The Nanos3-3′UTR Is Required for Germ Cell Specific NANOS3 Expression in Mouse Embryos

Background The regulation of gene expression via a 3′ untranslated region (UTR) plays essential roles in the discrimination of the germ cell lineage from somatic cells during embryogenesis. This is fundamental to the continuation of a species. Mouse NANOS3 is an essential protein required for the germ cell maintenance and is specifically expressed in these cells. However, the regulatory mechanisms that restrict the expression of this gene in the germ cells is largely unknown at present. Methodology/Principal Findings In our current study, we show that differences in the stability of Nanos3 mRNA between germ cells and somatic cells is brought about in a 3′UTR-dependent manner in mouse embryos. Although Nanos3 is transcribed in both cell lineages, it is efficiently translated only in the germ lineage. We also find that the translational suppression of NANOS3 in somatic cells is caused by a 3′UTR-mediated mRNA destabilizing mechanism. Surprisingly, even when under the control of the CAG promoter which induces strong ubiquitous transcription in both germ cells and somatic cells, the addition of the Nanos3-3′UTR sequence to the coding region of exogenous gene was effective in restricting protein expression in germ cells. Conclusions/Significance Our current study thus suggests that Nanos3-3′UTR has an essential role in translational control in the mouse embryo.


Introduction
The manner in which genes are regulated to produce the correct combination of proteins for every cell type remains a fundamental question in biology. In many cases, gene expression is primarily regulated via transcription under the control of enhancer and promoter sequences. However, it is now becoming clear that post-transcriptional regulation mediated via a 39 untranslated region (UTR) plays key roles in the control of mRNA stability and/or translation. A critical step in the establishment of elaborate germ cell lineages during early embryogenesis in nematodes, fly, fish and frog is the temporal and spatial regulation of several proteins via mechanisms that are dependent on the 39UTR of maternal mRNAs including nanos [1,2,3,4].
The nanos genes are evolutionarily conserved among many organisms and play important roles during germ cell development [5,6,7,8,9,10,11]. During germ cell specification in Drosophila, maternal nanos mRNA becomes localized in the germ plasm in the posterior part of the egg [12,13]. This localization is inefficient, however, and translational repression is therefore essential for the restricted production of Nanos protein in the posterior region. This repression is mediated by a 90-nucleotide translational control element (TCE) in the 39UTR of nanos mRNA [14,15,16,17,18] to which Smaug (Smg) or Glorund (Glo) bind [19,20,21]. On the other hand, the localization in the germ plasm and subsequent translational activation of nanos mRNA is regulated by the Oskar (Osk) protein via 39UTR-dependent mechanisms [19,22,23]. In Danio rerio, maternal nanos1 mRNA is also present in a whole oocyte, but only a portion is localized to the germ plasm and translated specifically in the PGC. The translation of the bulk of nanos1 mRNA in somatic cells is then rapidly degraded during embryogenesis. The regulation of nanos1 both in the PGC and somatic cells depends on three elements within the nanos1-39UTR: (1) a site required for its localization to the germ plasm [24]; (2) two miR430 sites responsible for mRNA degradation in somatic cells; and (3) the binding site for the Dead end 1(Dnd1) protein that is expressed only in the PGC and protects mRNA from miR430-dependent degradation [2,25].
In Mus musculus, primordial germ cells (PGCs) are induced from a population of pluripotent epiblast cells [26,27]. Following their induction, these PGC precursors translocate to the base of the allantois by E7.25 and once formed, migrate to the endoderm (E7.5), travel through the hindgut (from E8.0), dorsal mesentery and dorsal body wall, and reach the genital ridge at around E10.5 to E11.5. Following the sex differentiation of the somatic gonads, PGCs themselves differentiate into male or female germ cells at around E12.0 [28]. Three Nanos homologs (Nanos1-3) have been identified in mice, and Nanos2 and Nanos3 have been implicated in germ cell development [10]. NANOS2 is specifically expressed in the mouse male germ cells after their colonization of the gonads and is essential for their development. In our previous study, we reported that the Nanos2-39UTR promotes the efficient translation of this protein in the male germ cell after E13.5 via an unknown mechanism [29]. Nanos3 is expressed in the PGCs after their formation until shortly after their settlement in the gonads (E14.5 in male, E13.5 in female), and is re-expressed after birth in the testes [10]. Nanos3 knockout mice are thus sterile because of the loss of migrating PGCs during embryogenesis. These data suggest that NANOS3 plays an important role in the maintenance and survival of PGCs [10,30]. However, the regulatory mechanism of NANOS3 expression and the function of the Nanos3-39UTR had not been fully investigated as yet.
In our present report, we show that Nanos3 mRNA is transcribed in both germ cells and somatic cells, although NANOS3 protein is expressed specifically in germ cells. By applying a transgenic mouse strategy, we show that the translation of NANOS3 in somatic cells is suppressed via an mRNA destabilizing mechanism mediated by the Nanos3-39UTR.

Results
The Nanos3-39UTR Is Required for Suppression of Nanos3 Expression in Somatic Cells To elucidate whether the Nanos3-39UTR is involved in NANOS3 expression in mouse, we generated two BAC transgenic mouse lines; BAC-Nanos3-mRFP(Nos3-39UTR) containing the endogenous Na-nos3-39UTR ( Fig. 1A and E), and BAC-Nanos3-mRFP(BghpA) harboring an exogenous 39UTR, Bovine growth hormone poly(A) signal (BghpA; Fig. 1B). We first confirmed that either BAC-Nanos3-mRFP(Nos3-39UTR) or BAC-Nanos3-mRFP(BghpA) could rescue the Nanos3 2/2 embryonic phenotype (Fig. S1), indicating that this BAC construct contains regulatory elements that are sufficient to maintain endogenous NANOS3 expression and the NANOS3-mRFP protein was functional. The NANOS3-mRFP expression in germ cells in both transgenic embryos showed a similar pattern to the endogenous protein ( Fig. 2 and S2) exhibiting cytoplasmic localization as seen for NANOS2 [31]. The small difference between Nanos3 and Nanos3-mRFP was that the former was  clearly localized to the cytoplasm whereas the latter was less clear and slightly localized in nuclei also. It was consistent with a previous report [32].
Interestingly, however, in the embryo harboring BAC-Nanos3-mRFP(BghpA), the intensity of NANOS3-mRFP was gradually increased in the somatic tissues at later embryonic stage ( Fig. 2K-L). In the E14.5 male, a striped pattern was observed for BAC-Nanos3-mRFP(Nos3-39UTR) reflecting germ cell localization in the testis cords in gonads, whereas the pattern was unclear in BAC-Nanos3-mRFP(BghpA), indicating strong expression in the surrounding somatic tissues (compare Figs. 2H and 2L, 3A-D). In addition, the whole body of the BAC-Nanos3-mRFP(BghpA) embryos expressed NANOS3-mRFP (Fig. 3D). These results suggest that Nanos3 is transcribed in many embryonic tissues and that the Nanos3-39UTR is required to suppress translation in somatic tissues.
To evaluate the suppressive effects of 39UTR in somatic cells quantitatively, we compared the abundance of Nanos3-mRFP protein based on the intensities of mRFP signals in both the germ cells and surrounding somatic cells in each transgenic embryo from E7.5 to E13.5 using imageJ software ( Fig. 3E-F). The changes in the relative mRFP intensities in the germ cells were similar between the two transgenic lines. These were gradually decreased after E7.5, reached their lowest level at E11.5 and then rapidly increased from E12.5. Contrary to the data found in the germ cells, the mRFP intensities in the surrounding somatic cells showed a clear difference between the two lines. In the embryo harboring Nanos3-39UTR, the mRFP intensity was maintained at very low levels throughout embryogenesis. However, in the embryo containing the BghpA elements, this expression gradually increased from E10.5 and at E13.5 reached 60% of the intensity seen in the germ cells, although it was maintained at low levels at E7.5 and E9.5. These data suggest that the translation of NANOS3 is upregulated after E9.5 in somatic tissues and that the Nanos3-39UTR is required to suppress this activity.
The Accumulation of Somatic Nanos3 mRNA Is Suppressed by the Nanos3-39UTR We next examined the endogenous Nanos3 mRNA levels in somatic tissues between E9.5 and E13.5 by quantitative RT-PCR (qRT-PCR) and found transcripts even in somatic tissues ( Fig. 4A-C), consistent with the above data. The anterior half of the embryo at E9.5 (9.5A) and the kidney at E13.5 (13.5K) do not contain any germ cells, but Nanos3 mRNA was detected ( Fig. 4B-C). The level of Nanos3 mRNA in somatic tissue from E9.5 to E13.5 was maintained at very low levels compared with the gonads which containing many germ cells (Fig. 4C). Interestingly, Nanos3 expression was not detected in the anterior half of the embryo at E7.5 (E7.5A), suggesting the transcription of this gene is restricted to the PGCs when they are formed and may be slightly increased in the somatic cells at the later stage. It is consistent with previous reports, which include single-cell PCR analyses demonstrating Nanos3 expression exclusively in the PGCs [33] and our lineage study using Nanos3-cre [30].
In embryos harboring BAC-Nanos3-mRFP(Nos3-39UTR), the levels of mRFP at both stages was low, similar to Nanos3 mRNA in the wild-type embryo ( Fig. 4B and D). In contrast, in embryos harboring BAC-Nanos3-mRFP(BghpA), the relative mRFP levels became two-fold higher at E13.5 than those of BAC-Nanos3-mRFP(Nos3-39UTR) (Fig. 4D). Taken together, these data suggest that Nanos3 is transcribed in both germ cells and somatic tissues by at least E9.5 and that Nanos3-39UTR is required to suppress Nanos3 accumulation in somatic tissues. The Nanos3-39UTR Is Sufficient for the Establishment of

Germ-Cell Specific Expression Pattern in the Mouse Embryo
To further investigate whether the Nanos3-39UTR affects the transcription or stability of mRNA, we generated two additional transgenic mice. We utilized the CAG promoter, a known strong promoter-enhancer that drives the ubiquitous transcription of mRFP with either Nanos3-39UTR or BghpA (Fig. 1C-D). Surprisingly, Nanos3-39UTR proved to be effective in restricting the mRFP expression in the germ cells at E14.5 (Fig. 5A-B), whilst mRFP was always expressed ubiquitously with no pattern observed in the CAG-mRFP(BghpA) embryo (Fig. 5C). The same expression pattern was observed in CAG-lyn-mRFP(BghpA) transgenic embryo that had been previously established in our laboratory [34]. The lyn-mRFP is an mRFP that contains the lyn kinase at its N-terminus, which serves as a membrane localization signal but does not affect neither transcription and translation [34]. Therefore, we considered CAG-lyn-mRFP(BghpA) is compatible with CAG-mRFP(BghpA). By qRT-PCR analyses, we further revealed that the relative amounts of mRFP mRNA in somatic tissues (a kidney and a hind limb) were significantly lower than those in the gonads in CAG-mRFP(Nos3-39UTR) embryos whereas not significantly altered in CAG-lyn-mRFP(BghpA) (Fig. 5D and E). The results suggest that Nanos3-39UTR is sufficient to suppress protein expression in somatic cells by destabilizing mRNA and establishing a germ-cell specific expression pattern.
To determine the point at which Nanos3-39UTR begins to function and contribute to the establishment of the germ cell-specific pattern for NANOS3, we compared the mRFP expression patterns between CAG-mRFP(Nos3-39UTR) and CAG-lyn-mRFP(BghpA) transgenic embryos. In CAG-lyn-mRFP(BghpA), mRFP was expressed in all embryonic tissues at all stages (Fig. 6A-D). In CAG-mRFP(Nos3-39UTR) however, mRFP was expressed in all embryonic cells prior to germ cell formation at E7.0, similar to the profile found in the CAG-mRFP(BghpA) embryo (data not shown). After PGC formation, mRFP was still found to be expressed in most embryonic cells at E7.5, but in somatic cells this expression is gradually reduced (E9.5) and the germ-cell specific pattern is almost established by E11.5 ( Fig. 6E-H). At E12.5, mRFP expression in the germ cells became notably stronger than in the somatic cells of both male and female embryos ( Fig. 6I-J). This germ cell specific mRFP pattern was maintained until at least E16.5 at which stage endogenous Nanos3 expression is almost lost. These observations suggest that the Nanos3-39UTR might function in all embryonic cells from E7.5 to E16.5. It is possible also that the Nanos3-39UTR is involved in translational activation in germ cells. These different functions of this regulatory element in germ cells and in somatic cells might therefore contribute to the establishment of germ cell-specific NANOS3 protein expression.
It is noteworthy that the addition of the 39UTRs of other germ cell specific genes such as Nanos2 and Stella/PGC7 was not sufficient to establish germ cell-specific expression patterns i.e. the transgenic embryos CAG-mRFP(Nanos2-39UTR) and CAG-mRFP(Stella-39UTR) showed strong and ubiquitous mRFP expression (Fig. S3). This suggests that the function of Nanos3-39UTR is relatively unique in the mouse germ cell, unlike fly and nematodes in which 39UTRs of many genes each have significant responsibility for the temporal and spatial control of a specific protein in their germlines [4,35].

Discussion
The regulatory mechanisms underlying gene expression remains one of the most fundamental and significant themes in biology. In our current study, we analyzed the mechanisms underlying NANOS3 expression in vivo using BAC modification and transgenic technologies. Although Nanos3 is transcribed both in germ cells and in many somatic tissues, efficient translation of NANOS3 protein occurs only in germ cells.
It has been shown previously that the expression of maternal mRNAs depends on the corresponding 39UTR in many animal species. The 39UTRs of nanos homologs play essential roles in the respective mRNA localization, translation and degradation in C. elegans, Drosophila, and Zebrafish [1,9,36]. We have found in our present experiments that the Nanos3-39UTR of Mus musculus also has a regulatory function during embryogenesis, even though NANOS3 is transcribed zygotically and germ cell formation in mice is quite different from other animals. There have been several reported examples of 39UTR regulation of zygotic mRNA [37,38,39]. However, in all these cases, the transcripts were driven by tissue-specific promoters. Hence, Nanos3-39UTR is the first example of a regulator of zygotic mRNA that can establish a tissue-specific gene expression pattern even if the mRNA is transcribed by a ubiquitous promoter.
The mechanisms of 39UTR-dependent nanos mRNA regulation have been addressed previously in fishes and flies, in which miR430 and the Dnd1 protein, or the Smg, Glo and Osk proteins are involved in mRNA regulation via the nanos-39UTR. In mice, one ortholog of Dnd1 and two orthologs of Smg have now been identified [40,41]. We examined the possible effects of these proteins on the translation of an mRNA harboring the Nanos3-39UTR by a luciferase assay in the NIH3T3 cell line. The stability of luciferase mRNA is also affected by Nanos3-39UTR. However, the addition of both proteins did not result in any effects on reporter activities (data not shown). It is possible that they need cofactors which are not expressed in this cell line. It is also possible that the abundant expression of endogenous Smg in NIH3T3 cells caused no effect. In addition, the sequence of Nanos3-39UTR has almost no similarity to the 39UTRs of nanos orthologs and has no significant match with any miRNA target sites. Although several stem-loop structures have been predicted using the 'mfold' program (Zuker, 2003), these are not similar to the Drosophila TCE (data not shown). Hence, the mechanism of Nanos3-39UTR dependent regulation is still unclear and is an essential project for a future study.
The somatic expression of BAC-Nanos3-mRFP(BghpA) did not affect mouse development unlike in the case of fly and the biological significance of Nanos3-39UTR-dependent regulation also remains unclear. Since the regulation of gene transcription appears not always to be strict, Nanos3-39UTR may prevent the accumulation of waste materials in the cell by promoting mRNA degradation.

Mice
The methods used to generate Nanos3-L-39UTR (Nanos3 +/2 ) mice and their subsequent characterization has been previously described [10]. All mice were an MCH background (closed colony derived from an ICR strain, CREA, Japan).

Cloning of 39UTR Sequences
The Nanos3-39UTR was cloned by PCR using a DNA template prepared from the tail of a C57BL6/J mouse. The primers used were N3-stop-SalI-F1 and N3-39U-HindIII-R1.

Generation of Transgenic Mice
All BAC constructs were digested with Csp45I and PmacI (Takara) and then gel purified. Transgenic mice were then generated by the microinjection of DNA into fertilized eggs. The injected eggs were then transferred into the oviducts of pseudopregnant foster females. The genotypes of the mice or embryos were identified by PCR using isolated genomic DNA from the tail or yolk-sac. The primers used were as follows: RFP-F2 and N3-3U-R1 for BAC-Nanos3-mRFP(Nos3-39UTR) and BAC-DNos3-mRFP(Nos3-39 UTR); mRFP-F2 and bghpA-R2 for BAC-Nanos3-mRFP(BghpA) and BAC-DNos3-mRFP(BghpA). The primer sequences were described in Method S1.

Immunofluorescence
The mouse embryos and gonads were fixed in 4% PFA for 2 hours at 4uC and washed three times for 5 min each with PBS. After blocking with PBS containing 3% skim milk or 10% FBS for 1 hour at RT, samples were rinsed and incubated overnight with primary antibodies in PBS containing 0.1% TritonX-100 (PBS-Tr) at 4uC or RT. The following day, samples were washed 6 times for 15 min each in PBS-Tr and were incubated for 2 hours at RT with secondary antibodies in PBS-Tr. After the samples had been washed 6 times for 15 min each with PBS-Tr, they were mounted on MAS-coated slide glasses or a glass-bottom dish (Matsunami) and enclosed with PBS by manicure. The samples were then analyzed by confocal laser microscopy (Zeiss).

Evaluation of mRFP Intensity
Samples were fixed in 4%PFA and immunostained with anti-Oct4 and Alexa-488 conjugated donkey anti-mouse IgG to determine germ cells. Then the images were taken using confocal laser microscopy (Zeiss). mRFP intensity in the immunostained embryo did not show significant difference from the unfixed embryo (data not shown). The mRFP intensity of each cell was measured using imageJ software (NIH, Bethesda, MD) and the mean gray values were regarded as the intensity of cell. Three embryos for each stage were examined. All data were normalized using the intrinsic background intensity of wild type embryos at each embryonic stage. The intensity of the E7.5 PGCs was assigned a value of 10 and the data were plotted accordingly.