PRG-1 and 21U-RNAs interact to form the piRNA complex required for fertility in C. elegans

Abstract

In metazoans, Piwi-related Argonaute proteins have been linked to germline maintenance, and to a class of germline-enriched small RNAs termed piRNAs. Here we show that an abundant class of 21 nucleotide small RNAs (21U-RNAs) are expressed in the C. elegans germline, interact with the C. elegans Piwi family member PRG-1, and depend on PRG-1 activity for their accumulation. The PRG-1 protein is expressed throughout development and localizes to nuage-like structures called P granules. Although 21U-RNA loci share a conserved upstream sequence motif, the mature 21U-RNAs are not conserved and, with few exceptions, fail to exhibit complementarity or evidence for direct regulation of other expressed sequences. Our findings demonstrate that 21U-RNAs are the piRNAs of C. elegans and link this class of small RNAs and their associated Piwi Argonaute to the maintenance of temperature-dependent fertility.

Figure 1. 21U-RNAs can be distinguished from other RNA species by their lengths and upstream motif matches

(A) A schematic representation of the 21U-RNA upstream motif as described previously (Ambros et al., 2003; Ruby et al., 2006). (B) The frequency of 21nt RNA reads (blue) or unique sequences (pink) versus upstream motif score. A cut-off score of 7 (orange) was used to define the 21U-RNA population. (C) The distribution of 21U-RNA reads across chromosome IV. Reads were classified as 21U-RNAs by their motif scores and normalized read counts were summed for each non-overlapping 100kb bin (blue). (D) The upstream motif score predicts the magnitude of 21U-RNA expression. For each three-bit bin of motif scores, the number of reads was determined for every observed 21U-RNA locus whose motif score fell in that range. For each bin, the median read number was determined; error bars indicate the 25^th and 75^th percentiles. The number of loci in each bin is indicated by n. (E) Two 21U-RNA loci whose core upstream motifs are aligned (Blanchette et al., 2004). The core motif (green) and 21U-RNA (pink) are highlighted. The C. briggsae 21U-RNA was annotated based on the highest-scoring 5′ end given the conserved core motif position. The number of reads for the 21U-RNA species from C. elegans is given above, and the motif score for each 21U-RNA ortholog is provided.

Figure 2. 21U-RNAs are expressed in the C. elegans germ line

(A and B) Northern blot analysis of 21U-RNA-1 and 21U-RNA-3442 small RNAs. The SL1 precursor is shown as a loading control. In (A), RNA was generated from synchronized wild-type populations at distinct developmental stages. In (B) RNA generated from wild-type worms was compared to RNA obtained from strains with no germline: glp-4(bn2) and eft-3(q145); a male-only population: fog-2(q71); and a population of worms with no sperm: fem-1(hc17). (C) The expression profile for the bulk population of 21U-RNAs as determined by large-scale sequencing. The percent of reads from each library with 21U-RNA identity are shown. Light blue libraries were prepared for sequencing with Rnl2(1−249) ligase, dark blue with T4 RNA ligase 1 (see methods).

Figure 3. PRG-1 protein is expressed in the germ line and required for 21U-RNA accumulation

(A) Northern blot analysis of 21U-RNA-1, 21U-RNA-3442 and miR-66 expression in wild-type and mutant strains. In the left panel RNA generated from prg-1 and dcr-1 homozygous populations was analyzed. In right panel RNA samples obtained from the tm872 and pk2298 alleles of prg-1 and tm1094 and ok1328 alleles of prg-2 as well as from the double mutant prg-1(tm872) prg-2(tm1094) were analyzed. The SL1 precursor serves as a loading control. (B) Western blot analysis of the PRG-1 developmental expression profile. A tubulin specific antibody was used as a loading control. On the left panel, protein lysate generated from wild-type populations at distinct developmental stages was analyzed. On the right panel, protein lysates from wild type worms were compared to protein lysates from strains with no germline: glp-4(bn2); a male-only population; fog-2(q71) mutant worms; and a population of worms with no sperm: fem-1(hc17). Quantitative Real-Time PCR analysis of the expression of the prg-1/prg-2 mRNA is shown in the bottom panel. The PCR primers used recognize both prg-1 and prg-2Expression level is indicated relative to the levels obtained for actin (act-3) mRNA in each sample. (C-F) PRG-1 immunofluorescence (red) and DNA DAPI staining (blue) in dissected gonad arms from an adult hermaphrodite (C) and male (D), a two-cell embryo (E), and a 4-cell embryo (F). In (C and D) The mitotic (MPZ) and meiotic zones (Transition Zone plus Pachytene) are indicated, as are the proximal zones containing oocytes and sperm (respectively). (G) Dual immunofluorescence analysis of 3 oocytes in the proximal arm of a wild-type hermaphrodite gonad stained for PRG-1 and PGL-1 as indicated. Yellow represents overlap in the merged image (bottom panel).

Figure 4. PRG-1 interacts with and is required for the accumulation of all 21URNAs

(A) The frequency of 21nt RNA reads from wild-type young adults (blue) and prg-1(tm872) young adult (pink) versus upstream motif score. A cut-off score of 7 (orange) was used to define the 21U-RNA population. (B) 21U-RNAs are absent in glp-4(bn2) and prg-1(tm872) mutant worms. Percent reads from each library with 21U-RNA identity are shown. Histogram bars colored as in Figure 2C. (C) Endogenous siRNAs are absent in glp-4(bn2) but not prg-1(tm872) mutant worms. Percent reads with 5′ G nucleotides and complete antisense overlap with coding exons (Ambros et al., 2003) are shown. Histogram bars colored as in Figure 2C. (D) IP/Northern blot analysis of small RNAs in PRG-1 and GFP::ALG1/2 complexes. Immunoprecipitations were performed on lysates prepared from an otherwise wild-type transgenic strain carrying GFP-tagged ALG-1 and ALG-2. The top panels are Northern blots probed for associated small RNAs. The lower panels are Western blots probed as indicated. (E) Thin Layer Chromatography analysis of the first nucleotide of the small RNA population that co-immunoprecipitates with the PRG-1 protein. Bars show where the single nucleotides migrate. (F) The length and 5′ nucleotide distribution of reads from the Input (top) and PRG-1 co-IP (bottom) libraries. (G) The frequency of 21nt RNA reads from the Input (blue) and PRG-1 co-IP (red) libraries versus upstream motif score. Plotted as in Figure 4B. (H) The distribution of 21U-RNA reads from the PRG-1 co-IP library (red) versus the young adult wild-type library prepared with T4 RNA ligase 1 (see methods; blue). Reads were classified as 21U-RNAs by their motif scores and normalized read counts were summed for each non-overlapping 100kb bin.

Figure 5. PRG-1 exhibits a broad spectrum of germ line defects

(A) Brood size analysis of prg-1 and prg-2 mutant strains. The brood size of ’n’ individual animals for each strain was determined at 20°C and 25°C. Left and right lines represent highest and lowest values respectively. Left and right ends of each box represent the 75^th and 25^th percentile respectively, the diamond represents the average brood size and the vertical line inside the box represents the median value. (B - F) DAPI staining of excised gonads from wild-type, prg-1 and prg-2 strains (as indicated). Gonadal zones are indicated as in Figure 3.

Figure 6. prg-1 mutants exhibit surprisingly subtle changes in gene expression

(A) Gene expression was not preferentially affected in the 21U-rich portions of the C. elegans genome. Median values are shown for each of the indicated probe sets. Error bars indicate 25^th and 75^th percentile signal changes. The number of probes is indicated by ’n’. (B) The overall expression of some gene mountains was significantly altered in the prg-1 (tm872) mutant. All probes overlapping the exons of all genes from each mountain were considered; the median signal changes from each probe set are shown (error bars indicate the 25^th and 75^th percentiles). All mountains from (Kim et al., 2001) are shown whose median expression was increased or decreased by 0.4 log₂ units or more. The number of probes is indicated by n. (C) Small RNA depletion in the prg-1(tm872) mutant versus enrichment in the PRG-1 co-IP. The x axis indicates the ratio of read frequencies between the input versus PRG-1 co-IP libraries. The y axis indicates the ratio of antisense read frequencies between the wild-type and prg-1(tm872) mutant siRNA-enriched libraries. Each blue dot indicates the antisense read count for one gene whose wild type siRNA-enriched read count is ≥500. Each red dot indicates the read count for a 21U-RNA species with ≥200 reads from the young adult wild type library prepared with T4 RNA ligase 1 (see methods) and at least one read between the two libraries of each plot axis. (D) Changes to mRNA versus derivative siRNA levels per gene in the prg-1(tm872) mutants. Each point indicates a gene with ≥10 array probes and ≤500 antisense reads from the wt siRNA-enriched library overlapping annotated exons. The x axis is as in Figure 6A. The y axis is as in Figure 6C. (E) A schematic view of a full-length Tc3 transposon showing the inverted repeats (grey) and Tc3A transposase gene (red). The position of 21U-RNA-15703 is indicated by a red asterisk. (F) Density of reads mapping to the sense (blue) and antisense (orange) strands of the Tc3 element from Figure 6E. Reads per 50nt window are shown from the wild-type (top) and prg-1(tm872) mutant (bottom) siRNA-enriched libraries. Read counts are not normalized to the number of genomic matches. Dotted grey lines indicate 0.002% of each library. (G) Density of reads mapping to the sense (blue) and antisense (orange) strands of the Tc3 element from Figure 6E. Reads per 50nt window are shown from the input (top) and PRG-1 co-IP (bottom) libraries. Read counts are not normalized to the number of genomic matches. Dotted grey lines indicate 0.002% of each library. (H) qRT-PCR analysis of the expression of the TC3A mRNA. Primers recognizing TC3A mRNA were used in qRT-PCR on cDNA generated from wild-type, prg-1(tm872) and prg-1(pk2298) worms. Expression level is relative to actin (act-3) mRNA in each sample.

Figure 7. Models for 21U-RNA function

(A) Regulation of TC3 inverted repeats by PRG-1/21U-RNA-15703 (B) Regulation of germ line transcripts by imperfect base paring