TY - JOUR
AU - Wada, Masamitsu
AB - Abstract RNA interference (RNAi) has become a powerful tool for determining gene function and is used in a wide variety of organisms. Since it is necessary to generate double-stranded RNA (dsRNA) as an inducer for RNAi, preparation of RNAi-inducing constructs is somewhat cumbersome and time consuming, especially for the thousands of genes used in a genome-wide analysis. To overcome these problems, we have developed a more convenient gene-silencing method in the fern Adiantum using double-stranded DNA (dsDNA) as a model system for functional analysis in plants. Delivery of dsDNA fragments homologous to an endogenous gene into gametophytic cells can induce sequence-specific gene silencing. As it only requires dsDNA fragments homologous to a target gene, PCR-amplified fragments are enough to trigger gene silencing. Maximum gene silencing efficiencies of >90% have been achieved for transformed plants. In addition, simultaneous transfer of dsDNA fragments corresponding to multiple genes still has a silencing effect for individual genes. We term this approach ‘DNA interference’. (Received August 9, 2004; Accepted August 30, 2004) Introduction As a result of recent high-throughput sequencing projects for whole genomes and expressed sequence tags (ESTs), an impressive amount of sequence information for genes of unknown function is now available. Accordingly, reverse genetics suitable for high-throughput analysis becomes an increasingly important approach to identifying the role of these genes in the growth and development of plants. Recent advances in epigenetic gene silencing make this phenomenon possible to use as a reverse-genetics tool termed RNA interference (RNAi). The RNAi process is triggered by delivery of double-stranded RNA (dsRNA) into an organism or cell and results in sequence-specific mRNA degradation that effectively silences a target gene function (Hannon 2002). This degradation mechanism is versatile and evolutionarily conserved across the plant and animal kingdoms and is used as a tool for studying gene function in various experimental systems of seed plants, ferns and mosses (Waterhouse et al. 1998, Klink and Wolniak 2001, Bezanilla et al. 2003). Stable transformation of target plants with transgenes that encode self-complementary hairpin RNA (hpRNA) is a frequently used approach for generating dsRNA as an inducer of RNAi (Waterhouse et al. 1998, Chuang and Meyerowitz 2000). Generic vectors have been developed for this purpose and make it easier to prepare RNAi-inducing constructs than before (Wesley et al. 2001, Helliwell and Waterhouse 2003, Miki and Shimamoto 2004). Another approach for inducing RNAi is direct delivery of dsRNA or DNA constructs that encode hpRNA to plant cells and trigger transient suppression of gene function. Delivery of dsRNA or hpRNA-encoding plasmid by particle bombardment results in the reduced activity of the target gene in transformed cells of cereals (Schweizer et al. 2000) and moss (Bezanilla et al. 2003). Transient expression of hpRNA by Agrobacterium-mediated gene delivery also effectively induces RNAi in Nicotiana benthamiana (Johansen and Carrington 2001). In two fern species, Marsilea vestita and Ceratopteris richardii, RNAi has been found to occur by treating the dry spores with dsRNA at the time of imbibition (Klink and Wolniak 2001, Stout et al. 2003). In these currently used RNAi systems, cloning of target gene fragments into specific vectors is a commonly used process for generating dsRNA either in plants or in vitro. This procedure is somewhat onerous and requires time as the number of target genes grows. If an experimental system inducing gene silencing that omits these cloning steps could be developed, such a system would improve processing capabilities dramatically. Interestingly, in a few instances, it has been reported that promoterless double-stranded DNA (dsDNA) molecules can induce sequence-specific gene silencing in plants (Voinnet et al. 1998, Palauqui and Balzergue 1999, Rutherford et al. 2004). This phenomenon has been shown only in Nicotiana species (Voinnet et al. 1998, Palauqui and Balzergue 1999) and Ceratopteris (Rutherford et al. 2004) to date, and the efficiency of phenotype knockout for target genes was considerably lower than that induced by dsRNA. Consequently, development of this phenomenon into a simpler and more efficient method that could be applied to functional genomics has been eagerly awaited. In this study, we report that delivery of dsDNA fragments homologous to an endogenous gene efficiently cause a knockout phenotype of a corresponding gene in Adiantum. We term this phenomenon DNA interference (DNAi) by analogy with RNAi. Since PCR-amplified fragments are enough to trigger DNAi and simultaneously DNAi induction targeting multiple genes also effectively silences individual gene function, DNAi in Adiantum might facilitate high-throughput analysis of gene function in plants. Results and Discussion Introduction of promoterless dsDNAs cause knockout phenotypes of target genes in Adiantum Introduction of DNA fragments homologous to the sequence of the AcPHOT2 (Adiantum capillus-venerisPHOTOTROPIN2) gene (which encodes the photoreceptor controlling the chloroplast avoidance response under high intensity blue light in Adiantum; Kagawa et al. 2004) into Adiantum gametophytes, gave rise to the same phenotype as AcPHOT2-deficient mutants (Fig. 1A). This was achieved by introduction of plasmid DNA containing the promoterless, full-length AcPHOT2 cDNA coding sequence into young Adiantum gametophytes (three- or four-cell stage) by particle bombardment. A plasmid carrying hygromycin B phosphotransferase (hygr) gene driven by the cauliflower mosaic virus (CaMV) 35S promoter was simultaneously introduced as a selectable marker. Gametophytes with transient drug resistance were isolated within 9–14 d and after further incubation on non-selective medium, chloroplast relocation movement was analyzed. In the cells of gametophytes into which only the hygr gene had been introduced, the chloroplasts spread over the cell surface under low fluence-rate white light and moved to the anticlinal wall under high fluence-rate blue light, indicating that the chloroplast avoidance response occurred normally, similar to wild-type gametophytes (Fig. 1A). In contrast, in the cells of gametophytes into which both AcPHOT2 cDNA and the hygr gene had been introduced, the chloroplasts remained at the cell surface under high fluence-rate light (Fig. 1A). These observations indicate that the introduction of AcPHOT2 cDNA fragments prevents the chloroplast avoidance response under high fluence-rate light. We also investigated whether a DNA fragment derived from the AcPHY3 (A. capillus-veneris PHYTOCHROME3) gene, encoding the photoreceptor mediating red-light-induced phototropism and chloroplast relocation (Kawai et al. 2003), could lead to the same phenotype as a mutation in the AcPHY3 gene. A drug-resistant prothallus that was obtained by co-bombardment of plasmids containing promoterless, full-length AcPHY3 coding sequence and the hygr gene was cultured under unilateral red light to analyze red-light-induced phototropism. The growth direction of protonemata regenerated from the AcPHY3-introduced plant was variable, regardless of the direction of red light, while protonemata generated from a prothallus into which only the hygr gene had been introduced grew straight towards the red light source (Fig. 1B). In contrast, the blue light-mediated response was shown to be normal in AcPHY3-introduced gametophytes (data not shown), indicating that the defect in the phototropic response was specific to the red-light-mediated response. Chloroplast relocation was also examined in protonemal cells regenerated from drug-resistant prothalli. The cells were irradiated horizontally with polarized red or blue light that had a vertically vibrating electrical vector. In a protonemal cell of a hygr-introduced gametophyte, chloroplasts accumulated to the cell flank in response to both polarized red and blue light, as in the case of wild-type cells (Fig. 1C). However, in a cell of a gametophyte into which both AcPHY3 and hygr had been introduced, chloroplasts were distributed normally under polarized blue light, but randomly under polarized red light (Fig. 1C), indicating that no response was induced by red light. These results indicate that introduction of an AcPHY3 fragment caused the defective phenotype in both red-light-mediated phototropism and chloroplast relocation, which corresponds to the phenotype of AcPHY3-deficient mutants (Kadota and Wada 1999, Kawai et al. 2003). Introduction of plasmids containing a promoterless partial sequence of the AcFtsZ1 gene, which encodes the homolog of a chloroplast division-mediating protein (Osteryoung et al. 1998, Strepp et al. 1998) in Adiantum was also performed. Among drug-resistant gametophytes obtained by co-bombardment of the AcFtsZ1 sequence and the hygr gene, cells with abnormally large chloroplasts were frequently observed. Precise observation using a confocal laser scanning microscope revealed that sheet-like chloroplasts spread under the cell surface in these gametophytes, whereas gametophytes bombarded with the hygr gene alone possessed many normal, lens-shaped chloroplasts (Fig. 1D). These observations indicate that the partial fragments of the AcFtsZ1 gene can prevent chloroplast division, which is presumably mediated by the AcFtsZ1 protein. Taken together, these findings demonstrate that introduction of dsDNA fragments homologous to promoterless sequences of several genes causes epigenetic knockout phenotypes corresponding to the function of each gene in Adiantum. We call this phenomenon ‘DNAi’. Promoterless dsDNAs induce sequence-specific gene silencing To determine whether DNAi is caused by the altered transcript level of an endogenous gene, we investigated the transcript level of AcPHOT2 using reverse transcription-PCR (RT-PCR) in individual lines of gametophytes into which the promoterless, full-length AcPHOT2 cDNA fragment had been introduced. In these gametophytes, the defective phenotype in chloroplast avoidance response was confirmed. Complementary DNA was synthesized from total RNA extracted from gametophytes, which were further incubated on non-selective medium under continuous white light for 6 months. The transcript level of AcPHOT2 is shown as an amount of PCR product derived from AcPHOT2 relative to that of AcCRY2 (A. capillus-veneris CRYPTOCHROME2). The AcCRY2 gene was used as an internal standard since its transcript level is almost identical in various developmental stages and under different light treatments (Imaizumi et al. 2000). In these plants, the transcript level of AcPHOT2 was significantly lower than in plants transformed with the hygr gene alone, although both plants had equivalent levels of the closely related AcPHOT1 transcript (58.9% identical to AcPHOT2 at the nucleotide level; Nozue et al. 2000) (Fig. 2). A similar result of reduced transcript level induced by a specific promoterless DNA fragment was observed for the AcPHY3 gene (data not shown). These results show that the introduction of promoterless dsDNA fragments causes sequence-specific gene silencing, and consequently results in the knockout phenotypes of the target genes in Adiantum gametophytes. PCR-amplified dsDNAs, as well as linear and circular forms of plasmid dsDNAs, efficiently induce DNAi Next, we examined the efficiency of DNAi induced by both linear and circular forms of plasmids containing the promoterless AcPHOT2 cDNA sequence. After screening gametophytes for transient drug resistance, we investigated the percentage of gametophytes that showed little or no chloroplast avoidance response under high fluence-rate blue light. Approximately 98% of the gametophytes that were transformed with either linear or circular plasmids showed the defective phenotype in a chloroplast avoidance response assay (Table 1). We also tested the effect of PCR-amplified promoterless AcPHOT2 cDNA fragments. The PCR amplicons were as effective (98% defective phenotype) as plasmids amplified in Escherichia coli (Table 1). The efficiency of AcPHY3 PCR-amplified fragments was also examined. In 95% of the AcPHY3-introduced gametophytes, the chloroplast relocation movement induced by polarized red light was impaired (Table 1). These results indicate that DNAi in Adiantum appears to be highly efficient. Moreover, these results also demonstrate that PCR-amplified dsDNA fragments can serve as an effective and potent inducer of DNAi. Gene silencing mediated by PCR-amplified promoterless dsDNA has been reported previously in Nicotiana species (Voinnet et al. 1998, Palauqui and Balzergue 1999). When N. benthamiana plants stably transformed with a green fluorescent protein (GFP) gene were bombarded with PCR-amplified promoterless dsDNA fragments of the GFP gene, up to 41% of the plants showed suppressed GFP expression (Voinnet et al. 1998). Palauqui and Balzergue (1999) reported that 60% of N. tabacum plants bombarded with a PCR-amplified promoterless dsDNA fragment of a nitrate reductase gene showed a silencing effect in limited areas around the bombarded cells and this localized silencing spread systemically in 20% of the bombarded plants. Our results indicate that the efficiencies of DNAi induced by PCR-amplified dsDNA fragments in Adiantum are significantly higher than that of the other reports given above and can reach over 90%. At present, we are unable to assess whether gene silencing induced by PCR-amplified promoterless dsDNA is exceptionally efficient in Adiantum or not. One possibility is that the high efficiencies of DNAi in Adiantum gametophytes are attributed in part to the ease of observation of the phenotypic changes owing to their simple organization, which consists of uniform cells, not surrounded by any other tissue. It would appear that DNAi in Adiantum can provide sufficient efficiency for use in functional analyses. The ability to use PCR-amplified dsDNA fragments makes DNAi very much simpler and faster because of the removal of the need for a cloning step. Efficiency of DNAi depends on exon length in dsDNA fragments We further investigated the effect of dsDNA fragment length for DNAi. We tested a fragment containing the full-length cDNA coding sequence, fragments corresponding to the N- and C-terminal halves of the protein, and fragments corresponding to quarters of the protein (Fig. 3A). Similarly, the full length, halves, quarters and eighths of the genomic DNA comprising exons and introns were also tested (Fig. 3B). The results are summarized in Table 2. When approximately 1,500 bp and 850 bp fragments containing different parts of the AcPHOT2 cDNA were introduced, an average of 89.0% and 79.1%, respectively, of the drug-resistant plants showed the DNAi-induced phenotype. Individual cDNA fragments of about the same length gave similar results (Table 2), indicating that the sequence length rather than particular sequences (such as functional domain-encoded regions) accounts for the silencing efficiency as reported previously (Voinnet et al. 1998, Palauqui and Balzergue 1999). However, for genomic DNA, the efficiency varied with each fragment that had about the same length (Table 2). The efficiency was unlikely to depend on whether the fragments contained particular regions or not, as in the case of cDNA fragments. The efficiency depended, in part, on how much of the exon was included in each DNA fragment (Fig. 4). Additionally, a fragment containing only intron sequence (fragment g′′ of the genomic DNA) (see Fig. 3) had almost no effect on DNAi (Table 2). These observations indicate that silencing efficiency does not depend on whether the sequence is derived from cDNA or genomic DNA, but is influenced by how much coding sequence is included in the DNA fragments. Moreover, these results suggest that DNAi signal is likely to target a transcript derived from an endogenous gene. Further study of the mechanism of DNAi will allow us to assess this possibility. DNAi signal transmission Previous studies have shown that localized delivery of promoterless dsDNA molecules initiates the systemic spread of the silencing signal (Voinnet et al. 1998, Palauqui and Balzergue 1999, Rutherford et al. 2004). We investigated DNAi signal transmission in Adiantum gametophytes after the introduction of dsDNA fragments by bombardment. PCR-amplified fragments of promoterless, full-length AcPHOT2 cDNA and a plasmid carrying GFP marker gene as a visible marker were co-bombarded into prothallial cells. Three, 8 and 12 d after bombardment, prothalli with GFP-expressing cells were irradiated with high fluence-rate light to examine the DNAi effect. Impairment of the chloroplast avoidance response was observed in neighboring cells of the transformed cell on the 8th day, and spread over almost the whole prothalli on the 12th day (Fig. 5A). In a prothallus separated by band-like dead cells in the middle of the prothalli, the effect could only spread within the part that includes the transformed cell (Fig. 5B). In the cell of prothalli into which only the GFP gene was introduced, a normal response could be observed at any stage (Fig. 5C). These results indicate that the DNAi signal was transmitted through living cells and resulted in a systemic effect across the prothallus; however, the signal could not be transmitted through dead cells. Systemic signal transmission was also observed when the AcPHY3 gene was the target of DNAi (data not shown). In this case, the phenotypic changes are still observed in the gametophytes of the next generation, which were generated from self-fertilized sporophytes of two independent lines (data not shown). This observation indicates that the DNAi signal could be potentially transmitted to the next generation, although its mechanism has yet to be resolved. Availability of simultaneous induction of DNAi may facilitate high-throughput gene function analysis To obtain preliminary information towards applying the DNAi approach to high-throughput analyses of gene function, we examined the DNAi effects of multiple genes introduced simultaneously as a mixture of dsDNA fragments. Five kinds of dsDNA fragments of AcPHOT2, AcPHY3, AcPHY1, partial sequences of two myosin-homologous genes (Myo11–1 and Myo11–2), and the hygr gene (as a selectable marker) were bombarded into gametophytes as a mixture. Chloroplast relocation responses were examined in gametophytes selected by hygromycin resistance. In gametophytes in which DNAi of AcPHOT2 and AcPHY3 were induced independently, the defective chloroplast avoidance response and red-light-induced chloroplast relocation were observed, respectively (Fig. 6) as already shown above. When a mixture of dsDNA fragments homologous to the five genes was introduced, both the responses of chloroplast movement were shown to be impaired in each gametophyte selected (Fig. 6). Although we cannot assess the DNAi effects on the AcPHY1 gene and the myosin genes, as the precise function of these genes is unknown, the results show that simultaneous introduction of multiple, different dsDNA fragments is also effective in inducing silencing of multiply introduced genes in a single cell. This means that the technique of simultaneous introduction of dsDNA fragments corresponding to multiple genes could be useful for silencing genes belonging to multigene families that function redundantly. In addition, this method of multiple DNAi induction could be quite powerful for analyzing the functions of genes of unknown function accumulating as a result of EST and/or genome projects. Since an Adiantum EST database is now being established, a combination of EST sequences and DNAi should provide a comprehensive approach to studying gene function in Adiantum. The information accumulated by this approach will lead to significant insights into the molecular mechanism underlying the physiological responses of fern species, which is not sufficient so far. In particular, the availability of PCR-amplified dsDNA fragments makes this approach easy to use and may allow us to take a rapid step to elucidate the functions of genes of unknown function. Owing to its convenience, high efficiency and potential for silencing multiple genes simultaneously, Adiantum DNAi has the potential to be a model system for high-throughput, genome-wide studies towards understanding gene function in the growth and development of plants. Materials and Methods Plant materials Spores of A. capillus-veneris were harvested in the greenhouse of Tokyo Metropolitan University. Five milligrams of spores were sterilized with 10% (v/v) antiformin containing 0.1% (v/v) Triton X-100 and rinsed three times with sterilized deionized water. Spores suspended in sterilized, deionized water were spread evenly onto a cellophane sheet (4.5 cm in diameter) placed on the surface of solidified medium containing White’s basal salt mixture (W-0876; Sigma, St. Louis, MO, U.S.A.) and 0.5% agar in a 6 cm diameter Petri dish. After imbibition in darkness for 4 d, spores were cultured under unilateral red light (2.0–2.3 µmol m–2 s–1) for 3 d and subsequently grown under white light (20–30 µmol m–2 s–1) for 2 d. For analysis of DNAi signal transmission, 1.25 mg of spores were sown per dish and cultured under white light for 16 d to generate prothalli. The temperature was kept at 25°C. Bombarded DNA fragments With each bombardment, a plasmid carrying the hygromycin phosphotransferase gene (hygr) driven by the CaMV 35S promoter (pGEM-T Hyg; Imaizumi et al. 2002) was used as a selectable marker for DNA transfer. A modified GFP gene driven by the CaMV 35S promoter (psmRS-GFP; Davis and Vierstra 1998) was used as a visible maker. The AcPHOT2 cDNA fragment containing the full-length coding region (Kagawa et al. 2004) was cloned into pCR-Blunt II-TOPO (Invitrogen, Carlsbad, CA, U.S.A.) (pCR-cPHOT2), and both SmaI-digested (linear) and circular forms of pCR-cPHOT2 were used for bombardment. PCR-amplified fragments of AcPHOT2 cDNA were also used in the experiments. For this purpose, the full-length coding region of AcPHOT2 cDNA was amplified using pCR-cPHOT2 as a template with the following primers: 5′-TCTCTGCAGCGTTACCTTATAC-3′ and 5′-GGATTGATGAAGAAGAAGGC-3′. To obtain the full-length genomic DNA sequence of the AcPHOT2 gene, genomic PCR was performed using the same primer pair described above, and the amplified fragment was cloned into pCR-XL-TOPO (Invitrogen) (pCR-gPHOT2). To analyze the effect of introduced fragment length on DNAi, a series of AcPHOT2 fragments (see Fig. 3) was amplified from either pCR-cPHOT2 or pCR-gPHOT2. The full-length coding region of the AcPHY3 sequence (Nozue et al. 1998) cloned into the pGEM T-Easy vector (Promega, Madison, WI, U.S.A.) (pPHY3) was digested with both ApaI and SmaI and used for bombardment. The partial genomic sequence of AcFtsZ1 (829 bp; positions 1–829; Accession No. AB186124) cloned into pCR2.1-TOPO (Invitrogen) was also bombarded. For the study of DNAi using multiple gene fragments (see Fig. 6), PCR amplifications were performed on the following plasmids and primer pairs: as described above for the AcPHOT2 cDNA fragment (3,207 bp); on pPHY3 using 5′-GGAGAACAAGAGCAGTTTGCAGGAG-3′ and 5′-TCTCGTCCCGCATGACCTGTAG-3′ for the AcPHY3 fragment (3,090 bp); on pFP1 (Okamoto et al. 1993) using 5′-TCTGGAGGCTCTGGGAAAT-3′ and 5′-CTCTGTGAGCCAAAGGGAAC-3′ for the AcPHY1 cDNA fragment (3,307 bp). Plasmids containing partial sequences of Adiantum myosin cDNAs, Myo11–1 (1,524 bp; positions 400–1,923; AB185287) and Myo11–2 (1,228 bp; positions 546–1,773; AB185288), were also used for this multiple gene fragment transfer study. All plasmid DNAs were isolated using the Wizard Plus Maxipreps DNA Purification System (Promega). The enzyme-digested DNAs and the PCR-amplified fragments used for bombardment were further purified by phenol/chloroform extraction or using a PCR purification kit (Qiagen, Hilden, Germany), respectively. Particle bombardment and selection Gametophyte cultures were subjected to osmotic pretreatment by placing them on media containing 0.25 M mannitol for 1 h prior to bombardment performed with a Biolistic PDS-1000/He Particle Delivery System (Bio-Rad, Hercules, CA, U.S.A.). Delivery conditions were as described previously (Kawai et al. 2003). The amount of DNA delivered was 0.48 pmol per shot. After bombardment, gametophytes were incubated in darkness for 1 d and then cultured under white light for 4 d. Gametophytes were then transferred to selective medium (White’s medium containing 10 µg ml–1 hygromycin B) and cultured under white light for 9–14 d. After that, they were transferred to White’s medium without hygromycin B. Phenotypic analysis For analysis of chloroplast relocation movement and phototropism, gametophytes were placed onto the surface of White’s medium solidified with 0.5% agar in a 3 cm Petri dish and covered with a coverslip. Prior to the analysis of the chloroplast avoidance response, prothalli were kept in darkness for 3 d to test the ability of chloroplasts to move toward the anticlinal walls (the dark position) (Kagawa and Wada 1999). They were then irradiated with low fluence-rate white light to induce the chloroplast accumulation response, high fluence-rate blue or white light to induce the chloroplast avoidance response, and/or polarized red light to induce red-light-mediated chloroplast relocation. For analysis of red-light-induced phototropism, prothalli were cultured under unilateral red light to generate protonemal cells. These protonemal cells were further irradiated with white light (20–30 µmol m–2 s–1) for 6 h and then incubated in darkness for 2 d for subsequent use in chloroplast photorelocation analysis. Images were recorded using an Axioplan microscope (Zeiss, Oberkochen, Germany) system as described previously (Kawai et al. 2003). For determining DNAi of AcFtsZ1, the prothalli that had large chloroplasts were picked up under a binocular dissecting microscope and the fluorescence arising from chlorophyll was observed under a confocal laser scanning inverted microscope (LSM 410; Zeiss). The three-dimensional image was reconstructed using ImageJ software (http://rsb.info.nih.gov/ij/). Light sources To induce the chloroplast avoidance response, light-emitting diode blue light lamps (LED-mb; Eyela, Tokyo, Japan) or a metal halide lamp was used to obtain blue or white light, respectively. To induce the chloroplast accumulation response, fluorescent tubes (FL40SD or FL20SD; Toshiba Lighting and Technology Corp., Tokyo, Japan) were used to obtain white light. Red or blue light was obtained by passing light from a fluorescent tube of the same type through a red plastic filter (Shinkolite A, #102; Mitsubishi Rayon Corp., Tokyo, Japan) or a blue plastic film (#63; Ryudensha Corp., Tokyo, Japan), respectively. Polarized light was obtained by passing light through a linear polarizer (HN22; Polaroid Corp. of Japan, Tokyo, Japan). RT-PCR Total RNA was extracted from approximately 100 mg of prothalli using an RNeasy Plant Mini Kit (Qiagen). Samples were then reverse-transcribed from an oligo(dT) primer using the SUPERSCRIPT preamplification system for first-strand cDNA synthesis (Invitrogen). An aliquot of first-strand cDNA was used for PCR amplification with Ex Taq polymerase (Takara Bio, Otsu, Japan) and transcript-specific primers: 5′-GCAAAAGATTAGAGATGCCATCAG-3′ and 5′-CTTTACAGGTCGAAAATGTTTCAG-3′ for AcPHOT2; 5′-ATCACAGGATCTGGCCATAACTC-3′ and 5′-AGTACTCAATGGGATTGCTAGCC-3′ for AcPHOT1; and 5′-AGGCATTGGTCTCTAGCTATCG-3′ and 5′-CCTCATGTTACAACAACCCTACAGT-3′ for AcCRY2. The PCR schedule was 2 min at 94°C; 40 cycles of 15 s at 94°C, 30 s at 59°C and 30 s at 72°C; and then 2 min at 72°C. Equal volumes of amplified products were electrophoresed on an agarose gel and stained with ethidium bromide. The amount of each PCR product was quantified using the public domain National Institutes of Health (NIH) Image program (http://rsb.info.nih.gov/nih-image/). The amount of PCR products obtained using the AcPHOT2- and AcPHOT1-specific primers was normalized to that obtained using the AcCRY2-specific primers. In Fig. 2, the gel images in which loading volume for each lane was normalized to the amount of AcCRY2 product are shown. Acknowledgments We thank Dr. Jane Silverthorne and Dr. Steen Christensen for critical reading of the manuscript; Dr. Takatoshi Kagawa for pCR-cPHOT2; Dr. Fumio Takahashi for Myo11–1 and Myo11–2; Dr. Akeo Kadota and Dr. Yoshikatsu Sato for valuable suggestions; and Mr. Hidenori Tuboi, Mr. Noboru Yamada, Ms. Yoko Takahashi, Ms. Kayoko Hara, Mr. Eitetsu Sugiyama and Ms. Takako Yasuki for their technical assistance. This work was supported by BRAIN and Grant-in-Aid for Scientific Research (on Priority Areas, no. 13139203, and A, no. 13304061) from MEXT to M.W. and by a grant from JSPS Research Fellowships for Young Scientists to H.K.T. 3 These authors contributed equally to this work. 4 Present address: Department of Applied Biological Science, Graduate School of Agriculture, Tokyo University of Agriculture and Technology, Fuchuu, Tokyo, 183-8509 Japan 5 Corresponding author: E-mail, wada-masamitsu@c.metro-u.ac.jp; Fax, +81-426-77-2559. View largeDownload slide Fig. 1 Gene-specific phenotypes induced by DNAi. (A) Chloroplast relocation movement in a prothallus into which hygr was introduced (hygr) and, a prothallus into which both promoterless AcPHOT2 (SmaI restriction fragment in pCR-cPHOT2) and hygr were introduced (AcPHOT2 + hygr). The accumulation response was induced by irradiation of prothalli with low fluence-rate white light (weak light: 20–30 µmol m–2 s–1) for 1 d, and the avoidance response by irradiation with high fluence-rate blue light (strong light: 77–131 µmol m–2 s–1) for 1 h. The scale bar represents 20 µm. (B) Red light-induced phototropism in protonemal cells generated from a prothallus into which hygr was introduced (hygr) and a prothallus into which both promoterless AcPHY3 (ApaI–SmaI restriction fragment in pPHY3) and hygr were introduced (AcPHY3 + hygr). The prothalli were cultured under unilateral red light (2.0–10 µmol m–2 s–1) from the left for 14 d. The scale bar represents 200 µm. (C) Chloroplast relocation induced by polarized light in protonemal cells obtained by the same process as (B). The cells were horizontally irradiated for 3 h with vertically vibrating polarized red (1.7–2.7 µmol m–2 s–1) or blue (1.2–2.0 µmol m–2 s–1) light. The protonemal tip is toward the left. The scale bars represent 20 µm. (D) Chloroplasts of a prothallus into which hygr was introduced (hygr) and, a prothallus into which both promoterless AcFtsZ1 and hygr were introduced (AcFtsZ1 + hygr). The fluorescence arising from chlorophyll appears red. The scale bar represents 20 µm. View largeDownload slide Fig. 1 Gene-specific phenotypes induced by DNAi. (A) Chloroplast relocation movement in a prothallus into which hygr was introduced (hygr) and, a prothallus into which both promoterless AcPHOT2 (SmaI restriction fragment in pCR-cPHOT2) and hygr were introduced (AcPHOT2 + hygr). The accumulation response was induced by irradiation of prothalli with low fluence-rate white light (weak light: 20–30 µmol m–2 s–1) for 1 d, and the avoidance response by irradiation with high fluence-rate blue light (strong light: 77–131 µmol m–2 s–1) for 1 h. The scale bar represents 20 µm. (B) Red light-induced phototropism in protonemal cells generated from a prothallus into which hygr was introduced (hygr) and a prothallus into which both promoterless AcPHY3 (ApaI–SmaI restriction fragment in pPHY3) and hygr were introduced (AcPHY3 + hygr). The prothalli were cultured under unilateral red light (2.0–10 µmol m–2 s–1) from the left for 14 d. The scale bar represents 200 µm. (C) Chloroplast relocation induced by polarized light in protonemal cells obtained by the same process as (B). The cells were horizontally irradiated for 3 h with vertically vibrating polarized red (1.7–2.7 µmol m–2 s–1) or blue (1.2–2.0 µmol m–2 s–1) light. The protonemal tip is toward the left. The scale bars represent 20 µm. (D) Chloroplasts of a prothallus into which hygr was introduced (hygr) and, a prothallus into which both promoterless AcFtsZ1 and hygr were introduced (AcFtsZ1 + hygr). The fluorescence arising from chlorophyll appears red. The scale bar represents 20 µm. View largeDownload slide Fig. 2 Down-regulation of AcPHOT2 transcript level in prothalli into which promoterless AcPHOT2 (SmaI restriction fragment in pCR-cPHOT2) was introduced. RT-PCR analysis was performed using primers specific for AcPHOT2, AcPHOT1 and AcCRY2 in individual lines of prothalli into which the hygr gene had been introduced and into which both AcPHOT2 cDNA and the hygr gene had been introduced. The relative transcript level of each phototropin gene was calculated by normalizing the amount of PCR product to that of AcCRY2. View largeDownload slide Fig. 2 Down-regulation of AcPHOT2 transcript level in prothalli into which promoterless AcPHOT2 (SmaI restriction fragment in pCR-cPHOT2) was introduced. RT-PCR analysis was performed using primers specific for AcPHOT2, AcPHOT1 and AcCRY2 in individual lines of prothalli into which the hygr gene had been introduced and into which both AcPHOT2 cDNA and the hygr gene had been introduced. The relative transcript level of each phototropin gene was calculated by normalizing the amount of PCR product to that of AcCRY2. View largeDownload slide Fig. 3 PCR-amplified fragments corresponding to AcPHOT2 cDNA (A) and genomic DNA sequence (B) used for the experiment of the effect of dsDNA fragment length on DNAi. The corresponding region of each fragment is indicated by a thick line. Each fragment is labeled by a lower case letter. The black boxes in the schematic representations of cDNA (A) and gene structure (B) indicate the coding region. In (B), the boxes and the lines between the boxes indicate exons and introns, respectively. View largeDownload slide Fig. 3 PCR-amplified fragments corresponding to AcPHOT2 cDNA (A) and genomic DNA sequence (B) used for the experiment of the effect of dsDNA fragment length on DNAi. The corresponding region of each fragment is indicated by a thick line. Each fragment is labeled by a lower case letter. The black boxes in the schematic representations of cDNA (A) and gene structure (B) indicate the coding region. In (B), the boxes and the lines between the boxes indicate exons and introns, respectively. View largeDownload slide Fig. 4 Relationship between the total length of the introduced sequence corresponding to exon sequence and silencing efficiency. The silencing efficiency of dsDNA fragments homologous to genomic DNA and cDNA of the AcPHOT2 gene are represented by yellow squares and blue circles, respectively. Each point represents the efficiency for a given gene transfer, and was obtained from 3–64 individuals. View largeDownload slide Fig. 4 Relationship between the total length of the introduced sequence corresponding to exon sequence and silencing efficiency. The silencing efficiency of dsDNA fragments homologous to genomic DNA and cDNA of the AcPHOT2 gene are represented by yellow squares and blue circles, respectively. Each point represents the efficiency for a given gene transfer, and was obtained from 3–64 individuals. View largeDownload slide Fig. 5 The time course of DNAi signal transmission. Both promoterless AcPHOT2 cDNA (PCR-amplified fragment) and a GFP marker gene (A, B) or GFP gene only (C) were introduced into prothallial cells. The green fluorescence in the leftmost image indicates GFP expression in a cell into which DNAs were introduced. The red fluorescence in the image arises from chlorophyll. Three, 8 and 12 d after bombardment, the prothalli were irradiated with high fluence-rate white light (500–800 µmol m–2 s–1) for 3 h to analyze the chloroplast avoidance response. The scale bar represents 200 µm. View largeDownload slide Fig. 5 The time course of DNAi signal transmission. Both promoterless AcPHOT2 cDNA (PCR-amplified fragment) and a GFP marker gene (A, B) or GFP gene only (C) were introduced into prothallial cells. The green fluorescence in the leftmost image indicates GFP expression in a cell into which DNAs were introduced. The red fluorescence in the image arises from chlorophyll. Three, 8 and 12 d after bombardment, the prothalli were irradiated with high fluence-rate white light (500–800 µmol m–2 s–1) for 3 h to analyze the chloroplast avoidance response. The scale bar represents 200 µm. View largeDownload slide Fig. 6 Effect of DNAi with multiple gene fragments. The hygr gene (hygr), AcPHOT2 and hygr (AcPHOT2 + hygr), AcPHY3 and hygr (AcPHY3 + hygr), or AcPHOT2, AcPHY3, AcPHY1, Myo11–1, Myo11–2 and hygr (AcPHOT2 + AcPHY3 + 3 fragments + hygr) were introduced into prothalli. Chloroplast relocation movement was induced by irradiation of prothalli with low fluence-rate white light (weak light: 20–30 µmol m–2 s–1) for 30 min, high fluence-rate white light (strong light: 500–800 µmol m–2 s–1) for 1 h, and vertically vibrating polarized red light from the horizontal direction (polarized red light: 0.43–0.55 µmol m–2 s–1) for 1 h. The scale bar represents 20 µm. View largeDownload slide Fig. 6 Effect of DNAi with multiple gene fragments. The hygr gene (hygr), AcPHOT2 and hygr (AcPHOT2 + hygr), AcPHY3 and hygr (AcPHY3 + hygr), or AcPHOT2, AcPHY3, AcPHY1, Myo11–1, Myo11–2 and hygr (AcPHOT2 + AcPHY3 + 3 fragments + hygr) were introduced into prothalli. Chloroplast relocation movement was induced by irradiation of prothalli with low fluence-rate white light (weak light: 20–30 µmol m–2 s–1) for 30 min, high fluence-rate white light (strong light: 500–800 µmol m–2 s–1) for 1 h, and vertically vibrating polarized red light from the horizontal direction (polarized red light: 0.43–0.55 µmol m–2 s–1) for 1 h. The scale bar represents 20 µm. Table 1 Efficiency of DNAi induced by plasmids and PCR-amplified fragments Bombarded dsDNA fragment  Silencing efficiency (%) a  nb  AcPHOT2 cDNA fragment in linearized plasmid  98.1 ± 2.5  133  AcPHOT2 cDNA fragment in circular plasmid  98.1 ± 3.8  121  AcPHOT2 cDNA PCR-amplified fragment  98.1 ± 3.7  102  AcPHY3 cDNA PCR-amplified fragment  95.3 ± 1.1  107  Bombarded dsDNA fragment  Silencing efficiency (%) a  nb  AcPHOT2 cDNA fragment in linearized plasmid  98.1 ± 2.5  133  AcPHOT2 cDNA fragment in circular plasmid  98.1 ± 3.8  121  AcPHOT2 cDNA PCR-amplified fragment  98.1 ± 3.7  102  AcPHY3 cDNA PCR-amplified fragment  95.3 ± 1.1  107  a The silencing efficiency is defined as the percentage of individuals that showed a defect in chloroplast avoidance response (for AcPHOT2) or chloroplast relocation induced by polarized red light (for AcPHY3) in total prothalli examined. Each efficiency value is the mean ± standard deviation determined from two to four gene transfer events, each of which generated 13–55 prothalli for analysis. b Total number of drug-resistant prothalli used for analysis. View Large Table 2 Efficiency of DNAi induced by PCR-amplified dsDNA fragments corresponding to various regions of the AcPHOT2 gene Bombarded dsDNA a    Corresponding region b  Fragment length (bp)  Total exon (bp)  Silencing efficiency (%) c  nd  cDNA              Full length    –93 to +3,114  3,207  3,207  98.1 ± 3.7  102  1/2  a  –93 to +1,571  1,664  1,664  93.3 ± 10.3  138    b  +1,626 to +3,114  1,489  1,489  81.9 ± 8.8  65  1/4  a′  –93 to +762  855  855  80.4 ± 14.2  155    b′  +726 to +1,571  846  846  78.5 ± 9.9  163    c′  +1,626 to +2,472  847  847  72.1 ± 5.4  207    d′  +2,250 to +3,114  865  865  85.2 ± 7.9  237  Genomic DNA              Full length    –93 to +11,154  11,247  3,207  94.2 ± 7.0  109  1/2  a  –93 to +5,456  5,549  2,145  99.2 ± 1.7  116    b  +5,478 to +11,154  5,677  1,041  84.5 ± 11.4  187  1/4  a′  –93 to +2,728  2,821  1,682  96.7 ± 4.2  186    b′  +2,593 to +5,456  2,864  472  79.3 ± 4.0  178    c′  +5,478 to +8,337  2,860  723  72.6 ± 26.0  169    d′  +8,330 to +11,154  2,825  318  48.1 ± 10.1  68  1/8  a′′  –93 to +1,408  1,501  866  70.4 ± 17.5  167    b′′  +1,233 to +2,728   1,496  816  61.5 ± 17.4  113    c′′  +2,593 to +4,012  1,420  116  38.9 ± 21.6  175    d′′  +4,024 to +5,456  1,437  356  45.5 ± 23.4  150    e′′  +5,478 to +6,839  1,362  171  47.6 ± 7.1  153    f′′  +6,861 to +8,337  1,477  547  67.2 ± 11.7  129    g′′  +8,330 to +9,758  1,429  0  1.3 ± 2.1  136    h′′  +9,736 to +11,154  1,419  318  33.1 ± 20.1  175  hygr only (Control)          0  498  Bombarded dsDNA a    Corresponding region b  Fragment length (bp)  Total exon (bp)  Silencing efficiency (%) c  nd  cDNA              Full length    –93 to +3,114  3,207  3,207  98.1 ± 3.7  102  1/2  a  –93 to +1,571  1,664  1,664  93.3 ± 10.3  138    b  +1,626 to +3,114  1,489  1,489  81.9 ± 8.8  65  1/4  a′  –93 to +762  855  855  80.4 ± 14.2  155    b′  +726 to +1,571  846  846  78.5 ± 9.9  163    c′  +1,626 to +2,472  847  847  72.1 ± 5.4  207    d′  +2,250 to +3,114  865  865  85.2 ± 7.9  237  Genomic DNA              Full length    –93 to +11,154  11,247  3,207  94.2 ± 7.0  109  1/2  a  –93 to +5,456  5,549  2,145  99.2 ± 1.7  116    b  +5,478 to +11,154  5,677  1,041  84.5 ± 11.4  187  1/4  a′  –93 to +2,728  2,821  1,682  96.7 ± 4.2  186    b′  +2,593 to +5,456  2,864  472  79.3 ± 4.0  178    c′  +5,478 to +8,337  2,860  723  72.6 ± 26.0  169    d′  +8,330 to +11,154  2,825  318  48.1 ± 10.1  68  1/8  a′′  –93 to +1,408  1,501  866  70.4 ± 17.5  167    b′′  +1,233 to +2,728   1,496  816  61.5 ± 17.4  113    c′′  +2,593 to +4,012  1,420  116  38.9 ± 21.6  175    d′′  +4,024 to +5,456  1,437  356  45.5 ± 23.4  150    e′′  +5,478 to +6,839  1,362  171  47.6 ± 7.1  153    f′′  +6,861 to +8,337  1,477  547  67.2 ± 11.7  129    g′′  +8,330 to +9,758  1,429  0  1.3 ± 2.1  136    h′′  +9,736 to +11,154  1,419  318  33.1 ± 20.1  175  hygr only (Control)          0  498  a Each fragment is represented in Fig. 3. b The corresponding region of each fragment is indicated by the positions of both of its ends relative to the first nucleotide (+1 in Fig. 3) of the start codon. c The silencing efficiency is defined as the percentage of chloroplast avoidance response-deficient individuals in total prothalli examined. Each efficiency value is the mean ± standard deviation determined from three to seven gene transfer events, each of which generated 3–64 prothalli for analysis. d Total number of drug-resistant prothalli used for analysis. View Large Abbreviations CaMV cauliflower mosaic virus DNAi DNA interference dsDNA double-stranded DNA dsRNA double-stranded RNA EST expressed sequence tag GFP green fluorescent protein hpRNA hairpin RNA RNAi RNA interference RT-PCR reverse transcription-PCR. References Bezanilla, M., Pan, A. and Quatrano, R.S. ( 2003) RNA interference in the moss Physcomitrella patens. Plant Physiol.  133: 470–474. Google Scholar Chuang, C.-F. and Meyerowitz, E.M. ( 2000) Specific and heritable genetic interference by double-stranded RNA in Arabidopsis thaliana. Proc. Natl Acad. Sci. USA  97: 4985–4990. Google Scholar Davis, S.J. and Vierstra, R.D. ( 1998) Soluble, highly fluorescent variants of green fluorescent protein (GFP) for use in higher plants. Plant Mol. Biol.  36: 521–528. Google Scholar Hannon, G.J. ( 2002) RNA interference. Nature  418: 244–251. Google Scholar Helliwell, C. and Waterhouse, P. ( 2003) Constructs and methods for high-throughput gene silencing in plants. Methods  30: 289–295. Google Scholar Imaizumi, T., Kadota, A., Hasebe, M. and Wada, M. ( 2002) Cryptochrome light signals control development to suppress auxin sensitivity in the moss Physcomitrella patens. Plant Cell  14: 373–386. Google Scholar Imaizumi, T., Kanegae, T. and Wada, M. ( 2000) Cryptochrome nucleocytoplasmic distribution and gene expression are regulated by light quality in the fern Adiantum capillus-veneris. Plant Cell  12: 81–95. Google Scholar Johansen, L.K. and Carrington, J.C. ( 2001) Silencing on the spot. Induction and suppression of RNA silencing in the Agrobacterium-mediated transient expression system. Plant Physiol.  126: 930–938. Google Scholar Kadota, A. and Wada, M. ( 1999) Red light-aphototropic (rap) mutants lack red light-induced chloroplast relocation movement in the fern Adiantum capillus-veneris. Plant Cell Physiol.  40: 238–247. Google Scholar Kagawa, T., Kasahara, M., Abe, T., Yoshida, S. and Wada, M. ( 2004) Function analysis of phototropin2 using fern mutants deficient in blue light-induced chloroplast avoidance movement. Plant Cell Physiol.  45: 416–426. Google Scholar Kagawa, T. and Wada, M. ( 1999) Chloroplast-avoidance response induced by high-fluence blue light in prothallial cells of the fern Adiantum capillus-veneris as analyzed by microbeam irradiation. Plant Physiol.  119: 917–923. Google Scholar Kawai, H., Kanegae, T., Christensen, S., Kiyosue, T., Sato, Y., Imaizumi, T., Kadota, A. and Wada, M. ( 2003) Responses of ferns to red light are mediated by an unconventional photoreceptor. Nature  421: 287–290. Google Scholar Klink, V.P. and Wolniak, S.M. ( 2001) Centrin is necessary for the formation of the motile apparatus in spermatids of Marsilea. Mol. Biol. Cell  12: 761–776. Google Scholar Miki, D. and Shimamoto, K. ( 2004) Simple RNAi vectors for stable and transient suppression of gene function in rice. Plant Cell Physiol.  45: 490–495. Google Scholar Nozue, K., Christie, J.M., Kiyosue, T., Briggs, W.R. and Wada, M. ( 2000) Isolation and characterization of a fern phototropin (Accession No. AB037188), A putative blue-light photoreceptor for phototropism (PGR00–039). Plant Physiol.  122: 1457. Google Scholar Nozue, K., Kanegae, T., Imaizumi, T., Fukuda, S., Okamoto, H., Yeh, K.-C., Lagarias, J.C. and Wada, M. ( 1998) A phytochrome from the fern Adiantum with features of the putative photoreceptor NPH1. Proc. Natl Acad. Sci. USA  95: 15826–15830. Google Scholar Okamoto, H., Hirano, Y., Abe, H., Tomizawa, K., Furuya, M. and Wada, M. ( 1993) The deduced amino acid sequence of phytochrome from Adiantum includes consensus motifs present in phytochrome B from seed plants. Plant Cell Physiol.  34: 1329–1334. Google Scholar Osteryoung, K.W., Stokes, K.D., Rutherford, S.M., Percival, A.L. and Lee, W.Y. ( 1998) Chloroplast division in higher plants requires members of two functionally divergent gene families with homology to bacterial ftsZ. Plant Cell  10: 1991–2004. Google Scholar Palauqui, J.-C. and Balzergue, S. ( 1999) Activation of systemic acquired silencing by localised introduction of DNA. Curr. Biol.  9: 59–66. Google Scholar Rutherford, G., Tanurdzic, M., Hasebe, M. and Banks, J.A. ( 2004) A systemic gene silencing method suitable for high throughput, reverse genetic analyses of gene function in fern gametophytes. BMC Plant Biol.  4: 6. Google Scholar Schweizer, P., Pokorny, J., Schulze-Lefert, P. and Dudler, R. ( 2000) Double-stranded RNA interferes with gene function at the single-cell level in cereals. Plant J.  24: 895–903. Google Scholar Stout, S.C., Clark, G.B., Archer-Evans, S. and Roux, S.J. ( 2003) Rapid and efficient suppression of gene expression in a single-cell model system, Ceratopteris richardii. Plant Physiol.  131: 1165–1168. Google Scholar Strepp, R., Scholz, S., Kruse, S., Speth, V. and Reski, R. ( 1998) Plant nuclear gene knockout reveals a role in plastid division for the homolog of the bacterial cell division protein FtsZ, an ancestral tubulin. Proc. Natl Acad. Sci. USA  95: 4368–4373. Google Scholar Voinnet, O., Vain, P., Angell, S. and Baulcombe, D.C. ( 1998) Systemic spread of sequence-specific transgene RNA degradation in plants is initiated by localized introduction of ectopic promoterless DNA. Cell  95: 177–187. Google Scholar Waterhouse, P.M., Graham, M.W. and Wang, M.B. ( 1998) Virus resistance and gene silencing in plants can be induced by simultaneous expression of sense and antisense RNA. Proc. Natl Acad. Sci. USA  95: 13959–13964. Google Scholar Wesley, S.V., Helliwell, C.A., Smith, N.A., Wang, M.-B., Rouse, D.T., Liu, Q., Gooding, P.S., Singh, S.P., Abbott, D., Stoutjesdijk, P.A., Robinson, S.P., Gleave, A.P., Green, A.G. and Waterhouse, P.M. ( 2001) Construct design for efficient, effective and high-throughput gene silencing in plants. Plant J.  27: 581–590. Google Scholar
TI - DNA Interference: a Simple and Efficient Gene-Silencing System for High-Throughput Functional Analysis in the Fern Adiantum
JF - Plant and Cell Physiology
DO - 10.1093/pcp/pch186
DA - 2004-11-15
UR - https://www.deepdyve.com/lp/oxford-university-press/dna-interference-a-simple-and-efficient-gene-silencing-system-for-high-Fp6QdbYQ5z
SP - 1648
EP - 1657
VL - 45
IS - 11
DP - DeepDyve
ER -