Smaug regulates germ plasm assembly and primordial germ cell number in Drosophila embryos

Abstract

During Drosophila oogenesis, the Oskar (OSK) RNA binding protein (RBP) determines the amount of germ plasm that assembles at the posterior pole of the oocyte. Here, we identify mechanisms that subsequently regulate germ plasm assembly in the early embryo. We show that the Smaug (SMG) RBP is transported into the germ plasm of the early embryo where it accumulates in the germ granules. SMG binds to and represses translation of the osk messenger RNA (mRNA) as well as the bruno 1 (bru1) mRNA, which encodes an RBP that we show promotes germ plasm production. Loss of SMG or mutation of SMG's binding sites in the osk or bru1 mRNA results in excess translation of these transcripts in the germ plasm, accumulation of excess germ plasm, and budding of excess primordial germ cells (PGCs). Therefore, SMG triggers a posttranscriptional regulatory pathway that attenuates the amount of germ plasm in embryos to modulate the number of PGCs.

Fig. 1.. SMG protein is enriched in the germ plasm of early embryos.

(A) Live imaging of Venus-SMG shows that SMG protein is enriched in the germ plasm as early as NC 5. Still images were taken from movie S1. The time after imaging began is indicated at the top right of each still image in minutes and seconds. White arrowheads indicate the boundaries of Venus-SMG in the germ plasm in the first three images; in the fourth image, the arrowheads point to Venus-SMG associated with nuclei in budding PGCs. The transgene insertion was on the second chromosome (see table S1). Scale bar, 15 μm (applies to all panels). (B) Live imaging of mCherry-SMG (red) showing association of SMG with the spindle poles (green; detected with Jupiter-GFP) upon arrival of nuclei at the posterior (NC 8) and incorporation into the PGCs when they bud (NC 9). Still images were taken from movie S2. The mCherry-SMG transgene insertion was on the X chromosome (see table S1). Scale bar, 15 μm (applies to all panels).

Fig. 2.. SMG is a component of the germ granules.

(A) SMG co-localizes with VAS, a known component of germ granules. Panels are still images taken from movies of embryos co-expressing mCherry-SMG (red) and VAS-GFP (green). The boxed areas in the left-hand images are shown at higher magnification on the right. Scale bars, 5 μm. See associated movie S3. The mCherry-SMG transgene insertion was on the third chromosome (see table S1). (B) Immuno-electron microscopy shows that, in the pole cells, SMG (10 nm gold: small particles) is found in germ granules along with VAS (15 nm gold: large particles, indicated with white arrowheads). The boxed area in the left-hand image is shown at higher magnification in the center. In the posterior somatic cells (right), SMG (10 nm gold) is enriched in electron-dense, non–membrane-bound organelles. Scale bars, 0.2 μm (C) SMG and VAS also co-localize in the apical region of the posterior soma, representing germ granules that are not taken up into the PGCs when they bud. White arrowheads point to co-localization in a PGC; white arrows point to co-localization in the apical somatic cytoplasm. Scale bar, 5 μm (applies to all panels). See also associated fig. S1. The mCherry-SMG transgene insertion was on the X chromosome (see table S1).

Fig. 3.. Extra PGCs form in smg mutants.

(A) Box plots (median, 75th and 25th percentiles) showing the number of PGCs in embryos from a variety of wild-type or smg-mutant mothers. P values are from the Kruskal-Wallis one-way analysis of variance (ANOVA) followed by Dunn’s test for multiple comparisons; N values as follows: w¹¹¹⁸ = 79; Df/TM3 = 14; smg¹/Df = 69; smg¹/smg⁴⁷ = 43; smg⁴⁷/Df = 19; smg⁴⁷/smg⁴⁷ = 23. ns, not significant. (B) Box plot (median, 75th and 25th percentiles) showing the number of pole buds at NC 9 in embryos from a wild-type or smg-mutant mothers. P value is from the unpaired, one-sided Wilcoxon rank sum test. N values as follows: w¹¹¹⁸ = 19; smg¹/smg⁴⁷ = 18. (C) Box plot (median, 75th and 25th percentiles) showing the number of phosphohistone H3 (PH3)–positive PGCs in embryos from a wild-type or smg-mutant mothers. P value is from Bonferroni-corrected unpaired, two-sided Wilcoxon rank sum test. N values as follows: w¹¹¹⁸ = 44; smg¹/Df = 36; smg¹/smg⁴⁷ = 43. (D to K) Confocal images showing 4′,6-diamidino-2-phenylindole (DAPI; blue), VAS (green), and PH3 (red) in wild type (w¹¹¹⁸) (D to G) or in smg mutants (H to K). The smg-mutant genotype of the females from which the embryos shown in (H) to (K) were obtained was smg¹/Df(3L)Scf-R6. White arrowheads point to VAS-positive cells that are also PH3 positive. Note that, in smg mutants, the somatic nuclei continue to divide at this stage (17) and, therefore, are PH3-positive but VAS-negative. Three-dimensional reconstructions were used to distinguish the VAS-positive, PH3-positive PGC cells from the VAS-negative, PH3-positive somatic nuclei (see Materials and Methods). Scale bar, 25 μm [applies to (D) to (K)]. See table S2 for data and statistical tests.

Fig. 4.. Excess OSK protein is produced in the germ plasm of smg mutants.

(A) Anti-OSK IF in wild-type (left; w¹¹¹⁸) or smg-mutant (right) embryos. The latter were from smg-mutant mothers [genotype, smg¹/Df(3L)Scf-R6]. The top and middle sets of panels show the posterior pole of embryos before budding of the PGCs; the bottom set of panels shows the posterior pole of embryos during budding of the PGCs. Images are z-stack–projected confocal images of OSK (red) and DNA as visualized with pico-green (green). Scale bar in the low magnification image is 50 μm and that in the high magnification image is 25 μm. (B) Quantification of germ plasm OSK. N ≥ 4 for each genotype and stage. Data points represent individual embryos, means ± SD are shown. P values are from the unpaired, one-sided Wilcoxon rank sum test. See also associated figs. S2 and S3. Table S2 provides data and statistical tests. A.U., arbitrary units.

Fig. 5.. Excess VAS protein is produced in the germ plasm of smg mutants.

(A) Anti-VAS IF in wild-type (left; w¹¹¹⁸) or smg-mutant (right) embryos. The latter were from smg-mutant mothers (genotype, smg¹/smg⁴⁷). The top and middle sets of panels show the posterior pole of embryos before budding of the PGCs; the bottom set of panels shows the posterior pole of embryos during budding of the PGCs. Images are z-stack–projected confocal images of VAS (red) and DNA as visualized with pico-green (green). Scale bar in the low magnification image is 50 μm and that in the high magnification image is 25 μm. (B) Quantification of germ plasm VAS. N ≥ 3 for each genotype and stage. Data points represent individual embryos, means ± SD are shown. P values are from the unpaired, one-sided Wilcoxon rank sum test. See table S2 for data and statistical tests.

Fig. 6.. Mutation of SREs in the osk mRNA results in increased OSK protein and PGCs.

(A) Confocal microscope images of NC 5 embryos expressing osk^2xSRE(+) or osk^2xSRE(−) in an osk⁰/Df(osk) background. Red, OSK; blue, DAPI. In both genotypes, OSK is localized to the germ plasm. Transgenic OSK-HA was detected with anti-OSK antibody. Scale bar in the low magnification image is 100 μm and that in the high magnification image is 25 μm. (B) A representative Western blot of embryos expressing osk^2xSRE(+) or osk^2xSRE(−) in an osk⁰/Df(osk) background; wild-type (WT), w¹¹¹⁸. Transgenic OSK-HA was detected with an anti-HA antibody. (C) Quantification of OSK in the two genotypes described in (B). Values were normalized to the actin loading control and then to osk^2xSRE(+) at the 0- to 1-hour time point after egg laying (AEL); N = 5 for each genotype and stage. Means ± SD are shown. (D and E) Quantification of OSK (D) and VAS (E) in the germ plasm of the two genotypes described in (B). NC 5 to NC 6: OSK and VAS (N = 7) for both osk^2xSRE(+) and osk^2xSRE(−). NC 7 to NC 8: OSK (N = 7 and 6) and VAS (N = 7 and 4) for osk^2xSRE(+) or osk^2xSRE(−), respectively. Data points represent individual embryos, means ± SD are shown. (F) Box plots (median, 75th and 25th percentiles) of PGC numbers in embryos expressing osk^2xSRE(+) or osk^2xSRE(−) in an osk⁰/Df(osk) background; N = 13 and 18, respectively. Each dot or square represents the PGC count in a single embryo. P values in (C) to (F) are from the one-sided, unpaired Wilcoxon rank sum test. See table S2 for data and statistical tests.

Fig. 7.. SMG represses bru1 mRNA in the germ plasm.

(A) Anti-BRU1 IF in wild-type (left; w¹¹¹⁸) or smg-mutant (right) embryos. The latter were from smg-mutant mothers [genotype, smg¹/Df(3L)Scf-R6]. The top and middle sets of panels show the posterior pole of embryos before budding of the PGCs, and the bottom set of panels shows the posterior pole of embryos during budding of the PGCs. Images are z-stack–projected confocal images of BRU1 (red); DNA as visualized with pico-green (green). Scale bar in the low magnification image is 50 μm and that in the high magnification image is 25 μm. (B) Quantification of germ plasm BRU1. N ≥ 3 for each genotype and stage. Data points represent individual embryos, means ± SD are shown. P values are from the unpaired, one-sided Wilcoxon rank sum test. See also associated fig. S2. Table S2 provides data and statistical tests.

Fig. 8.. Mutation of SREs in the bru1 mRNA results in increased BRU1 protein and PGCs.

(A) Confocal microscope images of NC 7 embryos expressing bru1^5xSRE(+) or bru1^5xSRE(−). Red, BRU1; blue, DAPI. In both genotypes, BRU1 is localized to the germ plasm. Scale bar in the low magnification image is 100 μm and that in the high magnification image is 25 μm. (B) A representative Western blot of embryos expressing bru1^5xSRE(+) or bru1^5xSRE(−) in a wild-type background; WT, w¹¹¹⁸. (C) Quantification of BRU1 in the two genotypes described in (B). Values were normalized to the actin loading control and then to bru1^5xSRE(+) at the 0- to 1-hour time point. N = 4 for each genotype and stage. Means ± SD are shown. (D and E) Quantification of OSK (D) and VAS (E) in the germ plasm of the two genotypes described in (B). NC 5 to NC 6: OSK (N = 17 and 13) and VAS (N = 15 and 8) for bru1^5xSRE(+) or bru1^5xSRE(−), respectively. NC 7 to NC 8: OSK and VAS (N = 7) for both bru1^5xSRE(+) and bru1^5xSRE(−). NC 7 to NC 8: OSK (N = 10 and 14) and VAS (N = 11 and 7) for bru1^5xSRE(+) or bru1^5xSRE(−), respectively. Data points represent individual embryos, means ± SD are shown. (F) Box plots (median, 75th and 25th percentiles) of PGC numbers in embryos expressing bru1^5xSRE(+) or bru1^5xSRE(−); N = 55 for each genotype. Each dot or square represents the PGC count in a single embryo. P values in (C) and (D) are from the one-sided Wilcoxon rank sum test. See table S2 for data and statistical tests.

Fig. 9.. SMG represses gcl mRNA in the germ plasm.

(A) Anti-GCL IF in wild-type (left; w¹¹¹⁸) or smg-mutant (right) embryos. The latter were from smg-mutant mothers (genotype, smg¹/smg⁴⁷). The top and middle sets of panels show the posterior pole of embryos before budding of the PGCs, and the bottom set of panels shows the posterior pole of embryos during budding of the PGCs. Images are z-stack–projected confocal images of GCL (red); DNA as visualized with pico-green (green). Scale bar in the low magnification image is 50 μm and that in the high magnification image is 25 μm. (B) Quantification of germ plasm GCL. N ≥ 4 for each genotype and stage. Data points represent individual embryos, means ± SD are shown. P values are from the unpaired, one-sided Wilcoxon rank sum test. See also fig. S2. Table S2 provides data and statistical tests.

Fig. 10.. Pathway for regulation of PGC number in embryos.

SMG protein is synthesized and transported into the germ plasm of the early embryo. There, SMG represses translation of the osk and bru1 mRNAs (solid lines), attenuating synthesis of OSK protein and recruitment of the VAS protein (double-headed gray arrow), thus preventing additional germ plasm assembly in the embryo (gray arrow). Our data suggest that SMG may also repress translation of the gcl mRNA (dashed lines), which encodes a protein required for budding of the PGCs. By down-regulating germ plasm assembly and GCL, SMG modulates PGC numbers in the embryo (gray arrows).