RNAi silences genes by targeting mRNAs for degradation. However, a second mode by which RNAi effects gene silencing has emerged: by triggering chromatin modifications. Gu et al have analysed the pattern of a specific chromatin modification in response to exogenous double stranded RNA (dsRNA) in C. elegans and show that RNAi triggered chromatin modification is target gene specific and transgenerationally heritable.
The ability of exogenous dsRNAs to silence homologous target genes (RNA interference, RNAi) was discovered in the nematode worm, C. elegans, approximately fifteen years ago. Feeding worms bacteria expressing dsRNA, or bathing worms in dsRNA, has the ability to specifically block gene function and most surprisingly this effect in C. elegans is inherited for some generations. RNAi has become an incredibly useful technique in biology as it works to a greater or lesser extent throughout eukarya, and offers a simple and fast method for compromising gene action specifically. Uncovering the mechanisms by which RNAi works has also opened up huge new vistas on cellular function: namely the proliferation of newly identified classes of endogenous RNA molecules and the discovery of their crucial roles in cellular regulation.
In C. elegans the mechanism of RNAi can be divided into two phases: Firstly, dsRNA is cut into 20-30nt molecules (primary short interfering (si) RNAs) by the enzyme Dicer. These siRNAs complexed with Argonaute proteins recognise and target mRNAs for degradation. The second phase (only present in some organisms) is the de novo synthesis of secondary siRNAs by the primary siRNA/Argonaute mediated recruitment of RNA directed RNA polymerases that use the target mRNA as a template. Apart from cytoplasmic siRNA mediated mRNA degradation, RNAi also functions in the nucleus by siRNA/Argonaute complex interactions with nascent mRNAs and RNA polymerase, and by causing chromatin silencing by histone modifications.
Gu et al have used ChIP-seq (Chromatin immunoprecipitation followed by high throughput sequencing) to make genome wide assessments of the effects of RNAi on the extent of a specific histone modification associated with transcriptionally silenced chromatin (histone 3 lysine 9 trimethylation, H3K9me3). RNAi against the gene lin-15B caused an enrichment of H3K9me3 at the lin-15B locus that spread as far as 9kb from the trigger region (the sequence directly targeted by dsRNA) meaning that two neighbouring genes also showed higher H3K9me3 modifications. No other genomic locations showed H3K9me3 enrichment, meaning that RNAi effects on this chromatin modification are specific to the target gene and neighbouring loci. The same pattern was seen when this experiment was repeated with RNAi against three other genes.
RNAi using dsRNA directed at target sequences that are not transcribed into mRNA was unable to affect H3K9me3 levels, showing that interactions with target mRNAs are essential for RNAi triggered chromatin modification. Likewise, RNAi on worm strains defective for various argonaute proteins that are necessary for secondary siRNA biogenesis, failed to trigger H3K9me3 chromatin modifications.
RNAi mediated gene silencing can last for multiple generations in C. elegans, however it is unclear whether these heritable silencing effects are mediated by inherited siRNAs, by a chromatin based mechanism or by a combination of the two. Gu et al profiled H3K9me3 and small RNAs through three generations after dsRNA exposure. In the first generation of offspring, H3K9me3 enrichment occurred at a similar level as in the parental worms, although the level of siRNAs had fallen off drastically. The chromatin response was still present in the second generation but had fallen away to background levels by the third generation. These results suggest that H3K9me3 chromatin modifications induced by RNAi are transgenerationally inherited without a requirement for inherited siRNAs or trigger RNA, but are not conclusive.
These results are consistent with an emerging model in which secondary siRNA/ argonaute complexes, transported to the nucleus, direct chromatin silencing by interacting with nascent RNAs or with cognate DNA sequences. Histone modification is then propagated some distance from the trigger sequences. The finding with regard to heritability, are inconclusive and seem to be potentially at odds with a previously discussed paper regarding heritable antiviral response in C. elegans. No doubt we’ll be revisiting this subject matter soon.
Gu, S., Pak, J., Guang, S., Maniar, J., Kennedy, S., & Fire, A. (2012). Amplification of siRNA in Caenorhabditis elegans generates a transgenerational sequence-targeted histone H3 lysine 9 methylation footprint Nature Genetics DOI: 10.1038/ng.1039