We are searching data for your request:
Upon completion, a link will appear to access the found materials.
I know that the RISC (RNA-induced silencing complex) is one of the primary complexes involved in gene regulation through RNAi. What I want to know is, what role exactly does Dicer play in this complex?
Dicer is an endo-ribonuclease belonging to the RNAse-III class. Dicer is not a part of the RISC. It however helps in the formation of RISC by cleaving dsRNA or the stem of hairpin RNA on two ends which liberates a small dsRNA product. Then one of the strands is loaded into the RISC.
There are several reviews on this topic and there is a video as well which you can find on youtube.
Frontiers in Molecular Biosciences
The editor and reviewers' affiliations are the latest provided on their Loop research profiles and may not reflect their situation at the time of review.
- Download Article
- Download PDF
- XML (NLM)
- Reference Manager
- Simple TEXT file
Get full journal access for 1 year
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Get time limited or full article access on ReadCube.
All prices are NET prices.
Berkeley Scientists Get First Detailed Look at Dicer
BERKELEY, CA – Scientists have gotten their first detailed look at the molecular structure of an enzyme that Nature has been using for eons to help silence unwanted genetic messages. A team of researchers with Lawrence Berkeley National Laboratory (Berkeley Lab) and the University of California at Berkeley used x-ray crystallography at Berkeley Lab&rsquos Advanced Light Source (ALS) to determine the crystal structure of Dicer, an enzyme that plays a critical role in the process known as RNA interference. The Dicer enzyme is able to snip a double-stranded form of RNA into segments that can attach themselves to genes and block their activity.
Jennifer Doudna is a biochemist who holds joint appointments with Berkeley Lab, UC Berkeley and HHMI. She is a leading authority on RNA molecular structures.
&ldquoWith this crystal structure, we&rsquove learned that Dicer serves as a molecular ruler, with a clamp at one end and a cleaver at the other end a set distance away, that produces RNA fragments of an ideal size for gene-silencing,&rdquo said Jennifer Doudna, a biochemist who led this study. Doudna, a leading authority on RNA molecular structures, holds joint appointments with Berkeley Lab&rsquos Physical Biosciences Division, UC Berkeley&rsquos Department of Molecular and Cell Biology and Department of Chemistry. She&rsquos also an investigator with the Howard Hughes Medical Institute (HHMI).
&ldquoKnowing the structure of Dicer sets the stage for understanding how Dicer enzymes are involved in other phases of the RNA interference pathway,&rdquo Doudna said. &ldquoIn human cells, the evidence points to Dicer being part of a larger molecular complex that directs the RNA interference process. The core structure of Dicer has been highly conserved by evolution and could serve as a guide in redesigning the RNA molecules that direct specific gene-silencing pathways.&rdquo
RNA interference is an ancient gene-silencing process that plays a fundamental role in a number of important functions, including viral defense, chromatin remodeling, genome rearrangement, developmental timing, brain morphogenesis and stem cell maintenance. All of these RNA interference activities depend upon Dicer, so understanding this enzyme&rsquos molecular structure is a critical step.
The results of this research are reported in the January 13, 2006 edition of the journal Science in a paper entitled: Structural Basis for Double-Stranded RNA Processing by Dicer. Co-authoring the paper with Doudna were Ian MacRae, Kaihong Zhou, Fei Li, Adrian Repic, Angela Brooks, Zacheus Cande and Paul Adams.
RNA &mdash ribonucleic acid &mdash has long been known as a multipurpose biological workhorse, responsible for carrying DNA&rsquos genetic messages out from the nucleus of a living cell and using those messages to make specific proteins in a cell&rsquos cytoplasm. In 1998, however, scientists discovered that RNA can also block the synthesis of proteins from some of those genetic messages. This gene-silencing process is called RNA interference and it starts when a double-stranded segment of RNA encounters the enzyme Dicer. Double-stranded RNA (dsRNA) is formed from two single strands of RNA with complementary base sequences.
Dicer cleaves dsRNA into smaller fragments called short interfering RNAs (siRNAs) and microRNAs (miRNAs). Dicer then helps load these siRNA and miRNA fragments into a large multiprotein complex called RISC, for RNA-Induced Silencing Complex. RISC can seek out and capture messenger RNA (mRNA) molecules (the RNA that encodes the message of a gene) with a base sequence complementary to that of its siRNA or miRNA. This serves to either destroy the genetic message carried by the mRNA outright, or else block the subsequent synthesis of a protein.
Until now, it has not been known how Dicer is able to recognize dsRNA and cleave those molecules into products with lengths that are exactly what is needed to silence specific genes. Doudna and her co-authors were able to purify and crystallize a Dicer enzyme from Giardia intestinalis, a one-celled microscopic parasite that can infect the intestines of humans and animals. This Dicer enzyme in Giardia is identical to the core of a Dicer enzyme in higher eukaryotes, including humans, that cleaves dsRNA into lengths of about 25 bases.
A front-on view of a ribbon representation of Dicer shows the enzyme to resemble an axe with the RNA clamp at the handle (the PAZ domain) and the cleaver at the blade (RNase IIIa and IIIb). A flat connector area measuring 65 angstroms is the ruler portion that is used to measure out segments of 25 nucleotides (bases) in length. A segment of double-stranded RNA (blue) is shown passing through the Dicer enzyme.
In their Science paper, Doudna and her colleagues describe a front view of the structure as looking like an axe. On the handle end there is a domain that is known to bind to small RNA products, and on the blade end there is a domain that is able to cleave RNA. Between the clamp and the cleaver is a flat-surfaced region that carries a positive electrical charge. Doudna and her colleagues propose that this flat region binds to the negatively charged dsRNA like biological Velcro, enabling Dicer to measure out and snip specified lengths of siRNA.
&ldquoWhen you put the clamp, the flat area and the cleaver together, you get a pretty good idea as to how Dicer works,&rdquo Doudna said. &ldquoWe&rsquore now using this structural model to design experiments that might tell us what triggers Dicer into action.&rdquo
Different forms of the Dicer enzyme are known to produce different lengths of siRNA. Having identified the flat-surfaced positively charged region in Dicer as the &ldquoruler&rdquo portion of the enzyme, it may be possible to alter the length of a long connector helix within this domain to change the lengths of the resulting siRNA products.
&ldquoOne size does not fit all for Dicer, it makes dsRNA products that range from 21 to 30 base pairs in length or longer. We would like to see what happens when you take a natural Dicer and change the length of its helix,&rdquo Doudna said.
Determining Dicer&rsquos crystal structure was made possible through the unique crystallography capabilities of ALS Beamline 8.2.1 and 8.2.2, Doudna said. Funded through HHMI, beamlines 8.2.1 and 8.2.2 are powered by a superconducting bend magnet, an ideal source of x-rays for protein crystallography experiments. The &ldquosuperbend&rdquo magnet is used to extract x-rays from a relativistic beam of electrons circulating through the ALS storage ring at energies up to two billion electron volts.
X-ray beams at the ALS are typically a hundred million times brighter than those from the best x-ray tubes. When a beam of x-rays is sent through a crystal, the atoms in the crystal cause the x-rays to scatter, creating a diffraction pattern. This diffraction pattern can be translated by computer into 3-D images of the crystal.
The work reported in the Science paper by Doudna and her colleagues was supported by funding from the Department of Energy&rsquos Basic Energy Sciences program and the National Institutes of Health.
Our results indicate that Loqs and Dicer-1 form a complex that converts pre-miRNAs into mature miRNAs so how do they act together in pre-miRNA processing? Sequence comparison reveals that Loqs is a paralog of R2D2 (see Figure 1). Therefore, Loqs may play the molecular role of R2D2 for Dicer-1. R2D2 forms a stable heterodimeric complex with Dicer-2, while either protein alone seems to be unstable in vivo . In the absence of R2D2, Dicer-2 is still capable of efficiently processing long dsRNA into siRNAs. Therefore, the siRNA generating activity of Dicer-2 is not dependent upon R2D2. However, the resultant siRNAs are not effectively channeled into RISC in the absence of R2D2. The Dicer-2–R2D2 complex, but not Dicer-2 alone, binds to siRNA, which indicates that siRNA binding by the heterodimer is important for RISC entry [59,66]. In the case of Loqs, this protein alone is not capable of converting pre-miRNAs into mature miRNAs, but it clearly stimulates and directs the specific pre-miRNA processing activity of Dicer-1. Furthermore, knocking down Loqs markedly reduced the pre-miRNA processing activity in cytoplasmic lysates in vitro (see Figures 4C and 5B), but did not cause a significant reduction of the level of Dicer-1 protein (see Figure 2B) implying that Dicer-1 may largely depend on Loqs for its pre-miRNA processing activity. Thus, the molecular role of Loqs for Dicer-1 is not simply similar to that of R2D2 for Dicer-2.
It can be envisioned that Loqs may have one of several roles in pre-miRNA processing. Dicer-1 contains only one dsRBD, which may not be sufficient for strong interaction with and/or specific recognition of the pre-miRNA substrate (see Figure 6C and 6D). Loqs, containing three dsRBDs with no other identifiable domains being apparent, could provide the additional RNA-binding modules required for specific recognition of the pre-miRNA, and thereby stabilize pre-miRNA binding for Dicer-1. Loqs could also organize binding of Dicer-1 on the pre-miRNA, contributing to the specific positioning of the Dicer-1 cleavage site. Alternatively, since dsRBDs are known to not only bind dsRNAs but also mediate protein–protein interactions , Loqs may directly bind Dicer-1 through its dsRBDs. This protein–protein interaction may trigger a conformational change of Dicer-1 that facilitates either the formation of an intramolecular dimer of its two RNase III domains [50,68], which creates a pair of catalytic sites, or the handover of the Dicer-1 cleaved mature miRNAs to the RISC.
Sequence analysis revealed that protein activator of protein kinase dsRNA dependent (PKR) (PACT)  and HIV TAR RNA binding protein (TRBP)  in mammals bear 34% identity to Loqs, and share a highly similar domain structure with it (Figure 8). Both PACT and TRBP are thought to play a role in the regulation of translation through modulating PKR that also contains two dsRBDs [71–73]. PACT interacts with PKR and enhances the autophosphorylation of PKR , which in turn, phosphorylates the α subunit of eukaryotic translation initiation factor 2 (eIF2α) and leads to an inhibition of mRNA translation in response to viral infection and other stimuli. TRBP prevents PKR-mediated inhibition of protein synthesis through binding to PKR . Considered together, it will be important to find out Loqs' partners other than Dicer-1 for possible involvement of Loqs in miRNA-mediated translational regulation in Drosophila.
The dsRNA-binding motif (dsRBD) is indicated as a black box. These proteins contain three putative dsRBDs. Loqs shares ∼34% amino acid identity with TRBP and PACT and ∼31% identity with Xlrbpa (Xenopus laevis RNA-binding protein A). Sequence comparison between Loqs and its human and Xenopus homologs also showed a higher degree of amino acid conservation in dsRBDs including C-terminal non-canonical dsRBDs. TheXenopus homolog, Xlrbpa, of TRBP/PACT has been found to associate with ribosomes in the cytoplasm , as is the case for many RNAi factors including miRNAs [47,78−82].
How does the enzyme Dicer function in the RISC complex? - Biology
Created by Alexandra Asaro
The protein Dicer is a key processing component in the biogenesis of small interfering RNAs (siRNAs), including microRNAs (miRNAs). Dicer's essential function is to recognize, bind, and cleave longer segments of double stranded RNA (dsRNA) into approximately 25 base pair long dsRNA duplexes. (1) Protein-RNA interactions underlie the RNA hydrolysis function of Dicer. Overall, Dicer functions to convert stem-loop dsRNA into RNA duplex precursors of siRNA and miRNA. The Dicer RNA duplex product features a two nucleotide 3' overhang, characteristic of RNAaseIII cleavage. Once cleaved by Dicer, the 25 bp duplexes are separated and one of the RNA strands is incorporated into the RNA Induced Silencing Complex (RISC). The small RNA strand facilitates RNA interference, negative regulation of gene expression at the RNA level, by guiding the RISC to an appropriate mRNA target and binding to a specific region within the mRNA, thereby inducing translational inhibition and/or degradation of the mRNA. (1) Dicer's structure was initially illucidated in Giardia intestinalis.
Giardia Dicer's functions include recognition and binding of an RNA substrate with stem loop secondary structure and a 3' two nucleotide overhang. Dicer must bind this stem-loop RNA in a specific manner along its various domains so that the RNA is aptly positioned for cleavage. Cleavage of the RNA into a smaller dsRNA duplex involves conserved residues, homodimer formation between catalytic domains, and cation coordination. Studies involving bacterial RNase and more complex Dicer enzymes, such as mouse and human Dicer, can help to elucidate the structure of domains that Giardia Dicer shares with these related enzymes. Giardia Dicer contains a P AZ domain , which functions in specific dsRNA recognition and binding, and t wo ribonuclease III (RNase III) domains which carry out cleavage of the substrate dsRNA. (1) The PAZ and RNase domains are connected by a 65 angstrom long c onnector helix . Its length corresponds to the length of the 25 nucleotide dsRNA product. (1) An N-terminal domain, the p latform domain , which consists of three alpha helices and a beta-pleated sheet, underlies the alpha helical connector segment. The two RNase III domains are connected through a multi-alpha helix domain called the b ridging domain . (1) Overall, Dicer contains alpha helix, beta pleated sheet, turns, and random coil s econdary structure .
The PAZ domain, which includes an alpha helix and beta sheet, recognizes and binds specific dsRNAs with a 2 nucleotide 3' overhang. The recognized 3' overhang is bound by a binding site within the PAZ domain. To accomplish binding, residues in Dicer's PAZ binding domain form hydrogen bonds with the RNA's hydroxyl groups. (1) Studies by Zhang et. al. have confirmed that the PAZ domain of dicer preferentially binds dsRNA substrates with two nucleotide 3' overhangs. (2) The end of the RNA duplex bound at the PAZ domain contains the 3' end of one strand of the duplex, which binds to the PAZ domain's 3 ' overhang binding pocket (highlighted in yellow), and the 5' end of the other strand, which binds to a b asic residue-rich loop region in the PAZ domain. These basic residues may serve to stabilize the binding interaction at the 3' overhang binding pocket below. (1) With one end of the dsRNA substrate bound at the PAZ domain, the other end of the RNA duplex is positioned between the RNase IIIa and RNase IIIb domains, which contain segments of alpha helical structure.
Positioning of the RNA duplex requires conformational flexibility in Dicer. Upon binding the dsRNA, a conformational change in the protein allows for appropriate substrate bending and positioning within the catalytic region. (3) The PAZ domain shifts considerably upon RNA binding. This shift is facilitated by a p roline residue in the connector helix, proline-266, which functions as a hinge and creates a kink in the helix. The RNase domains also undergo a shift. The interface of the platform and RNase domains contains another hinge region that contributes to conformational movement. (3) Giardia Dicer is able to bind a variety of dsRNA molecules, even those with imperfect base pairing. Because mismatched base pairs in dsRNA can induce variable secondary structure loops and bulges, conformational flexibility of Dicer is key for its binding of diverse dsRNA molecules. (3) Another key element in substrate aligning is the p ositioning loop located in the RNase IIIa domain, which helps properly position the RNA duplex at the catalytic site. The loop lies below the catalytic interface of the two RNase subunits where the dsRNA binds. Studies in Dicer positioning loop mutants have shown the positioning loop to be needed to give the desired 25 nucleotide product after cleavage in contrast to the precision seen in normal Dicer, the mutants generated abberantly sized RNA products. (4)
Each of the RNase III domains contain four or five c onserved acidic amino acid residues . Within RNase IIIa, the residues are E407, D340, D404, and E336 and within RNase IIIb, the residues are E684, E723, D653, D720, and E649. (1) These negatively charged aspartate and glutamate residues facilitate the b inding of two metal cations per RNase domain. Side chains of the residues participate in hydrogen bonding and form a pocket in which ionic interactions occur between the metal cations and the negatively charged side chains. The metal cations, usually Mg2+ (although Mn2+ shown here) are catalytically required for the RNase III domains to carry out cleavage of phosphodiester bonds in the two strands of the RNA helix substrate. (1) The two RNase III domains' pairs of metal ions are separated by a distance, 17.5 angstroms, equal to the width of the dsRNA substrate. (1)
Studies in bacterial RNase III have helped elucidate the mechanism of RNase cleavage. The cations are required not only for cleavage but for proper binding of the RNA substrate in the RNase region. (1) The two RNase subunits form a catalytic valley and the RNA must be positioned in this valley for cleavage to occur. The cations prevent charge repulsion between negative residues in the catalytic valley and the negative phosphates lining the dsRNA backbone, thus allowing the substrate RNA duplex to bind stably within the valley. (5) Without the four Mg2+ cations bound, the RNA substrate can bind, but cannot bind in the catalytic valley and subsequently cannot be cleaved. (5) The catalytic valley is created by dimerization of the two RNase III domains homodimer formation in this region is required for Dicer's catalytic activity. (4) Hydrogen bonds mediate binding of the RNA substrate along the different domains of Dicer. Hydrogen bonding occurs between amino acid and nucleotide residues in the RNase catalytic region. (5) The mechanism of scissile bond cleavage is a phosphoryl group transfer reaction. It requires two Mg2+ cations for cleavage within each RNA strand. To cleave one scissile bond, the first Mg2+ cation lowers the pKa of a water molecule, allowing it to lose a proton and become an -OH. This -OH acts as a nucleophile to attack the phosphorous atom bound to the nucleotide which will be cleaved. A pentacovalent intermediate about the phosphorous atom forms upon this nucleophilic attack. The second Mg2+ interacts with the leaving group and facilitates the collapse of this intermediate by neutralizing the oxygen atom's negative charge. The intermediate collapses with the breaking of a phosphorous-oxygen bond to achieve RNA strand cleavage. (5) The mechanism reveals the significance of the Mg2+ cations for catalysis. Two of the cations are required to cleave one strand of the RNA, so to cleave the duplex RNA substrate, four cations in total must be bound.
Giardia Dicer shares structural similarity with the Nuclease Domain of Ribonuclase III from Mycobacterium Tuberculosis. Like Giardia, the Tuberculosis endoribonuclease domain consists of two RNase III domains, which each bind two metal cations through four conserved acidic residues. The distance between these domains fits the span of the dsRNA sugar-phosphate backbone and the cleavage of the dsRNA produces a 3' two nucleotide overhang. (6) All of these features are consistent with the Giardia domains, showing that the RNaseIII endoribonucleases are highly structurally and functionally conserved. Many other RNaseIII enzymes retain these characteristic features.In addition to structural similarity, Giardia Dicer has key sequence similarities with the Ribonuclease III C terminal domain of eukaryotic, bacterial and archeal RNase III proteins, as revealed by a BLAST query. 5 out of 5 residues that make up the RNase III active site, 3 out of 3 residues that compose the RNase III metal binding site, and 15 out of 15 residues that make up the RNase III dimerization interface mapped to Giardia Dicer. These sites allow both R Nase III enzymes and Dicer to bind and cleave dsRNA at the subunit interface catalytic site.
Giardia Dicer also shares structural components with Human Dicer. Human Dicer has the PAZ domain and two RNase III domains. However, the Human Dicer also has a Helicase domain, a dsRBD (double stranded RNA binding) domain, and a DUF283 domain (domain of unknown function), which resembles the Giardia platform domain. (1) Human Dicer binds its dsRNA substrate with higher affinity than does Giardia Dicer, which can be attributed to its additional features and more complex structure. (1) High structural similarity exists between Giardia and Human Dicer's connector alpha helices that span the region between the PAZ and RNase III domains. This alpha helix, in both Human and Giardia, consists of alternating hydrophobic and hydrophilic residues, and contains a kink-creating proline residue which is necessary to bind the dsRNA in such a way as to position one end between the RNase III domains. (1) The RNase III domains of Human Dicer resemble the Giardia RNase III domains. As in Giardia, each domain binds two metal cations and produces a 3' two nucleotide overhang upon substrate cleavage. (7) The RNase portion of Human Dicer is larger and more complex than that of Giardia Dicer, yet their RNase domains show significant structural similarity.
Mouse Dicer contains many of the same structural elements and domains as Giardia Dicer. One Giardia Dicer RNase domain can be s uperimposed on one Mouse Dicer RNase domain. One of the Mouse Dicer's RNase III domains contains an alpha helical region and a loop region not present in Giardia Dicer. Both Mouse and Human Dicers contain this extra helix in their RNase IIIb domains, which may allow for additional interactions with the dsRNA.(8) Study of the Mouse Dicer RNase III domains has identified a lysine residue (K-1790) present in this catalytic region. The lysine residue is a conserved element of bacterial and Giardia Dicers and is involved in the catalysis of phosphodiester bond cleavage.(8) The functional lysine in Mouse Dicer is K-1790. This critical lysine is conserved in Giardia Dicer and appears as K-400.
Human Argonaute protein's PAZ domain is very s imilar in structure and function to the PAZ domain of Giardia Dicer. Argonaute functions downstream of Dicer in the RNAi pathway it serves to bind double stranded siRNA within the RISC complex. (9) In both proteins, the PAZ domains function in dsRNA substrate binding. The RNA binding is dependent upon recognition of a 3' two nucleotide overhang in the duplex. Argonaute, however, lacks in its PAZ domain a Dicer-specific loop rich in basic residues, which functions in 5' end binding and may serve to mediate Dicer's overall RNA binding and release. (1) A Dali Server search revealed significant sequence similarity in the PAZ domains of Argonaute2 and Giardia Dicer. (11)
How RNAi works
Two types of small RNA molecules function in RNAi. The first are synthetic, short interfering RNA (siRNA) molecules that target mRNA cleavage, effectively knocking down the expression of a gene of interest. MicroRNA (miRNA) molecules, on the other hand, are naturally occurring single-stranded RNAs 19–22 nucleotides long, which regulate gene expression by binding to the 3’ untranslated regions (UTRs) of target mRNAs and inhibiting their translation (Ambros, 2004).
There are several ways to induce RNAi: synthetic molecules, RNAi vectors, and in vitro dicing (Figure 1, below). In mammalian cells, short pieces of dsRNA—short interfering RNA— initiate the specific degradation of a targeted cellular mRNA. In this process, the antisense strand of siRNA becomes part of a multiprotein complex, or RNA-induced silencing complex (RISC), which then identifies the corresponding mRNA and cleaves it at a specific site. Next, this cleaved message is targeted for degradation, which ultimately results in the loss of protein expression.
Figure 1: Methods of RNAi knockdown in mammalian cells.
Both RNA polymerase II and III transcribe miRNA-containing genes, generating long primary transcripts (pri-miRNAs) that are processed by the RNase III–type enzyme Drosha, yielding hairpin structures 70 to 90 bp in length (pre-miRNAs). Pre-miRNA hairpins are exported to the cytoplasm, where they are further processed by the RNase III protein Dicer into short double-stranded miRNA duplexes 19 to 22 nucleotides long. The miRNA duplex is recognized by the RNA-induced silencing complex (RISC), a multipleprotein nuclease complex, and one of the two strands, the guide strand, assists this protein complex in recognizing its cognate mRNA transcript. The RISC-miRNA complex often interacts with the 3’ UTR of target mRNAs at regions exhibiting imperfect sequence homology, inhibiting protein synthesis by a mechanism that has yet to be fully elucidated (Figure 2, below).
Plant miRNAs can bind to sequences on target mRNAs by exact or near-exact complementary base pairing and thereby direct cleavage and destruction of the mRNA (Rhoades et al., 2002 Chen, 2005). Similar to the mechanism employed in RNA interference (RNAi), the cleavage of a single phosphodiester bond on the target mRNA occurs between bases 10 and 11 (Elbashir et al., 2001). In contrast, nearly all animal miRNAs studied so far do not exhibit perfect complementarity to their mRNA targets, and seem to inhibit protein synthesis while retaining the stability of the mRNA target (Ambros, 2004). It has been suggested that transcripts may be regulated by multiple miRNAs, and an individual miRNA may target numerous transcripts. Research suggests that as many as one-third of human genes may be regulated by miRNAs (Lim et al., 2003). Although hundreds of miRNAs have been discovered in a variety of organisms, little is known about their cellular function. Several unique physical attributes of miRNAs, including their small size, lack of polyadenylated tails, and tendency to bind their mRNA targets with imperfect sequence homology, have made them elusive and challenging to study.
Figure 2: Biogenesis and function of miRNA. MicroRNA transcripts, generated by RNA polymerases II and III, are processed by the RNase III enzymes Drosha (nuclear) and Dicer (cytoplasmic), yielding 19–22 nucleotide miRNA duplexes. One of the two strands of the duplex is incorporated into the RISC complex, which regulates protein expression.
A simplified model for the RNAi pathway
A simplified model for the RNAi pathway is based on two steps, each involving ribonuclease enzyme. In the first step, the trigger RNA (either dsRNA or miRNA primary transcript) is processed into an short, interfering RNA (siRNA) by the RNase II enzymes Dicer and Drosha. In the second step, siRNAs are loaded into the effector complex RNA-induced silencing complex (RISC). The siRNA is unwound during RISC assembly and the single-stranded RNA hybridizes with mRNA target. Gene silencing is a result of nucleolytic degradation of the targeted mRNA by the RNase H enzyme Argonaute (Slicer). If the siRNA/mRNA duplex contains mismatches the mRNA is not cleaved. Rather, gene silencing is a result of translational inhibition.
Access to Document
Research output : Contribution to journal › Article › peer-review
T1 - Functional analysis of dicer-2 missense mutations in the siRNA pathway of Drosophila
N1 - Funding Information: We thank Qinghua Liu for the gift of antibodies. This work was supported by the Korea Research Foundation Grant funded by the Korean Government (MOEHRD, KRF-2006-331-C00213) and a Korea University Grant.
N2 - The Drosophila RNase III enzyme Dicer-2 processes double-stranded RNA (dsRNA) precursors into small interfering RNAs (siRNAs). It also interacts with the siRNA product and R2D2 protein to facilitate the assembly of an RNA-induced silencing complex (RISC) that mediates RNA interference. Here, we characterized six independent missense mutations in the dicer-2 gene. Four mutations (P8S, L188F, R269W, and P365L) in the DExH helicase domain reduced dsRNA processing activity. Two mutations were located within an RNase III domain. P1496L caused a loss of dsRNA processing activity comparable to a null dicer-2 mutation. A1453T strongly reduced both dsRNA processing and RISC activity, and decreased the levels of Dicer-2 and R2D2 proteins, suggesting that this mutation destabilizes Dicer-2. We also found that the carboxyl-terminal region of R2D2 is essential for Dicer-2 binding. These results provide further insight into the structure-function relationship of Dicer, which plays a critical role in the siRNA pathway.
AB - The Drosophila RNase III enzyme Dicer-2 processes double-stranded RNA (dsRNA) precursors into small interfering RNAs (siRNAs). It also interacts with the siRNA product and R2D2 protein to facilitate the assembly of an RNA-induced silencing complex (RISC) that mediates RNA interference. Here, we characterized six independent missense mutations in the dicer-2 gene. Four mutations (P8S, L188F, R269W, and P365L) in the DExH helicase domain reduced dsRNA processing activity. Two mutations were located within an RNase III domain. P1496L caused a loss of dsRNA processing activity comparable to a null dicer-2 mutation. A1453T strongly reduced both dsRNA processing and RISC activity, and decreased the levels of Dicer-2 and R2D2 proteins, suggesting that this mutation destabilizes Dicer-2. We also found that the carboxyl-terminal region of R2D2 is essential for Dicer-2 binding. These results provide further insight into the structure-function relationship of Dicer, which plays a critical role in the siRNA pathway.
Since the first discovery of the passenger-strand cleavage mechanism a decade ago, it has been widely recognized that slicer-dependent unwinding is a prerequisite for the assembly of highly complementary siRNAs into RISCs ( 10, 11, 13). Our collective results indicate that this assumption is typically valid, but not always, particularly for mammals and birds. Earlier biochemical studies relied mainly on target cleavage assays to understand the overall nature of RISC catalysis, which often tended to dismiss important differences during the early stages of RISC assembly. Through a careful re-examination of RISC assembly using a variety of several biochemical analyses, including duplex loading, slicer-dependent and -independent unwinding, and classical target cleavage assays, we established here that slicer-independent unwinding is a more prevalent mechanism for human RISC maturation than previously thought, not only for miRNA duplexes but also for highly complementary siRNAs as well. Aside from the main findings, there are several other findings that are important to discuss in detail.
Small RNA sorting and the Mg 2+ level in slicer-dependent unwinding
It has been previously recognized that only AGO2 can efficiently unwind siRNA duplexes via passenger-strand cleavage in humans ( 21), and this is reasonable in the sense that AGO2 is among the most advantageous human AGO proteins that can utilize its slicer-activity ( 42, 43) for RISC maturation (Figure 7E). Nonetheless, we showed that cleavage-deficient AGOs (1, 3, and 4) can also be programmed with siRNAs at the physiologically relevant temperature of humans, suggesting that slicer-independent unwinding is likely a common feature of human AGO proteins. These observations provided a natural explanation for why both miRNAs and siRNAs are found in all four human AGO proteins, irrespective of their sequences ( 44, 45). In contrast with weak small RNA sorting in humans, siRNA duplexes are specifically sorted into fly AGO2 ( 46, 47), which is essential for antiviral defense ( 48). The fly AGO2 should therefore acquire an additional strategy for siRNA maturation (i.e., passenger-strand cleavage), otherwise not efficient at their body temperature.
We showed that a certain level of Mg 2+ is required for slicer-dependent unwinding to occur efficiently (Figure 2A). While the free cytosolic Mg 2+ level is less than 1 mM under normal conditions ( 49), the total cellular Mg 2+ concentration can vary from 5 to 20 mM ( 50, 51), as most Mg 2+ ions are bound to proteins and negatively charged molecules ( 50, 51). Therefore, it is difficult to estimate the exact amount of Mg 2+ bound to AGO2, although previous studies typically used 1.5–5 mM Mg 2+ for the in vitro slicing assays ( 10, 11, 42, 43, 52). In addition, cytosolic Mg 2+ levels can be altered by ATP, which is capable of chelating Mg 2+ ions ( 53). In other words, a transient decrease in ATP levels may give rise to a sudden increase in the Mg 2+ concentration ( 53). During two steps of RISC assembly, duplex loading requires ATP hydrolysis, and subsequent cleavage of the passenger-strand requires a relatively high level of Mg 2+ . It is tempting to speculate that bursts of metabolic activity during pre-RISC formation may induce a transient decrease in ATP that results in an increase in Mg 2+ levels that allows for more efficient slicer-activity, although it is technically difficult to demonstrate such rapid and transient changes.
Slicer-independent unwinding in human RISCs
Although chaperone machinery-mediated duplex loading is relatively well understood ( 3–6), it is still remain elusive how the loaded duplex unwinds to form the mature RISC. The helicase model was proposed at the beginning of the RNAi field — an ATP-dependent helicase separates the two strands of the duplex before they are loaded into the AGO protein ( 30), although such an ‘unwindase’ has yet to be identified ( 2). Another model assumes that duplexes are loaded into the AGO protein, which itself can dissociate the two strands ( 2, 18, 19, 22). Our findings, combined with those of previous studies, point to the latter model as a more plausible scenario. It has been extensively demonstrated that AGO proteins receive duplexes, rather than single strands, during RISC assembly ( 4, 5, 10–13, 19, 21). Once a duplex is deeply buried within the AGO protein, it is difficult to comprehend how the duplex is further transferred from AGO proteins to other regulatory factors in an accessible form. We showed that the unwinding process was significantly influenced by intrinsic factors, such as temperature and Mg 2+ , both of which were closely related to duplex stability. Our functional analyses also indicated that the AGO protein itself is a key factor that is needed for duplex unwinding. Recent structural studies have corroborated the idea that target dissociation is strongly coupled to a profound conformational change in the AGO proteins, which results in a widening of the N-PAZ channel, thereby leading to a disruption of the base-pairing in the 3′ half of the guide ( 54, 55). We postulate that high temperature may favor a conformational change ( 56, 57) in the AGO protein that accelerates RISC maturation.
miRNAs are the most abundant and common endogenous substrates for human AGOs and they often have multiple mismatches and G-U wobble base pairs (or even bulges) that enable highly efficient unwinding (Figure 7E and Supplementary Figure S6E). In contrast, a highly complementary siRNA duplex requires either a certain temperature or AGO2-mediated cleavage of passenger-strands, which results in a drastic change in the thermodynamic profile of the duplex. siRNA unwinding by AGO2 depends upon two main factors with opposite effects: temperature and Mg 2+ (Figure 7E). At the physiological temperature of humans, siRNAs could also be incorporated into slicer-deficient AGOs (1, 3 and 4) to form the mature RISC (Figure 7E). In mammals, endogenous siRNAs can repress complementary mRNAs and transposons in mouse oocytes ( 58) and embryonic stem cells ( 59), although little is known about their biological role.
A thermodynamic perspective of RISC maturation
In light of our current findings, as well as those of previous reports, we envision that the overall process of RISC assembly has characteristics closely resembling those of a classical thermodynamic reaction (Supplementary Figure S7). Duplex loading requires the energy from ATP hydrolysis (i.e. Ea) to trigger a conformational change in the AGO proteins that is mediated by the Hsc70–Hsp90 chaperone machinery ( 3–6). In this regards, the pre-RISC can be considered to be a transition state during RISC assembly (Supplementary Figure S7). The mature RISC is thought to be the most stable and energetically favorable state (ΔH < 0) (Supplementary Figure S7). This assumption is supported by the fact that AGO proteins mostly co-purify with endogenous guide RNAs and were stable enough to be structurally resolved by crystallography ( 33, 34, 54, 60).
Because duplex unwinding is generally accompanied by an increase in entropy ( 61, 62) (ΔS > 0), the spontaneity (ΔG = ΔH − TΔS) of RISC maturation is then largely influenced by the temperature that contributes to the entropy of the system. Although this may seem to be plausible, we cannot warrant further speculation with our present biochemical data alone. Future studies combining structural analyses with biophysical techniques are necessary to reveal the structural basis for duplex unwinding, as well as its thermodynamics.
Homeotherms, such as mammals and birds, have a specific physiological adaptation that enables them to regulate their body temperature from 36 to 42°C ( 63). In contrast, poikilotherms, including worms, flies, plants, and fish, lack the means to generate heat ( 64). Therefore, the body temperatures of these animals tend to conform to their external environment, and temperature fluctuations may affect numerous aspects of their physiology, including enzyme function, muscle activity, and energy metabolism ( 64). Our results suggested that the last step of small RNA biogenesis could be largely influenced by such temperature changes, which may provide a means to fine-tune the expression of small RNAs in various living organisms.