H/ACA snRNP-dependent ribosome biogenesis regulates translation of polyglutamine proteins

Abstract

Stem cells in many systems, including Drosophila germline stem cells (GSCs), increase ribosome biogenesis and translation during terminal differentiation. Here, we show that the H/ACA small nuclear ribonucleoprotein (snRNP) complex that promotes pseudouridylation of ribosomal RNA (rRNA) and ribosome biogenesis is required for oocyte specification. Reducing ribosome levels during differentiation decreased the translation of a subset of messenger RNAs that are enriched for CAG trinucleotide repeats and encode polyglutamine-containing proteins, including differentiation factors such as RNA-binding Fox protein 1. Moreover, ribosomes were enriched at CAG repeats within transcripts during oogenesis. Increasing target of rapamycin (TOR) activity to elevate ribosome levels in H/ACA snRNP complex-depleted germlines suppressed the GSC differentiation defects, whereas germlines treated with the TOR inhibitor rapamycin had reduced levels of polyglutamine-containing proteins. Thus, ribosome biogenesis and ribosome levels can control stem cell differentiation via selective translation of CAG repeat-containing transcripts.

Fig. 1.. Pseudouridine is a critical modification required for oogenesis.

(A) Schematic of Drosophila ovariole (top) and germarium (bottom). (B) Heatmap of mass spectrometry analysis of RNA modifications at each developmental stage (as described in Materials and Methods). A cold-hot gradient covers relative abundances from 0 to 3%. The different colors also express variations of relative abundance from the GSC column with a P < 0.05 statistical significance. (C) Schematic of the H/ACA snRNP complex comprising CG4038 (Gar1) (dark blue), Nop60B (gray), NHP2 (light blue), and CG7637 (Nop10) (salmon). (D to I) Confocal image of germarium of UAS-Dcr2;nosGAL4 (D and E), Nop10 germline depletion (F and G), and Nop60B germline depletion (H and I) stained with anti1B1 (magenta) and anti-Vasa (green) antibodies. Images were taken at either 40× (germaria) or 20× (ovariole). The white arrow marks cyst defect. Scale bars, 20 μm. (J) Quantification of oogenesis defect phenotypes. Statistical analysis performed with Fisher’s exact test (n = 50, ***P < 0.0001).

Fig. 2.. The H/ACA snRNP complex is required for proper cyst differentiation and meiotic progression.

(A to E) UAS-Dcr2;nosGAL4;bam-GFP (A) and germline depletion of Nop10 (B), Nop60B (C), Gar1 (D), and NHP2 (E) stained with anti-1B1 (magenta), anti-GFP (blue), and anti-Rbfox1 (green). GFP and Rbfox1 are shown in gray scale. Yellow dashed lines outline cysts that are positive for GFP but have lower Rbfox1 levels for all images. Scale bars, 20 μm. (F) Quantification of oogenesis defect phenotypes per genotype. Statistical analysis performed with Fisher’s exact test (n = 50, ***P < 0.0001). (G) Quantification of Rbfox1 levels normalized to soma for Nop10, Nop60B, Gar1, or NHP2 germline depletion. Statistics were performed using Dunnett’s multiple comparisons test and post hoc test after one-way analysis of variance (ANOVA) (n = 20, ***P < 0.0001). A.U., arbitrary units.

Fig. 3.. The H/ACA snRNP complex is required for translation of meiotic mRNAs.

(A) Mass spectrometry analysis of rRNA from germline ribosomal pulldowns. Statistics was performed using t test. For each genotype, at least two biological replicates were analyzed, with three technical replicates each (***P < 0.0001). (B) Polysome traces for YWT (UAS-Dcr2;nosGAL4) (black) and Nop60B germline depletion (red). Nop60B is required for proper ribosome biogenesis as loss of Nop60B results in 40S and 60S defects and reduction of polysomes compared to control. (C) Venn diagram illustrating overlap of Nop60B-polysome ≤2-fold less association with the ribosome (n = 2, e < 2.87 × 10⁻¹⁹², hypergeometric test). Germarial stages consist of ovaries from young UAS-Dcr2;nosGAL4, while the cyst stages are from bam RNAi; hs-bam (enrichment described in Materials and Methods). (D) Significant biological process GO terms of shared lowly associated mRNAs in Nop60B. (E to G) In situ hybridization to Rbfox1 RNA (green/gray) and staining with anti-1B1 (magenta) and anti-Vasa (blue) antibodies in UAS-Dcr2;nosGAL4 (E) and Nop10 germline depletion(F) or Nop60B germline depletion(G). Yellow dashed lines outline Rbfox1 RNA. (H) Rbfox1 RNA levels in Nop10 and Nop60B germline depletions normalized to soma. Statistics were performed using Dunnett’s multiple comparisons test and post hoc test after one-way ANOVA (n = 10, ns, P > 0.9999 and P = 0.9792, respectively). (I and J) Germarium of UAS-Dcr2;nosGAL4 driving UAS-Rbfox1-RN-HA (I) and UAS-Dcr2;nosGAL4 driving Rbfox1-RN-HA in the background of Nop60B germline depletion (J). Germaria stained with anti-Vasa (blue) and anti-HA (magenta) antibodies. (K) HA levels in control and Nop60B germline depletion normalized to Vasa. Statistics were performed using unpaired t test (n = 17, ***P < 0.0001). Yellow dashed lines outline Rbfox1-RN-HA.

Fig. 4.. The H/ACA snRNP complex is required for translating polyQ proteins.

(A) CAG motif identified by MEME enriched in the CDS of mRNAs lowly associated with polysomes. (B) PolyQ motif identified by MEME that is enriched in the amino acid sequence of mRNAs lowly associated with the ribosome. (C) Schematic of the CAG reporter used. (D and E) UAS-poly41Q-HA driven in UAS-Dcr2;nosGAL4 (D) and in the backgorund of Nop10 germline depletion (E) stained with anti-HA (magenta/gray), anti-GFP (green), and anti-Vasa (blue) antibodies. Scale bar, 20 μm. (F) HA reporter quantification using unpaired t test (n = 15, ***P = 0.0007). (G) CAG motif identified by MEME detected by ribosome footprinting. (H) Ribosome footprint distribution on atx2 mRNA. (I and J) Nop60B (I) and Nop60B rescue by overexpression of the Tor pathway member Raptor (J) stained with anti-Vasa (green) and anti-1B1 (magenta). The arrow points at the egg chamber. (K) Quantification of Nop60B depletion phenotypic rescue by overexpression of the Tor pathway member Raptor in H/ACA box knockdown (>Nop60B RNAi, n = 91, 1.1% contained the first egg chamber, while for UAS-Nop60B RNAi;UAS-Raptor, n = 151, 9.9% contained the first egg chamber, Fisher’s exact test, *P = 0.0037). (L) Representative model showing that a sufficient level of ribosomes is required for translation of meiotic CAG containing mRNAs promoting terminal differentiation. Ribosome insufficiency reduces translation of meiotic CAG containing mRNAs, due to ribosome stalling or slowing, causing terminal differentiation failure.