DEV025114 Supplementary Material

Files in this Data Supplement:

• Supplemental Table S1 - Adobe PDF
• Supplemental Figure S1 -

  Fig. S1. endos does not appear to regulate insulin signaling in adult ovaries. To test whether endos mutants have reduced levels of insulin-like peptide secretion, we measured insulin pathway activation in the germline by using tGPH, a phosphoinositide 3-kinase (PI3K) kinase in vivo reporter (Britton et al., 2002). This reporter contains a fusion of the pleckstrin homology domain of the Drosophila GRP1 to GFP. Upon insulin receptor activation, tGPH binds to phosphatidylinositol (3,4,5)-trisphosphate and is recruited to the plasma membrane. The membrane enrichment of tGPH in control (A) and endos⁰⁰⁰⁰³ mutant (B) germ cells is comparable, suggesting that endos probably does not control insulin-like peptide secretion. Consistent with these results, expression of Endos in brain insulin-producing cells does not rescue the reduced rate of oogenesis of endos mutants. In addition, endos and Sur, the homolog of an ATP-dependent potassium channel regulatory subunit reported to bind to α-endosulfine in cultured cells (Bataille et al., 1999), do not show genetic interactions, although we have not examined whether or not they physically interact. A and B are at the same magnification. Scale bar: 20µm.

• Supplemental Figure S2 -

  Fig. S2. Cdk1 is required for meiotic maturation and for the generation of MPM2 phosphoepitopes in Drosophila. (A-D) DAPI-stained control oocytes in prophase I (A,B) or metaphase I (C,D). Arrows indicate non-exchange fourth chromosomes. (E-H) Oocytes from cdc2^E1-24/cdc2^B47 females at the restrictive temperature have a prolonged prophase I (E-G) and abnormal DNA morphology at stage 14 (H) (see Table S1 in the supplementary material). These phenotypes are similar to those observed in endos⁰⁰⁰⁰³ oocytes (I-L). Scale bar for A-L: 5 µm. (D′) Control dehydrated oocyte. cdc2 mutant oocytes (H′) fail to dehydrate, similar to endos mutant oocytes (L′). Scale bar for D′,H′,L′: 100 µm. (M) Western blotting using MPM2 antibodies, which recognize conserved phosphoepitopes generated by Cdk1 and Polo kinase in many species, and anti-Cdk1 (anti-PSTAIR) antibodies. Wild-type oocytes show high MPM2 levels, whereas both endos⁰⁰⁰⁰³ and cdc2^E1-24/cdc2^B47 have drastically reduced MPM2 levels (30 stage 14 oocytes per lane). Actin was used as a loading control.

  Supplemental Figure 3

• Supplemental Figure S3 -

  Fig. S3. Nuclear envelope breakdown is delayed in endos⁰⁰⁰⁰³ mutants. (A-P) Nomarski images of oocytes showing absent or present nuclear envelopes. (A-D) yw control oocytes showing intact nuclear envelope (A,B) during prophase I or absent nuclear envelope (C,D) after meiotic maturation. (E-H) endos⁰⁰⁰⁰³ oocytes showing intact nuclear envelope until late stage 13 (E-G), which is absent at stage 14 (H). (I-L) Delayed nuclear envelope breakdown of twine¹ oocytes. (M-P) Premature nuclear envelope breakdown of elgi¹ oocytes. Arrowheads indicate the nuclear envelope. (A′-P′) DAPI images for oocytes shown in A-P, respectively. Arrows indicate oocyte DNA. Scale bar: 10 µm. (Q) Percentage of oocytes showing absent nuclear envelope at different stages. Numbers above bars represent the total number of oocytes analyzed. *P<0.01; **P<0.001.

• Supplemental Figure S4 -

  Fig. S4. Live imaging of nuclear envelope breakdown in endos⁰⁰⁰⁰³ and twine¹ mutants. Control, endos⁰⁰⁰⁰³ and twine¹ stage 13 oocytes injected with Oligreen to label DNA (green) and rhodamine-labeled tubulin (red) were analyzed using live imaging. Still images corresponding to time points at every 5 minutes are shown, arbitrarily starting 5 minutes prior (−5′) to the initiation of nuclear envelope breakdown (0′, arrows). Nuclear envelope breakdown duration was measured as the elapsed time from the beginning of nuclear envelope ruffling (arrows) until entry of rhodamine-labeled tubulin into the nucleus (asterisks).

• Supplemental Figure S5 -

  Fig. S5. Meiotic defects of endos mutants cannot be rescued by Twine or Polo expression. (A) Western blot of control and endos⁰⁰⁰⁰³ ovaries expressing myc::twine fusion transgenes. The germline-specific nanos-Gal4::VP16 driver drives robust Myc::Twe expression from UAS-myc::twine #8 or #11 transgenes in control ovaries, but reduced levels are detected in endos⁰⁰⁰⁰³ mutants. (B) Rescue of meiotic defects and sterility of twine¹ mutants by nanos-Gal4::VP16-induced expression of the UAS-myc::twine #8 transgene, but failure of this functional transgene to rescue the endos⁰⁰⁰⁰³ mutant defects. (C) Western blot of control and endos⁰⁰⁰⁰³ mutant ovaries expressing a functional UASp-polo transgene (Xiang et al., 2007). Robust Polo expression is induced by the germline-specific nanos-Gal4::VP16 driver in control females, but reduced expression levels are induced in endos⁰⁰⁰⁰³ mutants. One pair of ovaries (enriched for stage 14 oocytes by starving females) per lane. Actin was used as a loading control in A,C. (D) Failure of nanos-Gal4::VP16-induced expression of the UASp-polo transgene to rescue the meiotic defects of endos⁰⁰⁰⁰³ mutants. In B,D, numbers above each bar indicate the number of oocytes analyzed; *P<0.001; control genotypes are shown in black, and experimental genotypes are shown in blue.

• Supplemental Figure S6 -

  Fig. S6. Polo kinase levels are markedly reduced in endos⁰⁰⁰⁰³ mutants. Polo western analysis of egg chambers at different developmental stages. Wild-type and elgi¹ ovarioles show low Polo expression from germarium to stage 12, and increased Polo expression at stages 13 and 14. twine¹ stage 13 and 14 oocytes show a slight reduction in Polo levels, whereas endos⁰⁰⁰⁰³ oocytes show a stronger reduction. One hundred egg chambers or oocytes per lane. Actin was used as a loading control.

• Supplemental Figure S7 -

  Fig. S7. elgi mRNA is expressed in the ovary and in other adult tissues. (A) RT-PCR analysis reveals that elgi and CG9014 have distinct but overlapping expression patterns in adult tissues. Specifically, elgi is strongly expressed in the ovary, but the ovarian expression of CG9014 is relatively weaker. (B) elgi¹ homozygous or hemizygous ovaries do not express any detectable elgi mRNA, suggesting that the elgi¹ allele is a molecular null. By contrast, ovaries from females carrying the elgi² allele express low levels of mutant elgi mRNA, predicted to encode a truncated protein lacking a functional RING finger domain (see Fig. 4D and Materials and methods). Samples were normalized against Rp49 as a control. Reverse transcription reactions were performed in presence (RT) or absence (−) of reverse transcriptase.