Mechanism Of Action

Biochemical and molecular evidence supports a 4-step model for the RNAi pathway: initiation step, effector step, target recognition step, and target cleavage step.

Initiation Step - In the current model, the initiation step is an ATP-dependent cleavage of the silent trigger or dsRNA precursor into 21-25 nucleotides siRNAs (Fig. 1) (32,35). The multidomain enzyme named Dicer in Drosophila, member of the RNAse III family, produces the siRNAs with the features of other RNAse III products, i.e. two single-stranded nucleotides on their 3' ends and a 5' monophosphate (30,35). However, in experiments on mammalian cells, when siRNAs is produced using T7 transcription the duplexes obtained contain 5' triphosphate termini but are nonetheless effective. It is therefore possible that either the 5' triphosphates are efficiently converted into 5' monophosphates or that, in mammalian cells, the requirement for 5' monophosphate ends is not as stringent as it is in Drosophila (36). Dicer contains several domains: an ATP-dependent RNA helicase domain, a Piwi/Argonaute/Zwille (PAZ) domain, two RNAse III domains in tandem, and a dsRNA-binding domain. This enzyme acts as a dimer and each pseudo monomer, due to its two RNAse III domains, is able to cleave dsRNA twice, therefore each Dicer protein can generate one siRNA by introducing 4 cuts into the dsRNA precursor (15). The requirement of Dicer for ATP is surprising in light of the lack of requirement for high-energy cofactors by other RNAse III. Even though no conclusion has been reached yet, it has been suggested that it is related to either the need to unwind RNA to create bulges prior to cleavage or the need to translocate the enzyme along the precursor. Other suggestions include regulating binding of Dicer to dsRNA, or modulating its enzymatic activity. The function of the 130 amino acid PAZ-domain is unknown but is surmised to allow both homo and heterodimerization (37). Proteins carrying this domain have been shown to be crucial in RNAi or PTGS. In worms, flies and humans, some proteins of the Argonaute family link the RNAi machinery with the ribosomal machinery. For example, the translation elongation initiation factor-2c, elF2C also carries both a PAZ-domain as well as a 300 C-terminus amino acid Piwi-domain equally of unknown function (37,38). Spacing differences between the two RNAse III domains in tandem, in the various Dicer homologs has led to speculation that it is a reason for species-specific size differences in siRNAs (35).

Even though it is still unclear if Dicer acts alone, in a larger complex or both, it is has been established that Dicer is also involved with the processing of developmental stRNAs from dsRNA stable hairpin precursors (39). Indeed, deletion of dcr-1, the C.elegans Dicer homolog, not only abrogates RNAi but it also causes misregulation of developmental timing as well as defects in oogenesis (39). A similar deletion in plants leads among others to the disruption of embryo development (10). In worms the severe consequences of DCR-1 deletion, i.e. the heterochronic phenotype, are due to its role in cleaving dsRNA precursors into stRNAs from the genes lin-4 and let-7 which functions ensure the orderly progression of development, respectively from larva stage L1 to L2 and from stage L4 to adult (40).

Small non-translated RNAs increasingly appear to perform a multitude of functions some of which are components of conserved pathways. Only recently, over one hundred potential small regulatory RNAs called microRNAs or miRNAs were identified in fly embryos, worms, and cultured human cells (31). Many developmental switches are still unknown and it is possible that some of the newly discovered miRNAs will also act as stRNAs at various developmental stages as previously seen with lin-4 and let-7 regulation. Nonetheless, given the sheer number of miRNAs identified, alternative roles are possible. These molecules may be involved in spatial development, stress response, cell cycle regulation or in as yet uncovered specialized functions. Some miRNAs could also turn out to be new siRNAs. Indeed in plants newly identified miRNAs appear to regulate genes in a manner similar to siRNAs by targeting open reading frames (ORFs) and silencing genes through mRNA degradation processes (41).

The production of siRNAs by Dicer results in symmetric duplexes while processing of dsRNAs hairpin precursor into stRNAs results in the stabilization of only one branch of the stem (10). What mechanism Dicer uses to determine which part of the precursor needs to be excised is yet another fascinating mystery. In this function a role is likely to be played by accessory factors associated with Dicer activity, a number of which has been identified in genetic screens. Worm mutants lacking Rde-1 cannot initiate RNAi but are otherwise normal, while mutants of the PAZ-domain-containing proteins ALG-1/ALG-2 (Argonaute-like gene) can carry out RNAi but suffer severe developmental defect due to a failure to generate appropriate levels of lin-4 and let-7 stRNAs (27). In Drosophila, two Argonaute homologues (dAgo-1 and dAgo-2) are necessary for RNAi, and in human cells, the Argonaute homologue elF2C co-purifies as a component of the RNAi-induced silencing complex (RISC; see below). Baring similarities to human elF2C, the homologues to Rde-1 in fungi and plants, respectively qde-2 and ago-1, are both necessary for gene silencing in somatic tissues of these organisms (42,43).

After cleavage of the dsRNA precursor by Dicer, the siRNA bound to the enzyme triggers the formation of the RNAi-induced silencing complex or RISC that was purified in Drosophila (13). A human RISC has not yet been characterized with certainty. However, elF2C is associated with an RNA-protein complex, the miRNA ribonucleoprotein particle (miRNP) of similar size as the Drosophila RISC (13). This complex in HeLa cells also contains the proteins Gemin4 and the putative DEAD-box RNA helicase Gemin3. Both proteins are already known to associate with the survival of motor neurons protein (SMN) in a complex that restructures nuclear RNPs. However such complex does not contain elF2C and Dicer. The role of

Gemin3/4 in the RNAi is unclear but the presence of helicase motifs in Gemin 3 raises the possibility of a role in the unwinding of miRNAs and siRNAs (44).

Amplification of Silencing - In order to explain the strong potency of RNAi, a model was suggested involving a number of rounds of physical amplification of the aberrant population of RNA triggers by an RNA-directed RNA polymerase (RdRP) (45). Indeed, initial siRNAs are apparently either not in sufficient amount or do not have the proper structure to trigger efficient RNAi in vivo. Experiments carried out in C.elegans and in Drosophila have shown that the way amplification of the initial population of siRNAs occurs requires hybridization of the target mRNA transcript

Figure 1: Model for RNAi Interference in mammalian celis. Role of Dicer and RISC



Figure 1: Model for RNAi Interference in mammalian celis. Role of Dicer and RISC


Was this article helpful?

0 0

Post a comment