Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A potyvirus P1 protein interacts with the Rieske Fe/S protein of its host

A potyvirus P1 protein interacts with the Rieske Fe/S protein of its host INTRODUCTION Potyviruses are single‐stranded positive‐sense RNA viruses with a single large open reading frame (ORF) predicted to produce a large polyprotein that is subsequently processed to ten mature proteins ( Fig. 1a ). The P1 protein is released from the N‐proximal end of the polyprotein through autoproteolytic cleavage by its own C‐proximal serine‐type proteinase domain ( Verchot ., 1991 ). The function of P1 remains unclear, but cleavage of P1 is essential for viral infectivity ( Moreno ., 1998 ; Verchot and Carrington, 1995 ). P1 enhanced HC‐Pro‐mediated suppression of RNA silencing ( Pruss ., 1997 ), but did not itself suppress RNA silencing ( Anandalakshmi ., 1998 ). Until now, there have been no reports of interactions between viral P1 and host proteins. 1 Diagram showing (a) the organization of the SMV‐P genome, (b) the regions of the P1 gene and (c) the regions of the P. ternata Rieske Fe/S protein used in yeast two‐hybrid tests. Soybean mosaic virus (SMV), a member of the genus Potyvirus , is the most important viral pathogen of soybeans world wide. Numerous strains and isolates have been identified based on differences in their biological and serological properties, but most of them could only infect some species in the family Leguminosae. Recently, an unusual strain of SMV (SMV‐P) was identified from the aroid Pinellia ternata (Thunb.) Ten. ex Breitenb. in China ( Chen ., 2004 ) and was later found to occur widely in aroid plants ( Shi ., 2005 ). This strain could be mechanically inoculated to some soybean cultivars, in which it induced weak symptoms. The sequence of SMV‐P was very similar to that of SMV soybean isolates (SMV‐S) except in the P1 region, where it was much more similar to the aroid pathogen Dasheen mosaic virus (DsMV) than to SMV‐S. We now report studies showing that SMV‐P P1 interacts with the host Rieske Fe/S protein. RESULTS Interactions between SMV‐P‐P1 and other viral proteins Yeast two‐hybrid (Y2H) screens were used to test for interactions between SMV‐P P1 and the ten mature proteins of the virus. The expression of bait and target genes in yeast was confirmed by Western blot ( Fig. 2 ). Each potential interaction was tested three times in independent co‐transformations with positive and negative controls. No detectable interactions were found between P1 and any of the mature viral proteins of SMV‐P. 2 Western blot analysis of SMV‐P proteins expressed as fusion proteins in Saccharomyces cerevisiae strain Y109 using (a) the pGBKT7 vector and a c‐Myc monoclonal antibody or (b) the pGADT7 vector and an HA‐Tag polyclonal antibody. 1, pGBKT7 vector; 2, pGBK‐SMV‐P‐P1 plasmid with predicted fusion protein of 57.8 kDa; 3, pGADT7 vector; 4–13, pGAD‐SMV‐P‐X plasmid where gene X is: 4, P1; 5, HC‐Pro; 6, P3; 7, 6K1; 8, CI; 9, 6K2; 10, NIa‐VPg; 11, NIa‐Pro; 12, NIb; 13, CP; the fusion proteins had calculated molecular weights of 57.8, 72.7, 60.5, 26.0, 92.0, 26.0, 42.4, 47.9, 80.4 and 50.0 kDa, respectively. The positions of protein size markers (kDa) are shown on the left. Interactions between SMV‐P‐P1 and P. ternata proteins Using expressed SMV‐P P1 as bait in Y2H screens, 13 independent interacting clones were identified after two separate screenings of a P. ternata cDNA expression library (2.1 × 10 6 independent clones). Each clone could induce the two reporter genes, allowing growth on quadruple dropout plates and expression of α‐galactosidase activity. Sequencing showed that ten of the 13 clones were identical and had a ORF predicted to encode a 23.7‐kDa protein closely related to the cytochrome b6/f complex Rieske Fe/S genes of plants (nearest match by BLAST to Fritillaria agrestis O49078; score 8.4e‐95), whereas the other three clones were sequence fragments without recognizable ORFs. A plasmid encoding the complete gene was isolated and retransformed back to yeast either alone or in combination with pGBK‐SMV‐P‐P1, Lamin C or the binding domain vector alone. The reporter genes were only induced in combination with pGBK‐SMV‐P‐P1. The Rieske Fe/S gene of P. ternata was 942 nt long excluding the poly(A) tail and contained an ORF of 678 nt. The ORF was predicted to produce a protein of 226 amino acids and, by comparison with related proteins, this includes a transit peptide of 53 amino acids at its N‐terminus ( Fig. 3 ). A polyclonal antiserum raised to the polypeptide was used in Western blot with proteins isolated from P. ternata chloroplasts and this indicated that the mature Rieske Fe/S protein had a molecular weight of 18.6 kDa, consistent with the sequence data (data not shown). 3 Predicted amino acid sequence of the Rieske Fe/S protein of Pinellia ternata aligned with the homologous sequences of Fritillaria agrestis (O49078), Arabidopsis thaliana (Q9ZR03), Glycine max (this study) and Zantedeschia aethiopica (this study). Identical residues are marked ‘.’ and alignment gaps as ‘–’. The interaction between SMV‐P P1 and the mature Rieske Fe/S protein (without transit peptide) of the host was confirmed by in vitro co‐immunoprecipitation of the two proteins expressed from the wheat germ extract system. Proteins of the expected sizes of 21.7 (HA‐Fe/S) and 41.2 kDa (c‐Myc‐P1) were obtained, but were not detected in the controls ( Fig. 4 ). 4 Colorimetric (BCIP/NBT) analysis of SMV‐P P1 co‐immunoprecipitates with the Rieske Fe/S protein of Pinellia ternata . 1, endogenous biotinylated proteins of the in vitro reaction without templates (negative control); 2, co‐immunoprecipitates of the biotinylated HA‐Fe/S and c‐Myc‐SMVP‐P1 proteins; 3, cMyc‐SMV‐P P1 immunoprecitated with anti‐HA; 4, HA‐Fe/S immunoprecipitated with anti‐c‐Myc. The sizes of the HA and c‐Myc epitope tags were 3.1 and 2.6 kDa, respectively. The positions of protein size markers (kDa) are shown on the left. Dissection of the P1‐Rieske F/S protein interaction Y2H assays using different parts of the SMV‐P P1 protein ( Fig. 1 ) and three sections of the Rieske Fe/S protein showed that only the N‐terminal part (amino acids 1–82) of the SMV‐P P1 was responsible for the interaction with the Rieske Fe/S protein and that amino acids 1–33 interacted only with the transit peptide, while amino acids 34–82 could interact with the entire Rieske Fe/S protein ( Table 1 ). 1 Interactions (N, negative; P, positive) determined from yeast two‐hybrid experiments between different regions of the viral (SMV‐P) P1 and host ( Pinellia ternata ) Rieske Fe/S proteins. Human lamin C is included as a non‐specific control Rieske Fe/S protein 1–53 54–128 129–226 54–226 129–226 1–128 1–226 SMV‐P P1 1–33 P N N N N P P 34–82 P P P P P P P 83–330 N N N N N N N 1–82 P P P P P P P 1–164 P P P P P P P 1–246 P P P P P P P 1–330 P P P P P P P 83–264 N N N N N N N 265–330 N N N N N N N 34–164 P P P P P P P Human lamin C 1–165 N N N N N N N P1–Rieske Fe/S protein interactions in other hosts Y2H assays were used to examine the interactions between the P1 protein of SMV‐P and the Rieske Fe/S proteins of other plants. The unknown Rieske Fe/S proteins were identified and sequenced after screening cDNA libraries using the Rieske Fe/S gene of P. ternata as a probe. SMV‐P P1 interacted moderately with the Rieske Fe/S proteins of Glycine max cv. Xudou 1 and Zantedeschia aethiopica (both hosts of the virus) and weakly with that of the non‐host Arabidopsis thaliana . In parallel tests, the P1 protein of a soybean (SMV‐S) isolate interacted weakly with the Rieske Fe/S protein of its host G. max , but not at all to those of the non‐hosts Z. aethiopica and P. ternata . The amino acid sequences of the Rieske Fe/S proteins from different plants were relatively conserved expect for the transit peptide ( Fig. 3 ). DISCUSSION No detectable interactions were found between P1 and any of the mature viral proteins of SMV‐P. This contrasts with the weak interaction reported between the P1 and CI proteins of Potato virus A ( Merits ., 1999 ), but is in agreement with results reported for Pea seed‐borne mosaic virus ( Guo ., 2001 ) and SMV strain G7H ( Kang ., 2004 ). By contrast, there was a strong and consistent interaction with the Rieske Fe/S protein of its natural host, Pinellia ternata . The different strengths of the interactions between SMV‐P P1 and the Rieske Fe/S proteins of host and non‐host plants suggests a role in host range determination. The Rieske Fe/S protein of the cytochrome b6/f complex is an indispensable component of the photosynthetic electron transport chain in chloroplasts. It is a polypeptide that faces the stroma with only a few N‐terminal residues and is anchored in the membrane with a single transmembrane helix ( Karnauchov ., 1997 ). The large C‐terminal hydrophilic domain is exposed in the lumenal space of the thylakoid membrane system and provides the ligands for the [2 Fe – 2 S] cluster ( Zhang ., 1996 ). In higher plants, the Rieske protein is encoded in the nucleus and synthesized in the cytosol as a precursor molecule with a transit peptide mediating the import of the protein into the chloroplast stroma ( Madueno ., 1992 ). The signal for the subsequent thylakoid translocation step is provided by the N terminal region of the mature polypeptide chain that comprises the membrane anchor ( Madueno ., 1994 ). As mosaic and mottle symptoms are often observed after virus infection, it is reasonable to assume that functional and structural changes of the chloroplasts cause the chlorosis, and with some viruses infection is known to be associated with changes in the chloroplasts. After infection by Squash mosaic virus (genus Comovirus ), the rate of carbon fixation changed and the number of chloroplasts decreased but there was no detectable effect on the ultrastructure of the chloroplasts ( Magyarosy ., 1973 ). With Tobacco mosaic virus (TMV, genus Tobamovirus ), infection induced an inhibition of photosystem II electron transport, disturbing the oxygen‐evolving complex ( Lehto ., 2003 ). The possibility that one or more viral products were imported into the chloroplast was one of the proposed mechanisms underlying virus pathogenicity ( Culver ., 1991 ) and there is evidence that some viral proteins localize to chloroplasts. For example, the coat protein (CP) of TMV and the second triple gene block protein, TGBp2, of Barley stripe mosaic virus (genus Hordeivirus ) both accumulate in chloroplasts of systemically infected leaves ( Reinero and Beachy, 1986 ; Torrance ., 2006 ). It has also been shown that a CP–green fluorescent protein fusion of Cucumber necrosis virus (genus Tombusvirus ) targets chloroplasts in Nicotiana benthamiana leaves ( Xiang ., 2006 ). An association between potyvirus CP proteins and chloroplasts, and some interactions between the CP and chloroplast proteins have previously been reported ( Gunasinghe and Berger, 1991 ; Jimenez ., 2006 ; McClintock ., 1998 ). However, the way by which potyviruses induce symptoms remains unclear and our results suggest that the interaction between P1 and the Rieske Fe/S protein may be significant. The TMV replicase protein can disrupt the nuclear localization and subsequent function of interacting Aux/IAA proteins ( Padmanabhan ., 2006 ) and, by analogy, the SMV‐P P1 may block the transit peptide of the Rieske Fe/S protein and thus retard the Fe/S transport to the chloroplast, so delaying the formation of the cytochrome b6/f complex. This may explain why symptoms are most easily seen in newly expanding leaves. Once inside the chloroplast P1 may further interact with the Rieske Fe/S protein to modify the functions of the cytochrome b6/f complex. The N terminus of the P1 protein is the most variable region of the potyvirus genome, both in sequence and in length ( Adams ., 2005 ) and shows evidence of extensive recombination that is believed to be significant for adaptation to different hosts ( Valli ., 2007 ). We suggest that the P1–Rieske Fe/S protein interactions may play an important role in this adaptation. EXPERIMENTAL PROCEDURES Virus sources and test plants Healthy Pinellia ternata , Glycine max (soybean cv. Xudou 1), Zantedeschia aethiopica (Calla Lily) and Arabidopsis thaliana plants were grown in the greenhouse. SMV‐P was maintained by mechanical inoculation to P. ternata . Y2H tests for the interactions between SMV‐P P1 and the ten mature proteins of SMV‐P Y2H tests were performed using the Matchmaker Two‐Hybrid system 3 (Clontech, Mountain View, USA) according to the manufacturer's protocols. pGAD‐SMV‐P‐X plasmids were constructed for each of the ten genes of SMV‐P ( Fig. 1a ) by inserting the respective gene into a pGADT7 vector. The pGBK‐SMV‐P‐P1 plasmid was constructed by insertion of the P1 gene of SMV‐P into a pGBKT7 vector. These plasmids were first verified by sequencing, then transformed into Saccharomyces cerevisiae AH109. Western blots were used to verify the expression of inserted genes using an HA‐Tag ployclonal antibody (Clontech). Simultaneous co‐transformations of pGBK‐SMV‐P‐P1 and pGAD‐SMV‐P‐X to yeast were performed by the lithium acetate method using the small‐scale yeast transformation protocol. The co‐transformants were plated on double dropout plates. The positive interactions in the transformed yeast were confirmed by assaying colonies of the transformants on quadruple dropout plates containing X‐α‐Gal. The Y2H assay was then repeated after swapping the inserts between the AD and BD vectors. The plasmids pGBKT7‐53 and pGADT7‐T were used together as a positive control or in combination with our Gal4 fusion constructs as negative controls. Construction of the P. ternata cDNA library Total RNA was extracted from fresh, healthy leaves of P. ternata using TRIzol (Invitrogen) according to the manufacturer's manual. cDNA libraries were then constructed as described in the Matchmaker™ Library Construction & Screening Kits User Manual (Clontech). Double‐stranded cDNA and pGAD‐Rec were transformed to S. cerevisiae AH109. Screening the cDNA library by yeast mating pGBK‐SMV‐P‐P1 was transformed to S. cerevisiae Y187 and the cDNA library was screened by yeast mating. Plasmids were prepared from positive clones and used to transform Escherichia coli strain JM109. Recovered plasmids were sequenced. Diploids were plated on triple dropout medium (low stringency selection), and positives were re‐plated on to quadruple dropout plates (high stringency) containing X‐α‐gal. The target gene was subsequently re‐inserted into the pGADT7 vector and the two‐hybrid assay was repeated. All the positive interactions were re‐tested using yeast mating to eliminate false positives. In vitro co‐immunoprecipitation Plasmids pGAD‐Pt‐Fe/S and pGBK‐SMV‐P‐P1 were purified using the Promega Wizard ® Plus SV DNA Isolation System (Promega) according to the manufacturer's protocols. The SMV‐P P1 and Rieske Fe/S proteins were then produced in vitro using the TNT T7 Coupled Wheat Germ Extract system (Promega) and labelled with Transcend™ tRNA (Promega) according to the manufacturer's protocols. The two biotinylated translation products HA‐Fe/S and c‐Myc‐SMVP‐P1 were allowed to interact and were co‐immunoprecipitated with Protein A using the Matchmaker™ Co‐IP Kit (Clontech). After electrophoretic separation and transfer to the nitrocellulose membrane, the in vitro co‐immunoprecipitation was detected using the Transcend™ Non‐Radioactive Translation Detection System (Promega). Controls were endogenous biotinylated proteins of the in vitro reaction without templates, cMyc‐SMV‐P P1 immunoprecitated with anti‐HA and HA‐Fe/S immunoprecipitated with anti‐c‐Myc. Identification of regions involved in the interaction between SMV‐P P1 and the Rieske Fe/S protein Truncated parts of the SMV‐P P1 ( Table 1 , Fig. 1b ) were inserted into a pGBKT7 vector. The Rieske Fe/S gene was divided into three parts according to the published structure and fragments were inserted into a pGADT7 vector ( Table 1 , Fig. 1c ). Plasmid pGBKT7‐lam (Clontech), which encodes a fusion of the DNA‐BD with human lamin C, was used as a non‐specific control. Lamin C neither forms complexes nor interacts with most other proteins. These plasmids were then used to identify the regions involved in the interaction between SMV‐P P1 and the Rieske Fe/S protein by Y2H assay. Identification of P1–Rieske Fe/S protein interactions with other plants The Rieske Fe/S genes of calla lily and soybean were obtained by screening the cDNA libraries using the Rieske Fe/S gene of P. ternata as a probe. The Rieske Fe/S gene of A. thaliana was amplified by PCR using specific primer pairs designed to the published sequence (AJ243702). These three Rieske Fe/S genes were inserted into a pGADT7 vector and Y2H assays performed as described above. The cDNA sequences of the Rieske Fe/S genes of G. max , P. ternata and Z. aethiopica have been deposited in the EMBL/GenBank/DDBJ databases with accession numbers AM498291–AM498293, respectively. ACKNOWLEDGEMENTS Y.S. was a PhD student at Zhejiang University. The project was supported by grants from the Zhejiang Provincial Natural Science Foundation of China (ZA0207), the National Natural Science Foundation of China (30470080, 30200008, 30225031) and the National Basic Research (973) Program (2006CB708209). Rothamsted Research receives grant‐aided support from the Biotechnology and Biological Sciences Research Council of the United Kingdom. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Molecular Plant Pathology Wiley

A potyvirus P1 protein interacts with the Rieske Fe/S protein of its host

Loading next page...
 
/lp/wiley/a-potyvirus-p1-protein-interacts-with-the-rieske-fe-s-protein-of-its-mQr0vjBUSe

References (27)

Publisher
Wiley
Copyright
Copyright © 2007 Wiley Subscription Services, Inc., A Wiley Company
ISSN
1464-6722
eISSN
1364-3703
DOI
10.1111/j.1364-3703.2007.00426.x
pmid
20507538
Publisher site
See Article on Publisher Site

Abstract

INTRODUCTION Potyviruses are single‐stranded positive‐sense RNA viruses with a single large open reading frame (ORF) predicted to produce a large polyprotein that is subsequently processed to ten mature proteins ( Fig. 1a ). The P1 protein is released from the N‐proximal end of the polyprotein through autoproteolytic cleavage by its own C‐proximal serine‐type proteinase domain ( Verchot ., 1991 ). The function of P1 remains unclear, but cleavage of P1 is essential for viral infectivity ( Moreno ., 1998 ; Verchot and Carrington, 1995 ). P1 enhanced HC‐Pro‐mediated suppression of RNA silencing ( Pruss ., 1997 ), but did not itself suppress RNA silencing ( Anandalakshmi ., 1998 ). Until now, there have been no reports of interactions between viral P1 and host proteins. 1 Diagram showing (a) the organization of the SMV‐P genome, (b) the regions of the P1 gene and (c) the regions of the P. ternata Rieske Fe/S protein used in yeast two‐hybrid tests. Soybean mosaic virus (SMV), a member of the genus Potyvirus , is the most important viral pathogen of soybeans world wide. Numerous strains and isolates have been identified based on differences in their biological and serological properties, but most of them could only infect some species in the family Leguminosae. Recently, an unusual strain of SMV (SMV‐P) was identified from the aroid Pinellia ternata (Thunb.) Ten. ex Breitenb. in China ( Chen ., 2004 ) and was later found to occur widely in aroid plants ( Shi ., 2005 ). This strain could be mechanically inoculated to some soybean cultivars, in which it induced weak symptoms. The sequence of SMV‐P was very similar to that of SMV soybean isolates (SMV‐S) except in the P1 region, where it was much more similar to the aroid pathogen Dasheen mosaic virus (DsMV) than to SMV‐S. We now report studies showing that SMV‐P P1 interacts with the host Rieske Fe/S protein. RESULTS Interactions between SMV‐P‐P1 and other viral proteins Yeast two‐hybrid (Y2H) screens were used to test for interactions between SMV‐P P1 and the ten mature proteins of the virus. The expression of bait and target genes in yeast was confirmed by Western blot ( Fig. 2 ). Each potential interaction was tested three times in independent co‐transformations with positive and negative controls. No detectable interactions were found between P1 and any of the mature viral proteins of SMV‐P. 2 Western blot analysis of SMV‐P proteins expressed as fusion proteins in Saccharomyces cerevisiae strain Y109 using (a) the pGBKT7 vector and a c‐Myc monoclonal antibody or (b) the pGADT7 vector and an HA‐Tag polyclonal antibody. 1, pGBKT7 vector; 2, pGBK‐SMV‐P‐P1 plasmid with predicted fusion protein of 57.8 kDa; 3, pGADT7 vector; 4–13, pGAD‐SMV‐P‐X plasmid where gene X is: 4, P1; 5, HC‐Pro; 6, P3; 7, 6K1; 8, CI; 9, 6K2; 10, NIa‐VPg; 11, NIa‐Pro; 12, NIb; 13, CP; the fusion proteins had calculated molecular weights of 57.8, 72.7, 60.5, 26.0, 92.0, 26.0, 42.4, 47.9, 80.4 and 50.0 kDa, respectively. The positions of protein size markers (kDa) are shown on the left. Interactions between SMV‐P‐P1 and P. ternata proteins Using expressed SMV‐P P1 as bait in Y2H screens, 13 independent interacting clones were identified after two separate screenings of a P. ternata cDNA expression library (2.1 × 10 6 independent clones). Each clone could induce the two reporter genes, allowing growth on quadruple dropout plates and expression of α‐galactosidase activity. Sequencing showed that ten of the 13 clones were identical and had a ORF predicted to encode a 23.7‐kDa protein closely related to the cytochrome b6/f complex Rieske Fe/S genes of plants (nearest match by BLAST to Fritillaria agrestis O49078; score 8.4e‐95), whereas the other three clones were sequence fragments without recognizable ORFs. A plasmid encoding the complete gene was isolated and retransformed back to yeast either alone or in combination with pGBK‐SMV‐P‐P1, Lamin C or the binding domain vector alone. The reporter genes were only induced in combination with pGBK‐SMV‐P‐P1. The Rieske Fe/S gene of P. ternata was 942 nt long excluding the poly(A) tail and contained an ORF of 678 nt. The ORF was predicted to produce a protein of 226 amino acids and, by comparison with related proteins, this includes a transit peptide of 53 amino acids at its N‐terminus ( Fig. 3 ). A polyclonal antiserum raised to the polypeptide was used in Western blot with proteins isolated from P. ternata chloroplasts and this indicated that the mature Rieske Fe/S protein had a molecular weight of 18.6 kDa, consistent with the sequence data (data not shown). 3 Predicted amino acid sequence of the Rieske Fe/S protein of Pinellia ternata aligned with the homologous sequences of Fritillaria agrestis (O49078), Arabidopsis thaliana (Q9ZR03), Glycine max (this study) and Zantedeschia aethiopica (this study). Identical residues are marked ‘.’ and alignment gaps as ‘–’. The interaction between SMV‐P P1 and the mature Rieske Fe/S protein (without transit peptide) of the host was confirmed by in vitro co‐immunoprecipitation of the two proteins expressed from the wheat germ extract system. Proteins of the expected sizes of 21.7 (HA‐Fe/S) and 41.2 kDa (c‐Myc‐P1) were obtained, but were not detected in the controls ( Fig. 4 ). 4 Colorimetric (BCIP/NBT) analysis of SMV‐P P1 co‐immunoprecipitates with the Rieske Fe/S protein of Pinellia ternata . 1, endogenous biotinylated proteins of the in vitro reaction without templates (negative control); 2, co‐immunoprecipitates of the biotinylated HA‐Fe/S and c‐Myc‐SMVP‐P1 proteins; 3, cMyc‐SMV‐P P1 immunoprecitated with anti‐HA; 4, HA‐Fe/S immunoprecipitated with anti‐c‐Myc. The sizes of the HA and c‐Myc epitope tags were 3.1 and 2.6 kDa, respectively. The positions of protein size markers (kDa) are shown on the left. Dissection of the P1‐Rieske F/S protein interaction Y2H assays using different parts of the SMV‐P P1 protein ( Fig. 1 ) and three sections of the Rieske Fe/S protein showed that only the N‐terminal part (amino acids 1–82) of the SMV‐P P1 was responsible for the interaction with the Rieske Fe/S protein and that amino acids 1–33 interacted only with the transit peptide, while amino acids 34–82 could interact with the entire Rieske Fe/S protein ( Table 1 ). 1 Interactions (N, negative; P, positive) determined from yeast two‐hybrid experiments between different regions of the viral (SMV‐P) P1 and host ( Pinellia ternata ) Rieske Fe/S proteins. Human lamin C is included as a non‐specific control Rieske Fe/S protein 1–53 54–128 129–226 54–226 129–226 1–128 1–226 SMV‐P P1 1–33 P N N N N P P 34–82 P P P P P P P 83–330 N N N N N N N 1–82 P P P P P P P 1–164 P P P P P P P 1–246 P P P P P P P 1–330 P P P P P P P 83–264 N N N N N N N 265–330 N N N N N N N 34–164 P P P P P P P Human lamin C 1–165 N N N N N N N P1–Rieske Fe/S protein interactions in other hosts Y2H assays were used to examine the interactions between the P1 protein of SMV‐P and the Rieske Fe/S proteins of other plants. The unknown Rieske Fe/S proteins were identified and sequenced after screening cDNA libraries using the Rieske Fe/S gene of P. ternata as a probe. SMV‐P P1 interacted moderately with the Rieske Fe/S proteins of Glycine max cv. Xudou 1 and Zantedeschia aethiopica (both hosts of the virus) and weakly with that of the non‐host Arabidopsis thaliana . In parallel tests, the P1 protein of a soybean (SMV‐S) isolate interacted weakly with the Rieske Fe/S protein of its host G. max , but not at all to those of the non‐hosts Z. aethiopica and P. ternata . The amino acid sequences of the Rieske Fe/S proteins from different plants were relatively conserved expect for the transit peptide ( Fig. 3 ). DISCUSSION No detectable interactions were found between P1 and any of the mature viral proteins of SMV‐P. This contrasts with the weak interaction reported between the P1 and CI proteins of Potato virus A ( Merits ., 1999 ), but is in agreement with results reported for Pea seed‐borne mosaic virus ( Guo ., 2001 ) and SMV strain G7H ( Kang ., 2004 ). By contrast, there was a strong and consistent interaction with the Rieske Fe/S protein of its natural host, Pinellia ternata . The different strengths of the interactions between SMV‐P P1 and the Rieske Fe/S proteins of host and non‐host plants suggests a role in host range determination. The Rieske Fe/S protein of the cytochrome b6/f complex is an indispensable component of the photosynthetic electron transport chain in chloroplasts. It is a polypeptide that faces the stroma with only a few N‐terminal residues and is anchored in the membrane with a single transmembrane helix ( Karnauchov ., 1997 ). The large C‐terminal hydrophilic domain is exposed in the lumenal space of the thylakoid membrane system and provides the ligands for the [2 Fe – 2 S] cluster ( Zhang ., 1996 ). In higher plants, the Rieske protein is encoded in the nucleus and synthesized in the cytosol as a precursor molecule with a transit peptide mediating the import of the protein into the chloroplast stroma ( Madueno ., 1992 ). The signal for the subsequent thylakoid translocation step is provided by the N terminal region of the mature polypeptide chain that comprises the membrane anchor ( Madueno ., 1994 ). As mosaic and mottle symptoms are often observed after virus infection, it is reasonable to assume that functional and structural changes of the chloroplasts cause the chlorosis, and with some viruses infection is known to be associated with changes in the chloroplasts. After infection by Squash mosaic virus (genus Comovirus ), the rate of carbon fixation changed and the number of chloroplasts decreased but there was no detectable effect on the ultrastructure of the chloroplasts ( Magyarosy ., 1973 ). With Tobacco mosaic virus (TMV, genus Tobamovirus ), infection induced an inhibition of photosystem II electron transport, disturbing the oxygen‐evolving complex ( Lehto ., 2003 ). The possibility that one or more viral products were imported into the chloroplast was one of the proposed mechanisms underlying virus pathogenicity ( Culver ., 1991 ) and there is evidence that some viral proteins localize to chloroplasts. For example, the coat protein (CP) of TMV and the second triple gene block protein, TGBp2, of Barley stripe mosaic virus (genus Hordeivirus ) both accumulate in chloroplasts of systemically infected leaves ( Reinero and Beachy, 1986 ; Torrance ., 2006 ). It has also been shown that a CP–green fluorescent protein fusion of Cucumber necrosis virus (genus Tombusvirus ) targets chloroplasts in Nicotiana benthamiana leaves ( Xiang ., 2006 ). An association between potyvirus CP proteins and chloroplasts, and some interactions between the CP and chloroplast proteins have previously been reported ( Gunasinghe and Berger, 1991 ; Jimenez ., 2006 ; McClintock ., 1998 ). However, the way by which potyviruses induce symptoms remains unclear and our results suggest that the interaction between P1 and the Rieske Fe/S protein may be significant. The TMV replicase protein can disrupt the nuclear localization and subsequent function of interacting Aux/IAA proteins ( Padmanabhan ., 2006 ) and, by analogy, the SMV‐P P1 may block the transit peptide of the Rieske Fe/S protein and thus retard the Fe/S transport to the chloroplast, so delaying the formation of the cytochrome b6/f complex. This may explain why symptoms are most easily seen in newly expanding leaves. Once inside the chloroplast P1 may further interact with the Rieske Fe/S protein to modify the functions of the cytochrome b6/f complex. The N terminus of the P1 protein is the most variable region of the potyvirus genome, both in sequence and in length ( Adams ., 2005 ) and shows evidence of extensive recombination that is believed to be significant for adaptation to different hosts ( Valli ., 2007 ). We suggest that the P1–Rieske Fe/S protein interactions may play an important role in this adaptation. EXPERIMENTAL PROCEDURES Virus sources and test plants Healthy Pinellia ternata , Glycine max (soybean cv. Xudou 1), Zantedeschia aethiopica (Calla Lily) and Arabidopsis thaliana plants were grown in the greenhouse. SMV‐P was maintained by mechanical inoculation to P. ternata . Y2H tests for the interactions between SMV‐P P1 and the ten mature proteins of SMV‐P Y2H tests were performed using the Matchmaker Two‐Hybrid system 3 (Clontech, Mountain View, USA) according to the manufacturer's protocols. pGAD‐SMV‐P‐X plasmids were constructed for each of the ten genes of SMV‐P ( Fig. 1a ) by inserting the respective gene into a pGADT7 vector. The pGBK‐SMV‐P‐P1 plasmid was constructed by insertion of the P1 gene of SMV‐P into a pGBKT7 vector. These plasmids were first verified by sequencing, then transformed into Saccharomyces cerevisiae AH109. Western blots were used to verify the expression of inserted genes using an HA‐Tag ployclonal antibody (Clontech). Simultaneous co‐transformations of pGBK‐SMV‐P‐P1 and pGAD‐SMV‐P‐X to yeast were performed by the lithium acetate method using the small‐scale yeast transformation protocol. The co‐transformants were plated on double dropout plates. The positive interactions in the transformed yeast were confirmed by assaying colonies of the transformants on quadruple dropout plates containing X‐α‐Gal. The Y2H assay was then repeated after swapping the inserts between the AD and BD vectors. The plasmids pGBKT7‐53 and pGADT7‐T were used together as a positive control or in combination with our Gal4 fusion constructs as negative controls. Construction of the P. ternata cDNA library Total RNA was extracted from fresh, healthy leaves of P. ternata using TRIzol (Invitrogen) according to the manufacturer's manual. cDNA libraries were then constructed as described in the Matchmaker™ Library Construction & Screening Kits User Manual (Clontech). Double‐stranded cDNA and pGAD‐Rec were transformed to S. cerevisiae AH109. Screening the cDNA library by yeast mating pGBK‐SMV‐P‐P1 was transformed to S. cerevisiae Y187 and the cDNA library was screened by yeast mating. Plasmids were prepared from positive clones and used to transform Escherichia coli strain JM109. Recovered plasmids were sequenced. Diploids were plated on triple dropout medium (low stringency selection), and positives were re‐plated on to quadruple dropout plates (high stringency) containing X‐α‐gal. The target gene was subsequently re‐inserted into the pGADT7 vector and the two‐hybrid assay was repeated. All the positive interactions were re‐tested using yeast mating to eliminate false positives. In vitro co‐immunoprecipitation Plasmids pGAD‐Pt‐Fe/S and pGBK‐SMV‐P‐P1 were purified using the Promega Wizard ® Plus SV DNA Isolation System (Promega) according to the manufacturer's protocols. The SMV‐P P1 and Rieske Fe/S proteins were then produced in vitro using the TNT T7 Coupled Wheat Germ Extract system (Promega) and labelled with Transcend™ tRNA (Promega) according to the manufacturer's protocols. The two biotinylated translation products HA‐Fe/S and c‐Myc‐SMVP‐P1 were allowed to interact and were co‐immunoprecipitated with Protein A using the Matchmaker™ Co‐IP Kit (Clontech). After electrophoretic separation and transfer to the nitrocellulose membrane, the in vitro co‐immunoprecipitation was detected using the Transcend™ Non‐Radioactive Translation Detection System (Promega). Controls were endogenous biotinylated proteins of the in vitro reaction without templates, cMyc‐SMV‐P P1 immunoprecitated with anti‐HA and HA‐Fe/S immunoprecipitated with anti‐c‐Myc. Identification of regions involved in the interaction between SMV‐P P1 and the Rieske Fe/S protein Truncated parts of the SMV‐P P1 ( Table 1 , Fig. 1b ) were inserted into a pGBKT7 vector. The Rieske Fe/S gene was divided into three parts according to the published structure and fragments were inserted into a pGADT7 vector ( Table 1 , Fig. 1c ). Plasmid pGBKT7‐lam (Clontech), which encodes a fusion of the DNA‐BD with human lamin C, was used as a non‐specific control. Lamin C neither forms complexes nor interacts with most other proteins. These plasmids were then used to identify the regions involved in the interaction between SMV‐P P1 and the Rieske Fe/S protein by Y2H assay. Identification of P1–Rieske Fe/S protein interactions with other plants The Rieske Fe/S genes of calla lily and soybean were obtained by screening the cDNA libraries using the Rieske Fe/S gene of P. ternata as a probe. The Rieske Fe/S gene of A. thaliana was amplified by PCR using specific primer pairs designed to the published sequence (AJ243702). These three Rieske Fe/S genes were inserted into a pGADT7 vector and Y2H assays performed as described above. The cDNA sequences of the Rieske Fe/S genes of G. max , P. ternata and Z. aethiopica have been deposited in the EMBL/GenBank/DDBJ databases with accession numbers AM498291–AM498293, respectively. ACKNOWLEDGEMENTS Y.S. was a PhD student at Zhejiang University. The project was supported by grants from the Zhejiang Provincial Natural Science Foundation of China (ZA0207), the National Natural Science Foundation of China (30470080, 30200008, 30225031) and the National Basic Research (973) Program (2006CB708209). Rothamsted Research receives grant‐aided support from the Biotechnology and Biological Sciences Research Council of the United Kingdom.

Journal

Molecular Plant PathologyWiley

Published: Nov 1, 2007

There are no references for this article.