piRNAs and piRNA-Dependent siRNAs Protect Conserved and Essential C. elegans Genes from Misrouting into the RNAi Pathway

Abstract

piRNAs silence foreign genes, such as transposons, to preserve genome integrity, but they also target endogenous mRNAs by mechanisms that are poorly understood. Caenorhabditis elegans piRNAs interact with both transposon and nontransposon mRNAs to initiate sustained silencing via the RNAi pathway. To assess the dysregulation of gene silencing caused by lack of piRNAs, we restored RNA silencing in RNAi-defective animals in the presence or absence of piRNAs. In the absence of piRNAs and a cellular memory of piRNA activity, essential and conserved genes are misrouted into the RNAi pathway to produce siRNAs that bind the nuclear Argonaute HRDE-1, resulting in dramatic defects in germ cell proliferation and function such that the animals are sterile. Inactivation of RNAi suppresses sterility, indicating that aberrant siRNAs produced in the absence of piRNAs target essential genes for silencing. Thus, by reanimating RNAi, we uncovered a role for piRNAs in protecting essential genes from RNA silencing.

Figure 1

Resetting endogenous RNAi in the absence of piRNAs causes sterility. (A) Schematic illustrating a mating-based approach used to reset endogenous RNAi. (B) The proportions of fertile and sterile animals after resetting RNAi in the presence or absence of piwi/prg-1 and associated piRNAs. (C) The proportions of fertile and sterile animals after resetting RNAi in the absence of piRNAs while treating with mut-16 RNAi to prevent reactivation of the RNAi pathway. P0-F1, RNAi treatment began at the P0 L4 larval stage and continued through the F1 generation. P0, RNAi treatment began at the P0 L4 larval stage and the F1 eggs were transferred off of RNAi. F1, RNAi treatment began at the F1 L1 larval stage. (D) The proportions of fertile and sterile F1 animals after crossing wild type males to mut-16−/− or prg-1−/− mut-16−/− hermaphrodites. (E) The proportions of fertile and sterile F1 animals after crossing mut-16−/− or prg-1−/− mut-16−/− males to prg-1−/− hermaphrodites. See also Figure S1.

Figure 2

Resetting endogenous RNAi in the absence of piRNAs disrupts germline development. (A-B) Germlines from adult animals showing modest (A) or severe (B) defects in germline proliferation and progression after RNAi was reset in the absence of piRNAs. Animals were dissected and immunostained for the synaptonemal complex component HTP-3, indicative of meiotic entry. DNA was stained with DAPI. Scale bars represent 20 μm. (C) Localization of the P granule component PGL-1 after resetting RNAi in the presence or absence of piRNAs. Scale bars represent 5 μm. All images are projections of 3D images following deconvolution.

Figure 3

High-throughput sequencing of small RNAs after resetting endogenous RNAi. (A-C) Each CSR-1 and WAGO target is represented as the number of miR-35 normalized siRNA reads (reads per 10,000 miR-35 reads) to account for differences in the numbers of germ cells between strains. (A) siRNA levels in the F1 progeny of prg-1 males crossed to prg-1 hermaphrodites (y-axis) and in the F1 progeny of a control cross between wild type animals (x-axis). (B) siRNA levels in the F1 progeny of animals in which RNAi was reset in the presence of piRNAs (y-axis) and in the F1 progeny of a control cross between wild type animals (x-axis). (C) siRNA levels in the F1 progeny of animals in which RNAi was reset in the absence of piRNAs (y-axis) and in the F1 progeny of a control cross between wild type animals (x-axis). See also Figure S2 and Table S1.

Figure 4

The nuclear Argonaute HRDE-1 binds siRNAs from essential genes after resetting endogenous RNAi in the absence of piRNAs. (A) The proportions of fertile and sterile animals after resetting RNAi in the absence of piRNAs while treating with hrde-1 RNAi or a mock RNAi control. Animals were treated with RNAi starting at the P0 L4 larval stage and continuing through the F1 generation. (B) Schematic illustrating the approach used to reset endogenous RNAi and then immunoprecipitate (IP) FLAG::HRDE-1 and sequence the associated small RNAs. Animals of the indicated genotype were selected for FLAG::HRDE-1 IP and small RNA sequencing. Animals in which RNAi was reset in the absence of piRNAs were treated with mut-16 RNAi to prevent efficient reactivation of RNAi while they were genotyped and expanded. The western blots display FLAG::HRDE-1 protein in the input (in) and IP. (C-E) Each CSR-1 and WAGO target is represented as the number of normalized siRNA reads in FLAG::HRDE-1 input (x-axis) and IP (y-axis) fractions after resetting RNAi in the presence (C and D) or absence (E) of piRNAs. (F) Enrichment or depletion of CSR-1 and WAGO class 22G-RNAs in FLAG::HRDE-1 IPs relative to the corresponding input fractions after resetting RNAi. (G) The pie charts display the proportions of total small RNA reads in FLAG::HRDE-1 input and IP fractions after resetting RNAi. (H) The proportions of conserved genes and genes required for fertility that produced siRNAs enriched by >2 fold in FLAG::HRDE-1 IPs relative to input fractions after resetting RNAi. (I) Model depicting the role of Piwi/PRG-1 in directing the proper mRNAs into the RNAi pathway. See also Figure S3 and Table S2.