The RNA-induced transcriptional silencing complex targets chromatin exclusively via interacting with nascent transcripts

Abstract

Small RNAs regulate chromatin modification and transcriptional gene silencing across the eukaryotic kingdom. Although these processes have been well studied, fundamental mechanistic aspects remain obscure. Specifically, it is unclear exactly how small RNA-loaded Argonaute protein complexes target chromatin to mediate silencing. Here, using fission yeast, we demonstrate that transcription of the target locus is essential for RNA-directed formation of heterochromatin. However, high transcriptional activity is inhibitory; thus, a transcriptional window exists that is optimal for silencing. We further found that pre-mRNA splicing is compatible with RNA-directed heterochromatin formation. However, the kinetics of pre-mRNA processing is critical. Introns close to the 5' end of a transcript that are rapidly spliced result in a bistable response whereby the target either remains euchromatic or becomes fully silenced. Together, our results discount siRNA-DNA base pairing in RNA-mediated heterochromatin formation, and the mechanistic insights further reveal guiding paradigms for the design of small RNA-directed chromatin silencing studies in multicellular organisms.

Keywords: RNAi; epigenetics; heterochromatin; nascent transcript; noncoding RNA.

Figure 1.

Transcription start site (TSS)-distal pre-mRNA splicing does not affect the stability of heterochromatin. (A) Schematic of the adh1⁺ promoter-driven ura4 hairpin construct. The double-stranded stem consists of complementary ura4⁺ ORF sequences (401–679; gray), while the loop encodes the cox4⁺ intron (pink) (Iida et al. 2008). Normalized 5′ ends of derived siRNAs are depicted below as reads per million (RPM). (B) Length histogram of generated siRNAs from A colored by their 5′ starting nucleotides. (C) Schematic representation of siRNA target constructs. The cox4⁺ intron sequence was inserted at position +700 nt of the ade6⁺ ORF in forward or reverse orientation (targets 1 and 2, respectively). A GFP fragment of the same size but without canonical splice sites was inserted at the same position (target 3). (D) RT–PCR to assess intron removal was performed with primers flanking the exon–exon junction (mb2202 and mb167). (E) Chromatin immunoprecipitation (ChIP) experiments with an antibody recognizing H3K9me2. Fold enrichments were normalized to adh1⁺ and cendg and are shown relative to the respective leo1⁺/no siRNA samples. Error bars indicate SD. n = 3 independent biological replicates. Two-tailed Student's t-test. (F) ade6⁺ mRNA levels were determined by quantitative RT–PCR. Values were normalized to act1⁺ mRNA and are shown relative to the respective leo1⁺/no siRNA samples. Error bars indicate SD. n = 3 independent biological replicates. Two-tailed Student's t-test. (G) Normalized siRNAs mapping to the spliced (target 1; blue) and unspliced (target 2; red) targets in leo1⁺ (top) and leo1Δ (bottom) cells. Primary siRNAs produced from the cox4⁺ intron in the hairpin (shown in A) are shaded in pink. Secondary siRNAs are generated solely at the target locus. (H) Length histogram of target mapping siRNAs, similar to B. (I) Browser screen shot depicting siRNAs mapping to the ade6⁺ target locus.

Figure 2.

TSS-proximal pre-mRNA splicing reduces the stability of heterochromatin. (A) Schematic representation of siRNA target constructs. The cox4⁺ intron (in forward and reverse orientation in targets 4 and 5, respectively) and GFP control sequence were inserted 46 nt after the TSS of the endogenous ade6⁺ gene. Numbers indicate the positions of primer pairs used in D. (B) RT–PCR to assess intron removal was performed with primers flanking the exon–exon junction (mb10007 and mb10008). (C) ChIP experiments with an antibody recognizing H3K9me2. Fold enrichments were normalized to adh1⁺ and cendg and are shown relative to the respective leo1⁺/no siRNA samples. Error bars indicate SD. n = 3 independent biological replicates. Two-tailed Student's t-test. (D) H3K9me2 enrichment across the ade6⁺ locus in cox4 siRNA-expressing leo1Δ cells. The primer pairs used are indicated in A. Error bars indicate SD. (E) ade6⁺ mRNA levels were determined by quantitative RT–PCR. Values were normalized to act1⁺ mRNA and are shown relative to the respective leo1⁺/no siRNA samples. Error bars indicate SD. n = 3 independent biological replicates. Two-tailed Student's t-test. (F) Normalized siRNA mapping to the spliced (target 4; blue) and unspliced (target 5; red) targets, similar to Figure 1G. (G) leo1Δ cells expressing the spliced siRNA target gene (target 4) and cox4 primary siRNAs were seeded on yeast extract (YE) plates. White and red originator colonies were spotted on yeast extract plates to assess the initiation and maintenance of silencing, respectively.

Figure 3.

High transcriptional activity counteracts RNAi-directed heterochromatin assembly. (A) Scheme depicting the endogenous ade6⁺ gene (left) or the thiamine-repressible nmt1 promoter (right). The siRNA target region is indicated by dashed lines. (B) ade6⁺ mRNA levels were determined by quantitative RT–PCR. Values were normalized to act1⁺ mRNA and are shown relative to leo1Δ cells grown in the presence of 20 µM thiamine. Error bars indicate SD. n = 3 independent biological replicates. (C) ChIP experiments with an antibody recognizing H3K9me2. Fold enrichments over clr4Δ cells are indicated. Error bars indicate SD. n = 3 independent biological replicates. (B,C) ChIP and quantitative RT–PCR experiments were performed with cells from the same culture. (D) Fivefold serial dilutions were spotted on adenine-limited PMG agar plates supplemented with thiamine at the concentrations indicated. Precultures were grown in YES medium, washed, and diluted in H₂O before spotting.

Figure 4.

Transcription is a prerequisite for heterochromatin formation. (A) Genome browser screenshot showing the insertion site for the ade6⁺ transgenes depicted in B. Normalized RNA-seq (green), small RNA-seq (gray), and H3K9me2 ChIP-seq (ChIP combined with high-throughput sequencing) (black) tracks from wild-type cells, including annotated genes (bottom), are indicated. (B) Schematic representation of the ade6⁺ transgenes inserted on chromosome III at the positions indicated in A. The green hairpin denotes primary ade6 siRNAs that are expressed from chromosome I (Kowalik et al. 2015). (C) The absolute number of ade6 RNA molecules per cell was determined by droplet digital PCR (ddPCR) in leo1⁺ cells not expressing ade6 siRNAs. n = 4 biological replicates. Average numbers of RNA molecules per cell are indicated in bold for each transgene, and standard deviation is shown in the brackets. (D–G) ChIP experiments were performed with an antibody recognizing H3K9me2 and the strains indicated. Enrichments were normalized to adh1⁺ and cendg and are shown relative to leo1⁺ cells that do not express ade6 siRNAs. n = 3 independent biological replicates. Error bars indicate SD. Absolute numbers of ade6 RNA molecules per cell were determined in one of the three ChIP replicates.