April 1, 2012;
Xenopus Nanos1 is required to prevent endoderm gene expression and apoptosis in primordial germ cells.
Nanos is expressed in multipotent cells, stem cells and primordial germ cells
(PGCs) of organisms as diverse as jellyfish and humans. It functions together with Pumilio
to translationally repress targeted mRNAs. Here we show by loss-of-function experiments that Xenopus Nanos1
is required to preserve PGC
fate. Morpholino knockdown of maternal Nanos1
resulted in a striking decrease in PGCs and a loss of germ cells
from the gonads. Lineage tracing and TUNEL staining reveal that Nanos1
-deficient PGCs fail to migrate out of the endoderm
. They appear to undergo apoptosis rather than convert to normal endoderm
. Whereas normal PGCs do not become transcriptionally active until neurula
-depleted PGCs prematurely exhibit a hyperphosphorylated RNA polymerase II C-terminal domain at the midblastula transition. Furthermore, they inappropriately express somatic genes characteristic of endoderm
regulated by maternal VegT
, including Xsox17α, Bix4
. We further demonstrate that Pumilio
specifically binds VegT
RNA in vitro and represses, along with Nanos1
translation within PGCs. Repressed VegT
RNA in wild-type PGCs is significantly less stable than VegT
-depleted PGCs. Our data indicate that maternal VegT
RNA is an authentic target of Nanos1
translational repression. We propose that Nanos1
functions to translationally repress RNAs that normally specify endoderm
and promote apoptosis, thus preserving the germline.
[+] show captions
Fig. 1. Nanos1-depleted Xenopus embryos are deficient in PGCs. (A) Nanos1-MO blocks the translation of endogenous nanos1. Confocal analysis showing the expression of Nanos1 (red) and Xiwi (germ plasm marker, green) at the eight-cell stage of embryos previously injected with Nanos1-Ctrl-MO or Nanos1-MO at the one-cell stage. Uninjected embryos served as controls. Arrows indicate Nanos1 signal. (B) Knockdown of Nanos1 results in loss of PGCs. WISH Xpat analysis of stage 37/38 wild-type (WT), Nanos1-Ctrl-MO- and Nanos1-MO-injected embryos. For the rescue, Nanos1-MO embryos were injected with nanos1-mut RNA. (C) Summary of PGC loss at different stages after Nanos1 knockdown. P-values by one-way ANOVA. Error bars indicate s.e. (D) Hematoxylin and Eosin-stained sections of gonads from WT (stage 54), Nanos1-Ctrl-MO (stage 55), Nanos1-MO (stage 54) and rescued (stage 54) embryos, Arrows indicate germ cells. Note that nanos1 morphants were completely rescued by nanos1-mut RNA injection. Scale bars: 130 μm in A; 100 μm in B; 20 μm in D.
Fig. 2. Nanos1-depleted PGCs enter apoptosis. (A) PGCs deficient in Nanos1 are not apoptotic throughout early development (stage 18). WISH Xpat (germ plasm marker, green) and TUNEL staining (red) of WT, Nanos1-Ctrl-MO- and Nanos1-MO-injected embryos. Arrows indicate TUNEL-positive cells in neural ectoderm. (B) TUNEL-positive PGCs are detected during tailbud stage (stage 28) in Nanos1-depleted embryos, but not in control embryos. Arrows indicate TUNEL-positive PGCs. Confocal images were taken from the region indicated in the diagram (red box). Scale bars: 100 μm in A; 150 μm in B.
Fig. 3. Nanos1-depleted PGCs fail to migrate from the endoderm. (A) Experimental design for lineage tracing of PGCs. MOs were injected at stage 1. Germline lineage tracer NLD mRNA was injected into germ plasm-containing blastomeres at the 32-cell stage. Embryos were allowed to develop until stage 46/47 and stained for lacZ expression. (B) Whole-mount lineage analysis at stage 46/47 showing lacZ-positive PGCs (blue) in WT, Nanos1-Ctrl-MO- and Nanos1-MO-injected embryos. PGCs migrating in the dorsal mesenchyme are circled (red). PGCs were not detected in Nanos1-depleted embryos in this region. Black arrows indicate wild-type ectopic PGCs in the gut. In the sagittal paraffin section of a WT embryo stained with Eosin, PGCs are indicated by blue arrows. Scale bars: 100 μm. (C) Summary of lineage traced PGCs in stage 46/47 tadpoles. Ectopic PGCs within the gut occurred in both control and experimental embryos. Statistical analysis was performed only for PGCs in the genital ridges (Student’s t-test for two-group analysis, one-way ANOVA for three-group analysis). Error bars indicate s.e. for the green bars.
Fig. 4. Nanos1-depleted PGCs exhibit CTD-PSer2 prematurely. (A) Confocal analysis of WT, Nanos1-Ctrl-MO- and Nanos1-MO-injected embryos at stage 11, showing double immunostaining with H5 monoclonal antibody (CTD-PSer2, green) and rabbit anti-Xiwi antibody (germ plasm, red). Merged images are shown at the top, with separate channels beneath. Images were taken from the endoderm core (see the diagram in Fig. 5). Scale bars: 50 μm. (B) PGCs immunopositive for H5 (which is specific for the CTD-PSer2 epitope) in Nanos1-MO (74%) as compared with WT (21%) and Nanos1-Ctrl-MO (16%) embryos. Differences between each group were highly significant by one-way ANOVA. The experiment was repeated three times.
Fig. 5. PGCs deficient in Nanos1 activity express endoderm-specific genes. (A,C) RT-PCR analysis of PGCs or endoderm cells isolated from the endoderm core at stage 11 (circled in red, surrounded by endomesoderm in blue) (A) and stage 15 (C). P, primordial germ cells; E, endoderm; WE, whole embryo. Xpat is a germ plasm marker, Xsox17α an endoderm marker, Bix4 an endomesoderm marker and Xbra a mesoderm marker. ODC provides an internal control. The asterisk marks endoderm cell contamination. (B) Confocal analysis of WT, Nanos1-Ctrl-MO- and Nanos1-MO-injected embryos at stage 11. Xsox17α RNA (red) was detected by WISH and PGCs (outlined by dashed line) were identified by Xiwi immunostaining (green). Merged images are shown at the top, with separate channels beneath. Images were taken from the endoderm core. Scale bars: 50 μm.
Fig. 6. Nanos1 represses VegT translation in PGCs. (A) RNA EMSA analysis shows that the Xenopus Pumilio1 RNA-binding domain (Xpum1 RBD) can bind biotin-labeled VegT PBE (1 ng) but not the mutant VegT PBE. This binding reaction was competed by unlabeled VegT PBE. (B) Experimental design for results shown in C. Venus-DEADSouth 3′UTR and DsRED-DEADSouth 3′UTR are germline lineage tracers. VegT PBE or its mutant was subcloned downstream of the DsRED open reading frame. Nanos1-MO was injected at stage 1. Germline lineage reporters were injected into four germ plasm-containing blastomeres at the 32-cell stage. Embryos were allowed to develop until stage 35 and the fluorescent images were taken from live embryos. (C) In vivo fluorescent reporter assay indicates that VegT PBE mediates translational repression of DsRED (red) in PGCs (green). Repression was lost in embryos bearing a PBE mutation or that were depleted of Nanos1. Reporters injected are noted across the top. This experiment was repeated three times. Arrows indicate DsRED signal detected in PGCs. Data from one experiment are presented beneath the images for each group. Scale bars: 50 μm.
Fig. 7. VegT RNA is stabilized in PGCs after Nanos1 knockdown. (A) VegT PBE mediates the degradation of the reporter message DsRED. PGCs were detected by either WISH for Xpat (top panels) or for DsRED antisense probe (bottom panels) in tailbud stage 31 embryos. Note that the DsRED reporter with a PBE was specifically degraded. Arrows indicate PGCs. (B) Real-time PCR analysis of VegT expression at stages 8 and 10. Expression of VegT in Nanos1 knockdown samples was normalized to that of wild-type samples. Real-time PCR was performed in duplicate, and experiments were repeated three times and showed a similar pattern. Error bars indicate s.e. P-value by unpaired Student’s t-test.
Fig. 8. Model for Nanos1 function in Xenopus PGCs. Xenopus Nanos1 preserves the germline fate through multiple functions: preventing endoderm gene expression by repressing VegT translation, blocking RNA Pol II transcription at MBT, and inhibiting apoptosis. PUM, Pumilio.