The E116 isolate of Dutch pea early-browning virus is a recombinant virus

The E116 isolate of Dutch pea early-browning virus is a recombinant virus

Virus Research 60 (1999) 87 – 94 The E116 isolate of Dutch pea early-browning virus is a recombinant virus Maud M. Swanson, Stuart A. MacFarlane * De...

322KB Sizes 0 Downloads 27 Views

Virus Research 60 (1999) 87 – 94

The E116 isolate of Dutch pea early-browning virus is a recombinant virus Maud M. Swanson, Stuart A. MacFarlane * Department of Virology, Scottish Crop Research Institute, In6ergowrie, Dundee DD2 5DA, UK Received 4 October 1998; received in revised form 22 December 1998; accepted 22 December 1998

Abstract The complete nucleotide sequence of RNA2 of the E116 isolate of Dutch pea early-browning virus (PEBV-D) was obtained from overlapping cDNA clones. The RNA was found to encode three open reading frames corresponding to, in 5% to 3% order, the coat protein, the 2b nematode transmission protein and the C-terminal part of the cysteine-rich 1b protein derived from RNA1. The 3% non-coding region of PEBV-D RNA2 was also shown to be derived from RNA1. This is the first demonstration that recombination of PEBV occurs in nature. Comparison of the amino acid sequences of the PEBV-D RNA2 proteins with those of British PEBV and several isolates of tobacco rattle virus reveals complex patterns of mixing of the genomes of these two viruses. © 1999 Elsevier Science B.V. All rights reserved. Keywords: Pea early-browning virus; RNA recombination; Tobravirus

Pea early-browning virus (PEBV) belongs to the tobravirus genus of plant viruses, which also includes tobacco rattle virus (TRV) and pepper ringspot virus (PRV). These viruses have a genome consisting of two, single-stranded, positive sense RNAs which are packaged in rodshaped particles (Harrison and Robinson, 1986).

* Corresponding author. Tel.: + 44-1382-562731; fax: +441382-562426. E-mail address: [email protected] (S.A. MacFarlane)

Sequencing studies have been carried out primarily with TRV, for which more than six isolates have been investigated, whereas partial sequence data is available for only two isolates of PEBV and a single isolate of PRV. The gene organisation of the larger genomic RNA (RNA1) of the three tobraviruses is very highly conserved (Hamilton et al., 1987; MacFarlane et al., 1989). Thus, RNA1 encodes two proteins involved in replication (the second of which is produced by readthrough translation of the termination codon of the 5% proximal, helicase-like gene). Immediately downstream is a gene encoding a putative

0168-1702/99/$ - see front matter © 1999 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 1 7 0 2 ( 9 9 ) 0 0 0 0 5 - 2

88

M.M. Swanson, S.A. MacFarlane / Virus Research 60 (1999) 87–94

cell-to-cell movement protein, which is followed by a gene encoding a cysteine-rich protein that, at least in PEBV, is involved in seed transmission of the virus (Wang et al., 1997). Hybridisation studies showed that, despite the similarities in gene organisation, RNA1 of each of the three tobraviruses share little, if any, nucleotide sequence identity. The RNA1 among a group of ten different TRV isolates was highly conserved but completely distinct from RNA1 of three isolates of PEBV and one isolate of PRV (Robinson and Harrison, 1985a). The situation is markedly different for RNA2 where, for TRV, hybridisation experiments showed that there was very little nucleotide sequence conservation between different isolates (Robinson, 1983; Robinson and Harrison, 1985a). Previously, the sequencing of RNA2 from five TRV isolates revealed that each encoded a virus coat protein gene but differed extensively in the remainder of the RNA. TRV PpK20 should perhaps be considered an isolate with the entire complement of genes, as RNA2 encodes the coat protein and two other proteins (40 and 33 kDa), one of which (40 kDa) has been shown to be necessary for nematode transmission of this isolate (Hernandez et al., 1995, 1997). RNA2 of two other TRV isolates, PSG and PLB (Cornelissen et al., 1986; Angenent et al., 1989), contains the coat protein gene followed by part of the 3% terminal sequence of TRV RNA1. RNA2 of TRV isolate ON is also a recombinant molecule, encoding the coat protein gene, a gene for a 27-kDa protein, a gene for a 9-kDa protein and 3% sequences derived from TRV RNA1, including part of the gene for the cysteine-rich, 16-kDa protein (Uhde et al., 1998). TRV TCM RNA2 has an even more complex structure comprising the 5% terminus derived from TRV RNA2, the central region probably derived from PEBV RNA2, and the 3% region derived from TRV RNA1 (Angenent et al., 1986). Recently RNA2 from two other TRV isolates (PaY4 and TpO1) has been sequenced (MacFarlane, unpublished). These isolates resemble PpK20 in some respects; they each encode the coat protein and, respectively, two or three other proteins but again have very little nucleotide sequence identity with one another or with PpK20.

In contrast to TRV, relatively little information about the sequence diversity of PEBV RNA2 is available. Isolates of PEBV have been grouped into three different serotypes corresponding to British PEBV (PEBV-B), Dutch PEBV (PEBV-D) and broad bean yellow band virus (PEBVBBYBV) (Russo et al., 1984). Hybridisation experiments showed that, as was the case with TRV, RNA1 of PEBV isolates of each serotype shared extensive nucleotide sequence similarities (Robinson and Harrison, 1985b). However, RNA2 of isolates of the three serotypes did not appear to be closely related, although two isolates of PEBV-B (SHE and SP5) were similar to one another. Comparison of the nucleotide sequence of RNA2 of PEBV isolates SP5 and TpA56, both PEBV-B serotypes, revealed almost complete (99.7%) identity (MacFarlane and Brown, 1995). The only other known (probable) PEBV sequences are those that are present in the central region of the TRV recombinant isolate TCM. The central portion of RNA2 of this virus encodes genes for a coat protein and a 29.1-kDa protein that hybridised to probes derived from a Dutch isolate of PEBV (Angenent et al., 1986). In order to understand the natural variation existing among PEBV isolates from the different serotype groups, we have cloned and sequenced the complete RNA2 from the E116 isolate of PEBV-D (Bos and van der Want, 1962). Unexpectedly, comparison of this sequence with that of other tobravirus isolates revealed that PEBV-D E116 is a recombinant isolate. This is the first demonstration that PEBV, like TRV, undergoes recombination in nature. PEBV isolate E116 was obtained originally from L. Bos (IPO-DLO, Wageningen, The Netherlands) as infected pea seed, and maintained at SCRI by limited passage in Nicotiana cle6elandii at SCRI. Virus was isolated from infected N. cle6elandii plants at 3 weeks post-inoculation as described previously (MacFarlane et al., 1989), except that polyethylene glycol-precipitated virus was not extracted with 1:1 chloroform:isobutanol before further centrifugation. Immunosorbent electron microscopy confirmed that the purified virus reacted strongly to antibody specific for PEBV-D but did not react to antibody specific to PEBV-B.

M.M. Swanson, S.A. MacFarlane / Virus Research 60 (1999) 87–94

Viral RNA was extracted from purified virus by digestion with proteinase K, followed by extraction with phenol and chloroform, and precipitation with an equal volume of 4 M LiCl (MacFarlane et al., 1989). Double-stranded cDNA was synthesised using a kit according to the manufacturer’s instruction (Pharmacia), except that first-strand synthesis was primed with oligonucleotide 158 (5% GGGCGTAATAACGCTTACG 3%), which is complementary to the 3% terminal nucleotides of all PEBV and TRV RNAs sequenced to date. The cDNA was size fractionated by passage through a Sephacryl S-300 spin column (Pharmacia), ligated to EcoRI/NotI linkers, and ligated into EcoRI-cut pBluescript (Stratagene). A clone (D6) containing the largest (: 2.4 kb) cDNA insert was chosen for further analysis. The terminal regions of clone D6 were sequenced using primers specific for the pBluescript plasmid, showing that one end of the clone was almost identical to sequences at the 3% terminus of PEBV-B RNAs 1 and 2. Consequently, an oligonucleotide primer (165), complementary to sequence at the other end of clone D6, was designed to allow cloning of the 5% terminus of PEBV-D RNA2 using ligation-anchored polymerase chain reaction (PCR) (Troutt et al., 1992). First-strand cDNA was synthesised from PEBVD RNA using oligonucleotide 158 and MMLV reverse transcriptase according to the manufacturer’s instructions (Life Technologies). The RNA was removed by incubation at 65°C for 30 min in 400 mM NaOH and, subsequently, the cDNA was purified by binding to silica (Nucleon EasiClene, Scotlab) and precipitation. T4 RNA polymerase was used to ligate a 3% blocked anchor primer (5% CCTATAGTGAGTCGTATTACCTGCAGCATGCCA 3%) to the 3% end of the firststrand cDNA. This region of the cDNA was PCR amplified with Taq polymerase (Gibco-BRL) using primer 165 (5% ATCATACGGCCAATGAC 3%) and a second primer (5% TGGCATGCTGCAGGTAATACGAC 3%), complementary to part of the anchor primer. The PCR reaction conditions were: 95°C for 4 min, 80oC for 2 min, followed by three cycles of 95°C for 45 s, 39°C for 45 s, 72°C for 1 min, then 25 cycles of 95°C for 45

89

s, 49°C for 45 s, and 72°C for 1 min. This procedure yielded a product of 370 nucleotides (nt), which was purified from an agarose gel and ligated into the PCR-cloning vector pT7Blue (Novagen) to produce clone DS-2. Sequencing reactions were carried out using a fluorescent dye terminator kit (Perkin Elmer), according to the manufacturer’s instructions, and the data were collected with an ABI 373 automatic sequencer. Clones D6 and DS-2 were fully sequenced on both strands using virus-specific primers that were designed during the course of the sequencing programme. Sequence analysis was carried out using Wisconsin Package programs (Genetics Computer Group). Assembly of the sequence derived from clones D6 and DS-2 showed that RNA2 of the E116 type isolate of PEBV-D is 2868 nucleotides (nt) in length. The 5% and 3% non-coding regions (NCRs) of PEBV-D RNA2 are 519 nt and 451 nt, respectively. The first six bases at the 5% terminus (AUAAAA) are identical to those of PEBV-B and most isolates of TRV (Hamilton et al., 1987). However, in the first 100 nt PEBV-D and PEBVB are only 76% identical whereas, for example, the same region in three different TRV isolates are more than 90% identical (MacFarlane, 1996). In the 3% NCR, PEBV-D RNA2 and RNA1 of PEBV-B are 97% identical (Fig. 1A). There are three significant open reading frames (ORFs) encoded by the viral RNA (Fig. 1B). The first ORF starts at nucleotide 520 (the eighth AUG codon from the 5% terminus of the RNA) and extends to nucleotide 1128. This ORF encodes the coat protein (CP), which has a calculated molecular mass of 21 857 Da and has amino acid similarity to the CP of PEBV-B and TRV (Table 1). The PEBV-D CP is most similar to that of the recombinant TRV isolate TCM, and is more similar to the CP of three isolates of TRV (ON, PaY4 and TpO1) than it is to the CP of PEBV-B. Robinson et al. (1987) proposed a system to name the proteins encoded by the internal genes of RNA1 of tobraviruses according to the relative size of the subgenomic RNA (sgRNA) from which they are expressed. Thus, the TRV 29-kDa movement protein would be referred to as the 1a protein and the 16-kDa protein would be referred

90

M.M. Swanson, S.A. MacFarlane / Virus Research 60 (1999) 87–94

Fig. 1. (A) Nucleotide sequence comparison of the 3% terminal regions of PEBV-D E116 RNA2 and PEBV-B RNA1. PEBV-D RNA2 is the upper sequence and PEBV-B is the lower sequence. The translation termination codon of the 1b (12-kDa) gene is highlighted in bold and underlined. Vertical lines indicate nucleotide identities, dots indicate gaps introduced to maximise the sequence homologies. (B) Genome organisation of RNA2 of PEBV-D E116. Open reading frames are boxed and include the gene name; D1b is truncated at the 5% end. Vertical lines below the genome map are spaced at 500-nt intervals. The 3% part of the sequence that is derived from RNA1 is underlined.

M.M. Swanson, S.A. MacFarlane / Virus Research 60 (1999) 87–94

91

Table 1 Percentage amino acid similaritiesa between the coat proteins of isolates of PEBV and TRV

TRV TCM TRV ON TRV PaY4 TRV TpO1 PEBV-B TpA56 PRV TRV PpK20 TRV PLB a b

PEBV-D E116

TRV TCM

TRV ON

TRV PaY4

TRV TpO1

92.6 89.6 89.6 79.0 59.4

88.6 88.2 76.5 60.5

93.8 79.0 58.6

80.6 59.6

60.2

63.1 53.9 53.9

61.0 53.4 52.4

61.0 56.9 55.5

60.5 56.4 56.0

61.7 57.1 55.6

PEBV-B TpA56

64.7 54.6 53.3

PRVb

TRV PpK20

63.1 64.0

94.1

Data were generated using the default parameters of the GCG Gap sequence comparison program. Bergh et al., 1985.

to as the 1b protein. For TRV PpK20 RNA2 all three proteins are thought to be expressed from sgRNAs, including the 5% proximal CP. Thus, the CP might also be referred to as the 2a protein, the second, 40K gene would encode the 2b protein and the 3% proximal 33K gene would encode the 2c protein. For RNA2 of PEBV-D E116, the second ORF (extending from nucleotide 1248 to nucleotide 2009) encodes the 2b protein with a calculated molecular mass of 28 885 Da. This protein has amino acid similarity with the 2b proteins of several other tobravirus isolates (Table 2). The third ORF extends from nucleotide 2185 to 2415 and encodes a protein of 77 amino acids that is homologous to the C-terminus of the PEBV-B RNA1-encoded 1b (12 kDa, cysteinerich) protein (Fig. 2). Thus, PEBV-D E116 is a recombinant virus, which lacks a homologue of the 3% proximal PEBV-B 2c gene but instead car-

ries the entire 3% NCR and 231 nt from the 3% part of the 3% proximal 1b gene of RNA1. It was previously noted that, although several TRV isolates were known to be natural recombinants in which the RNA2 included some sequences derived from TRV RNA1 and PEBV RNA2, there were no known recombinant isolates of PEBV (MacFarlane, 1997). A PEBV RNA2 recombinant containing coding and 3% non-coding sequences from TRV was generated in vitro and shown to replicate efficiently (Mueller et al., 1997), indicating that such recombinants could be expected to occur in nature. Although the mechanism of recombination in tobraviruses (specifically TRV) is not known, a motif (AUAAACAUU), which resembles the 5% terminal sequences of TRV genomic RNAs 1 and 2 (AUAAAACAUU), was found to occur only 17 bases upstream of the recombination junction in

Table 2 Percentage amino acid similaritiesa between the 2b nematode transmission proteins of isolates of PEBV and TRV

TRV TCM TRV ON TRV PaY4 PEBV-B TpA56 TRV TpO1 TRV PpK20 a

PEBV-D E116

TRV TCM

TRV ON

TRV PaY4

PEBV-B TpA56

TRV TpO1

89.3 78.1 77.3 58.5 52.7 51.0

80.1 79.3 55.6 53.3 51.6

98.3 60.6 57.4 53.0

59.8 56.4 52.6

67.3 44.7

41.7

Data were generated using the default parameters in the GCG Gap sequence comparison program.

92

M.M. Swanson, S.A. MacFarlane / Virus Research 60 (1999) 87–94

Fig. 2. Amino acid sequence comparison of the PEBV-B 12-kDa protein (upper sequence) and N-terminal-truncated 12-kDa protein encoded by PEBV-D RNA2 (lower sequence). Vertical lines indicate identical residues, colons indicate conservative substitutions.

TRV isolate PLB (Angenent et al., 1989). It was suggested that this sequence might act as a cryptic polymerase recognition site that would promote aberrant template switching during RNA2 replication. Two other recombinants (TCM and PSG) were found to have a related sequence (AUAAUUGUU), this time located just downstream of the recombination junction, and recombination of TRV isolate PpK20 produced strand switching within the sequence AUUUUAAUUU (Hernandez et al., 1996).

However, close examination of the sequences of different recombinant tobraviruses suggests, alternatively, that the recombination event in some isolates might be stimulated by internal, sgRNA promoter sequences. As described above, all of the genes encoded by RNA2 of the tobraviruses are thought to be expressed from 3% co-terminal sgRNAs. Tobravirus CP sgRNAs usually initiate about 400 nt downstream of the 5% terminus of RNA2 (Goulden et al., 1990). Transcription begins at the first adenine in the motif GCAUA, and

Fig. 3. Alignment of the 2b–2c intergenic region from different tobravirus isolates. PaY4, TpO1, TCM and PpK20 are TRV isolates. E116 is PEBV-D and TpA56 is PEBV-B. The termination codon of the upstream, 2b gene is highlighted by diagonal hatching, and the initiation codon of the downstream, 2c gene is highlighted in black. Grey highlighting identifies identical residues present in four or more isolates. The probable motif for initiation of the 2c sgRNA is boxed. The position in the PEBV-D E116 sequence where recombination has occurred is marked by an asterisk.

M.M. Swanson, S.A. MacFarlane / Virus Research 60 (1999) 87–94

for PEBV SP5 and TRV TCM the CP sgRNA has the terminal sequence AUAAUUAUACUG, whereas for TRV PSG it is AUAAAAUAAACUG and for TRV PpK20 it is AUAAAUAUAAUAC. These sequences have striking similarities with the 5% termini of the tobraviral genomic RNAs and also with the sequences flanking the recombination junctions described above. The TRV PpK20 recombinants R1 and R2 (Hernandez et al., 1996) have recombination junctions located only 40 nt and 42 nt, respectively, downstream of the CP sgRNA promoter. Fig. 3 shows an alignment of sequences immediately downstream of the 2b genes of PEBV-D E116, TRV TCM and several non-recombinant TRV isolates. This shows that RNA2 of PEBV-D E116 and TRV TCM were probably created by recombination that occurred only 16 nt and 84 nt, respectively, downstream of the putative 2c subgenomic promoter of their progenitor RNAs. This alignment also emphasises the sequence diversity among PEBV isolates that this study has revealed. Although the 2b–2c intergenic regions of PEBV-D and TRV TCM are most similar, this region of PEBV-D is more similar to all the other TRV isolates than it is to PEBV-B. It is also apparent from the comparisons of the CP and 2b proteins of the different viruses that PEBV-D most closely resembles TRV TCM. However, while the CP of PEBV-D is more similar to the CP of TRV ON, TRV PaY4 and TRV TpO1 than to that of PEBVB, the 2b protein of PEBV-D is more similar to the 2b protein of PEBV-B than to the 2b protein of TRV TpO1 (Tables 1 and 2). This clear evidence of extensive mixing of PEBV and TRV RNA2 sequences in field isolates of tobraviruses probably has biological significance. The genes encoded by RNA2 are known to be involved in the transmission of these viruses by vector nematodes (MacFarlane et al., 1996; Hernandez et al., 1997). Changing the combination of CP and 2b genes that are carried by a particular PEBV or TRV isolate probably will alter the species of nematode that acquires and transmits the virus, and may also change the range of host plants that the virus is transmitted to. The EMBL databank accession number for the sequence of PEBV-D E116 RNA2 is AJ006500.

93

Acknowledgements SCRI is grant-aided by the Scottish Office Agriculture, Environment and Fisheries Department.

References Angenent, G.C., Linthorst, H.J.M., van Belkum, A.F., Cornelissen, B.J.C., Bol, J.F., 1986. RNA2 of tobacco rattle virus strain TCM encodes an unexpected gene. Nucleic Acids Res. 14, 4673 – 4682. Angenent, G.C., Posthumus, E., Brederode, F.T., Bol, J.F., 1989. Genome structure of tobacco rattle virus strain PLB: further evidence on the occurrence of RNA recombination among tobraviruses. Virology 171, 271 – 274. Bergh, S.T., Koziel, M.G., Huang, S.-C., Thomas, R.A., Gilley, D.P., Siegel, A., 1985. The nucleotide sequence of tobacco rattle virus RNA2 (CAM strain). Nucleic Acids Res. 13, 8507 – 8518. Bos, L., van der Want, J.P.H., 1962. Early browning of pea, a disease caused by a soil- and seed-borne virus. Tijdschr. Plantenziekten 68, 368 – 390. Cornelissen, B.J.C., Linthorst, H.J.M., Brederode, F.T., Bol, J.F., 1986. Analysis of the genome structure of tobacco rattle virus strain PSG. Nucleic Acids Res. 14, 2157 – 2169. Goulden, M.G., Lomonossoff, G.P., Davies, J.W., Wood, K.R., 1990. The complete nucleotide sequence of PEBV RNA2 reveals the presence of a novel open reading frame and provides insights into the structure of tobraviral promoters. Nucleic Acids Res. 18, 4507 – 4512. Hamilton, W.D.O., Boccara, M., Robinson, D.J., Baulcombe, D.C., 1987. The complete nucleotide sequence of tobacco rattle virus RNA1. J. Gen. Virol. 68, 2563 – 2575. Harrison, B.D., Robinson, D.J., 1986. Tobraviruses. In: van Regenmortel, M.H.V., Fraenkel-Conrat, H. (Eds.), The Plant Viruses, vol. 2. Plenum Press, New York, pp. 339 – 369. Hernandez, C., Mathis, A., Brown, D.J.F., Bol, J.F., 1995. Sequence of RNA2 of a nematode-transmissible isolate of tobacco rattle virus. J. Gen. Virol. 76, 2847 – 2851. Hernandez, C., Carette, J.E., Brown, D.J.F., Bol, J.F., 1996. Serial passage of tobacco rattle virus under different selection conditions results in deletion of structural and nonstructural genes in RNA2. J. Virol. 70, 4933 – 4940. Hernandez, C., Visser, P.B., Brown, D.J.F., Bol, J.F., 1997. Transmission of tobacco rattle virus isolate PpK20 by its nematode vector requires one of the two non-structural genes in the viral RNA2. J. Gen. Virol. 78, 465 – 467. MacFarlane, S.A., 1996. Rapid cloning of uncharacterised tobacco rattle virus isolates using long template (LT) PCR. J. Virol. Methods 56, 91 – 98. MacFarlane, S.A., 1997. Natural recombination among plant virus genomes: evidence from tobraviruses. Semin. Virol. 8, 25 – 31.

94

M.M. Swanson, S.A. MacFarlane / Virus Research 60 (1999) 87–94

MacFarlane, S.A., Brown, D.J.F., 1995. Sequence comparison of RNA2 of nematode-transmissible and nematode-nontransmissible isolates of pea early-browning virus suggests that the gene encoding the 29 kDa protein may be involved in nematode transmission. J. Gen. Virol. 76, 1299–1304. MacFarlane, S.A., Taylor, S.C., King, D.I., Hughes, G., Davies, J.W., 1989. Pea early-browning virus RNA1 encodes four polypeptides including a putative zinc-finger protein. Nucleic Acids Res. 17, 2245–2259. MacFarlane, S.A., Wallis, C.V., Brown, D.J.F., 1996. Multiple virus genes involved in the nematode transmission of pea early-browning virus. Virology 219, 417–422. Mueller, A-M., Mooney, A., MacFarlane, S.A., 1997. Replication of in vitro tobravirus recombinants shows that the specificity of template recognition is determined by 5% non-coding but not 3% non-coding sequences. J. Gen. Virol. 78, 2085 – 2088. Robinson, D.J., 1983. RNA species of tobacco rattle virus strains and their nucleotide sequence relationships. J. Gen. Virol. 64, 657 – 665. Robinson, D.J., Harrison, B.D., 1985a. Unequal variation in the two genome parts of tobraviruses and evidence for the

.

existence of three separate viruses. J. Gen. Virol. 66, 171 – 176. Robinson, D.J., Harrison, B.D., 1985b. Evidence that broad bean yellow band virus is a new serotype of pea earlybrowning virus. J. Gen. Virol. 66, 2003 – 2009. Robinson, D.J., Hamilton, W.D.O., Harrison, B.D., Baulcombe, D.C., 1987. Two anomalous tobravirus isolates: evidence for RNA recombination in nature. J. Gen. Virol. 68, 2551 – 2561. Russo, M., Gallitelli, D., Vovlas, C., Savino, V., 1984. Properties of broad bean yellow band virus, a possible new tobravirus. Ann. Appl. Biol. 105, 223 – 230. Troutt, A.B., McHeyzer-Williams, M.G., Pulendran, B., Nossal, G.J.V., 1992. Ligation-anchored PCR: A simple amplification technique with single-sided specificity. Proc. Natl. Acad. Sci. USA 89, 9823 – 9825. Uhde, K., Koenig, R., Lesemann, D.E., 1998. An onion isolate of tobacco rattle virus: reactivity with an antiserum to Hypochoeris mosaic virus, a putative furovirus, and molecular analysis of its RNA2. Arch. Virol. 143, 1041 – 1053. Wang, D., MacFarlane, S.A., Maule, A.J., 1997. Viral determinants of pea early-browning virus seed transmission in pea. Virology 234, 112 – 117.