Matches in SemOpenAlex for { <https://semopenalex.org/work/W2042129286> ?p ?o ?g. }
- W2042129286 endingPage "20075" @default.
- W2042129286 startingPage "20069" @default.
- W2042129286 abstract "Pokeweed antiviral protein (PAP) is a ribosome-inactivating protein isolated from the pokeweed plant (Phytolacca americana) that inhibits the proliferation of several plant and animal viruses. We have shown previously that PAP and nontoxic mutants of PAP can directly depurinate brome mosaic virus (BMV) RNA in vitro, resulting in reduced viral protein translation. Here we expand on these initial studies and, using a barley protoplast system, demonstrate that recombinant PAP and nontoxic mutants isolated from E. coli are able to reduce the accumulation of BMV RNAs in vivo. Pretreatment of only BMV RNA3 with PAP prior to transfection of barley protoplasts reduced the accumulation of all BMV RNAs, with a more severe effect on subgenomic RNA4 levels. Using in vitro RNA synthesis assays, we show that a depurinated template causes the BMV replicase to stall at the template nucleotide adjacent to the missing base. These results provide new insight into the antiviral mechanism of PAP, namely that PAP depurination of BMV RNA impedes both RNA replication and subgenomic RNA transcription. These novel activities are distinct from the PAP-induced reduction of viral RNA translation and represent new targets for the inhibition of viral infection. Pokeweed antiviral protein (PAP) is a ribosome-inactivating protein isolated from the pokeweed plant (Phytolacca americana) that inhibits the proliferation of several plant and animal viruses. We have shown previously that PAP and nontoxic mutants of PAP can directly depurinate brome mosaic virus (BMV) RNA in vitro, resulting in reduced viral protein translation. Here we expand on these initial studies and, using a barley protoplast system, demonstrate that recombinant PAP and nontoxic mutants isolated from E. coli are able to reduce the accumulation of BMV RNAs in vivo. Pretreatment of only BMV RNA3 with PAP prior to transfection of barley protoplasts reduced the accumulation of all BMV RNAs, with a more severe effect on subgenomic RNA4 levels. Using in vitro RNA synthesis assays, we show that a depurinated template causes the BMV replicase to stall at the template nucleotide adjacent to the missing base. These results provide new insight into the antiviral mechanism of PAP, namely that PAP depurination of BMV RNA impedes both RNA replication and subgenomic RNA transcription. These novel activities are distinct from the PAP-induced reduction of viral RNA translation and represent new targets for the inhibition of viral infection. Pokeweed antiviral protein (PAP) 1The abbreviations used are: PAP, pokeweed antiviral protein; BMV, brome mosaic virus; PEG, polyethylene glycol; MES, 4-morpholineethanesulfonic acid. is a 29-kDa ribosome-inactivating protein of the pokeweed plant Phytolacca americana. Since its initial description as an antiviral agent against tobacco mosaic virus (1Duggar B.M. Armstrong J.K. Ann. Mo. Bot. Gard. 1925; 12: 359-366Crossref Google Scholar), PAP has been demonstrated to reduce the propagation of several plant and animal viruses, including potato virus X, HIV, and influenza (2Tumer N.E. Hwang D.J. Bonness M. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 3866-3871Crossref PubMed Scopus (130) Google Scholar, 3Zarling J.M. Moran P.A. Haffar O. Sias J. Richman D.D. Spina C.A. Myers D.E. Kuebelbeck V. Ledbetter J.A. Uckun F.M. Nature. 1990; 347: 92-95Crossref PubMed Scopus (195) Google Scholar, 4Tomlinson J.A. Walker V.M. Flewtett T.H. Barclay G.R. J. Gen. Virol. 1974; 22: 225-232Crossref PubMed Scopus (88) Google Scholar). It therefore holds promise as a broad-spectrum antiviral agent. Years after its initial discovery, the enzymatic activity of PAP was characterized as an N-glycosylase (5Endo Y. Tsurugi K. Lambert J.M. Biochem. Biophys. Res. Commun. 1988; 150: 1032-1036Crossref PubMed Scopus (177) Google Scholar). Like all ribosome-inactivating proteins, PAP efficiently removes a conserved adenine from the sarcin/ricin loop within domain VI of the large ribosomal RNA (6Endo Y. Tsurugi K. J. Biol. Chem. 1987; 262: 8128-8130Abstract Full Text PDF PubMed Google Scholar, 7Stirpe F. Bailey S. Miller S.P. Bodley J.W. Nucleic Acids Res. 1988; 16: 405-412Crossref Google Scholar). This depurination slows the elongation step of protein synthesis and is considered to be the reason for cytotoxicity of the protein (reviewed in Refs. 8Wang M. Hudak K.A. Genet. Eng. (N. Y.). 2003; 25: 143-161PubMed Google Scholar and 9Tumer N.E. Hudak K. Di R. Coetzer C. Wang P. Zoubenko O. Curr. Top. Microbiol. Immunol. 1999; 240: 139-158PubMed Google Scholar). The accompanying decline in cellular protein translation may cause local cell death and limit virus propagation (10Ready M.P. Brown D.T. Robertus J.D. Proc. Natl. Acad. Sci. U. S. A. 1986; 84: 5053-5056Crossref Scopus (118) Google Scholar). This model is supported by observations showing a positive correlation between ribosome depurination and inhibition of virus infection (11Taylor S. Massiah A. Lomonossoff G. Roberts L.M. Lord J.M. Hartley M. Plant J. 1994; 5: 827-835Crossref PubMed Scopus (106) Google Scholar). The accompanying decline in cellular protein translation, as a result of depurination, is often cited as the cause of antiviral activity. For example, reduction of poliovirus infection of HeLa cells incubated with PAP was attributed to inhibition of translation in virus-infected cells (12Ussery M.A. Irvin J.D. Hardesty B. Ann. N. Y. Acad. Sci. 1977; 284: 431-440Crossref PubMed Scopus (68) Google Scholar). In addition, inhibition of tobacco mosaic virus multiplication in tobacco protoplasts correlated well with PAP-mediated inhibition of translation (13Watanabe K. Kawasaki T. Sako N. Funatsu G. Biosci. Biotech. Biochem. 1997; 61: 994-997Crossref PubMed Scopus (19) Google Scholar). More recent results have revealed that many ribosome-inactivating proteins are capable of depurinating RNA substrates apart from the rRNA (14Bolognesi A. Polito L. Lubelli C. Barbieri L. Parente A. Stirpe F. J. Biol. Chem. 2002; 277: 13709-13716Abstract Full Text Full Text PDF PubMed Scopus (56) Google Scholar, 15Nicolas E. Beggs J.M. Haltiwanger B.M. Taraschi T.F. J. Biol. Chem. 1998; 273: 17216-17220Abstract Full Text Full Text PDF PubMed Scopus (57) Google Scholar, 16Wang P. Tumer N.E. Nucleic Acids Res. 1999; 27: 1900-1905Crossref PubMed Scopus (59) Google Scholar). Rajamohan et al. (17Rajamohan F. Venkatachalam T.K. Irvin J.D. Uckun F.M. Biochem. Biophys. Res. Commun. 1999; 260: 453-458Crossref PubMed Scopus (84) Google Scholar) showed that PAP removes both adenines and guanines from HIV-1 when incubated in vitro with the genomic viral RNA. In addition, Hudak et al. (18Hudak K.A. Wang P. Tumer N.E. RNA. 2000; 6: 369-380Crossref PubMed Scopus (92) Google Scholar) have shown that PAP and nontoxic PAP mutants depurinate brome mosaic virus (BMV) RNAs in vitro and that this depurination inhibits their translation in a cell-free system. Therefore, the direct depurination of viral RNAs by PAP may contribute to its antiviral activity. BMV is a model positive-strand RNA virus with a genome composed of three positive sense RNAs designated RNA1, RNA2, and RNA3. Each RNA is 5′-capped and contains a conserved 200-nucleotide tRNA-like structure at the 3′-end (reviewed in Refs. 19Ahlquist P. Curr. Opin. Genet. Dev. 1992; 2: 71-76Crossref PubMed Scopus (139) Google Scholar and 20Kao C.C. Sivakumaran K. Mol. Plant Pathol. 2000; 1: 91-97Crossref PubMed Scopus (83) Google Scholar). RNA1 is monocistronic and encodes a 1a protein containing an N-terminal domain with similarity to m7G methyltransferases involved in viral RNA capping and a C-terminal domain with similarity to RNA helicases (21Kong F. Sivakumaran K. Kao C.C. Virology. 1999; 259: 200-210Crossref PubMed Scopus (64) Google Scholar, 22Ahola T. den Boon J.A. Ahlquist P. J. Virol. 2000; 74: 8803-8811Crossref PubMed Scopus (87) Google Scholar). RNA2 is also monocistronic and encodes a 2a protein that has all of the motifs expected of RNA-dependent RNA polymerases (23Argos P. Nucleic Acids Res. 1988; 16: 9909-9916Crossref PubMed Scopus (298) Google Scholar). RNA3 is dicistronic and encodes a movement protein and a coat protein that is translated from a subgenomic RNA4 (24Miller W.A. Dreher T.W. Hall T.C. Nature. 1985; 313: 68-72Crossref PubMed Scopus (218) Google Scholar, 25Allison R. Thompson C. Ahlquist P. Proc. Natl. Acad. Sci. U. S. A. 1990; 87: 1820-1824Crossref PubMed Scopus (154) Google Scholar, 26Schmitz I. Rao A.L.N. Virology. 1996; 226: 281-293Crossref PubMed Scopus (79) Google Scholar). Synthesis of the RNAs therefore involves replication of negative and positive strand RNAs and transcription of subgenomic RNA. In this report, we expand on our initial in vitro studies to show evidence that PAP and nontoxic PAP mutants inhibit the replication and transcription of BMV RNAs in barley protoplasts. The inhibition caused by PAP is not due to ribosome depurination or decline of cellular translation. Rather, PAP and nontoxic mutants reduced the accumulation of BMV RNAs in protoplasts, by inhibiting both viral RNA replication and transcription. Furthermore, depurinated RNAs were shown to prevent efficient elongative RNA synthesis by the BMV replicase in vitro. Cloning and Expression of PAPs in E. coli—The mature form of wild-type PAP was amplified from pNT188, a yeast vector expressing the complete unprocessed form of PAP. The 5′ primer (CATGGATCCGTGAATACAATC) was designed to begin at Val23, and the 3′ primer (CCAAGCTTGTTAAGTTGTCTGACAGCTCCC) was designed to stop at Thr284, thereby amplifying the mature, processed form of the protein present in eukaryotes. The mutants of PAP, namely PAPx, PAPn, and PAPc, were amplified with the same primers, and all PCR products were cloned into the expression vector pET30a (Novagen) at NdeI and HindIII sites. All constructs were confirmed by DNA sequencing and transformed into BL21 cells. The overexpressed wild-type and mutant forms of PAP were purified by affinity chromatography on a Ni2+-nitrilotriacetic acid column. Fractions containing PAP were pooled and concentrated by filtration centrifugation with a 10-kDa cut-off filter (Amicon). Purified proteins were separated by 12% SDS-PAGE and stained with Coomassie Blue. RNase Activity Assay—To determine whether ribonucleases co-purified with preparations of PAP and PAP mutants from E. coli, an endoribonuclease assay was adapted from Bhardwaj et al. (27Bhardwaj K. Guarino L. Kao C.C. J. Virol. 2004; 78: 12218-12224Crossref PubMed Scopus (148) Google Scholar). A chemically synthesized RNA template (Dharmacon, Inc.) of 10 nucleotides was 5′-end-labeled with T4 polynucleotide kinase and [α-32P]ATP. Approximately 100 ng of RNA substrate was incubated with 50 ng of PAP or PAP mutants in 100 mm KCl, 50 mm Tris-HCl, pH 7.5, 4 mm MgCl2, and 1 mm dithiothreitol at 30 °C for 30 min. RNA incubated without protein was used as a negative control, and the endoribonuclease of the SARS coronavirus, Nsp15 (50 ng), was used as a positive control (27Bhardwaj K. Guarino L. Kao C.C. J. Virol. 2004; 78: 12218-12224Crossref PubMed Scopus (148) Google Scholar). The positive control endoribonuclease was incubated in the same buffer as the PAPs except that 4 mm MgCl2 was replaced with 5 mm MnCl2. Following incubation, samples were separated by a 7.5 m urea, 18% acrylamide gel. The gel was wrapped in plastic and quantification of radiolabeled bands was performed using a PhosphorImager (Amersham Biosciences). Isolation of Ribosomes and Primer Extension of rRNA—Ribosomes were isolated from barley leaves according to the method described by Tumer et al. (2Tumer N.E. Hwang D.J. Bonness M. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 3866-3871Crossref PubMed Scopus (130) Google Scholar). PAP and PAP mutants purified from E. coli (50 ng) were incubated with barley ribosomes (50 μg) in RIP buffer to a final volume of 100 μl for 30 min at 30 °C. rRNA was extracted, and depurination was assessed by primer extension as previously described (18Hudak K.A. Wang P. Tumer N.E. RNA. 2000; 6: 369-380Crossref PubMed Scopus (92) Google Scholar) using 500 ng of barley rRNA. A second primer, which anneals close to the 5′-end of the 28 S rRNA, was included in each sample for primer extension and served as an internal control for RNA loading (28Parikh B.A. Coetzer C. Tumer N.E. J. Biol. Chem. 2002; 277: 41428-41437Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar). Treatment of BMV RNAs with PAP and PAP Mutants—BMV RNAs (1 μg) isolated from viral particles were incubated with purified wild-type PAP (5, 10, 50, or 100 ng) in RIP buffer (60 mm KCl, 10 mm Tris-HCl, pH 7.4, 10 mm MgCl2) to a final volume of 100 μl for 30 min at 30 °C. Following incubation, PAP was removed from the RNAs by phenol/chloroform extraction, and the treated RNAs were precipitated in ethanol. The RNAs were resuspended in diethyl-pyrocarbonate-treated water and used to transfect isolated protoplasts of barley. BMV RNAs (1 μg) were also treated with PAP mutants (50 ng) as described for incubation with wild-type PAP. An in vitro generated transcript of BMV RNA3 (1 μg) was incubated with wild-type PAP (50 ng) as described for total BMV RNAs. Generation of Protoplasts and Inoculation with BMV RNAs—Protoplasts were generated by enzyme digestion of 7-day-old leaves of barley, and inoculation of protoplasts with BMV RNAs was done using PEG 1500 essentially according to the method of Kroner et al. (29Kroner P. Richards D. Traynor P. Ahlquist P. J. Virol. 1989; 63: 5302-5309Crossref PubMed Google Scholar). Briefly, PAP-treated BMV RNA samples (1 μg prepared as described above) were added to 4 × 105 protoplasts in 0.55 m mannitol and 8% PEG 1500. Protoplasts were incubated for 20 min at room temperature and then washed in 0.55 m mannitol and resuspended in 1 ml of incubation medium (10 mm CaCl2, 1 mm KNO3, 1 mm MgSO4, 0.2 mm KH2PO4, 1 μm KI, 0.1 μm CuSO4, 10% mannitol, 2% sucrose, 0.01% gentamycin sulfate). Protoplasts were incubated for 18 h at 27 °C and constant low light (165 μmol/m2/s). As a negative control, 0.5 μg of BMV RNAs were inoculated without the addition of PEG, and the positive control was 0.5 μg of BMV RNAs inoculated in the presence of PEG but without prior treatment with PAP. Incubation of Inoculated Protoplasts with PAP and PAP Mutants— Aliquots of 4 × 105 protoplasts were inoculated with 0.5 μg of BMV RNAs and incubated as described above in 1 ml of incubation medium for 30 min prior to the addition of PAP or mutant PAPs (1.0 μg). Protoplasts were incubated for an additional 18 h at 27 °C and constant low light (165 μmol/m2/s). To compare the effect of delayed PAP addition to protoplasts on the synthesis of positive versus negative strand BMV RNAs, aliquots of 4 × 105 protoplasts were transfected with 0.5 μg of BMV RNAs and incubated, as described, for 30 min or 3 h prior to the addition of 1.0 μg of PAP and subsequent incubation for a total of 18 h. To test whether PAP caused a decline in the stability of BMV RNAs in protoplasts, a pool of 2.4 × 106 protoplasts was transfected with 6.0 μg of in vitro transcript of BMV RNA1 and incubated in 6 ml of incubation medium for 30 min prior to the addition of 6.0 μg of PAP. Aliquots (1 ml) were removed at the indicated time points and analyzed for the presence of BMV RNA1 by Northern blot. Inoculated protoplasts without the addition of PAP were used as a negative control. Isolation of Protoplast RNA and Northern Blot Analysis—Following incubation of protoplasts, total RNA was isolated from these cells. Protoplasts were pelleted at 300 × g for 3 min and resuspended in guanidinium buffer (4 m guanidinium isothiocyanate, 50 mm β-mercaptoethanol, 20 mm MES, pH 7.0, 20 mm EDTA), and phenol/chloroform/isoamyl alcohol. Cells were vortexed and then centrifuged at 1,000 × g for 10 min. The aqueous layer was re-extracted with phenol/chloroform/isoamyl alcohol, and total RNA was precipitated by the addition of 0.2 volumes of 1 m acetic acid and 0.7 volumes of 100% ethanol. The precipitated RNA was pelleted by centrifugation and washed with 3 m NaOAc, washed again with 70% ethanol, air-dried, and resuspended in diethyl-pyrocarbonate-treated water. Equal amounts of total RNA per sample were separated in a 4.5% acrylamide, 7 m urea gel, transferred to nylon membrane (Amersham Biosciences), and probed with radiolabeled negative strand partial transcripts of BMV or 25 S rRNA. The probe for positive-strand BMV RNA was transcribed from pB3HE1 containing an ∼200-nucleotide fragment of the tRNA-like structure from the 3′-end of BMV RNA3 that is conserved in sequence in all BMV positive strand RNAs. The probe for negative strand BMV RNA was transcribed from the same plasmid but in the reverse direction. The 28 S rRNA probe was transcribed from pKH002, containing an ∼80-nucleotide fragment of the conserved sarcin/ricin loop region of the yeast 25 S rRNA. Hybridization of probes to total RNA was visualized by exposure of the blot to x-ray film. In Vivo [35S]Methionine Incorporation—Protein synthesis in protoplasts was assayed by the incorporation of radiolabeled methionine into protein. Protoplasts (2 × 105 cells/ml) transfected with BMV RNAs recovered for 30 min in incubation medium and then were incubated in 1 ml of incubation medium containing 1.0 μg of PAP or PAP mutants for 1 h at 25 °C. Protoplasts were then pulsed with 10 μCi of [35S]Methionine (1000 Ci/mmol; Amersham Biosciences), and 100-μl aliquots were removed at the times indicated. Protoplasts were pelleted by centrifugation at 1,000 × g, and 100 μl of 100% trichloroacetic acid was added to each aliquot. The preparation was incubated at 70 °C for 20 min followed by 10 min on ice. Trichloroacetic acid-insoluble material was filtered through 25-mm glass microfiber filters (Whatman GFC), washed with ice-cold 5% trichloroacetic acid and then ice-cold 95% ethanol. Filters were air-dried, and radioactivity was quantified by scintillation counting. BMV Replicase Assays—RNA templates used in replicase assays were purchased from Dharmacon Inc. (Boulder, CO). BMV replicase was isolated from infected barley leaves as described by Sun et al. (30Sun J. Adkins S. Faurote G. Kao C.C. Virology. 1996; 226: 1-12Crossref PubMed Scopus (79) Google Scholar). Replicase assays were performed essentially according to Adkins et al. (31Adkins S. Stawicki S. Faurote G. Siegel R. Kao C.C. RNA. 1998; 4: 455-470PubMed Google Scholar). Briefly, template RNA (0.5 pmol) and 7 μl of replicase were combined in reaction buffer (20 mm sodium glutamate, pH 8.2, 12 mm dithiothreitol, 4 mm MgCl2, 2 mm MnCl2, 500 μm GTP, 200 μm ATP, 200 μm UTP, 242 nm [α-32P]CTP (400 Ci/mmol; Amersham Biosciences), 0.5% Triton X-100) in a 40-μl final volume. Following incubation at 30 °C for 60 min, the reaction products were extracted with phenol/chloroform and precipitated in 6 volumes of 100% ethanol, 10 μg of glycogen, and a final concentration of 0.4 m ammonium acetate. Samples were resuspended in formamide loading buffer and separated on 12% acrylamide, 7 m urea gels. The amount of label incorporated into newly synthesized RNA was determined with a PhosphorImager and quantified using Amersham Biosciences software. To measure the rate of RNA synthesis over time, the same replicase reaction mixture was assembled as described above, and aliquots of 20 μl were removed at the indicated times. Reaction products were precipitated and analyzed as above. The percentages of synthesis were normalized for CMP incorporation relative to the control template analyzed in the same set of reactions. Synthesis of Mature PAP and PAP Mutants in E. coli—In pokeweed, PAP is first synthesized as a 313-amino acid-long precursor that is processed to produce the mature (262-amino acid) form of the protein, which has 22 and 29 amino acids cleaved from the N and C termini, respectively. Analysis of the in vitro activity of wild-type PAP has been facilitated by E. coli-expressed forms of the protein (32Rajamohan F. Engstrom C.R. Denton T.J. Engen L.A. Kourinov I. Uckun F.M. Protein Expression Purif. 1999; 16: 359-368Crossref PubMed Scopus (36) Google Scholar, 33Honjo E. Watanabe K. Biosci. Biotechnol. Biochem. 1999; 63: 1291-1294Crossref PubMed Scopus (10) Google Scholar). However, mutant forms of mature PAP derived from E. coli have not yet been prepared or analyzed. Expression of enzymatically active PAP and its respective mutants in E. coli required the synthesis of their fully processed forms, which would accurately mimic their synthesis in eukaryotic cells. The wild-type constructs and mutants with amino acid substitutions were designed to begin at Val23 and end with Thr284 to produce mature proteins of 262 amino acids. To facilitate purification, these proteins also included an N-terminal His6 tag. Three mutant forms of PAP were expressed in addition to wild-type PAP. PAPx is an active site mutant with a point mutation E176V that inactivates the glycosylase activity of this protein (34Hur Y. Hwang D.J. Zoubenko O. Coetzer C. Uckun F.M. Tumer N.E. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 8448-8452Crossref PubMed Scopus (56) Google Scholar). PAPn and PAPc contain a point mutation G75D and a termination codon in place of Trp259, respectively, and are nontoxic to yeast growth (34Hur Y. Hwang D.J. Zoubenko O. Coetzer C. Uckun F.M. Tumer N.E. Proc. Natl. Acad. Sci. U. S. A. 1995; 92: 8448-8452Crossref PubMed Scopus (56) Google Scholar). The purified proteins migrated according to their expected masses, with PAPc moving slightly faster due to the absence of 26 C-terminal residues. Importantly, the purified enzymes lacked detectable contaminating proteins (Fig. 1A) and, at the concentrations of PAP used in protoplast assays, had minimal ribonuclease activities (Fig. 1B). PAPx, which showed the highest level of contamination (7% degradation of template), did not affect the accumulation of BMV RNAs in protoplasts (Figs. 2 and 3). Although PAPn was not included in this analysis, it was expected to contain a similar level of contaminating nucleases, since all proteins were isolated in the same manner.Fig. 2Inhibition of BMV RNA accumulation in barley protoplasts by prior incubation with PAP and PAP mutants. A, BMV RNAs (1.0 μg) were incubated with PAP (5, 10, 50, or 100 ng), and following incubation, PAP was removed by phenol/chloroform extraction. The treated BMV RNAs were inoculated into protoplasts and allowed to replicate for 18 h. Total protoplast RNA was analyzed by Northern blot and probed for positive strand BMV RNAs. A protoplast sample without PEG (–PEG) and a sample without BMV RNAs (–BMV) were used as negative controls for inoculation. BMV RNAs incubated in buffer alone (0 PAP) prior to inoculation into protoplasts were used as a positive control for replication. Std, 300 ng of BMV RNAs loaded directly onto the gel. B, the same samples probed for 28 S rRNA as a loading control for total RNA. C, BMV RNAs (1.0 μg) were incubated with PAP, PAPx, PAPn, or PAPc (50 ng), and following incubation, PAPs were removed by phenol/chloroform extraction. The treated BMV RNAs were inoculated into protoplasts and allowed to replicate for 18 h. Total protoplast RNA was analyzed by Northern blot and probed for positive strand BMV RNAs. BMV RNAs incubated in buffer alone (–PAP) prior to inoculation into protoplasts were used as a positive control for replication. Std, 300 ng of BMV RNAs loaded directly onto the gel. D, the same samples probed for 28 S rRNA as a loading control for total RNA.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Fig. 3Effect of incubation of PAP and PAP mutants with barley protoplasts inoculated with BMV RNAs. A, protoplasts were inoculated with 0.5 μg of BMV RNAs (not treated initially with PAP) and incubated in 1 ml of incubation medium containing 1 μg of PAP or PAP mutants for 18 h. Total protoplast RNA was analyzed by Northern blot and probed for positive strand BMV RNAs. Protoplasts incubated without PAP (–PAP) were used as a positive control for BMV replication. Std, 300 ng of BMV RNAs loaded directly onto the gel. B, the same samples probed for 28 S rRNA as a loading control for total RNA. C, protoplasts were inoculated with 0.5 μg of BMV RNAs (not treated initially with PAP) and incubated in 1 ml of incubation medium for either 30 min or 3 h prior to the addition of PAP. Protoplasts were incubated for a total of 18 h. Total protoplast RNA was analyzed by Northern blot and probed for positive and negative strand BMV RNAs. Protoplasts incubated without PAP (–PAP) were used as a positive control for BMV replication. D, the same samples probed for 28 S rRNA as a loading control for total RNA.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Depurination of Barley Ribosomes by E. coli-expressed PAP and PAP Mutants—The mutants PAPx, PAPn, and PAPc do not depurinate ribosomes when expressed in tobacco or yeast (35Zoubenko O. Hudak K.A. Tumer N.E. Plant Mol. Biol. 2000; 44: 219-229Crossref PubMed Scopus (56) Google Scholar, 36Hudak K.A. Hammell A.B. Yasenchak J. Tumer N.E. Dinman J.D. Virology. 2001; 279: 292-301Crossref PubMed Scopus (20) Google Scholar). To determine whether mature PAP and PAP mutants expressed in E. coli were able to depurinate ribosomes, the proteins were incubated with ribosomes isolated from barley leaves, and primer extension analysis was performed on the rRNA to detect the missing purine residue. Barley is a host for BMV, the focus of this study. As shown in Fig. 1C, only minimal levels of depurination of the sarcin/ricin loop were evident in barley ribosomes not treated with PAP. This background depurination may be due to the endogenous ribosome-inactivating protein present in barley (37Chaudhry B. Muller U.F. Cameron Mills V. Gough S. Simpson D. Skriver K. Mundy J. Plant J. 1994; 6: 815-824Crossref PubMed Scopus (127) Google Scholar). However, incubation of ribosomes with PAP purified from pokeweed increased rRNA depurination 12-fold over background levels. Efficient levels of depurination were also observed for ribosomes incubated with mature PAP isolated from E. coli. In contrast, the nontoxic mutants PAPx, PAPn, and PAPc did not depurinate barley ribosomes above background levels. These data illustrate that the mature forms of PAP expressed in E. coli exhibit similar depurination properties as the corresponding proteins expressed in transgenic tobacco and yeast. Inhibition of BMV RNA Accumulation in Barley Protoplasts by Pretreatment with PAP—We have demonstrated previously that wild-type PAP depurinates BMV RNAs in vitro and inhibits the translation of these PAP-treated RNAs in a cell-free system (18Hudak K.A. Wang P. Tumer N.E. RNA. 2000; 6: 369-380Crossref PubMed Scopus (92) Google Scholar). To determine whether PAP treatment could affect BMV RNAs in vivo, the RNAs were incubated with wild-type PAP, followed by phenol/chloroform extraction and ethanol precipitation to remove PAP. Treated RNAs were then inoculated into barley protoplasts, and the amount of replication product that accumulated was monitored by Northern blot analysis (mutant BMV RNAs incapable of replication are not detected by this assay). Fig. 2A illustrates a decline in the level of BMV RNAs that correlated with treatment of increasing concentrations of PAP. These results indicate that prior incubation of BMV RNAs with PAP inhibits the accumulation of these RNAs in barley protoplasts. This analysis was repeated with 50 ng of nontoxic mutants of PAP, an amount that caused severe inhibition of accumulation with wild-type PAP (Fig. 2A). Fig. 2C shows that both PAPn and PAPc were able to efficiently inhibit accumulation of BMV, despite being inactive for rRNA depurination. The active site mutant PAPx did not inhibit BMV RNA accumulation, since the amount of viral RNA was indistinguishable from samples without PAP treatment. These results indicate that rRNA depurination and inhibition of BMV RNA accumulation both require PAP with a functional active site; however, each has different requirements as revealed by mutants PAPn and PAPc. Samples were also probed for 28 S rRNA as an indicator of total RNA loading (Fig. 2, B and D). Analysis of PAP Activity in Barley Protoplasts—To determine whether PAP could affect BMV RNA accumulation in vivo, without prior treatment of the viral RNAs, PAP or PAP mutants were added to the protoplast incubation medium 30 min after RNA transfection. PAP is known to be able to traverse protoplast membranes; thus, access of PAP to the cytosol was anticipated (13Watanabe K. Kawasaki T. Sako N. Funatsu G. Biosci. Biotech. Biochem. 1997; 61: 994-997Crossref PubMed Scopus (19) Google Scholar). Northern blot analysis of total protoplast RNA after an 18-h incubation shows a decline in the amount of BMV RNAs for those treated with PAP, PAPn, or PAPc (Fig. 3A). The pattern of BMV RNA accumulation was similar to that seen when the RNAs were incubated with PAP or PAP mutants prior to the transfection of BMV RNAs. Thus, in vitro and in vivo treatments with PAP have comparable effects on BMV accumulation levels in barley protoplasts. Samples were also probed for 28 S rRNA as a loading control for total RNA (Fig. 3B). Next, we examined whether PAP could selectively affect the replication of positive strand BMV RNAs after the initiation of negative strand RNA replication. Three hours after transfection, BMV translation and negative strand RNA replication are known to be well under way, but positive strand RNA synthesis is not detectable (38Hema M. Kao C.C. J. Virol. 2004; 78: 1169-1180Crossref PubMed Sco" @default.
- W2042129286 created "2016-06-24" @default.
- W2042129286 creator A5031852302 @default.
- W2042129286 creator A5054379040 @default.
- W2042129286 creator A5083880051 @default.
- W2042129286 date "2005-05-01" @default.
- W2042129286 modified "2023-09-27" @default.
- W2042129286 title "Pokeweed Antiviral Protein Inhibits Brome Mosaic Virus Replication in Plant Cells" @default.
- W2042129286 cites W150824067 @default.
- W2042129286 cites W1519055488 @default.
- W2042129286 cites W1533831149 @default.
- W2042129286 cites W1569531441 @default.
- W2042129286 cites W179979008 @default.
- W2042129286 cites W1917399449 @default.
- W2042129286 cites W1964689773 @default.
- W2042129286 cites W1976754433 @default.
- W2042129286 cites W1977863283 @default.
- W2042129286 cites W1978471475 @default.
- W2042129286 cites W1981176722 @default.
- W2042129286 cites W1984349537 @default.
- W2042129286 cites W1985067074 @default.
- W2042129286 cites W1987558980 @default.
- W2042129286 cites W1988473696 @default.
- W2042129286 cites W1991928942 @default.
- W2042129286 cites W1997099356 @default.
- W2042129286 cites W1997714902 @default.
- W2042129286 cites W1998144653 @default.
- W2042129286 cites W2006140229 @default.
- W2042129286 cites W2014067427 @default.
- W2042129286 cites W2017461061 @default.
- W2042129286 cites W2025930859 @default.
- W2042129286 cites W2027148086 @default.
- W2042129286 cites W2029816142 @default.
- W2042129286 cites W2039290476 @default.
- W2042129286 cites W2049315807 @default.
- W2042129286 cites W2049421697 @default.
- W2042129286 cites W2065061097 @default.
- W2042129286 cites W2072839681 @default.
- W2042129286 cites W2076101600 @default.
- W2042129286 cites W2081579862 @default.
- W2042129286 cites W2085825896 @default.
- W2042129286 cites W2089822958 @default.
- W2042129286 cites W2092527458 @default.
- W2042129286 cites W2092867032 @default.
- W2042129286 cites W2095433332 @default.
- W2042129286 cites W2097432692 @default.
- W2042129286 cites W2097602179 @default.
- W2042129286 cites W2111252388 @default.
- W2042129286 cites W2125714200 @default.
- W2042129286 cites W2127242992 @default.
- W2042129286 cites W2129468535 @default.
- W2042129286 cites W2131566401 @default.
- W2042129286 cites W2139636517 @default.
- W2042129286 cites W2146042887 @default.
- W2042129286 doi "https://doi.org/10.1074/jbc.m413452200" @default.
- W2042129286 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/15764597" @default.
- W2042129286 hasPublicationYear "2005" @default.
- W2042129286 type Work @default.
- W2042129286 sameAs 2042129286 @default.
- W2042129286 citedByCount "49" @default.
- W2042129286 countsByYear W20421292862012 @default.
- W2042129286 countsByYear W20421292862013 @default.
- W2042129286 countsByYear W20421292862014 @default.
- W2042129286 countsByYear W20421292862015 @default.
- W2042129286 countsByYear W20421292862016 @default.
- W2042129286 countsByYear W20421292862017 @default.
- W2042129286 countsByYear W20421292862018 @default.
- W2042129286 countsByYear W20421292862019 @default.
- W2042129286 countsByYear W20421292862020 @default.
- W2042129286 countsByYear W20421292862021 @default.
- W2042129286 countsByYear W20421292862022 @default.
- W2042129286 countsByYear W20421292862023 @default.
- W2042129286 crossrefType "journal-article" @default.
- W2042129286 hasAuthorship W2042129286A5031852302 @default.
- W2042129286 hasAuthorship W2042129286A5054379040 @default.
- W2042129286 hasAuthorship W2042129286A5083880051 @default.
- W2042129286 hasBestOaLocation W20421292861 @default.
- W2042129286 hasConcept C104317684 @default.
- W2042129286 hasConcept C109110057 @default.
- W2042129286 hasConcept C12590798 @default.
- W2042129286 hasConcept C140704245 @default.
- W2042129286 hasConcept C159047783 @default.
- W2042129286 hasConcept C2522874641 @default.
- W2042129286 hasConcept C2779535198 @default.
- W2042129286 hasConcept C41258723 @default.
- W2042129286 hasConcept C54355233 @default.
- W2042129286 hasConcept C67705224 @default.
- W2042129286 hasConcept C86803240 @default.
- W2042129286 hasConceptScore W2042129286C104317684 @default.
- W2042129286 hasConceptScore W2042129286C109110057 @default.
- W2042129286 hasConceptScore W2042129286C12590798 @default.
- W2042129286 hasConceptScore W2042129286C140704245 @default.
- W2042129286 hasConceptScore W2042129286C159047783 @default.
- W2042129286 hasConceptScore W2042129286C2522874641 @default.
- W2042129286 hasConceptScore W2042129286C2779535198 @default.
- W2042129286 hasConceptScore W2042129286C41258723 @default.
- W2042129286 hasConceptScore W2042129286C54355233 @default.
- W2042129286 hasConceptScore W2042129286C67705224 @default.