Inhibition of the Expression of the Human RNase P Protein Subunits Rpp21, Rpp25, Rpp29 by External Guide Sequences (EGSs) and siRNA

doi:10.1016/j.jmb.2004.06.006

J. Mol. Biol. (2004) 342, 1077–1083

COMMUNICATION

Inhibition of the Expression of the Human RNase P Protein Subunits Rpp21, Rpp25, Rpp29 by External Guide Sequences (EGSs) and siRNA Haifeng Zhang and Sidney Altman Department of Molecular Cellular and Developmental Biology, Yale University, New Haven, CT 06520, USA

External guide sequences (EGSs) and siRNAs were targeted individually to the mRNA of three of the protein subunits of human RNase P, Rpp21, Rpp25 and Rpp29.The production of each of the three targets was inhibited in every specific case. In addition, some of the remaining protein subunits were also inhibited by these specific EGSs and the siRNAs. These data, in general, confirm previous results on the inhibition of a subgroup of all the protein subunits with an EGS against Rpp38.The effect of EGSs is apparent in 24 hours after transfection but the effect of siRNAs, which is comparable to the EGS data in amounts of inhibition, takes at least 48 to 96 hours to become evident. No general understanding of the mechanism of action of the siRNAs, in terms of which portion of a target mRNA they bind to for function, was apparent from the design of those used here. q 2004 Elsevier Ltd. All rights reserved.

Keywords: HeLa cells; transfection; RNase P protein subunits; Rpp38

Several methods that use RNA have been employed to suppress gene expression in human cells. These include RNA interference (RNAi),1 and external guide sequences (EGSs).2 EGSs target complementary RNA for specific cleavage by RNase P. EGS oligonucleotides require an accessible single-stranded region on their target RNA and create a tRNA-like structure (called 3/4 EGS) in mammalian cells2 that is cleaved by the endogenous RNase P. This method has been successfully used to inhibit genes in bacteria, animal viruses and human HeLa cells in tissue culture.3 – 7 RNAi has been demonstrated in plants, nematodes, Drosophila, protozoa, and mammalian cells.8 – 11 An RNase III related enzyme, designated Dicer, digests a double-strand RNA into fragments of 21 –23 nt long12 and these small fragments, known as small interfering RNAs (siRNAs),1,11 are incorporated into an RNA-induced silencing complex (RISC),10 which generates the guide RNA

sequence that recognizes and facilitates cleavage of the target RNA. Human nuclear RNase P has at least ten protein subunits and one RNA subunit and is localized in nuclei.13 – 15 Previous work has shown that the inhibition of Rpp38 by an EGS against Rpp38 also decreases the amount of at least four other RNase P protein subunits, namely hpop5, Rpp21, Rpp25 and Rpp29, whereas the rest of the subunits were unchanged.5 We now report that the EGS method was used to inhibit specifically the RNase P subunits Rpp21, Rpp25 and Rpp29. The results are similar to those obtained with an EGS against Rpp38. siRNAs against Rpp21, Rpp25 and Rpp29 were also used to inhibit the same subunits in HeLa cells. EGS and siRNA technologies are similar in the amounts of inhibition they can produce, although the time taken for these effects are different. Both methods have comparable effectiveness. Design and testing EGSs and siRNAs

Present address: Haifeng Zhang, Department of Pathology, Yale University School of Medicine, New Haven, CT 06510, USA. Abbreviations used: EGSs, external guide sequences; RNAi, RNA interference; siRNAs, small interfering RNAs; RISC, RNA-induced silencing complex.

Computer RNA folding programs16 were used to assist the design of EGSs against Rpp21, Rpp25 and Rpp29. Sites were chosen in looped or partially looped regions to provide accessible regions in the target RNA. The targeting sites of one such

0022-2836/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.

1078

EGS and siRNA Inhibition of Human RNase P Subunits

Figure 1. Targeting sites of EGSs and siRNAs in RNase P protein subunits. (a) The cDNA coding sequences is shown for Rpp25. The siRNA sequences are in bold letters with their names above the sequences (21 nt) and the EGS sequences are underlined with their names under the sequences (13 nt). The first seven of the 13 nt of the EGS sequences complement with their target mRNA to form a tRNA-like acceptor stem structure. The following 2 nt are unpaired and the last 4 nt complement the target mRNA forming the tRNA D stem-like structure. (b) The proposed secondary structure of a complex of Rpp25 mRNA (the underlined target sequence) and its corresponding EGS, 25E263. An arrow indicates the site of cleavage by RNase P. Similar diagrams for Rpp21 and Rpp29 are available from the authors.

sequence, 25E263, are shown in Figure 1A, which also includes data for siRNAs and another EGS site. The underlined nucleotides of Rpp25 mRNA formed a presumed complex with 25E263 (Figure 1B). Standard methods were used to design the EGSs2 and siRNA.17 Experimental details are available from the authors. The internally labeled radioactive transcripts of Rpp21, Rpp25, Rpp29 (about 400 nt near the 50 end of the gene for Rpp21 and Rpp29; 722 nt for Rpp25) were mixed with partially purified human RNase P and corresponding EGSs, and the stopped reactions were loaded onto a polyacrylamide/urea gel.2 25E263 guides RNase P to cleave Rpp25 message to yield appropriately sized 50 and 30 cleavage products. RNase P cleavage of mRNA increases with increasing EGS dose, with 1 : 50 (mRNA to EGS) yielding the optimal cleavage (data not shown). Similar results were demonstrated with the fragments of Rpp21 and Rpp29 transcripts (data not shown). Vectors pmU6 or pEGFP were used to harbor EGSs and for transient transfections. Vector pBS/U6 was used for the generation in vivo of stem – loop siRNA. The coding sequences for the first set of siRNAs overlap with the corresponding EGS (see Figure 1) and all commence with GG as described.18 The second set of siRNAs were designed roughly following the current selection guidelines for siRNAs.17 The expression of siRNA in these plas-

mids is driven by pmU6 promoters and terminate at four or five T bases.19 The conditions for transfections were optimized with HeLa cells. Western blots of transfected cells Cells were harvested, 24 or 96 hours after transfection for EGSs and siRNA, respectively. In 24 well plates, 0.8 mg of DNA per well was used for EGS transfection and 0.2 mg of DNA per well for siRNA transfection. At 96 hours, the cell numbers were decreased by 25% in 25I358 transfected cells but not for the other siRNAs. There were no phenotype changes that could be observed visually. Lysates from transfected cells were separated by SDS-PAGE and transferred onto membranes for immunoblotting. The results for EGSs (24 hours after transfection) and siRNA (96 hours after transfection) are shown in Table 1. All the EGSs and siRNAs succeeded in inhibiting the expression of the specific RNAs they were designed to target (Table 1). In addition, several EGSs also inhibited specific other protein subunits as has been reported.5 We note, however, that Rpp40 was not inhibited at all, as had also been reported.5 Not every RNase P subunit was assayed in our experiments. Certainly, the EGS 21E193 was the most efficient inhibitor of several genes besides Rpp21. The other EGSs also inhibited two or three protein subunits of those tested, including the one they

1079

EGS and siRNA Inhibition of Human RNase P Subunits

Table 1. Relative amounts of protein subunits of HeLa RNase P after transient transfection with EGSs or siRNAs EGS/control or siRNA/control Gene product Actin Rpp21 Rpp25 Rpp29 Rpp30 Rpp38 Rpp40

21E193

25E263

29E210

21I182

21I259

25I263

25I358

29I198

29I547

1 0.2 0.17 0.3 0.44 0.42 1.4

1 0.8 0.52 1.2 0.91 0.68 0.93

1 1.1 0.82 0.68 0.65 0.55 0.83

1 0.39 0.92 0.89 0.8 0.62 0.9

1 0.5 0.86 0.98 1.1 0.54 1.2

1 0.54 0.7 0.81 0.6 0.69 1.1

1 0.89 0.04 1.23 0.9 0.8 1.5

1 0.21 0.38 0.39 0.7 0.04 1

1 0.63 1.18 0.57 0.7 0.65 1.3

Western assays were done 24 hours after transfection of HeLa cells for EGSs and 96 hours after transfection for siRNAs. The results are quantified after normalization for actin. The controls for 21E193 were pmU6 transfected samples and the controls for 25E263 and 29E210 were pGmU6 transfected samples. Characters in bold type indicate significant experimental values of inhibition. The controls for siRNAs were pGL2C transfected samples. Ordinary characters are at the limit of sensitivity for detection of inhibition. The numbers are averages of at least two experiments. Methods for Western blots are given by Kovrigina et al.5

targeted. Rpp38 was most affected among the protein subunits that were not actually targeted. We note that the various siRNAs also had a variety of effects. For example, 29I198 was successful in inhibiting Rpp29, its target, and the other proteins tested, aside from Rpp40. The rest of the siRNAs had lesser effects, with 25I358 only inhibiting its own target at a high efficiency. The inhibitory rates for all targets varied between . 90% to about 20% (inhibition numbers lower than 20% were not regarded as significant). Several of siRNAs overlapped with looped regions against which EGSs were designed (21I182, 25I263, 29I198), but some did not (see Figure 2). An examination of these observations did not reveal any global understanding of how the siRNAs functioned with respect to accessibility to looped regions, sequence placement, etc. siRNAs that are targeted to looped regions do inhibit their proposed protein targeted subunits, but so do those that are targeted to basepaired structures. We also tested an EGS (25E358;

Figure 1) made against one of the siRNA sequences without a preliminary assay of its ability to find a sequence susceptible to RNase T1 cleavage: this EGS did not work at all. Long double-stranded RNAi To test whether long double-stranded RNAi can effectively inhibit the expression of the subunits of RNase P, an Rpp25 sequence (326 bp) starting from the first translated codon was cloned between opposing pmU6 promoters. There are four T bases in the middle of the antisense strand (as well as five T bases at either end of the sequences, thereby allowing pol III promoter transcription termination). The sense transcripts, therefore, contain 326 nt and the antisense transcripts contain 101 nt of Rpp25 coding region (Figure 3). The plasmids were transfected into HeLa cells and the quantification of Western blots after 96 hours transfection are shown in Table 2. The transfection of this

Figure 2. Computer-folded secondary structures of a target RNA. The length of sequences for folding of Rpp25 is approximately 320 nt: 150 nt precedes the siRNA. The siRNA sequences start from nucleotide 151 and are 21 nt in length. Structures of 25I263, and 25I358 and the corresponding EGSs are shown in the Figure. Similar diagrams for Rpp21 and Rpp29 are available from the authors.

1080

25U6U6 construct, which harbors the 326 bp between the opposing mU6 promoters, successfully inhibited the expression of Rpp25 but not that of Rpp30, Rpp38 and Rpp40. This inhibition of Rpp25 by 25U6U6 is quite similar to that of 25I358, in that both constructs inhibit Rpp25 by 80% but much less, or not at all, for other subunits of RNase P tested here. siRNA or RNAi from long double-stranded RNA requires, in our hands, 96 hours to achieve significant inhibition of their targets. Other results The deliberate inhibition of one of the subunits of RNase P, Rpp38, by an EGS also results in other subunits reduced in protein and mRNA levels (e.g. Rpp21, 25 and Rpp29 and hpop5) in transient transfection5 and in stable mammalian cell lines that express an EGS (E. Kovrigina, unpublished results: the inducible U6 promoter is stable for at least three months in these EGS expressing cells). EGSs directed at three other mRNAs of subunits of RNase P subunits were used in this study. The inhibition of their corresponding targets also concurrently inhibited several other subunits of RNase P indirectly, similar to those reported with EGSRpp38.5 The data we show in Table 1 may be due to the natural half-lives of each protein species if all proteins are degraded at some characteristic rate after RNase P is partly disabled. We presume, from our previous experiments with EGSRpp38, that the mRNAs were also degraded for those proteins of which total amounts were not inhibited. The different subunits might also have distinct susceptibility to a degradative protease that may affect their life-time in vivo. Another similar experiment, in which an EGS against Rpp40 (one of the genes not inhibited by the four other EGSs used in our previous experiments) was employed, failed to give the nature of the specific inhibition we report here (E. Pfund, unpublished results). siRNAs gave similar levels of inhibition to the EGSs we used (Table 1). Aside from a reported longer-lasting effect of siRNAs,1 both methods were quite comparable in their inhibition levels (plus or minus 20% of the same reading) and the time needed for their design. There have been no observations of non-specific inhibition of EGS

EGS and siRNA Inhibition of Human RNase P Subunits

Table 2. Relative amounts of protein subunits of HeLa RNase P after transient transfection with long doublestranded RNA Gene product 25U6U6/control

Control

Rpp25

Rpp30

Rpp38

Rpp40

1

0.14

0.8

0.91

0.9

The data were obtained at 96 hours after transient transfection of HeLa cells. The results are quantified after normalization to actin. The controls were pGL2C transfected samples. Bold characters indicate significant results. The numbers are averages for two experiments.

sequences, aside from an indirect effect on lamin A, another nucleolar protein that behaves similarly to indirect results on non-targeted RNase P subunits. However, non-specific effects of other siRNAs due to direct hydrogen bonding of the siRNA and the non-specific target have been noted.20 – 23 The exact method of inhibition to be used, EGS or siRNA, will depend to some extent on the temporal distribution of effects on genes since, in other aspects, these technologies are nearly identical. We also tried double EGS constructs with two individual EGSs against two of the susceptible gene products (e.g. Rpp21 and Rpp25) to increase the effectiveness of our inhibition assay. This combination did work effectively with flu virus RNAs in cells stably expressing the EGSs.4 However, in our transfection assay, the combined inhibitory effect of the double EGSs was marginally greater (20%) than the single EGS experiments in one combination (E. Kovrigina, unpublished results). Double constructs with two of the same EGSs against a single susceptible gene also did not work effectively. In transient transfection, we have found that the expression of more than one EGS generally depletes the overall effect of the EGSs. We did not check an M1 RNA-EGS construct, which gives better inhibition in eukaryotic cells than does our tRNA-like EGSs.24 The use of an EGS acts quickly and specifically to change the phenotype of certain cellular characteristics. For example, EGS against Rpp38 resulted in the reduction of the Rpp38 protein level within four hours after transfection of HeLa cells.5 On the contrary, hairpin siRNAs encoded in the pBS/U6 vector used in this study acts on mature mRNA after processing in the Dicer and

Figure 3. A representation of the long double-stranded RNA generated by 25U6U6 plasmid in vivo. The opposing mU6 promoters were constructed in Bluescript vector. In all, 326 nt of Rpp25 coding sequence starting from ATG was inserted between PstI and ApaI with five T bases at the end of the transcripts. There is a stretch of four T bases in the midst of the antisense sequence. The transcript of the sense strand contains 326 nt of Rpp25 sequence and short flanking sequences. The transcription of the antisense strand stops at four T bases, and the resulting transcript is 101 nt of Rpp25 sequence.

1081

EGS and siRNA Inhibition of Human RNase P Subunits

RISC complexes1,11 and shows no effect at least until 48 hours after transfection. The temporal studies were repeated several times. The predicted secondary structure of a target mRNA16 using about 320 nt centered with the siRNA targeting site, is shown in Figure 2. The siRNA sequence started from nucleotide 151 and consists of 21 nt. The siRNAs proceed across loop regions (25I263; Figure 2). For example, the other three siRNAs have similar positions in their mRNAs with 21I259 starting from an open loop and 25I358 and 29I547 move across a loop region. Most of the sequences in the other siRNAs are in paired stem regions. Long double-stranded (ds)RNA has been reported to elicit interferon responses and nonspecific shut down of protein translation.25 We did not examine these possibilities. Here, we generated a long sense strand containing 326 nt of Rpp25 and an antisense strand containing 101 nt of Rpp25. The sequence of 25I358 is located inside this 101 nt region. The vector we used contained opposing pmU6 promoters to yield long dsRNA. A similar construct has recently been reported.26 Opposing mU6 promoters are also clearly useful for gene silencing in HeLa cells. Several of the previous experiments with EGSs had specific value as demonstrations of the utility of this technology.3,4,6,7 We inhibited the expression of human RNase P protein subunits as the means of providing information regarding the regulation of these genes.5 These new results demonstrate the utility of the EGS technology for basic research purposes. Construction of EGSs against Rpp21, Rpp25 and Rpp29 A DNA fragment that contains the Rpp25 cDNA adjacent to a T7 promoter was obtained by PCR using the full-length Rpp25 cDNA clone as a template (ATCC 1053991). Primers used were: T7Rpp25, TAATACGACTCACTATAGGCGTCTGC GCTG; Rpp25RIR, AAGCTTGAATTCAGGCCGT CTGATCC. The PCR fragment was subcloned into pUC19 digested with SmaI. The 3/4 EGSs, designed using theoretical substrate structures drawn with the Mfold program,16 were obtained by PCR using a plasmid encoding the gene for precursor to E. coli tRNATyr (pTyr) as a template using the following pairs of primers: Rpp21E193-50 , GCTGCAGTGGGAGCAGACTCTA AATCTG; Rpp21E193-30 , AAGCTTTAAAAATGGT GTGTCCTGAAGGATTCGAACCTTC; Rpp25E26350 , GCTGCAGGCCAAGCAGACTCTAAATCTG; Rpp25E263-30 , AAGCTTTAAAAATGGTGGAACCT GAAGGATTCGAACCTTC; Rpp29E210-50 , GCTG CAGGGTGAGCAGACTCTAAATCTG; Rpp29E21030 , AAGCTTTAAAAATGGTGGAGTACGAAGGA TTCGAACCTTC. The PCR products were cloned into pUC19 and the plasmids were digested with PstI and HindIII, and subcloned under T7 promoters for transcription in vitro, and then into pmU6

(a gift from R. Reddy, Baylor College of Medicine, Waco, TX) for experiments in tissue culture. The EGSs were designated as 21E193, 25E263 and 29E210, which directed RNase P against Rpp21, Rpp25 and Rpp29, respectively. A double EGS construct was manufactured as follows: 3/4 EGS DNA plasmids (21E193, 25E263 and 29E210) were digested with BamHI. This was followed by Klenow DNA polymerase treatment and the product was then digested with SacI. To get the insert fragment, the same 3/4 EGS DNA plasmids were digested with HincII and SacI. The vector fragments and the insert fragments were purified by gel electrophoresis and ligated together to generate a plasmid that contains two copies of a particular EGS, each with its own pmU6 promoter. These double EGS constructs were designated 21E193/193, 25E263/263 and 29E210/210. Similar double EGS construct were also made with 21E193/25E263, 21E193/29E210 and 25E263/ 29E210 (E. Kovrigina, personal communication). Transfection efficiency To assess transfection efficiency, the pEGFP-C1 vector was modified to contain the mU6 promoter and the downstream EGS sequence as follows: pEGFP-C1 was digested with BglII and BamHI, and ligated together to yield pEGFP-C1/DBglII/ BamHI. This plasmid was further digested with BclI and HincII, and was gel-purified to yield the pEGFP-C1/DBglII/BamHI//BclI/ HincII large fragment. pmU6 and the three EGSs in pmU6 vectors were digested with KpnI (blunted) and BamHI to obtain the fragments of mU6 promoter or mU6 promoter plus the corresponding EGSs. These fragments were ligated to pEGFP-C1/ DBglII/BamHI//BclI/HincII to obtain the pEGFP constructs containing the mU6 promoter or mU6 promoter plus the corresponding downstream EGSs sequences designated as mU6, 21E193, 25E263 and 29E210, respectively. Construction of siRNAs against Rpp21, Rpp25 and Rpp29 To generate the plasmids for stem –loop siRNAs (6 nt hairpin with a 21 bp stem),18 for example, the plasmid for siRNA against luciferase, the following primers are annealed together and inserted between ApaI (blunted) and EcoRI sites of the pBSU6 vector in which a mouse U6 promoter has been inserted into Bluescript vector (gift of Yang Shi, Harvard Medical School, Boston, MA) as a negative control using the following pair of primers: GL2CF, ACGTACGCGGAATACTTCGAACTC GAGTTCGAAGTATTCCGCGTACGTCTTTTTG; GL2CR, AATTCAAAAAGACGTACGCGGAATAC TTCGAACTCGAGTT CGAAGTATTCCGCGTACGT. The siRNA directed against luciferase is designated as GL2C.17 The plasmids for siRNAs against RNase P protein subunits were constructed in a similar fashion

1082

and their sense strands are listed below. In these plasmids, 6 bp followed the sense strand (CTCGAG) and that was followed by the antisense strand of the sense strand so that the complete product folded back on itself to form the 21 bp doublestranded molecule: 21I182-sense, GGCCGCCCAT TGTGTCCTTGC; 21I259-sense, TTGCGAAGCGG CTCGTCTTGC; 25I263-sense, GGAACCTGATGG CCTTCGCCA; 25I358-sense, CGTGCGCCGAGAT CCTCAAGC; 29I198-sense, GGCGGTGGTCCTGG AGT ACTT; 29I547-sense, GAAGACCGCCTGAAA GTTATC. The corresponding siRNAs generated were designated as 21I182, 21I259 against Rpp21; 25I263, 25I358 against Rpp25; and 29I198, 29I547 against Rpp29. Sequences of the plasmids were confirmed by sequencing from both directions. The sequences 21I182, 25I263 and 29I198 are located in the same region of mRNAs as corresponding EGS targeting sites, namely 21E193, 25E263 and 29E210. The remaining three are targeted at other regions and are designed according to siRNA guidelines,17 i.e. selection of the target sequence from the open reading frame of a given cDNA sequence, preferably 50– 100 nt downstream of the start codon; the 21 nt target sequence follow the dinucleotides AA or NA; the GC content of the sequence is preferably around 50% and less than 70%; BLAST search the selected siRNA sequence against EST libraries or mRNA sequences of the respective organism to ensure that only a single gene is targeted. Double pmU6 promoters in the opposing direction were used for expression of a long doublestranded RNA. pmU6 vector was digested with SacI and EcoRI, and the fragment that contains mouse U6 promoter was inserted into the digested pBS/U6 vector and resulted in a pBS/U6U6 vector with double mU6 promoters in the opposing directions. Approximately 320 bp starting from and downstream of ATG of Rpp25 were cloned into this vector between Pst I and Apa I sites. This fragment was amplified from an Rpp25 cDNA clone from ATCC (clone 1053991). The primers used are as follows: P25U6U6F, AACTGCAGAAAAATGG AGAACTTCCGT AAGG; P25u6u6R, TACGGGCC CAAAAAGGCTCTGCCACACCTCG. The long, double-stranded construct was designated as 25U6U6. The double-stranded RNA contains a 326 nt sense strand and a 101 nt antisense strand of Rpp25 as a consequence of the four T bases at the end of the antisense strand (Figure 3). Transcription in vitro Rpp25 mRNA under the control of the T7 promoter was transcribed from pUC19 harboring these genes after digestion of the plasmids with EcoRI. The 3/4 EGS was prepared from the appropriate plasmid which had been linearized by DraI digestion. The transcription reactions were carried out as described,27 and transcripts were purified by using Sephadex G-50 Quick Spin columns (Roche Applied Science).

EGS and siRNA Inhibition of Human RNase P Subunits

Assay for cleavage by RNase P Assays were performed in 10 ml with 50 mM Tris– HCl (pH 7.5), 10 mM MgCl2, 100 mM NH4Cl with 2000 cpm of internally radiolabeled Rpp25 mRNA and EGS RNA in the presence of the DEAE fraction of partly purified human RNase P (2 ml) as described.27 Reactions were incubated for one hour at 37 8C. Samples were electrophoresed in a 5% (w/v) polyacrylamide/7 M urea gel. Human RNase P was partially purified through the DEAE Sepharose step from HeLa cells according to established protocols,28 and was a gift from Dr Yong Li (Yale University). Cell culture and transfection HeLa cells were maintained in DMEM supplemented with 10% (v/v) FBS, 100 units/ml of penicillin, and 100 mg/ml of streptomycin.17 Cells at 90% confluence were used for EGSs transfection in 24 well plates with 0.8 mg of plasmid DNA per well and using the Lipofectamine 2000 (Invitrogen) according to the manufacture’s protocol. Cells at 50% confluence were used for siRNA transfection in 24 well plates with 0.2 mg of plasmid DNA per well. Northern blots The RNAs from HeLa cells were obtained using Trizol reagents (Gibco) following the manufacturer’s instruction. For Northern analysis, 10 mg of total RNA was loaded onto a 2% (w/v) agarose gel. Northern hybridizations were performed as described.3 Oligonucleotides complementary to 25E263 (TGGTGGAACCTGAAGGAT), 21E193 (TGGTGTGTCCTGAAGGAT) and 29E210 (TGGT GGAGTACGAAGGAT) were used to detect EGSs expression. All oligonucleotides were end-labeled with T4 polynucleotide kinase (New England Biolabs) and [g-32P]ATP (Amersham Pharmacia). Western blots Hela cells were detached from the bottom surface of wells with 2 £ trypsin –EDTA (0.1% (w/v) trypsin, 1 mM EDTA), and inactivated with complete DMEM. The pellets were washed once with PBS and boiled in gel loading buffer. These samples were separated in an SDS/12.5% polyacrylamide gel, transferred to a nitrocellulose membrane (Schleicher & Schuell), and incubated with antibodies against RNase P subunits as described.13 To calculate Western results, the blots were illuminated by an Enhanced Chemiluminescent kit (ECL; Amersham Biosciences). The films were scanned and quantified by Image Gauge. The amounts of the subunits of RNase P in the corresponding lanes as well as the control lanes were normalized to the amount of actin, which was also detected after stripping the same membrane. The relative amounts of the subunits of RNase P in the

1083

EGS and siRNA Inhibition of Human RNase P Subunits

presence of EGS or siRNA were obtained after dividing the normalized values (see above) by the controls for either pmU6/pGmU6 for EGSs, or the siRNA directed against luciferase.

12.

13.

Acknowledgements We thank Drs Cecilia Guerrier-Takada, Elizaveta Kovrigina, Yong Li, Taijiao Jiang, Jeffrey Mckinney and Donna Wesolowski, and other members of our laboratory for discussions and other support. We thank Dr Rong Ju for discussions. We thank Drs E. Kovrigina and E. Pfund for performing some experiments. This research was supported by a grant from the National Institutes of Health (USA), GM19422, to S.A.

14. 15. 16. 17.

18.

References 1. Elbashir, S. M., Harborth, J., Lendeckel, W., Yalcin, A., Weber, K. & Tuschl, T. (2001). Duplexes of 21nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature, 411, 494– 498. 2. Guerrier-Takada, C. & Altman, S. (2000). Inactivation of gene expression using ribonuclease P and external guide sequences. Methods Enzymol. 313, 442– 456. 3. Guerrier-Takada, C., Li, Y. & Altman, S. (1995). Artificial regulation of gene expression in Escherichia coli by RNase P. Proc. Natl Acad. Sci. USA, 92, 11115 – 11119. 4. Plehn-Dujovich, D. & Altman, S. (1998). Effective inhibition of influenza virus production in cultured cells by external guide sequences and ribonuclease P. Proc. Natl Acad. Sci. USA, 95, 7327– 7332. 5. Kovrigina, E., Wesolowski, D. & Altman, S. (2003). Coordinate inhibition of expression of several genes for protein subunits of human nuclear RNase P. Proc. Natl Acad. Sci. USA, 100, 1598 –1602. 6. Kraus, G., Geffin, R., Spruill, G., Young, A. K., Seivright, R. & Cardona, D. (2002). Cross-clade inhibition of HIV-1 replication and cytopathology using sequence-specific RNase P molecules. Proc. Natl Acad. Sci. USA, 99, 3406– 3411. 7. Kim, K., Umamoto, S., Trang, P., Hai, R. & Liu, F. (2004). Intracellular expression of engineered RNase P ribozymes effectively blocks gene expression and replication of human cytomegalovirus. RNA, 10, 438–447. 8. Fire, A., Xu, S., Montgomery, M. K., Kostas, S. A., Driver, S. E. & Mello, C. C. (1998). Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature, 391, 806– 811. 9. Fire, A. (1999). RNA-triggered gene silencing. Trends Genet. 15, 358– 363. 10. Hammond, S. M., Caudy, A. A. & Hannon, G. J. (2001). Post-transcriptional gene silencing by double-stranded RNA. Nature Rev. Genet. 2, 110 – 119. 11. Elbashir, S. M., Lendeckel, W. & Tuschl, T. (2001).

19.

20.

21. 22.

23. 24.

25. 26.

27. 28.

RNA interference is mediated by 21- and 22-nucleotide RNAs. Genes Dev. 15, 188– 200. Bernstein, E., Caudy, A. A., Hammond, S. M. & Hannon, G. J. (2001). Role for a bidentate ribonuclease in the initiation step of RNA interference. Nature, 409, 363– 366. Eder, P. S., Kekuda, R., Stolc, V. & Altman, S. (1997). Characterization of two scleroderma autoimmune antigens that copurify with human ribonuclease P. Proc. Natl Acad. Sci. USA, 94, 1101– 1106. Jarrous, N., Eder, P. S., Wesolowski, D. & Altman, S. (1999). Rpp14 and Rpp29, two protein subunits of human ribonuclease P. RNA, 5, 153– 157. Jarrous, N. & Altman, S. (2001). Human ribonuclease P. Methods Enzymol. 342, 93 – 100. Zuker, M. (2003). Mfold web server for nucleic acid folding and hybridization prediction. Nucl. Acids Res. 31, 3406– 3415. Elbashir, S. M., Harborth, J., Weber, K. & Tuschl, T. (2002). Analysis of gene function in somatic mammalian cells using small interfering RNAs. Methods, 26, 199 –213. Sui, G., Soohoo, C., Affar, el. B., Gay, F., Shi, Y., Forrester, W. C. & Shi, Y. (2002). A DNA vector-based RNAi technology to suppress gene expression in mammalian cells. Proc. Natl Acad. Sci. USA, 99, 5515–5520. Das, G., Henning, D., Wright, D. & Reddy, R. (1988). Upstream regulatory elements are necessary and sufficient for transcription of a U6 RNA gene by RNA polymerase III. EMBO J. 7, 503– 512. Jackson, A. L., Bartz, S. R., Schelter, J., Kobayashi, S. V., Burchard, J., Mao, M. et al. (2003). Expression profiling reveals off-target gene regulation by RNAi. Nature Biotechnol. 21, 635– 637. Saxena, S., Jonsson, Z. O. & Dutta, A. (2003). Small RNAs with imperfect match to endogenous mRNA repress translation. J. Biol. Chem. 278, 44312 –44319. Sledz, C. A., Holko, M., de Veer, M. J., Silverman, R. H. & Williams, B. R. G. (2003). Activation of the interferon system by short-interfering RNAs. Nature Cell Biol. 5, 834– 839. Scacheri, P. C., Rozenblatt-Rosen, O., Caplen, N. J., Wolfsberg, T. G., Umayam, L., Lee, J. C. et al. (2004). Proc. Natl Acad. Sci. USA, 101, 1892– 1897. Zou, H., Chan, K., Trang, P. & Liu, F. (2004). General design and construction of RNase P ribozymes for gene-targeting applications. Methods Mol. Biol. 252, 385 –398. Stark, G. R., Kerr, I. M., Williams, B. R., Silverman, R. H. & Schreiber, R. D. (1998). How cells respond to interferons. Annu. Rev. Biochem. 67, 227– 264. Tran, N., Cairns, M. J., Dawes, I. W. & Arndt, G. M. (2003). Expressing functional siRNAs in mammalian cells using convergent transcription. BMC Biotechnol. 3, 21– 27. Yuan, Y., Hwang, E.-S. & Altman, S. (1992). Targeted cleavage of messenger RNA by human RNase P. Proc. Natl Acad. Sci. USA, 89, 8006– 8010. Bartkiewicz, M., Gold, H., & Altman, S. (1989). Identification and characterization of an RNA molecule that copurifies with RNase P activity from HeLa cells. Genes Dev. 3, 488– 499.

Edited by J. Karn (Received 13 April 2004; received in revised form 23 May 2004; accepted 3 June 2004)