|Home | About | Journals | Submit | Contact Us | Français|
RNA interference (RNAi) is a form of posttranscriptional gene silencing mediated by microRNA (miRNA) and small interfering RNA (siRNA). In Drosophila melanogaster, the RNase III enzymes Dicer-1 and Dicer-2 generate miRNA and siRNA, respectively. We describe the methods for the expression, purification, and analysis of recombinant Dicer-1 and Dicer-2 enzymes. Our studies demonstrate that Dicer-1 and Dicer-2 display different substrate specificities and ATP requirements.
RNA interference (RNAi) is a form of posttranscriptional gene silencing mediated by 21- to 25-nucleotide microRNA (miRNA) and small interfering RNA (siRNA). These tiny regulatory RNAs control many important biological processes, such as development, antiviral defense, heterochromatin formation, and maintenance of genomic stability (1–4). It follows that misregulation of miRNAs has been linked to human diseases, such as cancer (5,6). Moreover, siRNAs or miRNAs have been widely used as a powerful gene-silencing tool for functional genomic analyses in multiple model organisms, including humans (1,3).
In principle, miRNAs and siRNAs can be viewed as two parallel branches of the RNAi pathway, which consists of initiation and effector steps. In the initiation step, siRNAs are derived from mostly exogenous long dsRNAs, whereas miRNAs originate from endogenous (~ 60–70 nt) stem-loop precursor miRNAs (pre-miRNAs). Both siRNAs and miRNAs are produced as 21- to 25-nt duplexes with 2-nt 3′-overhangs and 5′-phosphate and 3′-hydroxyl termini. While siRNAs are perfectly complementary, miRNAs frequently contain mismatches, bulges, or G:U wobble base pairs. In the effector step, nascent siRNA and miRNA duplexes are assembled into the respective RNA-induced silencing complexes that are simply referred to as siRISC and miRISC. In RISCs, single-stranded siRNA or miRNA functions as the guide RNA to direct sequence-specific cleavage and/or translational repression of complementary mRNA (1,2,7,8).
Both siRNAs and miRNAs are produced by Dicers, a family of large (~ 200 kDa) multidomain RNaseIII enzymes that exist in most eukaryotic organisms (9). A typical Dicer contains a putative RNA helicase domain, a DUF (domain of unknown function) 283 domain, and a PAZ domain at the amino (N)-terminus as well as two RNase III domains and a dsRNA-binding domain at the carboxyl (C)-terminus. The tandem RNase III domains of Dicer form an intramolecular dimer to make a pair of cuts on dsRNA, creating a characteristic 2-nt 3′-overhang (10,11). In most organisms, such as C. elegans and humans, a single Dicer generates both siRNAs and miRNAs. However, two Dicer enzymes, Dicer-1 and Dicer-2, have been identified in the Drosophila genome (10,12,13). Both genetic and biochemical studies have implicated that Dicer-1 and Dicer-2 are involved in the biogenesis of miRNAs and siRNAs, respectively (14–18).
Dicer enzymes do not function alone. We have previously purified the siRNA-generating enzyme from Drosophila S2 cells and found that it consisted of Dicer-2 and R2D2 (15). R2D2 contains two dsRNA-binding domains (R2) and forms a heterodimeric complex with Dicer-2 (D2). Although R2D2 does not regulate siRNA production, R2D2 and Dicer-2 coordinately bind siRNA and facilitate its incorporation into the effector siRISC complex (15,19). An R2D2-like protein, Loquacious (Loqs, also known as R3D1), has recently been identified as a cofactor for Dicer-1 in the miRNA pathway. The loqs gene encodes at least two alternatively spliced proteins, Loqs-L (long) and Loqs-S (short). Both Loqs isoforms contain three putative dsRNA-binding domains. Loqs-L forms a stable complex with Dicer-1 and greatly enhances its miRNA-generating activity. It remains uncertain if the Dicer-1/Loqs complex facilitates miRNA loading onto the miRISC complex (16–18).
Therefore, the Dicer-1/Loqs-L and Dicer-2/R2D2 complexes function in parallel to generate miRNAs and siRNAs in Drosophila cells. The functional differences of the two Dicer complexes can be explained by either that Dicer-1 and Dicer-2 possess distinct biochemical activities or that Loqs-L and R2D2 help define the functional specificities for Dicer-1 and Dicer-2 (16). Here we describe the detailed protocols for the expression, purification, and analysis of recombinant Dicer-1 and Dicer-2 enzymes. Our reconstitution studies indicate that Dicer-1 and Dicer-2 are biochemically distinct enzymes with different substrate specificities and ATP requirements.
Both recombinant Dicer-1 and Dicer-2 enzymes are expressed in insect cells using the BAC-to-BAC® Baculovirus Expression System (15,16). Thus, for simplicity, we will refer to both Dicer-1 and Dicer-2 as “Dicer” when not specified in the sections below. The cDNAs of Dicer-1 and Dicer-2 are cloned by reverse transcription from total RNA of Drosophila S2 cells using the RLM-RACE kit from Ambion (Cat# 1700). To simplify purification of recombinant Dicers, polyhistidine (His)-tags were added by polymerase chain reaction (PCR) to the N- and C-termini of Dicers. We found that the addition of double His-tags greatly enhanced the expression levels of Dicer viruses. After verified by sequencing, the double His-tagged Dicer cDNAs were subcloned into the pFastBac™ 1 vector. The pFastBac-Dicer plasmids were transformed into E. coli DH10Bac competent cells to generate recombinant bacmids. The recombinant Dicer bacmids were isolated from positive bacteria transformants and verified by PCR that they contained the full-length Dicer cDNA (Fig. 1). The protocols for this section are described in detail in the manufacturer's instruction manual and will not be repeated here.
Both Dicer-1 and Dicer-2 recombinant proteins are highly purified by nickel affinity purification followed by SP- and Q-Sepharose chromatography (Fig. 3). Nickel affinity purification is based on the double His-tags of recombinant Dicers, which is the same for Dicer-1 and Dicer-2. However, the column behaviors of Dicer-1 and Dicer-2 are different for the SP- and Q-Sepharose chromatography. On the SP-Sepharose column, Dicer-1 flows through, whereas Dicer-2 binds and is eluted at 180–220 mM of sodium chloride. On the Q-Sepharose column, both Dicer-1 and Dicer-2 bind but are respectively eluted at 350–380 mM and 260–300 mM sodium chloride. All purification steps are performed at 4 °C on an ACTA FPLC machine purchased from Amersham. After the multistep purification, we can typically get ~ 0.6–1.0-mg recombinant Dicer proteins with more than 90% purity from 1.2 L of insect cell culture (Fig. 3).
Genetic studies have suggested that Dicer-1 and Dicer-2 are respectively involved in miRNA and siRNA production (14). By biochemical fractionation, it has been shown that the Dicer-1/Loqs-L complex processes pre-miRNA into mature miRNA, whereas the Dicer-2/R2D2 complex cleaves long dsRNA into siRNA (15–18). The functional specificities of the two Dicer complexes can be explained by the fact that Dicer-1 and Dicer-2 possess different biochemical activities or that Loqs-L and R2D2 help define the functional specificities for their respective Dicer partners. Here we compare the abilities of purified recombinant Dicer-1 and Dicer-2 enzymes to process radiolabeled pre-miRNA or long dsRNA in the absence or presence of ATP (see Notes 4.3). Our studies indicate that, despite sharing extensive sequence homology, Drosophila Dicer-1 and Dicer-2 enzymes display different substrate specificities and ATP requirements. While Dicer-1 prefers pre-miRNA as its ideal substrate, Dicer-2 is a much better enzyme for processing dsRNA (Fig. 4). Furthermore, Dicer-1 produces miRNA or siRNA independent of ATP, whereas Dicer-2 absolutely requires ATP hydrolysis for efficient siRNA production (Fig. 5).
Make 90 μL of the master mixture for 10 reactions with the following: 10 μL of 10X pre-miRNA processing buffer, 10 μL of 10 mM ATP, 5 μL of SUPERase•In™, 10 μL of pre-miRNA probe (4 × 104 cpm/μL), and 55 μL of nuclease-free water. For the rest of the procedure, follow steps 2–5 as above.
We thank Dr. Zain Paroo for critical reading of the manuscript and Feng Jiang for technical assistance. Q. L. is a W. A. “Tex” Moncrief Jr. Scholar in Medical Research and a Damon Runyon Scholar supported by the Damon Runyon Cancer Research Foundation (DRS-43). The work is also supported by a Welch grant (I-1608).