(A) Analysis of protein expression in wild-type and transgenic strains (as indicated). The blot was probed with anti-GFP (GFP::CSR-1), anti-CSR-1 (Native CSR-1) and anti-α-tubulin (α-tubulin) antibodies (as indicated). The neSi9 gfp::csr-1(RNAe) strain was generated by crossing neSi9 gfp::csr-1 to neSi10 gfp::csr-1(RNAe). The neSi8 gfp::csr-1 strain was generated by crossing neSi8 gfp::csr-1(RNAe) to rde-3.
(B, C) qPCR analysis of gfp::csr-1 mRNA, pre-mRNA, and H3K9me3 levels in silent (blue) and expressed (red) transgenic lines. The strains and probes used are indicated in (D). In (B) gfp::csr-1 expression was normalized to the clp-3 mRNA. The data is shown as fold-change between the expressed and silent gfp::csr-1 alleles. Error bars represent the standard deviation for two experimental replicates. In (C), error bars indicate the standard deviation for three experimental replicates.

prg-1(tm872) mutant worms injected with the gfp::cdk-1 construct (top right) give rise to MosSCI lines that express GFP::CDK-1 (P0, top left). The micrographs show the expression status of GFP::CDK-1 in oocyte nuclei (arrowheads) before (P0) and after outcrossing to wild type (F1 and F2 panels), and after segregating homozygous prg-1(+) and prg-1(−) strains for several generations (F3–F10 panels). More than 10 worms were examined per generation. Results are detailed in the text.

(A) Schematic showing the density of 22G-RNAs targeting GFP in neSi8 gfp::csr-1(RNAe) worms, as described in the legend of Figure 3J. Scale bar indicates 20 reads per million. The positions of several 21U-RNAs that could base pair with mismatches to the gfp sequence are indicated below the gene diagram. Five major 22G hotspots (numbered boxes) are enlarged to show the base pairing between the candidate 21U-RNA and gfp, as well as the density of 22G-RNAs at single-nucleotide resolution. Each 21U-RNA has at most two G:U pairs within the seed region (nts 2–8th, yellow highlight), and at most 3 non-seed mismatches (nts 9–21st).
(B) Model for the allelic interactions between transgenes observed in this study.

(A, B) Fluorescence micrographs of adult hermaphrodite germ lines from (A) neSi9 gfp::csr-1, GFP positive and, (B) neSi8 gfp::csr-1(RNAe), GFP negative transgenic lines. GFP::CSR-1 is expressed prominently in the peri-nuclear P-granules in the syncytial germ line (dashed outline) and is also visible in the cytoplasm of maturing oocytes.
(C, D) Schematic diagrams illustrating the results of genetic crosses between expressed (green) and silenced (gray) gfp::csr-1 transgenic lines (>100 animals scored per generation after F2). In (C) neSi8 gfp::csr-1(RNAe) hermaphrodites were mated with neSi9 gfp::csr-1 males. In (D) neSi10 gfp::csr-1(RNAe) hermaphrodites, integrated on chromosome IV (LGIV), were mated to neSi9 gfp::csr-1 males, integrated on LGII (LGII). In the F2 generation, the neSi9 gfp::csr-1 allele was segregated away from neSi10 and propagated for 8 more generations.

(A–F) Fluorescence microscopy of transgene desilencing in wago mutant backgrounds. The transgenes used were neSi8 gfp::csr-1(RNAe), which localizes to P-granules when expressed (indicated by arrow in A and B), and neSi11 gfp::cdk-1(RNAe), which is most prominent in oocyte nuclei (indicated by arrowheads in C–F).
(G) WAGO-9 is a germ-line expressed nuclear Argonaute. Fluorescence micrograph of GFP::WAGO-9 in the adult hermaphrodite germ line. The dashed lines in the micrograph indicate the position of the syncytial germ line.
(H) WAGO-9-associated small RNAs overlap extensively with WAGO-1 small RNAs. The plot shows the enrichment of 22G-RNAs in FLAG::WAGO-9 IP relative to input. Each point in the graph corresponds to previously identified WAGO-1 (blue) and CSR-1 (red) target genes. The x- and y-axes represent the number of 22Gs (log₂ scale) targeting each gene in the Input and WAGO-9 IP samples, respectively. The diagonal lines signify 2-fold enrichment (upper), identity (middle), and 2-fold depletion (lower) of 22G-RNAs in the WAGO-9 IP.
(I) Phylogenetic tree of WAGOs, CSR-1 and RDE-1. Adapted from (Yigit et al., 2006).
(J) Small RNA density along the gfp and cdk-1 coding regions of wild-type and indicated transgenic lines. Vertical bars represent the 5′ nt of a small RNA, and the height of each bar indicates the number of reads that start at that position. The strand is represented by color; sense (light blue) and antisense (pink). Scale bar indicates 10 reads per million. Strain neSi12 cdk-1::gfp(RNAe) was generated by crossing neSi12 cdk-1::gfp to neSi11 gfp::cdk-1(RNAe).

(A) Schematic illustrating the cross between neSi11 gfp::cdk-1(RNAe) and teIs1 oma-1::gfp. The micrographs show the expression status of GFP::CDK-1 in oocyte nuclei (arrowhead) when expressed and OMA-1::GFP in the oocyte cytoplasm. The dashed circles (top left) show the position of GFP-negative oocyte nuclei in the neSi11 gfp::cdk-1(RNAe) strain. The cartoon below each micrograph indicates whether the transgene is expressed (green) or silent (gray).
(B–C) Schematics illustrating crosses between neIs2 gfp::wrm-1 males and (B) neSi11 gfp::cdk-1(RNAe) or (C) neSi12 cdk-1::gfp(RNAe) hermaphrodites. After each cross the two transgenes were either maintained together or allowed to segregate away from each other. The GFP::WRM-1 signal is very weak and was scored periodically during the analysis. The percentage of GFP+ worms indicates the expression of the CDK-1 fusion proteins.