Two sets of RNAi components are required for heterochromatin formation in trans triggered by truncated transgenes

Across kingdoms, RNA interference (RNAi) has been shown to control gene expression at the transcriptional- or the post-transcriptional level. Here, we describe a mechanism which involves both aspects: truncated transgenes, which fail to produce intact mRNA, induce siRNA accumulation and silencing of homologous loci in trans in the ciliate Paramecium. We show that silencing is achieved by co-transcriptional silencing, associated with repressive histone marks at the endogenous gene. This is accompanied by secondary siRNA accumulation, strictly limited to the open reading frame of the remote locus. Our data shows that in this mechanism, heterochromatic marks depend on a variety of RNAi components. These include RDR3 and PTIWI14 as well as a second set of components, which are also involved in post-transcriptional silencing: RDR2, PTIWI13, DCR1 and CID2. Our data indicates differential processing of nascent un-spliced and long, spliced transcripts thus suggesting a hitherto-unrecognized functional interaction between post-transcriptional and co-transcriptional RNAi. Both sets of RNAi components are required for efficient trans-acting RNAi at the chromatin level and our data indicates similar mechanisms contributing to genome wide regulation of gene expression by epigenetic mechanisms.


Preparing libraries of 5'-tri-and mono-phosphorylated small RNAs
For capturing 5'-tri-and mono-phosphorylated RNAs in the same small RNA library, gel-purified small RNA extracts of 50 µg total RNA (see main text) were treated with the Acid Pyrophosphatase Cap-Clip™ (CELLSCRIPT, Madison, Wisconsin) prior to library construction. It removes a pyrophosphate from 5'-triposphorylated and 5'capped RNAs (m7GpppG or others), leaving a 5'-monophosphate. The reaction was set up with 1x Cap-Clip™ reaction buffer, 20U Murine RNase-Inhibitor (NEB, Frankfurt, Germany) and 15U Cap-Clip™ enzyme, and incubated at 37°C for 2.5 hours. After purification with phenol (pH 4) and precipitation with ethanol-sodium acetate and 10.5 µg glycogen, the small RNA was dissolved in nuclease-free water and used for library construction as described (see main text). Two control libraries were prepared, one with small RNAs treated identically, but without adding the Cap-Clip™ enzyme, and one with untreated small RNAs.
In order to verify successful Cap-Clip™ treatment, another aliquot of small RNA was supplemented with 200ng of an oligonucleotide mix (see below) and treated as described, using 20U Cap-Clip™. After phenol purification and precipitation, the reaction product was dissolved in water, and two third were further treated with 1U of Terminator™ 5´-monophosphate-dependent exonuclease (Epicentre, Madison, Wisconsin), in 1x reaction buffer A and 20U Murine RNase-Inhibitor for 70 minutes at 30°C. Both samples were precipitated, run on a 17.5% denaturing ureapolyacrylamide gel and oligonucleotides were visualized by SybrGold staining.
The oligonucleotide mix was prepared as follows: A 5'-triphosphorylated RNA oligo (21nt) was synthesized in vitro using annealed DNA oligonucleotide templates containing a T7 polymerase promoter sequence (Fermentas High Yield Transcript Aid T7 Kit (Thermo Fisher Scientific, Waltham, Massachusetts), according to the manufacturer's instructions). After DNase I treatment the product was gel-purified and mixed in equal proportions with a gel-purified, commercial 5'monophosphorylated RNA (21nt) and a 5'-OH DNA oligo (42nt).
Isolation and preparation of intact chromosomes of Paramecium cells was carried out to determine transgene and endogenous copy number by Southern blots. 200,000 cells were starved in Volvic ® water to digest bacterial DNA. Cells were centrifuged and resuspended in 0.5M EDTA pH 9, 1% Sarcosyl, 1% SDS, 0.25mg/ml proteinase K followed by incubation at 55°C over night. After extraction with Tris-buffered phenol pH 8, the aqueous phase was dialysed against TE buffer for minimum two days. After this, a final concentration of 20ng/µl RNase A was added for 20 min. After additional phenol extraction, DNA was loaded on a 1% agarose gel and Southern blotted according to standard procedures (including depurination to guarantee for efficient blotting of large chromosomes). Labelling of PCR products and hybridisations (at 60°C) were carried out as described for Northern blots using the same probe for the ND169 gene covering the full orf.

Strand-specific RT-PCR
8 µg of total RNA were treated with 2.5U of RNase-free DNase I (Qiagen, Hilden, Germany) in1x buffer RDD for 20 minutes at room temperature and then purified by acid phenol (pH 4) and precipitation with ethanol-sodium acetate. To ensure strandspecificity of the reverse transcription, cDNA synthesis was performed using primers fused 5' to the artificial anchor sequence GACTGGAGCACGAGGACACTGA (according to (7), modified). 500 ng DNase I-treated total RNA were reverse transcribed as follows: the RNA was denatured for 3 minutes at 95°C in presence of primers and dNTPs and then chilled on ice, in order to separate sense and antisense strands. Target transcripts were reverse transcribed according the supplier's instructions using 10 pmol of each primer (ND169 sense or antisense-and GAPDH (GSPATG00016902001) sense-specific), 200U of Maxima ® Reverse Transcriptase (Thermo Fisher Scientific, Waltham, Massachusetts ) and 20U of Murine RNase-Inhibitor in 1x RT buffer for 30 minutes at 57°C. The enzyme was heat-denatured for 5 minutes at 85°C and the cDNA was diluted 1:1 with nuclease-free water. 1 µl of cDNA were used in a 25 µl PCR reaction. PCRs were performed with Q5® High-Fidelity DNA Polymerase (NEB, Frankfurt, Germany) according to the supplier's instructions. An anchor-specific forward primer and a ND169-specific reverse primer were used for amplification of antisense transcripts; a ND169-specific forward primer and an anchor-specific reverse primer were used for amplification of sense transcripts. GAPDH transcripts were amplified in a separate reaction using GAPDHspecific forward-and reverse primers.