|
|
Genetics. 2008 October; 180(2): 785–796. doi: 10.1534/genetics.108.093310. | PMCID: PMC2567380 |
Copyright © 2008 by the Genetics Society of America ADBP-1 Regulates an ADAR RNA-Editing Enzyme to Antagonize RNA-Interference-Mediated Gene Silencing in Caenorhabditis elegans Hiromitsu Ohta,* Manabi Fujiwara,*1 Yasumi Ohshima,† and Takeshi Ishihara* *Department of Biology, Graduate School of Science, Kyushu University, Fukuoka 812-8581, Japan and †Department of Applied Life Science, Faculty of Biotechnology and Life Science, Sojo University, Kumamoto 860-0082, Japan Received June 30, 2008; Accepted July 22, 2008. Small interfering RNAs (siRNAs) and microRNAs (miRNAs) mediate gene silencing through evolutionarily conserved pathways. In Caenorhabditis elegans, the siRNA/miRNA pathways are also known to affect transgene expression. To identify genes that regulate the efficiencies of the siRNA/miRNA pathways, we used the expression level of a transgene as an indicator of gene silencing and isolated a transgene-silencing mutant, adbp-1 (ADR-2 binding protein). The adbp-1 mutation caused transgene silencing in hypodermal and intestinal cells in a cell-autonomous manner, depending on the RNA interference (RNAi) machinery. The adbp-1 gene encodes a protein with no conserved domains that is localized in the nucleus. Yeast two-hybrid screening and co-immunoprecipitation analysis demonstrated that ADBP-1 physically interacts with ADR-2, an RNA-editing enzyme from the ADAR (adenosine deaminase acting on dsRNA) family. In the adbp-1 mutant, as previously shown in adr-2 mutants, A-to-I RNA editing was not detected, suggesting that ADBP-1 is required for the RNA-editing activity of ADR-2. We found that ADBP-1 facilitates the nuclear localization of ADR-2. ADBP-1 may regulate ADR-2 activity and the consequent RNA editing and thereby antagonize RNAi-mediated transgene silencing in C. elegans. RNA interference (RNAi) is an evolutionary conserved gene-silencing process. During RNAi, double-stranded RNA molecules (dsRNAs) are processed into ~21-nt small interfering RNAs (siRNAs) by the Dicer family of endoribonucleases. These small RNAs are incorporated into RNA-induced silencing complexes and guide the sequence-specific cleavage of complementary target RNAs ( Ketting et al. 2001). RNAi can be elicited by exogenous dsRNAs supplied from outside the cell or by endogenous dsRNAs such as bidirectional transcripts and transcripts with extensive secondary structure. The RNAi-related gene-silencing mechanism is mediated by microRNAs (miRNAs), which are generated from short fold-back hairpin gene products. miRNAs are similar in length to siRNAs and regulate many important aspects of development and physiology in multicellular eukaryotes ( Carthew 2006; Cerutti and Casas-Mollano 2006). Caenorhabditis elegans contains numerous siRNAs and miRNAs, suggesting that, by RNAi and related mechanisms, gene regulation occurs on a genomic scale ( Ambros et al. 2003). These siRNAi/miRNA pathways are known to interact with each other in vivo, competing for shared components ( Duchaine et al. 2006). Therefore, the changes in RNAi efficiency can influence the expression of a great number of genes, and the efficiency must be controlled precisely ( Lee et al. 2006). RNAi has been shown to be involved in the protection of host genomes against invasion by repetitive sequences, such as viruses, transposons, and repetitive transgenes ( Ketting et al. 1999; Li et al. 2002; Sarot et al. 2004; Robert et al. 2005). In the germline of C. elegans, expression of repetitive transgenes is repressed. This germline–transgene silencing process depends on some RNAi machinery genes, including mut-7, rde-2, and rde-3 ( Tabara et al. 1999; Robert et al. 2005). It was demonstrated that the germline–transgene silencing is also mediated by chromatin proteins such as the Polycomb group of the transcriptional repressors MES-2 and MES-6 and their interacting protein, MES-4 ( Kelly and Fire 1998; Fong et al. 2002). The transgene-silencing effect is less obvious in the somatic cells of C. elegans, although it can be induced in the RNAi-hyperactivated condition, such as in eri-1 and rrf-3 mutants ( Simmer et al. 2002; Kim et al. 2005; Duchaine et al. 2006). The somatic transgene silencing is likely to be a similar process to the germline–transgene silencing because it also depends on mut-7 and mes-4 ( Kim et al. 2005). It is predicted that dsRNAs are generated from aberrant transcripts of the repetitive transgenes and induce transgene silencing through RNAi and chromatin remodeling ( Robert et al. 2005). The somatic transgene silencing is also induced by mutations of the ADAR ( adenosine deaminases acting on ds RNA) genes, adr-1 and adr-2, in C. elegans. ADARs consist of a family of RNA-editing enzymes, which catalyze the hydrolytic deamination of adenosine to inosine in dsRNAs (A-to-I editing) and disrupt A–U pairs to create less stable I–U wobble pairs in dsRNAs ( Wagner et al. 1989). It is likely that ADARs edit transgene-derived dsRNAs and thus antagonize RNAi of the transgene, resulting in the expression of the transgenes in wild-type C. elegans somatic cells ( Knight and Bass 2002). Interestingly, in other species, ADARs have been also reported to antagonize the RNAi pathways ( Nishikura 2006). For example, in Drosophila, dsRNAs that are excessively edited by ADARs become resistant to Dicer, resulting in reduced RNAi efficiency ( Scadden and Smith 2001). Moreover, miRNA precursors can be edited by ADARs ( Blow et al. 2006; Yang et al. 2006; Kawahara et al. 2007). The edited miRNA precursors are degraded by Tudor-SN, a ribonuclease specific for inosine-containing dsRNAs, and affect the developmental gene expression patterns ( Yang et al. 2006). Here we report a new C. elegans mutant adbp-1, which was identified due to decreased expression levels of the transgenes in the somatic cells. adbp-1 encodes a novel protein that is localized in the nucleus. We found that ADBP-1 physically interacts with the ADAR protein ADR-2, facilitates the nuclear localization of ADR-2, and plays the crucial role for RNA editing. Our experimental results suggest that ADBP-1, acting together with an ADAR, controls the transgene expression by antagonizing the RNAi pathway. Furthermore, our results raise a possibility that ADBP-1 might restrict the RNA-editing activity to the nucleus and function for the target selection of RNA editing and RNAi. Strains and culture of C. elegans: Worms were grown on NGM plates seeded with an OP50 Escherichia coli strain using standard methods ( Brenner 1974). The wild-type C. elegans strains used in this work were Bristol (N2) and CB4856. The following mutations were used in this work: LGII, adbp-1(qj1); LGIII, adr-2(ok735); LGIV, eri-1(mg366); and LGV, rde-1(ne219). We verified the mutation of adr-2(ok735), a deletion allele in which most of the coding region is deleted, by PCR. Mutant screening: Extrachromosomal simple arrays of lon-1p gfp and sra-6pgfp (as an injected marker) were integrated into a chromosome using UV irradiation, followed by outcrossing with N2 animals three times. The line carrying an integrated copy was mutagenized with 50 m m EMS. P0 worms were transferred to new plates and allowed to segregate self-progeny. The animals with reduced levels of lon-1pgfp expression were isolated from F 2 synchronized worms at the L4 stage. Screening of ~12,000 haploid genomes yielded 10 candidate mutant lines. Because cross progenies with wild-type animals restored the normal lon-1pgfp expression level, the phenotype in all lines was not due to a change in the transgene but to a recessive mutation in the genome. Among the candidate mutant lines, we further analyzed the qj1 line, as the transgene-silencing phenotype was relatively strong compared to the other lines. None of the other lines failed to complement the qj1 locus. Quantitative RT–PCR: Trizol (Invitrogen) was used to prepare total RNA from worm pellets harvested from 1 liter of liquid culture. Poly(A)+ RNA was purified twice using an mRNA purification kit (GE Healthcare). First-strand cDNA was synthesized from poly(A)+ mRNA using a NotI d(T) 25 primer (5′-AACTGGAAGAATTCGCGGCCGCAGGAAT25-3′) and SuperScript III reverse transcriptase (Invitrogen). Quantitative PCR experiments were performed as described in the manual using SYBR Green PCR and RT–PCR reagents (Applied Biosystems) and a Gene Amp 5700 sequence detection system (Applied Biosystems). lon-1 mRNA levels were normalized to those of the act-1 mRNA. Genetic mapping and PCR fragment rescue of adbp-1 mutants: Genetic mapping was performed using the snip–SNP method ( Wicks et al. 2001). PCR products used in the rescue experiments were amplified from wild-type genomic DNA. Germline transformations were performed according to the methods described previously ( Mello et al. 1991). PCR product (20 ng/μl) was co-injected with ttx-3pmRFP marker DNA at 50 ng/μl into adbp-1(qj1) mutant animals. Plasmid construction: To construct lon-1pgfp, the predicted lon-1 promoter region (3.0-kb sequence upstream of lon-1) was cloned into the pPD95.77 GFP expression vector (a gift from A. Fire). To construct the GFP reporters of adbp-1, a genomic region containing the adbp-1 gene, the upstream gene VW02B12L.3 in the operon, and the 3.0 kb of upstream sequence was amplified from the Y70H7 YAC clone using PCR and ligated into pPD95.77. For promoter-gfp fusion, gfp was fused to the initiation codon of adbp-1. For translational gfp fusions, gfp was fused just before the termination codon of adbp-1. For the tissue-specific expression of adbp-1, full-length cDNA was obtained from the yk1745h09 EST clone (a gift from Yuji Kohara) and inserted into the pPD49.26 expression vector (a gift from A. Fire). The vha-7 promoter (for the main body hypodermis), the nhr-72 promoter (for seam cells), or the myo-2 promoter (for the pharynx) were inserted into the adbp-1 expression vector. Each promoter was obtained from N2 genomic DNA using PCRs ( vha-7 and nhr-72) or by the digestion of pPD96.48 ( myo-2). For the expression of adbp-1 in the nervous system, gfp was replaced with adbp-1 cDNA in the H20pgfp pan-neuronal gfp expression construct ( Shioi et al. 2001). Two-hybrid screening: To construct a plasmid encoding the bait (pGBKT7-adbp-1) for the yeast two-hybrid screen, a full-length adbp-1 cDNA was obtained from yk1745h09 and inserted into the pGBKT7 vector (Clontech). Two-hybrid screens with ADBP-1 as the bait were performed in the Saccharomyces cerevisiae strain AH109. A C. elegans cDNA library (a gift from R. Barstead) was used for the screening. To screen for positive clones, the transformants were streaked onto the −Ade/−His/−Leu/−Trp plates and examined for the expression of the ADE2 and HIS3 reporters. RNA interference assay: RNAi clones were obtained from the C. elegans RNAi library (Geneservice). RNAi by feeding was performed as described previously ( Timmons and Fire 1998). The following clones were used in this work: Geneservice ID I-3N01, I-5F19, II-7B03, II-7F12, III-4C08, III-5O20, and III-6N01. Cell culture, transfection, immunoprecipitation, and Western blotting: For the co-immunoprecipitation assays, Flag-tagged adbp-1 (ADBP-1 Flag) and Myc-tagged adr-2 (ADR-2 Myc) were expressed under the control of mammalian promoters. For the ADBP-1 Flag expression construct, the sequence encoding 3× Flag was fused to the 3′ terminus of full-length adbp-1 cDNA using a PCR-generated XbaI site and inserted into the pcDNA 3.1 Zeo(+) expression vector (Invitrogen). For the ADR-2 Myc expression construct, full-length adr-2 cDNA, obtained from the yk1255b05 EST clone (a gift from Yuji Kohara), was inserted into the pcDNA3.1/Myc-His expression vector (Invitrogen). To obtain stronger expression, the region encoding ADR-2 Myc was transferred to another expression vector, pCAGGS-2 ( Niwa et al. 1991). HEK293T cells were cultured in D-MEM (Sigma) containing 10% fetal bovine serum. HEK293T cells were transfected with ADBP-1 Flag and/or ADR-2 Myc expression constructs using Lipofectamine 2000 (Invitrogen). After transfection, cells were cultured for 48 hr and whole-cell lysates were prepared using radio immunoprecipitation assay buffer containing complete mini protease inhibitor cocktail (Roche). The lysates were incubated with anti-Flag (Sigma) or anti-Myc antibodies (Invitrogen) and the immunocomplexes were collected with protein A–Sepharose CL-4B (GE Healthcare). The blotted membranes were incubated with anti-Myc (Invitrogen) or anti-Flag (Sigma) primary antibodies and anti-mouse IgG and HRP-linked whole secondary antibody from sheep (GE Healthcare) before visualization of the immunoreactive proteins using an ECL advanced Western blotting detection kit (GE Healthcare). Fluorescence intensity analysis of lon-1pgfp: The fluorescence images of L4-stage hermaphrodites harboring the Is[lon-1pgfp] transgene were captured with an Axioplan 2 microscope (Zeiss) equipped with an AxioCam CCD camera (Zeiss) using the same magnification and the same exposure time. These images were analyzed with the Image-Pro Plus image analysis program (Media Cybernetics). By using a line profile tool, the fluorescence intensities of the three body regions at the posterior end of the pharynx, the vulva, and the posterior end of the intestine were measured and averaged for the fluorescence intensity of the body of each animal. The score of each strain was the average measurement of 20–30 animals. Chemotaxis assay: The chemotaxis assay was based on assays developed previously ( Bargmann and Horvitz 1991; Bargmann et al. 1993). Odorant dilutions for the chemotaxis assays were as follows: diacetyl, 10 −3 and 10 −4; 2-butanone, 10 −2 and 10 −3; and benzaldehyde, 10 −2, 10 −3, and 10 −4. Amplification and sequencing of ADAR substrates: cDNA was prepared as described above. cDNA corresponding to ADAR substrates (9A, 16G, 36A, and laminin-γ; Morse and Bass 1999; Morse et al. 2002) was amplified in two rounds of PCRs. PCR products were sequenced using the PCR primers from the second amplification. Primers: The primers used in this study were the following. For quantitative RT–PCR, act-1 sense primer (5′-GAGCACGGTATCGTCACCAA-3′); act-1 antisense primer (5′-TGTGATGCCAGATCTTCTCCAT-3′); lon-1 sense primer (5′-TTCTGACGCTATCTCGATCT-3′); lon-1 antisense primer (5′-CTTGGATTATGGACCTCGTT-3′); for plasmid construction, vha-7 sense primer (5′-AAAACTGCAGCGACAGGAAATTGTGAGAAG-3′); vha-7 antisense primer (5′-AAAACTGCAGCAGATTACGTCGTTGGTGGA-3′); nhr-72 sense primer (5′-TTTTCTGCAGCCTGCTCTGATACAGCTATC-3′); nhr-72 antisense primer (5′-TTTTCTGCAGGAGTGAAATTCGGCGACTCA-3′); for amplification of ADAR substrates—first PCR primer, 9A sense primer (5′-ATATCATCGATGCAGCTCGG-3′); 16G sense primer (5′-CGCTTTCGTTCCGAATCTTG-3′); 36A sense primer(5′-GTTTTCAAACCAGTGGAGCC-3′); laminin-γ sense primer (5′-GAGATTCCACCGAAGCTGTT-3′); oligo(dT) tagged sequence antisense primer (5′-TCGCGGCCGCAGGAA-3′); for amplification and sequencing of ADAR substrates—second PCR primer, 9A sense primer (5′-GGCACTTCAATTTGCACACC-3′); 9A antisense primer (5′-AAATGACCTAAAGGTGGTTA-3′); 16G sense primer (5′-ACGTGTGTGGACATTTACGG-3′); 16G antisense primer (5′-AAAGAAATATCTTCGTTTGC-3′); 36A sense primer (5′-AATGGATTCGGTGTCCGAAC-3′); 36A antisense primer (5′-ACAGGCTAAAAGCTCAGCAC-3′); laminin-γ sense primer (5′-AAATGGCTACTGAAGCGGTC-3′); laminin-γ antisense primer (5′-GAGATTCCACCGAAGCTGTT-3′). Isolation of the transgene-silencing mutant adbp-1: To study the regulation of the siRNA/miRNA pathways in C. elegans, we screened for mutants that showed altered expression levels of repetitive transgenes. In C. elegans, injection of one or a few of DNA plasmids into the gonad results in formation of a highly repetitive transgene, referred to as a simple array because the injected DNA forms long tandemly arranged arrays ( Stinchcomb et al. 1985). We used a simple array of a lon-1pgfp transgene as an indicator, which is a fusion of the lon-1 promoter and gfp. The lon-1pgfp transgene gives strong GFP expression in the main-body hypodermal cells (hyp 7 and some other hypodermal cells in the tail and head) and intestine of L1 larvae through adulthood, which enabled us to detect changes in the expression levels easily. First, we produced a transgenic worm stably expressing the lon-1pgfp transgene, which was integrated into one of the chromosomes. After mutagenizing the worms with ethyl methansulfonate (EMS), we isolated one mutant ( qj1) that showed a significant reduction in the GFP level (). When the mutant animals were crossed with wild-type animals, heterozygote progenies showed a normal level of GFP expression, indicating that the altered GFP expression level was a result of a recessive mutation in an endogenous gene and not a change in the integrated transgene. We named this mutant adbp-1 (see below). The adbp-1(qj1) mutant was fertile and appeared healthy. The brood size of the adbp-1 mutant was similar to those of wild-type animals (data not shown), and no other obvious phenotypes were observed. The lon-1 gene encodes a secreted protein with homology to members of the pathogenesis-related-protein superfamily, and mutations in lon-1 cause a long-body-size phenotype ( Maduzia et al. 2002). If the adbp-1 mutation affected the expression of the endogenous lon-1 locus, the adbp-1(qj1) mutant should have exhibited the long-body phenotype. The mutant, however, did not show any obvious changes in its body length (data not shown). Furthermore, we quantified the level of lon-1 transcripts using quantitative RT–PCRs and found that the levels of lon-1 transcripts were similar in adbp-1(qj1) mutants and wild-type worms (). These data indicate that the adbp-1(qj1) mutation did not affect the expression level of the endogenous lon-1 gene. Therefore, adbp-1 is likely to be involved in the regulation of the expression of the transgene, but not in the regulation of the expression of the endogenous gene. The adbp-1 mutation induces silencing on various transgenes in the hypodermis and intestine: The silencing effect of adbp-1(qj1) was not limited to the integrated array of the lon-1pgfp transgene, because we observed a similar reduction of GFP expression from an extrachromosomal array of lon-1pgfp in the adbp-1(qj1) mutant (data not shown). Moreover, we examined whether other transgenes expressed in the hypodermis and intestine were also affected by the adbp-1 mutation. As shown in , we introduced extrachromosomal arrays of various gfp reporter constructs into adbp-1 and wild-type animals and compared the resulting GFP expression levels. We observed reduced GFP expression in adbp-1 mutant animals carrying the vha-7pgfp transgene, which drives GFP expression in the main-body hypodermal cells. The expression of nhr-68pgfp, which drives GFP expression in the main-body hypodermal cells, the hypodermal seam cells, and the intestine, was also affected by the adbp-1 mutation; the GFP expression levels in all these tissues were reduced. Next, we examined transgenes expressed in other tissues. The expression levels of GFP in body-wall muscle and pharyngeal muscle, driven by myo-3pgfp and myo-2pgfp, respectively, were only mildly affected by the adbp-1 mutation. The expression of GFP in neuronal tissues driven by H20pgfp appeared to be unaffected by the adbp-1 mutation, although we did not closely examine the GFP expression levels in individual neurons. Taken together, the results show that the adbp-1 mutation affected the expression of various transgenes expressed in the hypodermis and intestine, although the adbp-1 mutation did not appear to markedly affect transgenes expressed in muscles and neurons. Transgene silencing in the adbp-1 mutant depends on the repetitive context of the transgene: In C. elegans, transgenes are usually highly repetitive simple arrays, which are produced by injection of one or a few of DNA plasmids into animals. All the transgenes used in this study were the simple arrays unless noted. If the composition of the injected DNA is more complex, the resulting arrays are less repetitive and are referred to as complex arrays. To examine whether transgene silencing in adbp-1 mutant animals depended on the repetitive context of the simple arrays, we produced complex arrays by injecting nhr-68pgfp with an excess of other genomic DNA fragments. As mentioned above, the simple array of nhr-68pgfp was strongly silenced in the adbp-1(qj1) mutant. On the other hand, all three adbp-1(qj1) lines carrying the independent complex arrays of nhr-68pgfp showed strong GFP expression as was observed in wild-type animals (). Thus, transgene silencing in the adbp-1(qj1) mutant seemed to be caused by simple arrays but not by complex arrays and is likely to be dependent on the repetitive context of the transgenes. Molecular identification of adbp-1: The adbp-1 mutation was mapped to a small region (350 kb) on chromosome II between two SNP markers on the F37H8 and B0334 cosmids using standard SNP mapping methods. Because none of the cosmids in this region that we examined rescued the decreased lon-1pgfp expression in adbp-1 animals, eight genes located at the joints of the cosmid contigs were good candidates for the responsible gene. In an attempt to phenocopy the adbp-1 mutant, we conducted RNAi analyses using these eight genes. Strikingly, RNAi of the VW02B12L.4 gene caused a reduction in lon-1pgfp expression (). Because the VW02B12L.4 gene composed an operon and shared a promoter with the upstream gene VW02B12L.3, we injected a PCR-amplified fragment containing the ~3-kb promoter region and these two genes (PCR fragment A). This fragment rescued the decreased lon-1pgfp expression in adbp-1 animals (). When an ~200-bp deletion was made in the coding region of the upstream gene in the cloned PCR-amplified fragment (PCR fragment AΔ), the construct still rescued the decreased lon-1pgfp expression in adbp-1 animals (). In addition, we showed that the expression of VW02B12L.4 under the control of extrinsic promoters was able to rescue the adbp-1 phenotype (see below). VW02B12L.4 gene encodes a 217-amino-acid protein with no conserved domains and no apparent nuclear localization signal. Sequencing of the VW02B12L.4 locus in the adbp-1(qj1) mutant revealed a nonsense mutation in the middle of the coding region (Q119STOP) (). On the basis of these findings, we concluded that VW02B12L.4 was the adbp-1 gene. A BLAST search with the amino-acid sequence identified one ADBP-1 homolog in C. elegans and two homologs in C. briggsae, but no homologs in other organisms. The C. elegans homolog T17A3.9 showed 57% amino-acid identity and 70% amino-acid similarity with ADBP-1 (). Although feeding-mediated RNAi of the T17A3.9 gene did not cause a reduction in expression of transgenes (data not shown), mutant analyses are required to elucidate its function. ADBP-1 localizes in nuclei throughout development: To analyze the expression of adbp-1, we made two GFP reporter constructs. One was a promoter- gfp fusion; gfp was fused with an ~4-kb genomic fragment containing the promoter region (~3 kb), an upstream gene in the operon, and the ATG initiation codon of adbp-1 (). The second construct was a translational gfp fusion; gfp was fused in frame with an ~5-kb genomic fragment containing the promoter region (~3 kb), an upstream gene, and the whole coding region of adbp-1 except for the termination codon (). The promoter- gfp fusion construct drove GFP expression in the main-body hypodermal cells, the hypodermal seam cells, pharynx, intestine, and some neurons, starting at an embryonic stage and continuing through young adulthood (). The levels of GFP expression were weaker in older adults. The translational gfp fusion construct directed GFP expression in the same tissues, but GFP was strongly localized in the nuclei (). Because the translational fusion construct, whose reporter gene was converted into mRFP from gfp, rescued the decreased lon-1pgfp expression in adbp-1(qj1) mutant animals (data not shown), endogenous adbp-1 is likely to be expressed in at least some of the tissues in which we observed GFP expression. ADBP-1 functions in a cell-autonomous manner: adbp-1 is expressed in the hypodermis and other tissues. We then asked whether the expression of lon-1pgfp in the main-body hypodermal cells is controlled by ADBP-1 acting in the same cells. adbp-1 cDNA was fused to various promoters, which resulted in expression of the protein in various tissues, including the main-body hypodermal cells ( vha-7 promoter), pharynx ( myo-2 promoter), seam cells ( nhr-72 promoter), and neurons ( H20 pan-neuronal promoter). We introduced these constructs into adbp-1(qj1) mutant animals bearing an integrated array of lon-1pgfp. As shown in , the ADBP-1 expression by the vha-7 promoter restored the expression level of lon-1pgfp in the main-body hypodermal cells. On the other hand, the adbp-1 constructs with the other promoters, whose expressions in the expected tissues were confirmed by fusing mRFP cDNA to the 3′-end of the adbp-1 cDNA (data not shown), did not restore the lon-1pgfp expression level (). These data suggest that ADBP-1 functions in a cell-autonomous manner in the main-body hypodermal cell to regulate the hypodermal expression of the lon-1pgfp transgene. ADBP-1 interacts with ADR-2: To elucidate the molecular function of ADBP-1, we performed a yeast two-hybrid screen to identify interacting proteins. A screen of a C. elegans cDNA library yielded 10 positive clones when full-length ADBP-1 was used as the bait. Among these positive clones, two independent clones encoded N-terminal portions of the ADR-2 protein, indicating that ADR-2 was potentially interacting with ADBP-1. ADR-2 is one of the ADAR proteins, which are a family of RNA-editing enzymes in C. elegans ( Tonkin et al. 2002). Interestingly, Li et al. (2004) also reported a possible interaction between ADR-2 and ADBP-1 in their systematic yeast two-hybrid analyses for the interactome map of C. elegans. To confirm this interaction, we performed co-immunoprecipitation experiments. We transiently expressed Flag-tagged ADBP-1 and Myc-tagged ADR-2 in HEK293T cells. Using the cell lysates, ADBP-1 was immunoprecipitated with anti-Flag antibodies. We found that ADR-2 specifically co-immunoprecipitated with ADBP-1 by immunoblotting with anti-Myc antibodies (, top, lane 8). In a similar manner, we immunoprecipitated ADR-2 with anti-Myc antibodies and observed that ADBP-1 specifically co-immunoprecipitated with ADR-2 (, bottom, lane 12). These results suggest that ADBP-1 is likely to bind to ADR-2 in vivo. adbp-1 (ADR-2 binding protein) was named on the basis of this result. Further interaction analyses using worms may reveal more dynamic association of ADBP-1 and ADR-2 under the temporal or spatial regulation in C. elegans. Genetic interactions between adbp-1 and adr-2: Interestingly, a mutation in the adr-2 gene has been reported to cause a transgene-silencing phenotype ( Knight and Bass 2002). When the lon-1pgfp transgene was introduced into the adr-2(ok735) mutant strain, GFP expression decreased to a level similar to that observed in the adbp-1 mutant (). The transgene-silencing phenotype in the adr-2 mutant appeared to result from a defect in the same genetic pathway that includes adbp-1, because adbp-1; adr-2 double-mutant animals did not exhibit a further reduction in GFP expression compared to each single mutant (). We also examined whether adr-2 produced a similar tissue-specific transgene-silencing phenotype as was observed with adbp-1. When various gfp reporter constructs were introduced into the adr-2(ok735) mutant, we observed that adr-2 caused a marked reduction in GFP expression in the hypodermis and intestine, whereas only minor changes were detected for GFP expression in muscle and neurons (). These silencing patterns were similar to those caused by adbp-1(qj1), suggesting that ADBP-1 and ADR-2 have similar functions and supporting the idea that these proteins physically interact. Because ADBP-1 and ADR-2 appeared to act together, we asked if the other phenotypes observed in the adr-2 mutant would also be present in the adbp-1 mutant. Because it has been reported that the adr-2 mutant is defective for chemotaxis ( Tonkin et al. 2002), we analyzed the chemotactic responses of the adbp-1 animals. Although we assayed the animals with multiple odorants at various concentrations, we did not detect any significant defects in the adbp-1 mutants (). However, we detected only slight chemotactic defects in the adr-2 in our study (). RNAi processes mediate transgene silencing in adbp-1: The somatic transgene-silencing phenotype in the known mutants including adr-2, rrf-3, and eri-1 was suppressed by mutations in the RNAi machinery genes rde-1 and dcr-1, as consistent with the model that transgene silencing is caused through RNAi of the transgene ( Kim et al. 2005). In the adbp-1 mutant, we found that a rde-1 mutation and RNAi of dcr-1 also restored the expression of lon-1pgfp (; data not shown). Thus, transgene silencing in the adbp-1 mutant also depended on the RNAi machinery. We, however, found that the gene disruption of mes-4, a chromatin-remodeling gene, did not suppress the somatic transgene-silencing phenotype in the adbp-1 mutant, while the transgene silencing in the rrf-3, eri-1 mutants was suppressed by mes-4 gene disruption ( Kim et al. 2005; data not shown). Knight and Bass (2002) demonstrated that transgene-derived dsRNAs were subject to A-to-I editing in wild-type animals and proposed a model in which ADR-2 edits transgene-derived dsRNAs and prevents RNAi of the transgene in wild-type animals. In the adr-2 mutant, unedited dsRNA may trigger RNAi of the transgene. In the adbp-1 mutant, RNAi-inducible dsRNA also appears to exist, because the transgene silencing in adbp-1 was mediated by the RNAi machinery. Knight and Bass (2002) used transgenes of simple arrays and predicted that aberrant transcripts from the sequences tandemly arranged in opposing orientations in the simple arrays may result in the formation of dsRNA. This is supported by our results, because all three independent nhr-68pgfp transgenes of complex arrays, which had less repetitive context, were not silenced in the adbp-1 mutant (). We predicted that if dsRNAs generated from simple arrays caused transgene silencing, they may serve as trans-acting elements to suppress expression from complex arrays. To test this hypothesis, we produced an adbp-1(qj1) line bearing both a simple array ( lon-1pgfp) and a complex array ( nhr-68pgfp), both of which carried the same gfp gene (). When the simple array was introduced, GFP expression from the complex array () in the adbp-1 mutant was indeed reduced (). This result strongly supports the model proposed by Knight and Bass (2002). Interestingly, as shown in , we observed the GFP reduction in the main-body hypodermal cells and the intestine but not in the seam cells. In the main-body hypodermal cells and intestine, both lon-1pgfp and nhr-68pgfp were directed to express, whereas in the seam cells only nhr-68pgfp was directed to express. Therefore, dsRNA is likely to suppress the transgene expression cell autonomously. The adr-2 and adbp-1 mutants did not show RNAi hypersensitivities even in the hypodermis: ADR-2 probably edits dsRNAs produced from transgenes and antagonizes RNAi of the transgenes. On the basis of the results of the physical interaction assays and the genetic analyses, we predicted that ADBP-1 acts together with ADR-2 in the process. We asked whether ADR-2 and ADBP-1 also antagonize RNAi induced by exogenously introduced dsRNAs. If this is the case, adbp-1 and adr-2 mutants may show RNAi-hypersensitive phenotypes. In a previous study, the adr-2 mutant showed no enhancement of RNAi induced by injection of dsRNA ( Knight and Bass 2002). Since the previous analysis targeted mainly the unc-22 gene, which was expressed in the muscle, we intended to examine the RNAi hypersensitivity by targeting genes expressed in the hypodermis, where the transgene silencing was most obvious. For the RNAi targets, we chose four genes that are required for normal development or behavior and act in the hypodermis ( hmr-1, bli-1, and bli-5) or the muscle ( unc-15) ( Table 1). We fed the worms E. coli that expressed dsRNA specific for the targeted genes. As expected, the RNAi-defective rde-1 mutant and the RNAi-hypersensitive eri-1 mutant showed lower and higher sensitivities to RNAi of each tested gene, respectively ( Table 1). adr-2(ok735) and adbp-1(qj1) animals, however, showed only a small enhancement of hmr-1 RNAi and no significant changes in RNAi of other genes, compared to wild-type responses. Thus, ADR-2 and ADBP-1 do not appear to interfere with the siRNA pathway triggered by exogenous dsRNA even in the hypodermis. | TABLE 1The RNAi sensitivities of adbp-1 and adr-2 mutants |
ADBP-1 is required for the deaminase activity of ADR-2: Because ADBP-1 appeared to function with ADR-2, we hypothesized that ADBP-1 may be required for the deamination of adenosine by ADR-2. Although C. elegans has two ADAR homologs, ADR-1 and ADR-2, ADR-2 appears to play a major role in the deamination of adenosine because a deaminase activity was not detected in an adr-2 mutant, whereas a partial deaminase activity remained in an adr-1 mutant ( Tonkin et al. 2002). In C. elegans, 10 ADAR substrates have been identified ( Morse and Bass 1999; Morse et al. 2002); they are edited at stem-loop structures located in noncoding regions. We examined four of these ADAR substrates to test whether A-to-I editing occurred in the adbp-1 animals. We isolated poly(A) + RNA from the adbp-1(qj1) mutant and wild-type animals and prepared cDNA. The RNA-editing sites were amplified in PCRs and the entire population of PCR products was directly sequenced. As previously reported, the results should correctly reflect the editing states of each substrate ( Morse et al. 2002). Because inosine pairs with cytidine, inosine is represented by guanosine in the corresponding cDNA sequence. Sequencing electropherograms from wild-type cDNA showed that the A peaks (green in ) overlapped with the G peaks (black in ), indicating that the RNA editing occurred as predicted. On the other hand, in electropherograms from adbp-1(qj1) mutant cDNA, no A peaks overlapped with the G peaks for each of the four substrates (). Furthermore, we showed that the lack of A-to-I editing in the adbp-1(qj1) mutant was rescued by introduction of a wild-type copy of the adbp-1 gene (). This result clearly indicates that no RNA editing occurred in these substrates in the adbp-1 mutant and that ADBP-1 plays the crucial role for the deaminase activity of ADR-2. ADBP-1 facilitates the nuclear localization of ADR-2: How does ADBP-1 control the deaminase activity of ADR-2? We analyzed the subcellular localization pattern of ADR-2 by expressing full-length adr-2 cDNA fused with gfp under the control of the hypodermal promoter ( vha-7padr-2gfp). ADR-2 GFP accumulated in the nucleus and was also observed in the cytoplasm of hypodermal cells (, WT). It is known that certain ADARs in other species, including mammalian ADAR1, also showed a similar subcellular localization pattern ( Desterro et al. 2003). Because ADBP-1 was predominantly localized in the nucleus, the nuclear localization of ADR-2 is consistent with its observed physical interaction with ADBP-1. We asked if the nuclear localization of ADR-2 depended on ADBP-1 by examining the localization pattern of ADR-2 in the adbp-1(qj1) mutant background. Because of transgene silencing, GFP expression from the vha-7padr-2gfp transgene was reduced in the adbp-1 mutant, but the low-level expression of GFP observed at high magnification allowed us to examine the subcellular localization pattern. We found an apparent reduction of GFP fluorescence in the nucleus of the adbp-1 mutant; ADR-2 GFP was not accumulated in the nucleus, but distributed evenly in the nucleus and the cytoplasm (data not shown). This may not be simply due to the reduction of the GFP expression level because in another transgene-silencing mutant, eri-1, we observed the nuclear accumulation of ADR-2 GFP (data not shown). We further analyzed the ADR-2 GFP subcellular localization patterns by suppressing the transgene-silencing phenotype of the adbp-1 mutant by a rde-1 mutation. In the adbp-1(qj1);rde-1(ne219) double mutant, the expression level of ADR-2 GFP was comparable to those in wild-type and in rde-1(ne219) animals, but the nuclear accumulation was again less obvious than these control lines (). The fluorescence intensity in the nucleus was almost the same as that in the cytoplasmic region in the double mutant, while the fluorescence intensities in the nucleus were significantly higher than those in the cytoplasm in the control lines (). These results suggest that one of the ADBP-1 roles may be to facilitate the nuclear localization of ADR-2 in the hypodermis. In this article, we identified ADBP-1, an ADAR-binding protein, in a genetic screen for mutations causing transgene silencing and found that it affects the siRNA pathway through RNA editing. Thus, our results demonstrate that the screen of transgene-silencing mutants is a successful way to identify the genes involved in the miRNA/siRNA pathways, as was shown in previous studies ( Hsieh et al. 1999; Wang et al. 2005; Cui et al. 2006). ADBP-1 is required for RNA editing in C. elegans: In this study, we identified the novel protein ADBP-1, which plays a crucial role in RNA editing through binding to the RNA-editing enzyme ADR-2. ADR-2 is a major ADAR protein in C. elegans. RNA editing mediated by ADARs is an important biological process, because it creates diversity among the transcripts by recoding genes or changing characteristics of the transcripts, such as their stability ( Nishikura 2006). We have shown that adbp-1 and adr-2 mutant animals exhibit similar tissue-specific, transgene-silencing phenotypes. Double-mutant analysis suggested that these genes function in the same genetic pathway. By using co-immunoprecipitation analysis, we have demonstrated that ADBP-1 directly interacts with ADR-2 in vivo. Most importantly, we have shown that A-to-I editing was not detected in substrate mRNAs isolated from the adbp-1 mutant. Thus, our results clearly indicate that ADBP-1 functions in the control of ADR-2. The importance of RNA editing by ADAR proteins was first recognized when it was found that RNA editing causes codon changes and consequent changes in protein function ( Higuchi et al. 1993; Burns et al. 1997; Reenan et al. 2000). Recent studies, however, have revealed that editing of noncoding RNAs plays various roles in organisms, including modulation of splicing sites, regulation of mRNA degradation, and regulation of miRNA ( Nishikura 2006; Kawahara et al. 2007). In C. elegans, several ADAR substrates have been identified by purifying inosine-containing poly(A) + RNA molecules ( Morse and Bass 1999; Morse et al. 2002). All of these editing sites were found in noncoding regions and the functions of these modifications are still unknown. A previous report ( Tonkin et al. 2002) and our observations suggest that adr-2 and adbp-1 are not essential for the viability of the worm; these mutants are healthy and fertile with normal brood sizes. The function of ADR-2 in chemotactic behavior was reported previously ( Tonkin et al. 2002), although we were not able to reproduce the magnitude of the defect in our chemotactic analysis. Further studies using adr-2 and adbp-1 mutants would reveal the biological roles of RNA editing in C. elegans. Transgene silencing is caused through the siRNA pathway in the RNA-editing defective mutants: We have shown that the transgene-silencing phenotype of the adbp-1 mutant depended on RNAi machinery genes. The transgene-silencing phenotype of the adr-2 mutant also depends on RNAi machinery genes ( Knight and Bass 2002). These results suggest that a dsRNA intermediate is involved in the process of transgene silencing. Knight and Bass (2002) demonstrated that dsRNAs are produced from transgenes and that A-to-I editing occurs in these RNA molecules. They proposed a model in which RNA editing prevents the transgene-generated dsRNAs from causing RNAi of the transgenes in wild-type animals. In the adr-2 and adbp-1 mutants, defects in RNA editing may allow the transgene-generated dsRNA to initiate RNAi of the transgene. In this article, we have shown that adbp-1 acts cell autonomously in the hypodermis to maintain transgene expression, which is consistent with the proposed model. Our experimental result suggests that transgene silencing due to the adbp-1 mutation is most obvious with the transgenes expressed in the hypodermis and intestine. adbp-1, however, appears to be expressed not only in the hypodermis and intestine but also in other tissues, including the pharynx and some neurons where we did not observe strong transgene silencing. We do not know the reason for such a tissue-specific property of transgene silencing in adbp-1, but it might reflect the expression site of adr-2, which has not been reported yet. We have also shown that transgenes of simple arrays, which had a highly repetitive context, but not transgenes of complex arrays, were silenced. Interestingly, the expression from a complex array was suppressed in tissues in which transgene silencing was induced by a simple array containing the same gene. RNAi of transgenes appears to occur only in the cells in which the dsRNA is produced from the repetitive transgenes, although some dsRNAs can be transferred from cells to cells in C. elegans ( Feinberg and Hunter 2003). The molecular mechanism of the regulation of RNA-editing activity by ADBP-1: How does ADBP-1 control the ADR-2 activity? ADBP-1 did not contain any conserved motifs. However, we found that the accumulation of ADR-2 GFP in the nucleus was reduced in the adbp-1 mutant. This result indicates that ADBP-1 supports the ADR-2 localization in the nucleus. ADBP-1 predominantly localized in the nucleus, whereas ADR-2 was distributed in both the nucleus and the cytoplasm. By binding to ADR-2, ADBP-1 may retain ADR-2 in the nucleus or support its entry into the nucleus and thereby control the deamination by ADR-2 in the nucleus. In the adbp-1 mutant, however, a certain level of ADR-2 was still observed in the nucleus, whereas no RNA editing was detected at least in the examined substrates. Therefore, we speculated that ADBP-1 may have additional functions in promoting the ADR-2 activity. The adr-2 cDNA clones, which were isolated in our yeast two-hybrid screen using ADBP-1 as the bait, encoded the N-terminal portion of ADR-2 that contains the dsRNA-binding domain. Binding of ADBP-1 to the dsRNA-binding domain may enhance the ADR-2 activity through, for example, the stabilization of the binding between dsRNA substrates and ADR-2. Because the adr-2 and adbp-1 mutants did not exhibit hypersensitivities in our feeding RNAi analysis, no antagonistic effects of RNA editing were observed in RNAi that was induced by exogenously introduced dsRNA. This may mean that these exogenously introduced dsRNAs are not edited in wild-type animals. These dsRNAs were likely to exist in the cytoplasm but not in the nucleus. Because ADBP-1 is predominantly localized in the nucleus, ADR-2 in the cytoplasm might be inactive and unable to edit these dsRNAs. One interesting possibility is that the role of ADBP-1 might be to restrict the ADR-2 activity to the nucleus and to allow the cytoplasmic RNAi processes to continue undisturbed. This may be important for the worms because in C. elegans RNAi was proven to function as an immune mechanism to RNA viruses that produce dsRNAs in the cytoplasm ( Wilkins et al. 2005). On the other hand, in mammals it was reported that ADAR1 acts in the cytoplasm and indeed inhibits RNAi ( Yang et al. 2005). Inhibition of RNAi by the cytoplasmic ADAR could be beneficial for organisms in which more sophisticated antiviral defense mechanisms, such as an interferon response, have developed ( Nishikura 2006). The fact that we could find no ADBP-1 homologs in other organisms, including mammals, might reflect the importance of cytoplasmic RNA editing in these organisms. In this article, we have reported a crucial RNA-editing regulator, ADBP-1. ADBP-1, acting together with ADR-2, controls transgene expression by antagonizing the siRNA pathway in C. elegans. It would be interesting to examine whether ADBP-1, ADR-2, and consequent RNA editing are also involved in the regulation of the miRNA pathways. ADBP-1 was shown to regulate the nuclear localization of ADR-2. In other organisms, the spatial control of RNA-editing activities would also be important for appropriate gene regulation. Although the roles and the regulatory mechanism of RNA editing in C. elegans require further studies, our results have provided insight into gene expression control through RNAi and RNA editing. We thank the Caenorhabditis elegans Genetics Center for worm strains, the Sanger Center for cosmids and YAC clones, Andy Fire for the various vectors, Bob Barstead for the two-hybrid library, Jun-ichi Miyazaki for the pCAGGS-2 expression vector, and Yuji Kohara for the cDNA clones. We also thank Tokio Tani, Isao Katsura, and Oliver Hobert for comments on the manuscript. This work was supported by Grants-in-Aid for Scientific Research (B) (18370004 to T.I.) for Young Scientists (B) (15770002 to M.F.) on Priority Areas-Molecular Brain Science (17024044 to T.I.) of the Ministry of Education, Culture, Sports, Science and Technology of Japan, as well as other grants from the Uehara Foundation (T.I.) and the Asahi Glass Foundation (T.I.). - Ambros, V., R. C. Lee, A. Lavanway, P. T. Williams and D. Jewell, 2003. MicroRNAs and other tiny endogenous RNAs in C. elegans. Curr. Biol. 13 807–818. [PubMed]
- Bargmann, C. I., and H. R. Horvitz, 1991. Chemosensory neurons with overlapping functions direct chemotaxis to multiple chemicals in C. elegans. Neuron 7 729–742. [PubMed]
- Bargmann, C. I., E. Hartwieg and H. R. Horvitz, 1993. Odorant-selective genes and neurons mediate olfaction in C. elegans. Cell 74 515–527. [PubMed]
- Blow, M. J., R. J. Grocock, S. Van Dongen, A. J. Enright, E. Dicks et al., 2006. RNA editing of human microRNAs. Genome Biol. 7 R27. [PubMed]
- Brenner, S., 1974. The genetics of Caenorhabditis elegans. Genetics 77 71–94. [PubMed]
- Burns, C. M., H. Chu, S. M. Rueter, L. K. Hutchinson, H. Canton et al., 1997. Regulation of serotonin-2C receptor G-protein coupling by RNA editing. Nature 387 303–308. [PubMed]
- Carthew, R. W., 2006. Gene regulation by microRNAs. Curr. Opin. Genet. Dev. 16 203–208. [PubMed]
- Cerutti, H., and J. A. Casas-Mollano, 2006. On the origin and functions of RNA-mediated silencing: from protists to man. Curr. Genet. 50 81–99. [PubMed]
- Cui, M., E. B. Kim and M. Han, 2006. Diverse chromatin remodeling genes antagonize the Rb-involved SynMuv pathways in C. elegans. PLoS Genet. 2 e74. [PubMed]
- Desterro, J. M., L. P. Keegan, M. Lafarga, M. T. Berciano, M. O'Connell et al., 2003. Dynamic association of RNA-editing enzymes with the nucleolus. J. Cell Sci. 116 1805–1818. [PubMed]
- Duchaine, T. F., J. A. Wohlschlegel, S. Kennedy, Y. Bei, D. Conte, Jr. et al., 2006. Functional proteomics reveals the biochemical niche of C. elegans DCR-1 in multiple small-RNA-mediated pathways. Cell 124 343–354. [PubMed]
- Feinberg, E. H., and C. P. Hunter, 2003. Transport of dsRNA into cells by the transmembrane protein SID-1. Science 301 1545–1547. [PubMed]
- Fong, Y., L. Bender, W. Wang and S. Strome, 2002. Regulation of the different chromatin states of autosomes and X chromosomes in the germ line of C. elegans. Science 296 2235–2238. [PubMed]
- Higuchi, M., F. N. Single, M. Kohler, B. Sommer, R. Sprengel et al., 1993. RNA editing of AMPA receptor subunit GluR-B: a base-paired intron-exon structure determines position and efficiency. Cell 75 1361–1370. [PubMed]
- Hsieh, J., J. Liu, S. A. Kostas, C. Chang, P. W. Sternberg et al., 1999. The RING finger/B-box factor TAM-1 and a retinoblastoma-like protein LIN-35 modulate context-dependent gene silencing in Caenorhabditis elegans. Genes Dev. 13 2958–2970. [PubMed]
- Kawahara, Y., B. Zinshteyn, P. Sethupathy, H. Iizasa, A. G. Hatzigeorgiou et al., 2007. Redirection of silencing targets by adenosine-to-inosine editing of miRNAs. Science 315 1137–1140. [PubMed]
- Kelly, W. G., and A. Fire, 1998. Chromatin silencing and the maintenance of a functional germline in Caenorhabditis elegans. Development 125 2451–2456. [PubMed]
- Ketting, R. F., T. H. Haverkamp, H. G. Van Luenen and R. H. Plasterk, 1999. Mut-7 of C. elegans, required for transposon silencing and RNA interference, is a homolog of Werner syndrome helicase and RNaseD. Cell 99 133–141. [PubMed]
- Ketting, R. F., S. E. Fischer, E. Bernstein, T. Sijen, G. J. Hannon et al., 2001. Dicer functions in RNA interference and in synthesis of small RNA involved in developmental timing in C. elegans. Genes Dev. 15 2654–2659. [PubMed]
- Kim, J. K., H. W. Gabel, R. S. Kamath, M. Tewari, A. Pasquinelli et al., 2005. Functional genomic analysis of RNA interference in C. elegans. Science 308 1164–1167. [PubMed]
- Knight, S. W., and B. L. Bass, 2002. The role of RNA editing by ADARs in RNAi. Mol. Cell 10 809–817. [PubMed]
- Lee, R. C., C. M. Hammell and V. Ambros, 2006. Interacting endogenous and exogenous RNAi pathways in Caenorhabditis elegans. RNA 12 589–597. [PubMed]
- Li, H., W. X. Li and S. W. Ding, 2002. Induction and suppression of RNA silencing by an animal virus. Science 296 1319–1321. [PubMed]
- Li, S., C. M. Armstrong, N. Bertin, H. Ge, S. Milstein et al., 2004. A map of the interactome network of the metazoan C. elegans. Science 303 540–543. [PubMed]
- Maduzia, L. L., T. L. Gumienny, C. M. Zimmerman, H. Wang, P. Shetgiri et al., 2002. lon-1 regulates Caenorhabditis elegans body size downstream of the dbl-1 TGF beta signaling pathway. Dev. Biol. 246 418–428. [PubMed]
- Mello, C. C., J. M. Kramer, D. Stinchcomb and V. Ambros, 1991. Efficient gene transfer in C.elegans: extrachromosomal maintenance and integration of transforming sequences. EMBO J. 10 3959–3970. [PubMed]
- Morse, D. P., and B. L. Bass, 1999. Long RNA hairpins that contain inosine are present in Caenorhabditis elegans poly(A)+ RNA. Proc. Natl. Acad. Sci. USA 96 6048–6053. [PubMed]
- Morse, D. P., P. J. Aruscavage and B. L. Bass, 2002. RNA hairpins in noncoding regions of human brain and Caenorhabditis elegans mRNA are edited by adenosine deaminases that act on RNA. Proc. Natl. Acad. Sci. USA 99 7906–7911. [PubMed]
- Nishikura, K., 2006. Editor meets silencer: crosstalk between RNA editing and RNA interference. Nat. Rev. Mol. Cell Biol. 7 919–931. [PubMed]
- Niwa, H., K. Yamamura and J. Miyazaki, 1991. Efficient selection for high-expression transfectants with a novel eukaryotic vector. Gene 108 193–199. [PubMed]
- Reenan, R. A., C. J. Hanrahan and G. Barry, 2000. The mle(napts) RNA helicase mutation in Drosophila results in a splicing catastrophe of the para Na+ channel transcript in a region of RNA editing. Neuron 25 139–149. [PubMed]
- Robert, V. J., T. Sijen, J. Van Wolfswinkel and R. H. Plasterk, 2005. Chromatin and RNAi factors protect the C. elegans germline against repetitive sequences. Genes Dev. 19 782–787. [PubMed]
- Sarot, E., G. Payen-Groschene, A. Bucheton and A. Pelisson, 2004. Evidence for a piwi-dependent RNA silencing of the gypsy endogenous retrovirus by the Drosophila melanogaster flamenco gene. Genetics 166 1313–1321. [PubMed]
- Scadden, A. D., and C. W. Smith, 2001. Specific cleavage of hyper-edited dsRNAs. EMBO J. 20 4243–4252. [PubMed]
- Shioi, G., M. Shoji, M. Nakamura, T. Ishihara, I. Katsura et al., 2001. Mutations affecting nerve attachment of Caenorhabditis elegans. Genetics 157 1611–1622. [PubMed]
- Simmer, F., M. Tijsterman, S. Parrish, S. P. Koushika, M. L. Nonet et al., 2002. Loss of the putative RNA-directed RNA polymerase RRF-3 makes C. elegans hypersensitive to RNAi. Curr. Biol. 12 1317–1319. [PubMed]
- Stinchcomb, D. T., J. E. Shaw, S. H. Carr and D. Hirsh, 1985. Extrachromosomal DNA transformation of Caenorhabditis elegans. Mol. Cell. Biol. 5 3484–3496. [PubMed]
- Tabara, H., M. Sarkissian, W. G. Kelly, J. Fleenor, A. Grishok et al., 1999. The rde-1 gene, RNA interference, and transposon silencing in C. elegans. Cell 99 123–132. [PubMed]
- Timmons, L., and A. Fire, 1998. Specific interference by ingested dsRNA. Nature 395 854. [PubMed]
- Tonkin, L. A., L. Saccomanno, D. P. Morse, T. Brodigan, M. Krause et al., 2002. RNA editing by ADARs is important for normal behavior in Caenorhabditis elegans. EMBO J. 21 6025–6035. [PubMed]
- Wagner, R. W., J. E. Smith, B. S. Cooperman and K. Nishikura, 1989. A double-stranded RNA unwinding activity introduces structural alterations by means of adenosine to inosine conversions in mammalian cells and Xenopus eggs. Proc. Natl. Acad. Sci. USA 86 2647–2651. [PubMed]
- Wang, D., S. Kennedy, D. Conte, Jr., J. K. Kim, H. W. Gabel et al., 2005. Somatic misexpression of germline P granules and enhanced RNA interference in retinoblastoma pathway mutants. Nature 436 593–597. [PubMed]
- Wicks, S. R., R. T. Yeh, W. R. Gish, R. H. Waterston and R. H. Plasterk, 2001. Rapid gene mapping in Caenorhabditis elegans using a high density polymorphism map. Nat. Genet. 28 160–164. [PubMed]
- Wilkins, C., R. Dishongh, S. C. Moore, M. A. Whitt, M. Chow et al., 2005. RNA interference is an antiviral defence mechanism in Caenorhabditis elegans. Nature 436 1044–1047. [PubMed]
- Yang, W., Q. Wang, K. L. Howell, J. T. Lee, D. S. Cho et al., 2005. ADAR1 RNA deaminase limits short interfering RNA efficacy in mammalian cells. J. Biol. Chem. 280 3946–3953. [PubMed]
- Yang, W., T. P. Chendrimada, Q. Wang, M. Higuchi, P. H. Seeburg et al., 2006. Modulation of microRNA processing and expression through RNA editing by ADAR deaminases. Nat. Struct. Mol. Biol. 13 13–21. [PubMed]
|
This article has been 
