Research ArticleDEVELOPMENTAL BIOLOGY

Dlp-mediated Hh and Wnt signaling interdependence is critical in the niche for germline stem cell progeny differentiation

See allHide authors and affiliations

Science Advances  13 May 2020:
Vol. 6, no. 20, eaaz0480
DOI: 10.1126/sciadv.aaz0480
  • Fig. 1 Hh and Wnt signaling are interdependent in the differentiation niche.

    The germaria are labeled for PZ1444-LacZ expression to visualize IGS cells (two indicated by arrowheads) and cap cells (broken ovals), while DAPI staining identifies all nuclei. (A to D) Merged confocal images of germaria showing that the expression of both fz3-RFP (A) and ptc-GFP (B) is significantly decreased in smoKD and dshKD IGS cells 2 days after knocking down compared with the control (lucKD) (C and D) quantification results on fz3-RFP or ptc-GFP intensities normalized to LacZ in IGS cells, respectively; n = IGS cells number. (E to H) Merged FISH (green) and immunostaining (LacZ, red) confocal images showing that fz3 (E) or ptc (F) mRNA expression levels are significantly reduced in dshKD and smoKD IGS cells (G and H: quantification results on fz3 and ptc mRNA levels based on the fluorescence intensities normalized to LacZ, respectively; n = germarial number). Scale bars, 10 μm (all images at the same scale). In this study, all the quantitative data are shown as means ± SEM, whereas P values are determined by the two-sided Student’s t test (***P ≤ 0.001; **P ≤ 0.01).

  • Fig. 2 Dlp overexpression sufficiently inhibits Hh and Wnt signaling and promotes BMP signaling.

    The germaria (A, E, and F) are labeled for PZ1444-LacZ to visualize IGS cells (two by arrowhead) and cap cells (broken ovals), while DAPI staining identifies all nuclei. (A) Merged confocal images of germaria showing that Dlp protein (green) levels are significantly up-regulated in the dshKD1 and smoKD1 IGS cells 2 days after knockdown (B) quantification results normalized to LacZ; n = germarial number. (C) dlp overexpression (dlpOE) in IGS cells causes the accumulation of significantly more spectrosome-containing undifferentiated SGCs (only two in control and four in dlpOE indicated by arrows) in 5- and 10-day-old germaria but have no or a little impact on GSCs (highlighted by broken ovals) (D) CB/SGC and GSC quantification results; n = germarial number. (E to H) ptc-GFP and fz3-RFP expression is drastically down-regulated in dlpOE IGS cells (arrowheads) 2 days after overexpression compared with the control (G and H: quantification results on fz3-RFP or ptc-GFP intensities normalized to LacZ in each IGS cell; n = germarial number). (I to K) Merged confocal images showing that dlpOE germaria accumulate more bam-GFP–negative, pMad-positive, and Dad-LacZ–positive SGCs (some by arrows in J and K). (L to N) Inactivating one copy of dpp by dpphr27/+ or dpphr56 significantly decreases both pMad-positive SGCs (L) and GSC accumulation (M) caused by dlpOE without any obvious effect on GSC numbers (N: CB/SGC and GSC quantification results; n = germarial number). Scale bars, 10 μm. (***P ≤ 0.001; *P ≤ 0.05).

  • Fig. 3 Dlp repression is responsible for Hh/Wnt signaling interdependence in IGS cells and for GSC progeny differentiation.

    (A and B) dlp knockdown in IGS cells can significantly and drastically rescue the expression of ptc-GFP and fz3-RFP in dshKD and smoKD IGS cells (arrowheads), respectively (broken ovals highlight cap cells; C) quantification results on ptc-GFP and fz3-RFP fluorescence intensities normalized to LacZ; n = IGS cell numbers. (D and E) dlp knockdown in IGS cells can significantly decrease the accumulation of SGCs (only some of them by arrowheads) caused by dshKD and smoKD (E: CB/SGC quantification results; n = germarial number). Scale bars, 10 μm. (***P ≤ 0.001; **P ≤ 0.01).

  • Fig. 4 Canonical Hh and Wnt signaling repress dlp transcription via the dlp2.1.5 genomic region in the second intron.

    (A) Diagram of the dlp genomic regions showing dlp2.1 and dlp2.1.5 regions driving GFP expression, which recapitulates dlp mRNA and protein expression in the germarium (please see fig. S5 for details). (B) Immunostaining with anti-GFP (green) and anti-Dlp (red) showing dlp2.1.5-GFP has similar expression pattern with endogenous Dlp, which has very low level at IGS cells but high level at late-stage somatic cells. (C) Single confocal cross-sectional images of germaria showing that dlp2.1.5-GFP expression is up-regulated in dshKD1, smoKD1, panKD, ciKD, and dlpOE IGS cells compared with the control (lucKD) (anterior germarial regions highlighted by squares are shown at a high magnification). Scale bars, 20 μm in (B), 10 μm in (C).

  • Fig. 5 Ci and Pan bind to multiple sites in the dlp2.1.5 genomic region for directly repressing dlp expression in IGS cells.

    (A) EMSA results showing that GST-Ci-ZNF binds to four sites strongly and additional few sites weakly (green), while GST-Pan-HMG binds to three sites strongly and one site weakly (blue) (* and ** indicate weak and strong sites, respectively). (B) ChIP-qPCR results show that IGS-expressed Flag-Ci or Flag-Pan is associated in vivo with the 800 bp of dlp2.1.5 (P values compared with background control act5C). Mouse IgG antibodies were used as a negative IP control. (***P ≤ 0.001). (C) Summary diagram displaying Pan and Ci binding sites/regions in the dlp2.1.5 region based on EMSA and ChIP results (A and B), as well as the expression-activating regions based on the deletion results in fig. S6. (D) Mutating all the strong binding sites for Pan, Ci, or both causes a moderate GFP up-regulation in IGS cells compared with dlp2.1.5-GFP (note: since these mutations also decrease dlp2.1.5-GFP expression in follicle cells, GFP expression in follicle cells are normalized to that in normal dlp2.1.5-GFP). Scale bars, 10 μm.

  • Fig. 6 Hh/Wnt pathway downstream transcription factors Ci and Pan can recruit Croc, which subsequently recruit Eggless/SetDB1, to the dlp2.1.5 region.

    (A) Compared with WT, dlp2.1.5-GFP expression is up-regulated in eggKD and crocKD IGS cells (brackets) 5 days after knockdown. (B and C) Fourteen-day croc knockdown in IGS cells causes the accumulation of SGCs (arrowheads) (C: CB/SGC and GSC quantification results). n.s., no significance. (D) Summary diagram showing the Croc binding sites in the dlp2.1.5 region based on EMSA and ChIP results in fig. S8 (H to J). (E) ChIP-qPCR results show that the binding ability of Croc to dlp2.1.5 is significantly reduced in smoKD or dshKD IGS cells compared with WT. (F and G) In S2 cells, CiPKA-Myc and Pan-HA can bring down Croc-Flag. CiPKA is a noncleavable active full-length Ci, whereas ArmS10-Myc is the active Arm protein, which can bind to Pan for nuclear import in the absence of Wnt signaling. (H) In S2 cells, Croc-HA can bring down Croc-Flag, indicative of potential dimerization or oligomerization. (I) In S2 cells, Croc-HA can bring down Egg-Flag (IgG as a negative control: *, a nonspecific protein recognized by the anti-HA antibody). (J and K) Coexpression of Croc-HA can significantly increase Egg-Flag protein levels in S2 cells (empty plasmid used to normalize total transfected DNA; K: quantification results on Egg-Flag and Croc-HA levels). (L) Schematic diagram explaining how Hh/Wnt signaling–mediated direct Dlp repression maintains their interdependence and prevents BMP signaling, thereby promoting GSC progeny differentiation. Scale bars, 10 μm. (***P ≤ 0.001).

Supplementary Materials

  • Supplementary Materials

    Dlp-mediated Hh and Wnt signaling interdependence is critical in the niche for germline stem cell progeny differentiation

    Renjun Tu, Bo Duan, Xiaoqing Song, Ting Xie

    Download Supplement

    The PDF file includes:

    • Figs. S1 to S8

    Other Supplementary Material for this manuscript includes the following:

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article