Biochemical characterization of human Dicer and its interacting partner TRBP

Over the recent years, RNA interference (RNAi) has emerged as a powerful \nmethod to study the role of individual genes. However, the mechanism \nunderlying the gene silencing by the double-stranded RNA (dsRNA) is still not \nfully understood. RNAi is initiated when dsRNA is processed by RNase III \nendonuclease Dicer into short interfering RNAs (siRNAs), of 21 to 22 nucleotides \nin length. SiRNAs are then incorporated into RNA-induced Silencing Complex \n(RISC) that by base-pairing targets messenger RNA for degradation. Dicer is \nalso involved in processing of precursors of the small regulatory RNA species, \nmicroRNAs (miRNAs). MiRNAs are encoded in the genome and are implicated in \ngene expression regulation in various cellular processes. After maturation by \nDicer, miRNAs are incorporated into RISC-like complexes that in animals \nimperfectly base-pair with the target mRNA and lead to inhibition of translation. \nThis thesis focuses on Dicer, the central protein involved in both RNAi and \nmiRNA pathways. \nDetailed study of the ribonuclease activity of human Dicer and its \nancestral prototype, bacterial RNase III, are described in the first experimental \nchapter of this thesis. The common model for dsRNA cleavage by the RNase IIIclass \nenzymes is proposed. The use of mutagenesis to investigate the catalysis \nrevealed that Dicer and bacterial RNase III contain a single compound catalytic \ncenter. Both RNase III domains of Dicer contribute to the dsRNA cleavage \nreaction. The results obtained in this study have proved the then-accepted model \nof RNase III catalysis to be inadequate. We demonstrated that instead of the two \n \ncatalytic centers as proposed in the old model, both E. coli RNase III and Dicer \ncontain one compound catalytic center that generates products with 2-nt 3’ \noverhangs. In silico modeling of the dsRNA substrate into 3D crystal structure \ncoordinates of the bacterial RNase III offered additional support to our \ninterpretation. Together with other data, a new model was proposed according to \nwhich Dicer functions as an intramolecular pseudodimer of its two RNase III \ndomains, assisted by the flanking RNA binding domains, PAZ and dsRBD. \nSecond chapter describes dsRNA binding domain (dsRBD)-containing \nprotein, TRBP, that was found to associate with Dicer in mammalian cells and in \nvitro. We show that TRBP is required for optimal RNA silencing mediated by \nsiRNAs and endogenous miRNAs, and that it is involved in efficient processing of \npre-miRNAs. Since TRBP had previously been described as the inhibitor of the \ninterferon-induced double-stranded RNA-regulated protein kinase PKR, the \nTRBP-Dicer interaction raises a possibility of the connection between RNAi and \ninterferon-PKR pathways. \nDsRNA binding properties of human Dicer dsRBD are described in the \nthird chapter. We have found that this domain has the propensity to bind dsRNAs \nof different lengths. Surprisingly, it displays hardly detectable affinity for siRNAs. \nThis observation suggests that the dsRBD might be involved in substrate binding \nduring Dicer cleavage reaction and take part in substrate/product discrimination \npreventing the enzyme from sequestering its own product. \nThe two supplementary chapters contain the work performed in \ncollaboration with other laboratories. We show that like its Drosophila counterpart, \n \nhuman Dicer is able to form complexes with siRNAs both in vitro and in vivo. \nThese results indicate that also in mammals Dicer could function downstream of \nthe dsRNA cleavage step and could take part in RISC assembly. The other \nsupplementary chapter describes RNAi connection with chromatoid bodies \nduring spermatogenesis. We show that Dicer and components of the RISC-like \ncomplex (Ago and miRNA) are concentrated in chromatoid body. We also \ndemonstrate that Dicer directly interacts with the RNA helicase MVH (mouse \nVasa homolog) that is the germ-line specific chromatoid body component. Our \nfindings suggest that the chromatoid body might function as a subcellular \nconcentration site for the miRNA pathway components during spermatogenesis.