Cytoplasmic vs. Nuclear RNAi in Caenorhabditis elegans

ABSTRACT Experimentally and/or intrinsically delivered dsRNAs can trigger complex set of mechanisms, acting in multiple cell compartments and tissues throughout developmental time. Several methods are available for effective delivery of dsRNA to Caenorhabditis elegans, including ingestion-based feeding and soaking methods, injection of dsRNAs, and transgene-mediated transcription of dsRNA in vivo. Feeding is a more commonly used dsRNA-delivery strategy, and the RNAi mechanisms that respond to ingested dsRNAs are better elaborated than the responses to other dsRNA delivery methods. However, cellular mechanisms that are triggered by experimental delivery of dsRNAs also respond to foreign nucleic acids, such as viral and transposon sequences, and these mechanisms are active in the nucleus as well as the cytoplasm. In ingestion-based delivery methods, the dsRNA molecules enter cells from environmental sources, whereas when transgene-based delivery methods are used, dsRNA molecules are generated in the nucleus. Thus, RNAi mechanisms are activated by dsRNAs that enter the cytoplasm from opposite directions. We are investigating RNAi mechanisms with respect to the compartment in which the dsRNAs are generated and the directionality of their movement into the cytoplasm. Our genetics-based studies indicate that, even though intrinsically derived dsRNAs are produced in the nucleus, they depend on cytoplasmic RNAi machinery. In keeping with this model, NRDE-3, an argonaute protein predicted to shuttle silencing RNAs into the nucleus, had no effect on the RNAi responses from RNAi-proficient transgenes configured as extra-chromosomal arrays. (However, modest effects were observed in nrde-3 mutants when the same dsRNA-expressing sequences were present as single-copy-integrants.) We observed a surprising result that transgenes with negligible RNAi activity in wild-type animals gained RNAi activity in nrde-3 as well as hrde-1 mutants, and from these results, we infer that such dsRNA-expressing transgenes are themselves susceptible to NRDE-3 and HRDE-1 mediated silencing, as has previously been observed for mRNA-encoding genes in endogenous chromosomes and transgenes. However, while transgene-based RNAi is independent of MUT-7, MUT-16 is required for an RNAi phenocopy. A variety of factors can influence the RNAi competency of transgenes, including copy number of dsRNA-expressing cassettes, transgene design, temperature, maternal effects, background mutations and inter-compartmental transport