Project 7. R-loop function in nuclear RNA interference.

Small RNA-guided gene regulatory pathways represent a widely conserved mode of control of gene expression and genome stability. In essence, these pathways use small RNA molecules to guide proteins of the Argonaute family to specific targets, which in eukaryotic organisms is RNA. Upon target recognition, the Argonaute protein triggers a cascade of effects that lead to repression of gene activity, such as destabilization of the RNA and/or repression of translation. Even though many of the small RNA-guided mechanisms act in the cytoplasm, some have been convincingly shown to act in the nucleus. Such cases have been demonstrated in fungi, plants, insects, nematodes and mammals. The downstream effects of nuclear Argonautes involve changes of chromatin at loci with homology to the small RNAs. In C. elegans, one major small RNA pathway is mediated by the Argonaute protein HRDE-1. HRDE-1 triggers H3K9 trimethylation at targeted loci, and loss of this activity results in a gradual decline in germ cell function over the course of generations, and a loss of heritability of RNA interference (RNAi). Previous work from our group has shown HRDE-1 uses inherited small RNAs from the parents to trigger gene silencing in the zygote. Transcription of the targeted loci is required for nuclear small RNA activity. In fact, it is the nascent transcript, and not the DNA itself, that is recognized by nuclear Argonaute proteins. In S. pombe, it has been elegantly shown that the release of the nascent transcript has to be delayed, or inhibited in order for a locus to be a target for nuclear RNAi. Clearly, this constellation opens up the possibility that R-loops play a role in these processes (Figure 1).

Figure 1. Schematic depicting potential relationship between R-loops and nuclear RNAi. An R-loop could ‘hold’ the nascent transcript, making it a better, but could also shield it for recognition by small RNAs, inhibiting nuclear RNAi. Figure created with BioRender.

Indeed, the R-loop resolving protein Aquarius (AQR, or in C. elegans EMB-4) has been shown to be involved in nuclear RNAi, but the exact molecular role of AQR has remained unclear. This project will focus on the role of AQR in nuclear RNAi, with a special focus on R-loops. C. elegans is an excellent animal model for these studies, having a well-characterized nuclear RNAi pathway in its germ cells, a relatively small genome, fast genetics, and well-established genome editing tools. R-loop-specific questions that we aim to resolve are:

  1. Are R-loops enriched at known HRDE-1 targeted loci?
  2. Does experimentally induced nuclear RNAi induce R-loops?
  3. Do R-loops enhance or repress HRDE-1 activity?
  4. What is the precise molecular function of AQR at HRDE-1 targeted loci?

This last question is likely to start being addressed by the second PhD student. We will start by testing the S. pombe model, in which release of nascent transcripts suppresses nuclear RNAi. This is clearly of direct relevance to the proposed R-loop studies, but has not been tested rigorously in animal systems, yet. To do this, we will test if ribozyme-mediated release of the nascent transcript reduces HRDE-1’s ability to induce silencing. In this context, we will also probe the role of PAF1C in nuclear RNAi. PAFC is a protein complex involved in nascent transcript release, and has been shown to affect nuclear RNAi in S. pombe. Next, we will generate conditional alleles of hrde-1 and emb-4/aqr in order to allow timed depletion of these proteins in germ cells. Using these tools, we will then map R-loops in wild-type and hrde-1 and emb-4 mutant conditions. In these experiments we will also make use of available transgenes that are, or are not silenced by hrde-1, allowing careful comparison of R-loop formation at a specific locus, in the absence or presence of nuclear RNAi. We will aim to address strand specificity of the R-loops by also probing for ssDNA breaks, which can be expected to be elevated in the displaced ssDNA of the R-loop. In order to address the effect of R-loops on nuclear RNAi, we will overexpress the C. elegans RNaseH1 enzyme, RNH-1.0, in order to globally reduce R-loop levels. We also aim to refine this approach by engineering strains in which RNH-1.0 is tethered to specific loci via a DNA binding domain, for instance using the lacO/lacI system, to affect R-loop stability only at specific loci.