The Drosophila ribosome protein S5 paralog RpS5b promotes germ cell and follicle cell differentiation during oogenesis

Abstract

Emerging evidence suggests that ribosome heterogeneity may have important functional consequences in the translation of specific mRNAs within different cell types and under various conditions. Ribosome heterogeneity comes in many forms, including post-translational modification of ribosome proteins (RPs), absence of specific RPs and inclusion of different RP paralogs. The Drosophila genome encodes two RpS5 paralogs: RpS5a and RpS5b. While RpS5a is ubiquitously expressed, RpS5b exhibits enriched expression in the reproductive system. Deletion of RpS5b results in female sterility marked by developmental arrest of egg chambers at stages 7-8, disruption of vitellogenesis and posterior follicle cell (PFC) hyperplasia. While transgenic rescue experiments suggest functional redundancy between RpS5a and RpS5b, molecular, biochemical and ribo-seq experiments indicate that RpS5b mutants display increased rRNA transcription and RP production, accompanied by increased protein synthesis. Loss of RpS5b results in microtubule-based defects and in mislocalization of Delta and Mindbomb1, leading to failure of Notch pathway activation in PFCs. Together, our results indicate that germ cell-specific expression of RpS5b promotes proper egg chamber development by ensuring the homeostasis of functional ribosomes.

Keywords: Notch-Delta pathway; Ribosomal protein paralog; RpS5b; rDNA transcription.

Fig. 1.

The RpS5b mutant showed increased rRNA transcription. (A) Stage 7 egg chambers stained for Fibrillarin (red), Udd (green) and DNA (blue). RpS5b mutant nucleoli display elevated and more punctate Udd localization compared with control samples. Scale bar: 10 µm. (B) Western blots probed for Udd and β-Tubulin. The RpS5b mutant samples exhibit increased Udd levels. (C) Levels of Udd relative to β-Tubulin (n=3). Significance was determined using an unpaired Student's t-test (*P<0.1, ***P<0.001). (D) Schematic representation of ribosomal RNA (rRNA) processing. Probes used for northern blot analysis are color coded. (E) Northern blot analysis of wild-type and RpS5b mutant ovaries. 5S and ethidium bromide (EtBr) staining of mature 18S and 28S rRNA served as references. The RpS5b mutant ovaries expressed more pre-rRNA and intermediate processed rRNA transcripts compared with control samples. (F) Control and mutant egg chambers pulse labeled with Br-UTP for 5 min in the presence of α-amanitin stained for Br-UTP (red) and DNA (blue). Scale bar: 10 µm. (G) Quantification of immunofluorescence intensity. Significance was determined using an unpaired Student's t-test (***P<0.001).

Fig. 2.

Ribosome biogenesis is upregulated in the absence of RpS5b. (A) Schematic of the ribosome profiling experiment. (B) Gene ontology enrichment analysis of cellular components for upregulated transcripts. Ribosome-related genes (red bars) showed upregulated translational efficiency in RpS5b mutant ovaries compared with wild type. (C) Volcano plot of translational efficiencies among ribosomal proteins. Significantly changed genes are indicated as red dots. x and y axes are log₂ fold change (log₂FC) and negative value of log₁₀ P values (-log₁₀pvalue), respectively. (D) Representative ribosomal protein genes. Integrated genomics viewer (IGV) showing RNA-seq and ribosome footprint reads on indicated ribosomal proteins of control and RpS5b mutant samples.

Fig. 3.

Global translation rate is increased in the RpS5b mutant. (A) Western blot analysis of puromycin incorporation assay. Ponceau S staining was used as a loading control. RpS5b mutants have increased puromycin labeling, reflecting increased translation. (B) Plot of sucrose gradient fraction experiment showing UV absorbance at 254 nm of control (black line) and RpS5b mutant (red line) ovaries. Loss of RpS5b is accompanied by a modest increase in polysome peaks. (C) Box plots showing distribution of translation efficiencies in control and mutant samples. Significance was determined with the Wilcoxon signed rank test with continuity correction. ****P<2.2e-16.

Fig. 4.

RpS5b mutant germ cells display defects in yolk granule accumulation. (A) Volcano plot of translation efficiencies. Significantly changed genes are indicated with red dots. (B) Read maps of RNA-seq and footprints for Yolkless. (C) Translation efficiency of Yolkless. Significance was determined using the Benjamini and Hochberg method. ****P=2.75e-14. (D) DIC images of yolk granules in control w¹¹¹⁸ and RpS5b mutant oocytes. Stage 8 egg chambers in control ovaries start to accumulate yolk granules in the oocyte (arrow), but the mutant ovaries do not. Scale bars: 20 µm. (E) Auto-fluorescence images of yolk granules in control w¹¹¹⁸ and RpS5b mutant oocytes. Scale bars: 20 µm.

Fig. 5.

Posterior follicle cells in RpS5b mutant egg chambers display defects in Notch-Delta signaling. (A) Ovaries immunostained using anti-pH3 antibody (red) to label cells in mitosis. DNA was stained with DAPI (green). In control ovaries, follicle cells generally stop dividing at stage 6, but RpS5b mutant follicle cells still had pH3-positive cells at the posterior region after stage 7 (white arrows), indicating ongoing mitotic divisions. Scale bar: 20 µm. The images represent a stack of three optical slices. (B) Quantification of pH3-positive posterior follicle cells between control and RpS5b mutants (w¹¹¹⁸ n=84; RpS5b mutant n=76). (C) Ovaries immunostained using anti-Hnt antibody. Unlike control egg chambers, RpS5b mutant posterior follicle cells do not express Hnt (white arrows). Scale bar: 20 µm. (D) Ovaries immunostained using anti-Cut antibody. RpS5b mutant posterior follicle cells exhibit prolonged expression of Cut after stage 7 (white arrows). Scale bar: 20 µm.

Fig. 6.

Delta activation is disrupted in RpS5b mutant egg chambers. (A) Read maps of RNA-seq and footprints for Delta and Mindbomb1. (B) Ovaries immunostained using anti-Delta antibody. In control egg chambers, Delta accumulates at the cell junction between the oocyte and follicle cells (white arrow). By contrast, Delta does not localize to this junction and accumulates in the oocyte cytoplasm in the absence of RpS5b (arrowhead). Scale bars: 20 µm. (C,D) Ovaries were immunostained using anti-NICD (C) and anti-NECD (D) antibody. DNA is labeled with DAPI. Scale bars: 20 µm. (E) Ovaries were immunostained using anti-Mindbomb1 (Mib1) antibody (green). Mib1 localizes to the posterior region in control oocytes (arrow) but not in mutant oocytes (arrowhead). Scale bar: 20 µm.

Fig. 7.

Notch activation in the PFCs rescues multilayered follicle cell phenotype. (A) Ovaries stained using DAPI to label DNA before and after NICD expression. NICD expression partially rescued the multi-layered posterior follicle cell phenotype. Scale bars: 20 µm. (B) Quantification of the ovarioles that have single-layered posterior follicle cells. Number of ovarioles (n) are indicated above the bar graphs. (C) Schematic summarizing RpS5b mutant phenotype during the egg chamber development. Wild-type germ cells express both RpS5a and RpS5b to maintain normal ribosome biogenesis and microtubule organization. However, the absence of RpS5b results in elevated rDNA transcription and increases in the translation efficiency of several RP transcripts, resulting the upregulation of global translation. In addition, translation of genes associated with microtubule function is downregulated upon loss of RpS5b, leading to their mis-organization. As a result, Delta and Mindbomb1 protein do not localize to the interface between the oocyte and posterior follicle cells, resulting in a failure to activate Notch signaling. In turn, disruption of the Notch pathway leads to abnormal cell divisions within posterior follicle cells after stage 6 of egg chamber development.