PETISCO is a novel protein complex required for 21U RNA biogenesis and embryonic viability

Abstract

Piwi proteins are important for germ cell development in most animals. These proteins are guided to specific targets by small guide RNAs, referred to as piRNAs or 21U RNAs in Caenorhabditis elegans In this organism, even though genetic screens have uncovered 21U RNA biogenesis factors, little is known about how these factors interact or what they do. Based on the previously identified 21U biogenesis factor PID-1 (piRNA-induced silencing-defective 1), we here define a novel protein complex, PETISCO (PID-3, ERH-2, TOFU-6, and IFE-3 small RNA complex), that is required for 21U RNA biogenesis. PETISCO contains both potential 5' cap and 5' phosphate RNA-binding domains and interacts with capped 21U precursor RNA. We resolved the architecture of PETISCO and revealed a second function for PETISCO in embryonic development. This essential function of PETISCO is mediated not by PID-1 but by the novel protein TOST-1 (twenty-one U pathway antagonist). In contrast, TOST-1 is not essential for 21U RNA biogenesis. Both PID-1 and TOST-1 interact directly with ERH-2 using a conserved sequence motif. Finally, our data suggest a role for TOST-1:PETISCO in SL1 homeostasis in the early embryo. Our work describes a key complex for 21U RNA processing in C. elegans and strengthens the view that 21U RNA biogenesis is built on an snRNA-related pathway.

Keywords: 21U RNA; C. elegans; IFE-3; MID domain; PID-1; PID-3; PRG-1; TOFU-6; enhancer of rudimentary; piRNA.

Figure 1.

PID-1 interactors reside in P granules. (A) Volcano plot representing label-free proteomic quantification of PID-1 immunoprecipitations from nongravid adult extracts. Immunoprecipitations were performed and analyzed in quadruplicates. The X-axis represents the median fold enrichment of individual proteins in wild-type (WT) versus the pid-1(xf14) mutant strain. The Y-axis indicates −log₁₀(P-value) of observed enrichments. Dashed lines represent thresholds at P = 0.05 and twofold enrichment. Blue data points represent values out of scale. Red and green data points represent values above and below threshold, respectively. (B,C) Expression pattern and localization of tagged PETISCO components and P-granule marker PGL-1. Endogenous promoters and 3′ untranslated regions were used for PETISCO proteins. Proteins and observable tags are indicated. (B) Immunostaining images were acquired with a laser scanning confocal microscope. (C) Live worm images were acquired under a wide-field fluorescent microscope. Scale bars, 5 µm. The contrast of the images has been enhanced.

Figure 2.

PID-1 interactors form a novel protein complex, PETISCO. (A–D) Volcano plot representing label-free proteomic quantification of quadruplicate 3xFlag::mCherry::IFE-3;ife-3(xf101) (A), PID-3::mCherry::Myc;pid-3(tm2417) (B), TOFU-6::GFP::HA;tofu-6(it20) (C), and ERH-2::GFP::OLLAS;erh-2(xf168) (D) immunoprecipitations from nongravid adult extracts. The X-axis represents the median fold enrichment of individual proteins in the control (wild-type [WT]) versus the transgenic strain. The Y-axis indicates −log₁₀(P-value) of observed enrichments. Dashed lines represent thresholds at P = 0.05 and twofold enrichment. Blue data points represent values out of scale. Red and green data points represent above and below threshold, respectively. (E) Size exclusion chromatography of nongravid adult worm extracts of the indicated strains. Fractions were collected and probed for HA, GFP, and PID-1, respectively. The approximate molecular weights of the fractions are indicated. pid-1(xf35) extract was used as probing control. The red arrowhead indicates full-length TOFU-6::GFP, and the white arrowhead indicates PID-1. pid-1(xf35) extract was used as negative control for PID-1. (F) Venn diagram summarizing significant interactions in PETISCO protein immunoprecipitations. The asterisk represents protein found significantly enriched in only one experiment of 3xFlag::mCherry::IFE-3;ife-3(xf101) immunoprecipitation.

Figure 3.

PETISCO architecture. (A–D) Y2H interaction assays of PETISCO subunits in low-stringency (TRP⁻LEU⁻HIS⁻) or high-stringency (TRP⁻LEU⁻HIS⁻ADE⁻) medium as indicated. Interactions were screened in both Y2H orientations. (A) Full-length proteins. (B) TOFU-6 and individual domains tested for interaction with full-length IFE-3. (C) Interactions between PID-3 and TOFU-6. (D) Interaction between PID-3 and ERH-2. For details on domains and other selection conditions, see Supplemental Figures S1C and S5.

Figure 4.

PETISCO is required for 21U RNA biogenesis. (A) Wide-field fluorescent microscopy of adult hermaphrodites carrying the GFPH2B-21U sensor transgene in a sensitized background. Worms were subjected to RNAi via feeding (targets are indicated) from L1 larval stage to adulthood. Empty RNAi vector served as a negative control. Scale bar, 10 µm. (B) Quantitative RT-PCR of the 21U sensor transgene in the adult worm populations of A. Values were obtained from experimental triplicates and technical duplicates, normalized to pmp-3 mRNA levels. The significance of sample versus control was tested with Dunnett's post hoc test. (***) P-value < 0.001; (**) P-value < 0.01; (*) P-value < 0.05. Error bars represent standard deviation. (C) Pairwise differential gene expression of type I 21U RNAs and miRNAs in pid-3(tm2417), erh-2(xf168), and ife-3(xf102) versus the respective controls. Samples consisted of nongravid adult individuals. Red and blue dots indicate up-regulated and down-regulated transcripts, respectively.

Figure 5.

PID-1 and TOST-1 define different PETISCO functions. (A) Alignment of a short region of PID-1 and TOST-1 homologs from different nematodes. A conserved arginine residue was found to be mutated in pid-1(xf14), as indicated. A consensus sequence is presented below the alignment. Alignment was performed with Muscle version 3.8 (Edgar 2004), and representation was performed with ESPrit version 3.0 (Robert and Gouet 2014). (B) Y2H interaction assay of PID-1, TOST-1, and PID-1/TOST-1 carrying the corresponding arginine mutation found in pid-1(xf14). High-stringency plates (TRP⁻LEU⁻HIS⁻ADE⁻) were used. For other conditions, see Supplemental Figure S4D. (C) Global levels of type I 21U RNA precursors in wild-type (N2) and pid-1(xf35) nongravid adults. Values were obtained from CIP-RppH-treated smRNA libraries. (RPM) Reads per million mapped reads. Individual data points of three independent replicates are shown, and the horizontal bar represents the mean. Significance was tested with Student's t-test, and P-values are indicated in the graph. (D) Pairwise differential gene expression of type I 21U RNAs and miRNAs in pid-1(xf35) and tost-1(xf194) versus wild-type (N2) control. Samples consisted of gravid adult individuals. Red and blue dots indicate up-regulated and down-regulated transcripts, respectively.

Figure 6.

PETISCO interacts with 21U precursors and with SL1 snRNA. (A) Fold enrichments of 21U RNA precursors in mock (N2), 3xFlag::mCherry::IFE-3;ife-3(xf101), and PID-3::mCherry::Myc;pid-3(tm2417) RIPs over paired inputs in nongravid adult worms. The left column displays RppH-treated samples, and the right column shows nontreated samples. Individual data points of three or more independent replicates are shown, and the horizontal bar represents the mean. The significance of sample versus mock RIP was tested with Dunnett's post hoc test, and P-values are indicated in the graph. (B) Schematic representation of the SL1 RNA. Green arrows represent sequence borders of the previously SL1-derived 21U RNA. (C) Fold enrichments of SL1 RNA and 5S rRNA in mock (N2), 3xFlag::mCherry::IFE-3;ife-3(xf101), and PID-3::mCherry::Myc;pid-3(tm2417) RIPs over paired inputs in nongravid adult worms. The left column displays RppH-treated samples, and the right column shows nontreated samples. Individual data points of three or more independent replicates are shown, and the horizontal bar represents the mean. The significance of sample versus mock RIP was tested with Dunnett's post hoc test, and P-values are indicated in the graph. (D) Coverage profile, normalized to paired input of SL1 of the data displayed in C. Colors below SL1 RNA correspond to scaled colors represented in B.

Figure 7.

Schematic representation of PETISCO function and conservation in nematodes. (A) Heat map representing the conservation of PETISCO orthologs in nematodes according to BLASTp score versus C. elegans. ERH-1, PRDE-1, and PRG-1 are included for comparison. (B) A schematic of the proposed PETISCO function displaying its dual function: one in 21U biogenesis and another needed for embryonic survival. The latter may be connected to splice leader homeostasis. The m7G capped transcripts interact with the IFE-3. The presence of a 5′P-binding activity in PID-3 may reflect 5′ end processing of transcripts bound by PETISCO. The nuclease responsible has not yet been identified. ERH-2 serves as an anchor for PID-1 or TOST-1 driving PETISCO function toward either 21U RNA biogenesis or its embryonic survival function, respectively.