Small RNAs have long been known to be involved in specifying and stabilizing distinct chromatin states. Only recently has it become evident that small RNAs have the capacity to trigger epigenetic gene silencing.Once initiated by primary small RNAs, the repressed states can be propagated across multiple generations in different model organisms, commonly through the amplification of secondary small RNAs in stable feedback loops, while the primary small RNAs become dispensable.Small RNAs have also been demonstrated to direct the deposition of phenotypically plastic epigenetic marks, which mediate repression under specific conditions only.These observations lend support to the hypothesis that small RNAs might be involved in sensing environmental conditions and triggering epigenetic gene expression changes that may lead to increased population fitness of an organism that lives in a dynamic habitat.

Summary Small RNAs trigger the formation of epialleles that are silenced across generations. Consequently, RNA-directed epimutagenesis is associated with persistent gene repression. Here, we demonstrate that small interfering RNA-induced epimutations in fission yeast are still inherited even when the silenced gene is reactivated, and descendants can reinstate the silencing phenotype that only occurred in their ancestors. This process is mediated by the deposition of a phenotypically neutral molecular mark composed of tri-methylated histone H3 lysine 9 (H3K9me3). Its stable propagation is coupled to RNAi and requires maximal binding affinity of the Clr4/Suvar39 chromodomain to H3K9me3. In wild-type cells, this mark has no visible impact on transcription but causes gene silencing if RNA polymerase-associated factor 1 complex (Paf1C) activity is impaired. In sum, our results reveal a distinct form of epigenetic memory in which cells acquire heritable, transcriptionally active epialleles that confer gene silencing upon modulation of Paf1C.

Adenine auxotrophy is a commonly used non-selective genetic marker in yeast research. It allows investigators to easily visualize and quantify various genetic and epigenetic events by simply reading out colony color. However, manual counting of large numbers of colonies is extremely time-consuming, difficult to reproduce and possibly inaccurate. Using cutting-edge neural networks, we have developed a fully automated pipeline for colony segmentation and classification, which speeds up white/red colony quantification 100-fold over manual counting by an experienced researcher. Our approach uses readily available training data and can be smoothly integrated into existing protocols, vastly speeding up screening assays and increasing the statistical power of experiments that employ adenine auxotrophy.