Biochemistry and modeling of human Dicer, a key protein involved in RNA interference

Five years ago, an unexpected discovery opened a whole new paradigm of biology – RNA interference (RNAi). From the simple notion that dsRNA, when introduced into various organisms, can specificly inhibit expression of homologous endogenous gene, the RNA interference has expanded into a wide range of gene regulatory pathways of great biological importance. At the same time, RNAi started to be widely used as powerful experimental tool for probing gene function in almost any organism. The research on RNAi is moving forward at high speed, both at the mechanistic level and as a tool. Genetic and biochemical studies in various systems have revealed much information about the mechanism of RNAi. It is now well established that dsRNAs is processed by a nuclease Dicer into short dsRNAs varing in length from 21 to 25 nt, named siRNAs, which in turn are incorporated into the RNA induced silencing complex (RISC) to target mRNA degradation. Identification of siRNAs led to the discovery of a whole new class of regulatory small RNAs of similar size, named microRNAs (miRNAs), which have diverse biological functions. Hundreds of miRNAs were cloned, and their functions are being investigated. The single stranded miRNAs are also processed by Dicer from miRNA precursors and incorporated into a complex similar, if not identical, to RISC. In animals, miRNAs imperfectly base-pair with mRNA leading to translational repression. Dicer, a central protein of the RNAi and miRNA pathways is a focus of the study presented in this thesis. A full length human Dicer cDNA was cloned and protein overexpressed in the baculovirus system and purified. Its processing activity was demonstrated using both dsRNA and pre-miRNAs as substrates. Detailed study of the RNase III-like activity of Dicer, its biochemical properties and a model of its function are described in two experimental chapters of this thesis. This thesis is divided into three major chapters followed by a short general discussion. Chaptercontains a general introduction to RNA interference. It describes a history of RNAi discovery, summarizes what is known about the RNAi mechanism in general, and also about the species-specific differences. The mechanistic aspects of the miRNA pathway are also described. An overview of all important proteins involved in RNAi is presented. Finally, a summary of RNAi as a tool for reverse genetics is provided. Chapterdescribes the characterization of the purified recombinant human Dicer protein. In vitro experiments showed that the purified protein cleaves dsRNAs into ~22 nucleotide siRNAs. This was a first direct evidence that Dicer indeed has RNase III-like nuclease activity. Accumulation of processing intermediates of discrete sizes, and experiments performed with substrates
containing modified ends, indicated that Dicer preferentially cleaves dsRNAs at their termini.
Binding of the enzyme to the substrate could be uncoupled from the cleavage step by omitting
Mg2+ or performing the reaction at 4oC. Activity of the recombinant Dicer, and of the endogenous
protein present in mammalian cells extracts, was stimulated by limited proteolysis, and the
proteolysed enzyme became active at 4oC. Cleavage of dsRNA by purified Dicer and the
endogenous enzyme was ATP independent, in contrast to results obtained in Drosophila and C.
elegans. Additional experiments suggested that if ATP participates in the Dicer reaction in
mammalian cells, it might be involved in the product release needed for the multiple turnover of the
enzyme.
Chapter 3 describes the mutagenesis study of the human Dicer RNase III domains, which
revealed that Dicer contains a single compound catalytic center. Both RNase III domains in Dicer
contribute to the dsRNA cleavage reaction. The Dicer mutagenesis study was initiated whether a
model of dsRNA cleavage originating from an X-ray structural study of the Aquifex aeolicus RNase
III also applies to Dicer. Mutants containing changes in residues implicated in the catalysis in both
Dicer RNase III domains were prepared to study their effect on RNA processing. Our results were
in conflict with the bacterial Rnase III model and all speculated Dicer model. We have further
mutated the catalytic residues of the E. coli RNase III and tested their effect on processing of
different substrates. The results are consistent with those obtained with Dicer mutants. More
specifically, our results indicate that instead the two catalytic centers proposed previously, both
enzymes contain only one catalytic center, generating products with 2-nt 3’ overhangs. Together
with other data, a new model was proposed according to which Dicer functions as an intramolecular
dimer of its two RNase III domains, assisted by the flanking RNA binding domains, PAZ
and dsRBD.