# Sivamani E, $ Qu R.
# Department of Plant, Soil and Environmental Sciences, Montana State University, Bozeman, Montana, MT 59717-0312, USA.
$ Department of Crop Science, Box 7620, North Carolina State University, Raleigh, NC 27695-7620.
R. Qu, Department of Crop Science, Box 7620, North Carolina State University, Raleigh, NC
Telephone: 919-515-7616 Fax 919-515-7959 email: firstname.lastname@example.org
Recombinant plasmids containing cDNA from the wheat streak mosaic virus (WSMV) isolated from Montana were constructed and nucleotide sequences were determined. A 1485 nucleotide sequence containing the putative full-length nuclear inclusion protein b (NIb) and flanking regions was further studied. The deduced amino acid sequence of the NIb gene was compared to homologues from two aphid transmitted (PVY and TEV) and one mite transmitted (BrSMV) potyviruses. The WSMV-NIb protein showed 44-66% identity with corresponding proteins of the other potyviruses, with the highest degree of conservation with the mite transmitted BrSMV. The predicted WSMV-NIb protein product is 48.3 kDa. However, the WSMV-NIb cDNA in vitro translation product migrated at approximately 42 kDa.
Wheat streak mosaic virus (WSMV) was first found in wheat in the USA by McKinney in 1937 (McKinney, 1937). The virus causes a severe mosaic and stunting of winter wheat, barley, oats, maize rye and certain millets. It is a serious threat to winter and spring wheat in the USA (Pfannenstiel and Niblett, 1978) and Canada (Shahwan and Hill, 1984). Yield losses due to WSMV cost US wheat producers millions of dollars per year. The virus is known to occur in North and South America, Jordan, Romania, the former Yugoslavia and Russia (Slykhuis, 1967). WSMV has been reported to infect 116 species of 34 genera of the family Gramineae (Edwardson and Christie, 1991). WSMV is a member of Rymovirus genus of the family Potyviridae (Shukla et al., 1994). The virus is morphologically identical to other potyviruses, having long flexuous particles (700 X 13 nm) with a single stranded RNA genome of approximately 8.5 kb (Brakke, 1971). WSMV differs from other typical potyviruses by its transmissibility through mites (Hollings and Brunt, 1981) and its considerably larger (about 45 kDa) capsid protein (Niblett et al., 1991). Among the mite transmitted potyviruses (rymoviruses), a complete nucleotide sequence has been published only for brome streak mosaic virus (BrSMV) (Gotz and Maiss, 1995). Partial sequences, mainly comprising the coat protein (CP) region of other mite transmitted viruses such as, ryegrass mosaic virus (RGMV), Agropyron mosaic virus (AgMV) and Hordeum mosaic virus (HoMV), have also been published (Salm et al., 1996 a and b). A partial nucleotide sequence of the WSMV genome containing part of the large nuclear inclusion protein b (NIb), the coat protein and the 3' non-coding region was reported earlier (Niblett et al., 1991).
The NIb protein of potyviruses is believed to be an RNA dependent RNA polymerase (RdRp). RdRp has been shown to be essential for replicase function of plus-, minus- and double stranded RNA viruses (Ishihama and Nagata, 1988). Among potyvirus gene products, the most conserved region is in the RdRp (NIb) gene (Shukla et al., 1994). The replicase genes of many plant viruses including potyviruses have been found to be valuable in conferring resistance to virus infection when introduced and expressed in transgenic plants (Carr and Zaitlin, 1993). In this paper we report the complete nucleotide sequence of the WSMV-NIb, comparison of its deduced amino acid sequence with three other potyviruses and in vitro expression of the cDNA.
Ten-day-old wheat seedlings (cv. Rambo) were mechanically inoculated with a Montana isolate of WSMV previously purified from infected wheat leaves collected in Conrad, Montana. Ten days post inoculation, virus particles of WSMV and genomic RNA were purified from symptomatic tissue using a procedure from S. A. Lommel (personal communication). Two ug of total viral RNA were used for cDNA synthesis, using the Universal RibocloneR cDNA synthesis system (Promega, USA) with an oligo dT15 primer to initiate first strand synthesis. cDNA was ligated into EcoR I linearized pUC119 and transformed in E.coli DH5alpha cells. Transformants were screened by colony hybridization using a 32P labelled 1.3 kb WSMV-coat protein coding sequence from pSOG27 (plasmid kindly provided by Dr. L. Crossland, CIBA-Geigy, USA). Among the recombinant plasmids obtained, two plasmids named p5 and p37 contained insert sizes larger than 2.5 kb. Clone p37 (Figure 1) was selected for further study and the fragment resulting from digestion with Hind III and BamHI was subcloned. From this subclone, many other clones ranging from 100-1000 bp were obtained and used for sequencing.
DNA sequencing was performed using the SequenaseTM version 2.0 DNA sequencing kit (USB-Amersham, USA) and/or at the DNA sequencing/synthesis facility at Iowa State University, Ames, Iowa, USA. All sequences were determined on both strands and all subcloning restriction sites were overlapped. Sequence data were analysed with GCG- Wisconsin Sequence Analysis Software (Genetics Computer Group, Inc., Wisconsin, USA) and BOXSHADE/DOS 2.7 (Kay Hoffmann, Bioinformatics Group, ISREC, Switzerland). The 1485 nucleotide sequence data reported in this paper is deposited in the GenBank under the accession number U67937.
The putative amino- and carboxyl- termini of the NIb gene were derived by comparing the sequence data with other published potyviral sequences (see below). The following forward and reverse oligonucleotide primers were designed to contain translation start or stop codons (italicized) and a BamHI restriction enzyme site (underlined) in order to PCR amplify the WSMV-NIb cDNA for in vitro expression studies.
Forward WSMV-NIb: 5' CGGGATCCAACAATGAGTCTCCAAATGACG 3';
Reverse WSMV-NIb: 5' CGCGGATCCTTACTGAATAGCCTTCGGTTC 3'.
Polymerase chain reaction was performed using Taq polymerase (Promega, USA) and plasmid DNA from WSMV-cDNA clone p37 to obtain a 1.3 kb PCR product of the WSMV-NIb gene.
For in vitro expression, the WSMV-NIb cDNA synthesized by PCR was digested with BamHI and cloned into pBlueScript SK+ (Stratagene, USA) under control of the T7 promoter. In vitro transcription and translation of the WSMV-NIb cDNA was performed using the TNTR coupled reticulocyte lysate system (Promega, USA) according to manufacturer's instructions. The translated products were analysed and autoradiographed in 12.5% and/or 15% SDS-PAGE protein gels as described by Laemmli (1970).
Complete or partial genomic sequences have been reported for a number of potyviruses. The potyvirus genome encodes a single large polyprotein from which functional proteins are produced by cleavage at specific sites (Shukla et al., 1994). Comparison of the 427 deduced amino acid sequence encoded by the ORF of WSMV from clone p37 with the corresponding sequences of the polyproteins of other potyviruses indicated that the putative product contains the entire presumptive NIb protein flanked by four amino acid residues of the C-terminus of the NIa protein and an amino acid residue of the N-terminus of the coat protein (Figure 2).
A previous comparison of the NIa-NIb junction (Shukla et al., 1991) revealed a simple group-specific motif V-X-X-Q-(A, S, G or V) that is common to all potyviruses. Such a motif was found (V-S-W-Q-S) near the amino terminus of the deduced amino acid sequence of clone p37 between positions 54 and 58 (Figure 1). Therefore the Q/S at this region (positions 57/58; Figure 1) is assigned as the cleavage site to yield the N terminus of the WSMV-NIb protein. Similar to earlier observations (Niblett et al., 1991), five potential sites for hydrolysis by a viral encoded proteinase at the NIb-CP junction have also been found (Figure 1). Since WSMV contains two major coat proteins of 42-47 kDa and 36 kDa respectively (Brakke, 1971, Niblett et al., 1991 and our unpublished data) we have assumed the first Q/S (at positions 479/480; Figure 1) is the primary cleavage site between NIb and CP that will yield a protein of 46 kDa, as observed for the larger CP of WSMV (Niblett et al., 1991 and Zagula et al., 1992). The second Q/S site is very close to the putative NIb-CP junctions found in all other potyviruses with known sequences, including that of the mite transmitted BrSMV, and may also be cleaved to yield a CP of 36 kDa. Since WSMV has larger CP, its NIb protein could be substantially smaller (~48 kDa) than those of other potyviruses (~58 kDa). However, it cannot be excluded that WSMV may have major NIb proteins of sizes 48 and 58 kDa as a result of cleavages at either Q/S site.
On the basis of sequence similarity with other positive strand RNA viruses, the NIb of potyviruses was proposed to be an RNA dependent RNA polymerase (Riechmann et al., 1992). One of the most important consensus motifs (S-(T)-G-X-X-X-T-X-X-X-N-S-(T)-(18 - 37 A. A.)-G-D-D) is highly conserved in a variety of animal and plant positive stranded RNA viral RNA-dependent polymerases (Kamer and Argos, 1984), and is present in the deduced WSMV-NIb protein in a position similar to that of other potyviruses starting at position 358 (Figure 2).
The deduced amino acid sequence of the putative NIb of WSMV (positions 54-495; Figure 2) showed 64.0%, 62.6% and 77.6% similarity to PVY (positions 2272-2798; Robaglia et al., 1989), TEV (positions 2276-2793; Allison et al., 1986) (aphid transmitted viruses) and BrSMV (positions 2272-2828; Gotz and Maiss, 1995) (mite transmitted), respectively (Figure 2). Percent identity with the same three viruses was 44.1%, 45.5% and 66.2% respectively. The 3' end 407 bp (nucleotides 1031 - 1438) of the nucleotide sequence of p37 WSMV-NIb was 99.5 % identical to nucleotides 1 to 408 of the published sequence of WSMV (Niblett et al., 1991).
In the published partial NIb sequence of WSMV (Niblett et al., 1991), a stretch of 24 amino acid residues in front of the GDD motif seems to be unrelated to the NIb sequences of potyviruses documented so far and does not match the sequence of WSMV-NIb determined by us (Figure 2). These 24 amino acids were most likely the result of reading in a different frame (a single nucleotide shift). One of the four important consensus sequences of RdRp, namely G-N-N-S-G-Q-P-S-T-V-V-D (Poch et al., 1989) is present in this region (Figure 2).
Based on the putative cleavage sites, the WSMV-NIb protein is composed of 423 amino acids and the expected mass is 48.3 kDa. However, when the in vitro translated WSMV-NIb gene product was analysed in SDS-PAGE gels, it was found that the major product migrated to the position of about 42 kDa with additional proteins at 40 kDa and 35 kDa (Figure 3a). We conducted a time course in vitro translation experiment to determine if a precursor of the expected size (48 kDa) was initially synthesized and later reduced by proteolytic digestion to the size we observed. However, there was no precursor of larger size observed in the early time points of in vitro translation (Figure 3b), indicating that the 42 kDa protein was not a product of unknown proteolytic activity in the system. We have also sequenced the clone pBS-WSMV-NIb, used for the in vitro translation from both ends, and found that the sequence was identical to that of clone p37. The size difference could be due to the hydrophobic nature of this protein (about 48% of the WSMV-NIb protein comprises hydrophobic amino acid residues) which may bind excess SDS, thereby causing anomalous mobility (See and Jackowski, 1990).
Transformation experiments are being carried out to engineer and express the WSMV-NIb gene in wheat cultivars and more than ten independent transgenic lines have been obtained. These transgenic lines are to be tested for resistance against WSMV inoculations.
This project was funded in part by NSF grant OSR - 9350546 to the Montana Agricultural Experiment Station. The authors are grateful to Drs. S. A. Lommel, T. Carroll, W. Dyer, M. Young, S. Filichkin and F. Albert for critical suggestions and help. The authors also thank Dr. L. Crossland for plasmid pSOG27 and Ms. S. Brumfield and Mr. C. Brey for providing a Montana isolate of WSMV.
Gotz R, Maiss E, 1995. The complete nucleotide sequence and genome organization of the mite-transmitted brome streak mosaic rymovirus in comparison with those of potyviruses. Journal of General Virology 76, 2035-2042.
Niblett CL, Zagula KR, Calvert LA, Kendall TL, Stark DM, Smith CE, Beachy RN, Lommel SA, 1991. cDNA cloning and nucleotide sequence of the wheat streak mosaic virus capsid protein gene. Journal of General Virology 72, 499-504.
Salm SN, Rey MEC, French R, 1996a. Molecular cloning and nucleotide sequencing of a South African isolate of ryegrass mosaic virus and in vitro expression of the coat protein gene. Archives of Virology 141, 185-195.
Salm SN, Rey MEC, Robertson NL, French R, Rabenstein F, Schubert J, 1996b. Molecular cloning and nucleotide sequencing of the partial genomes of Agropyron and Hordeum mosaic viruses, two members of the Rymovirus genus in the taxonomic family Potyviridae. Archives of Virology 141, 2115-2127.
See YP, Jackowski G, 1990. Estimating molecular weights of polypeptides by SDS gel electrophoresis. In: Creighton TE (ed) Protein structure, a practical approach. IRL Press at Oxford University Press, England, 1-21.
Zagula KR, Niblett CL, Robertson NL, French R, Lommel SA, 1992. Potyviridae: genus rymovirus. In: Barnett OW (ed) Potyvirus taxonomy. Springer, Wien and New York, Archives of Virology (supplement) 5, 269-276.