DEADSouth protein localizes to germ plasm and is required for the development of primordial germ cells in Xenopus laevis

Abstract

DEADSouth mRNA is a component of germ plasm in Xenopus laevis and encodes a DDX25 DEAD-box RNA helicase. To determine the intracellular localization of DEADSouth protein, we injected mRNA encoding DEADSouth tagged with mCherry fluorescent protein into fertilized eggs from transgenic Xenopus expressing EGFP fused with a mitochondrial targeting signal. The DEADSouth-mCherry fusion protein was localized to the germ plasm, a mitochondria-rich region in primordial germ cells (PGCs). DEADSouth overexpression resulted in a reduction of PGC numbers after stage 20. Conversely, DEADSouth knockdown using an antisense locked nucleic acid gapmer inhibited movement of the germ plasm from the cortex to the perinuclear region, resulting in inhibition of PGC division at stage 12 and a decrease in PGC numbers at later stages. The knockdown phenotype was rescued by intact DEADSouth mRNA, but not mutant mRNA encoding inactive DEADSouth helicase. Surprisingly, it was also rescued by mouse vasa homolog and Xenopus vasa-like gene 1 mRNAs that encode DDX4 RNA helicases. The rescue was dependent on the 3' untranslated region (3'UTR) of DEADSouth mRNA, which was used for PGC-specific expression. The 3'UTR contributed to localization of the injected mRNA to the germ plasm, resulting in effective localization of DEADSouth protein. These results demonstrate that localization of DEADSouth helicase to the germ plasm is required for proper PGC development in Xenopus laevis.

Keywords: Germline; LNA gapmer; RNA helicase.

Fig. 1.. DEADSouth-mCherry fusion protein is localized to the germ plasm in PGCs.

mRNA of DEADSouth-mCherry-DS 3′UTR (A–E) or mCherry-DS 3′UTR (F–J) was injected at the vegetal pole of fertilized eggs by crossing wild-type male and mito-EGFP transgenic female Xenopus. PGCs were isolated from embryos at stage 12, fixed and stained with Hoechst 33342. Signals for mCherry, germ plasm (mitochondria) and nuclei are red, green and blue, respectively. Scale bars: 10 µm.

Fig. 2.. DEADSouth gene overexpression decreases the number of PGCs.

(A–C) Representative examples of stage 32 embryos injected with mRNA for v-DS only, v-DS plus DEADSouth-DS 3′UTR (DS-full) and v-DS plus nonsense DEADSouth-DS 3′UTR (DS-stop), respectively. Anterior is toward the left. Scale bar: 1 mm. (A′–C′) High magnification of the areas indicated in (A–C), respectively. (D) The number of PGCs per embryo at stage 32 injected with the indicated mRNAs. PGC numbers were determined by externally counting from both sides. (E) The number of PGCs per embryo at the indicated stage without/with DS-full mRNA. Embryos were dissociated, and EGFP-positive and large cells were counted as PGCs. ‘N’ and ‘n’ indicate the number of experiments and total embryos examined, respectively. The P-value was calculated by the Student's t-test. Error bars indicate s.d. (F) Apoptotic PGCs in DEADSouth-overexpressing embryos at stage 28. PGCs were observed after dissociating the embryonic region (red box) immunostained for Xdazl and caspase-3. Arrowheads and arrows indicate Xdazl-positive/caspase-3-negative cells (non-apoptotic PGCs) and Xdazl positive/caspase-3-positive cells (apoptotic PGCs), respectively. Scale bar: 100 µm.

Fig. 3.. An LNA gapmer causes degradation of targeted DEADSouth mRNA.

(A) Sequence alignment of DEADSouth mRNA (nucleotide position 29–61, accession no. AF190623), antisense and mismatch LNA gapmers and synthesized DEADSouth mRNA (DS-full). Upper letters indicate the amino acid sequence. LNA modifications and mismatch bases are underlined and in red, respectively. The sequence of DS-full mRNA was predicted from that of the template DNA. (B) Quantification of DEADSouth and Xpat mRNAs in stage 8 and 32 embryos injected with LNA gapmers. The amounts of these mRNAs were determined by real-time RT-PCR, normalized to EF1α mRNA levels as an internal control, and shown as relative amounts to the mRNA level in embryos injected with mismatch LNA gapmer. P-values and s.d. are indicated.

Fig. 4.. The number of PGCs is decreased by knockdown of the DEADSouth gene.

(A–D) Representative examples of stage 32 embryos injected with the mismatch LNA gapmer, antisense LNA gapmer, antisense LNA gapmer plus DS-full and antisense LNA gapmer plus mutant DEADSouth-DS 3′UTR (DS-AAA), in addition to v-DS as a PGC-tracer, respectively. Scale bar: 1 mm. (A′–D′) High magnification of the areas indicated in (A–D), respectively. (E) The number of PGCs per embryo at stage 32 injected with the indicated LNA gapmer(s) and mRNAs. PGC numbers were determined by externally counting from both sides. (F) The number of PGCs per embryo at the indicated stages with mismatch or antisense LNA gapmers. The injected mito-EGFP embryos were dissociated, and EGFP-positive and large cells were counted as PGCs. N and n indicate the number of experiments and total embryos examined, respectively. P-values were calculated by the one-tailed t test. Error bars indicate s.d.

Fig. 5.. Knockdown of the DEADSouth gene affects PGC division and translocation to germ plasm.

(A) Size distribution of PGCs at stages 12, 20 and 32 from embryos injected with mismatch or antisense LNA gapmers. All mito-EGFP- (at stages 12 and 20) and v-DS-labeled (at stage 32) PGCs from the injected embryos were isolated to measure their diameter. Total PGC numbers after each injection are shown as 100%. ‘n’ indicates total PGC numbers. Arrows indicate average diameters from the indicated experiments. (B) Localization of the germ plasm in PGCs isolated from mito-EGFP embryos at stages 7 and 12, which were injected with mismatch or antisense LNA gapmers. Scale bar: 100 µm. Note that at stage 12, PGCs injected with the antisense LNA gapmer are relatively large and contain the germ plasm beneath the cell membrane, compared with perinuclear localization of the germ plasm in PGCs with the mismatch LNA gapmer. (C) Ratio of PGCs with three localization patterns of germ plasm from stage 12 embryos injected with mismatch or antisense LNA gapmers. According to the localization patterns, 120 and 124 PGCs with mismatch or antisense gapmers, respectively, were classified into three groups; cortex, perinucleus and intermediate (cortex-perinucleus) shown at the bottom of the panel.

Fig. 6.. Knockdown of the DEADSouth gene is rescued by another type of DEAD-box RNA helicase.

(A) Rescue by injection of Mvh-DS 3′UTR mRNA. (B) Rescue by injection of XVLG1 and its dependency on the 3′UTR. At stage 32, PGCs were externally counted on both sides of embryos injected with the indicated mRNAs and v-DS mRNA as a PGC tracer, in addition to mismatch or antisense LNA gapmers. ‘N’ and ‘n’ indicate the number of experiments and total embryos examined, respectively. P-values and s.d. are indicated.

Fig. 7.. DEADSouth 3′UTR localizes the mRNA to the germ plasm.

Localization of Cy3-labeled XVLG1Δ-DS 3′UTR mRNA (A–C) and Cy3-labeled XVLG1Δ-XVLG1 3′UTR mRNA (D–F) in PGCs at stage 7. PGCs were isolated from mito-EGFP embryos injected with these mRNAs and observed for Cy3 signals (mRNA, in red) and germ plasm (mitochondria, in green). Note that the Cy3 signal is superimposed on the germ plasm in (A–C). Scale bar: 10 µm.