type control ovaries (Fig. 6E,F). Gcnf-mutant and wild-type ovarian germ cells were indistinguishable from each other in staining and structure. These results indicate that meiotic initiation and early meiotic prophase proceed normally in ovarian germ cells lacking GCNF.
Figure 6
Figure 6
Gcnf mutants undergo meiosis normally.
Lastly, we tested whether oogenesis was affected in Gcnf mutants. The Y-box protein MSY2 (also called YBX2) is expressed in the cytoplasm of maturing oocytes during the diplotene stage of meiosis I [21], [22]. We assayed for MSY2 expression in wild-type and Gcnf-mutant ovaries at E16.5 by immunofluorescence, identifying germ cells by expression of GCNA. All germ cells were positive for MSY2 in Gcnf-mutant and wild-type ovaries (Fig. 7), indicating that early oogenesis is unaffected by the absence of Gcnf. Taken together, these results indicate that Gcnf is not required for the early steps of meiosis and oogenesis.
Figure 7
Figure 7
Gcnf mutants initiate oogenesis.
The repression of pluripotency genes in ovarian germ cells around the time of meiotic initiation is a significant aspect of their development, and yet very little is known about this event. Dazl (Deleted in azoospermia-like) has been shown to be required for primordial germ cells to differentiate following their migration to the gonads. Subsequent downstream events, including down-regulation of pluripotency markers, are disrupted in the absence of Dazl [9]. Because Dazl controls a broad differentiation event several days before pluripotency marker repression, this is likely an indirect effect of Dazl function. We set out to determine what molecular regulators might mediate Dazl's effects in the repression of Oct4 and Nanog.
Gcnf represented a good candidate for this role because it is a known direct repressor of Oct4 and Nanog in somatic cells, both in vivo and in vitro [10], [11]. Here we showed that Gcnf is expressed in ovarian germ cells at about the time that Oct4 and Nanog are down-regulated in these cells. However, when we conditionally disrupted the LBD of Gcnf in the fetus, we found that, despite its role in Oct4 repression during gastrulation, Gcnf is not required for the regulation of Oct4 or Nanog in ovarian germ cells. In addition, Gcnf appears to be dispensable for the initiation of meiosis and oogenesis in fetal ovarian germ cells.
We have considered and ruled out several alternative explanations for our findings. One formal possibility is that ablation of Gcnf function was inefficient with our conditional allele, but the available evidence does not support this explanation. Each embryo provided a pair of ovaries, of which one ovary was used for RNA extraction and RT-PCR, while the other was processed for paraffin sectioning and immunostaining. Quantitative RT-PCR on individual mutant ovaries showed an average of less than 1.5% of exon 7 transcript remaining, indicating that exon 7 was deleted in the vast majority of germ cells.
We demonstrated that, in wild-type embryos, Gcnf transcript is expressed in germ cells and somatic cells at E9.5, becomes nearly undetectable in both tissue types at E10.5, and is expressed specifically in germ cells at E14.5. Because we induced the exon 7 deletion at E10.5, it is theoretically possible that small amounts of intact GCNF protein expressed at E9.5 persisted in germ cells until E14.5, when Oct4 and Nanog are down-regulated. However, the germ cell population expands from about 1,000 cells at E10.5 to more than 25,000 cells at E13.5 [23], which would have severely diluted any remaining GCNF protein. Therefore, it seems unlikely that enough GCNF protein would remain in every germ cell to effectively carry out its function.
Another possibility is that the DNA binding domain (DBD) is translated and functional in the absence of the ligand-binding domain (LBD). Given the modular structure of nuclear receptors, including GCNF, it is quite possible that the N-terminal portion of the protein, including the DBD, is produced and properly folded. Unfortunately, we were unable to examine whether the DBD was present in ovarian germ cells due to the lack of appropriate antibodies. However, GCNF has been shown to interact with SMRT and NCoR, two well-characterized nuclear co-repressors [11], [12]. Both of these co-repressors interact specifically with the LBD of GCNF, which is typical of their interactions with other nuclear receptors as well. In addition, we demonstrated that the LBD is required for GCNF function earlier in somatic development, and that in its absence mutants display a phenotype remarkably similar to Gcnf mutants lacking the DBD. Both mutants demonstrate a failure to turn from the lordotic to fetal position, a failure to completely close the neural tube, and truncated somitogenesis, as well as highly elevated Oct4 expression at E9.5 [11], [15], [16]. Although some of these defects can be observed in other mutants, taken together they suggest that disruption of the LBD does in fact abrogate GCNF function and that the LBD is required during Gcnf's role in earlier development, when Gcnf is thought to be required for posterior patterning [15]. Therefore, although we cannot formally rule out the possibility, it is unlikely that the DBD represses Oct4 and Nanog in the absence of the LBD during female germ cell development.
A more plausible explanation for our findings is that Oct4 and Nanog are regulated by different enhancers in the germline and the gastrulating embryonic soma. Indeed, there is published evidence of such enhancer-controlled tissue specificity in the case of Oct4. Yeom and colleagues demonstrated that there are two functionally separable enhancer elements present in the Oct4 promoter, one of which is more proximal to and the other more distal to the transcriptional start site [24]. The proximal enhancer (PE) drives expression of Oct4 in the epiblast and epiblast-derived cells such as embryonal carcinoma cells. The PE is required for expression of Oct4 in the cells of the epiblast around the time of gastrulation. However, the distal enhancer (DE) is sufficient for expression of Oct4 in the pre-implantation embryo as well as in the germline, even in the absence of the PE.
It is unclear how these enhancer elements are differentially regulated, but it seems likely that different transcription factors bind to the two enhancers. GCNF is thought to perform its repressive function by competing with activating nuclear receptors for binding sites in promoters, and it has been shown to bind to response elements in the proximal promoter of Oct4 near, though not within, the PE [11], [25]. It is possible that, during gastrulation, GCNF's proximity to the PE disrupts the activity of transcriptional activators that normally bind there, resulting in the GCNF-dependent repression of Oct4 observed in gastrulating embryos and ES cells [11]. In germ cells, on the other hand, GCNF may bind to the proximal promoter but may not be sufficient to repress Oct4 due to its distance from the DE. In this case, other factors, perhaps binding in or near the DE, would be responsible for repressing Oct4 in the germline.
Our experiments indicate that, although the LBD of Gcnf is required for repression of Oct4 and Nanog in somatic cells, it is not required for repression of these pluripotency factors in ovarian germ cells. The finding that pluripotency genes are differentially regulated in somatic and germ cells is intriguing, and it will be interesting to understand what factors contribute to this disparity. It may be due to differential regulation of enhancer elements in the promoters of these genes in somatic versus germ cells. Additionally, we cannot rule out the possibility that other factors act redundantly or in conjunction with Gcnf in this context. Finally, it will be important to determine what factors mediate Dazl's effects in the repression of pluripotency factors in fetal ovarian germ cells, and whether this repression of pluripotency factors is required for the initiation of meiosis and oogenesis.
Ethics statement
All experiments involving mice conformed to ethical principles and guidelines approved by the Committee on Animal Care at the Massachusetts Institute of Technology (IACUC number: 0711-075-14).
High throughput mRNA sequencing
The Gcnf expression data shown in Fig. 1A were kindly provided by Jacob Mueller and Mark Gill, who are conducting an analysis of fetal gonadal gene expression (personal communication). In brief, total RNA from E12.5, E14.5, and E16.5 gonads from wild-type and W/Wv embryos was processed using the GLOBINclear Kit (Ambion, AM1981) to remove hemoglobin mRNA. cDNA libraries for sequencing were prepared using the TruSeq RNA Sample Prep kit (Illumina). 36-base-pair reads were sequenced on an Illumina GAII sequencer. Reads were aligned to Mus musculus genomic sequence (mm9) using Tophat [26] and reads-per-million were estimated using custom Perl script (available upon request).
Single molecule fluorescence in situ hybridization
All studies were performed on an inbred C57BL/6 background. C57BL/6 Gcnfgt/+ targeted ES cells were obtained from the EUCOMM repository. The sequence of the targeting construct is available at: ES cell clones were injected into Balb/c blastocysts and transferred to pseudopregnant CD1 females to generate chimeras. Animals were genotyped by PCR using primers that amplify the region containing the 3′ LoxP site (fwd: TGCAAGGCCAGGTCTTTAAC; rev: AGGAGCCCTTCCAAGTTACC). Gcnffl/+ animals were generated by mating Gcnfgt/+ heterozygotes to mice carrying ACTB:FLPe [17] to delete the gene trap and neomycin resistance cassette. Offspring were genotyped by PCR for the 3′ LoxP site as described above, as well as with primers that flank the region containing the gene trap and neomycin resistance cassette (fwd: TGGGGCACTGTAACAAGACC; rev: TGACAATCTCCTACTTGCCCTACC). Primers for detecting the presence of Ubc-Cre-ERT2 were as follows: fwd: GATAGTGAAACAGGGGCAATGGTGC; rev: TAGAGTATGGGGGGCTCAGCATCC. Gcnf+/+, GcnfΔ/+, and GcnfΔ/Δ embryos were distinguished using forward primers in exon 7 (TGGAGCAACCATGGTGACAG) or overlapping the 3′ loxP site (ACGAAGTTATGGTCTGAGCTC), together with a common reverse primer downstream of the loxP site (GAGCCCTTCCAAGTTACCTC).
Gcnf-mutant embryos
To obtain GcnfΔ/Δ mutant embryos, GcnfΔ/+ heterozygous males and females were mated. To establish timed matings, female mice were housed with male mice overnight. Noon of the day when a vaginal plug was evident was considered E0.5. Embryos were dissected into PBS at E9.5, and yolk sacs were reserved for genotyping. To obtain Gcnffl/fl; Ubc-Cre-ERT2 mutants, Gcnffl/+ females were mated to Gcnf fl/+; Ubc-Cre-ERT2 males. Embryonic gonads and mesonephroi were dissected into PBS at E15.5 and E16.5, with tail samples reserved for genotyping. The sexual identity of the gonads was determined by scoring the presence or absence of testicular cords.
Tamoxifen administration
To induce recombination of the Gcnffl allele, tamoxifen (Sigma) was dissolved at 20 mg/mL in corn oil, and administered to adult males or pregnant females via intraperitoneal injection at 5 mg tamoxifen/40 g mouse. Each pregnant female was injected twice, at E10.5 and E11.5. Adult males were injected once per day for four consecutive days.
Fetal ovaries were fixed overnight at 4°C in 4% paraformaldehyde in PBS, embedded in paraffin, and sectioned. Slides were dewaxed, rehydrated, heated in Antigen Retrieval Buffer 1 (Spring Bioscience) for 8 min, and blocked for 30 min in 2.5% horse serum (colorimetric) or 3% goat serum (fluorescence). Slides were then incubated with primary antibody for 1 h, diluted as follows: anti-NANOG (Bethyl Laboratories, IHC-00205) 1[ratio]200; anti-OCT4 (BD Transduction Labs, 611203) 1[ratio]100; anti-MVH (Abcam, ab13840): 1[ratio]250; anti-SYCP3 (Santa Cruz, sc-74569) 1[ratio]200; anti-γH2AX (Millipore, 05-636) 1[ratio]200; anti-MSY2 (a gift from Richard Schultz, University of Pennsylvania) 1[ratio]200; anti-GCNA (a gift from George Enders, University of Kansas) undiluted. For colorimetric detection, slides were incubated with rabbit ImmPress reagent (Vector Labs) and developed using ImmPACT DAB substrate (Vector Labs), then counterstained with hematoxylin, dehydrated, and mounted using Permount (Fisher Scientific). For fluorescence detection, slides were incubated with FITC-conjugated anti-rat and Rhodamine-conjugated anti-rabbit or anti-mouse (Jackson ImmunoResearch Laboratories) at 1[ratio]250 dilution, then mounted with VECTASHIELD mounting media with DAPI (Vector Labs). Deconvolved images of SYCP3 immunostaining were obtained using a DeltaVision Elite imaging system (GE Healthcare).
Quantitative RT-PCR
To examine Oct4 expression in GcnfΔ/Δ and wild-type embryos, quantitative RT-PCR (qRT-PCR) was performed on whole embryo samples. To examine Gcnf expression in Gcnffl/fl; Ubc-Cre-ERT2TAM and wild-type ovaries, qRT-PCR was performed on ovaries dissected away from the mesonephroi. In both cases, samples were submerged in TRIzol reagent (Invitrogen) and stored at 80°C. After genotyping of the yolk sac or tail, total RNA was prepared according to the manufacturer's instructions. cDNA was transcribed from 100–200 ng of RNA using SuperScript III (Invitrogen) according to the manufacturer's instructions. Quantitative PCR was performed on cDNA using SYBR Green Core PCR reagents (Applied Biosystems) on an ABI9700 fast real-time PCR machine (Applied Biosystems). Results were analyzed using the delta-delta Ct method with Hprt (hypoxanthine-guanine phosphoribosyltransferase) or Actb (actin, beta) as a normalization control. RT-PCR primers were as follows:
We thank Mary Goodheart for performing blastocyst injections; Jacob Mueller and Mark Gill for unpublished RNA sequencing data; George Enders and Richard Schultz for antibodies; J Philipp Junker, Alexander van Oudenaarden, and Shirleen Soh for assistance with microscopy; and Tsutomu Endo, Gregoriy Dokshin, Jennifer Hughes, Jacob Mueller, Katherine Romer, and Shirleen Soh for helpful discussions and comments on the manuscript.
Funding Statement
This work was supported by Howard Hughes Medical Institute and by National Institutes of Health/National Human Genome Research Institute (2R01HG00257-20). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
1. Nichols J, Zevnik B, Anastassiadis K, Niwa H, Klewe-Nebenius D, et al. (1998) Formation of pluripotent stem cells in the mammalian embryo depends on the POU transcription factor Oct4. Cell 95: 379–391. [PubMed]
2. Mitsui K, Tokuzawa Y, Itoh H, Segawa K, Murakami M, et al. (2003) The homeoprotein Nanog is required for maintenance of pluripotency in mouse epiblast and ES cells. Cell 113: 631–642. [PubMed]
3. Sato M, Kimura T, Kurokawa K, Fujita Y, Abe K, et al. (2002) Identification of PGC7, a new gene expressed specifically in preimplantation embryos and germ cells. Mechanisms of development 113: 91–94. [PubMed]
4. Yabuta Y, Kurimoto K, Ohinata Y, Seki Y, Saitou M (2006) Gene expression dynamics during germline specification in mice identified by quantitative single-cell gene expression profiling. Biology of reproduction 75: 705–716. [PubMed]
5. Pesce M, Wang X, Wolgemuth DJ, Scholer H (1998) Differential expression of the Oct-4 transcription factor during mouse germ cell differentiation. Mechanisms of development 71: 89–98. [PubMed]
6. Rosner MH, Vigano MA, Ozato K, Timmons PM, Poirier F, et al. (1990) A POU-domain transcription factor in early stem cells and germ cells of the mammalian embryo. Nature 345: 686–692. [PubMed]
7. Kehler J, Tolkunova E, Koschorz B, Pesce M, Gentile L, et al. (2004) Oct4 is required for primordial germ cell survival. EMBO reports 5: 1078–1083. [PubMed]
8. Yamaguchi S, Kurimoto K, Yabuta Y, Sasaki H, Nakatsuji N, et al. (2009) Conditional knockdown of Nanog induces apoptotic cell death in mouse migrating primordial germ cells. Development 136: 4011–4020. [PubMed]
9. Gill ME, Hu YC, Lin Y, Page DC (2011) Licensing of gametogenesis, dependent on RNA binding protein DAZL, as a gateway to sexual differentiation of fetal germ cells. Proceedings of the National Academy of Sciences of the United States of America 108: 7443–7448. [PubMed]
10. Gu P, LeMenuet D, Chung AC, Mancini M, Wheeler DA, et al. (2005) Orphan nuclear receptor GCNF is required for the repression of pluripotency genes during retinoic acid-induced embryonic stem cell differentiation. Molecular and cellular biology 25: 8507–8519. [PMC free article] [PubMed]
11. Fuhrmann G, Chung AC, Jackson KJ, Hummelke G, Baniahmad A, et al. (2001) Mouse germline restriction of Oct4 expression by germ cell nuclear factor. Developmental cell 1: 377–387. [PubMed]
12. Yan Z, Jetten AM (2000) Characterization of the repressor function of the nuclear orphan receptor retinoid receptor-related testis-associated receptor/germ cell nuclear factor. The Journal of biological chemistry 275: 35077–35085. [PubMed]
13. Yamaguchi S, Kimura H, Tada M, Nakatsuji N, Tada T (2005) Nanog expression in mouse germ cell development. Gene expression patterns : GEP 5: 639–646. [PubMed]
14. Susens U, Aguiluz JB, Evans RM, Borgmeyer U (1997) The germ cell nuclear factor mGCNF is expressed in the developing nervous system. Developmental neuroscience 19: 410–420. [PubMed]
15. Chung AC, Katz D, Pereira FA, Jackson KJ, DeMayo FJ, et al. (2001) Loss of orphan receptor germ cell nuclear factor function results in ectopic development of the tail bud and a novel posterior truncation. Molecular and cellular biology 21: 663–677. [PMC free article] [PubMed]
16. Lan ZJ, Chung AC, Xu X, DeMayo FJ, Cooney AJ (2002) The embryonic function of germ cell nuclear factor is dependent on the DNA binding domain. The Journal of biological chemistry 277: 50660–50667. [PubMed]
17. Rodriguez CI, Buchholz F, Galloway J, Sequerra R, Kasper J, et al. (2000) High-efficiency deleter mice show that FLPe is an alternative to Cre-loxP. Nature genetics 25: 139–140. [PubMed]
18. Ruzankina Y, Pinzon-Guzman C, Asare A, Ong T, Pontano L, et al. (2007) Deletion of the developmentally essential gene ATR in adult mice leads to age-related phenotypes and stem cell loss. Cell stem cell 1: 113–126. [PMC free article] [PubMed]
19. Toyooka Y, Tsunekawa N, Takahashi Y, Matsui Y, Satoh M, et al. (2000) Expression and intracellular localization of mouse Vasa-homologue protein during germ cell development. Mechanisms of development 93: 139–149. [PubMed]
20. Prieto I, Tease C, Pezzi N, Buesa JM, Ortega S, et al. (2004) Cohesin component dynamics during meiotic prophase I in mammalian oocytes. Chromosome research : an international journal on the molecular, supramolecular and evolutionary aspects of chromosome biology 12: 197–213. [PubMed]
21. Yang J, Medvedev S, Yu J, Tang LC, Agno JE, et al. (2005) Absence of the DNA-/RNA-binding protein MSY2 results in male and female infertility. Proceedings of the National Academy of Sciences of the United States of America 102: 5755–5760. [PubMed]
22. Yu J, Hecht NB, Schultz RM (2001) Expression of MSY2 in mouse oocytes and preimplantation embryos. Biology of reproduction 65: 1260–1270. [PubMed]
23. Tam PP, Snow MH (1981) Proliferation and migration of primordial germ cells during compensatory growth in mouse embryos. Journal of embryology and experimental morphology 64: 133–147. [PubMed]
24. Yeom YI, Fuhrmann G, Ovitt CE, Brehm A, Ohbo K, et al. (1996) Germline regulatory element of Oct-4 specific for the totipotent cycle of embryonal cells. Development 122: 881–894. [PubMed]
25. Gu P, Goodwin B, Chung AC, Xu X, Wheeler DA, et al. (2005) Orphan nuclear receptor LRH-1 is required to maintain Oct4 expression at the epiblast stage of embryonic development. Molecular and cellular biology 25: 3492–3505. [PMC free article] [PubMed]
26. Trapnell C, Pachter L, Salzberg SL (2009) TopHat: discovering splice junctions with RNA-Seq. Bioinformatics 25: 1105–1111. [PMC free article] [PubMed]
27. Raj A, van den Bogaard P, Rifkin SA, van Oudenaarden A, Tyagi S (2008) Imaging individual mRNA molecules using multiple singly labeled probes. Nature methods 5: 877–879. [PMC free article] [PubMed]
Articles from PLoS ONE are provided here courtesy of
Public Library of Science