- Email: [email protected]
Viral Determinants of Pea Early Browning Virus Seed Transmission in Pea
VIROLOGY
234, 112–117 (1997) VY978637
ARTICLE NO.
Viral Determinants of Pea Early Browning Virus Seed Transmission in Pea Daowen Wang,*,1 Stuart A. MacFarlane,† and Andrew J. Maule*,2 *Department of Virus Research, John Innes Centre, Norwich Research Park, Colney, Norwich NR4 7UH, United Kingdom; and †Department of Virology, Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, United Kingdom Received February 14, 1997; returned to author for revision March 10, 1997; accepted May 16, 1997 Pea early browning virus (PEBV) is a member of the genus, Tobravirus. It is transmitted by soil-inhabiting trichodorid nematodes and through seeds from diseased plants. By introducing mutations into the PEBV genome, we have studied the viral determinants of seed transmission in pea. Neither deleting a portion of the genome containing the three nonstructural genes in RNA2 nor the interuption of any of the three genes individually prevented PEBV seed transmission. However, a comparison of two PEBV isolates indicated a minor role for RNA2 or its products. In contrast, the removal of the coding sequence of the 12K gene in RNA1 almost completely abolished viral seed transmission. The virus lacking the 12K gene caused more severe symptoms on leaves and pods, and accumulated to a higher level than the wild-type virus in both types of tissues. However, the 12K deletion mutant accumulated poorly in anthers and carpels, and could not be detected in pollen grains and ovules. These results suggest that the 12K gene is involved in the infection of the gametic cells and hence the seed transmission of PEBV in pea. q 1997 Academic Press
INTRODUCTION
1969; Shepherd, 1972). This was confirmed later for nepoviruses (Hanada and Harrison, 1977) and cucumoviruses (Hampton and Francki, 1992) when highly and weakly seed-transmissible isolates were compared. In these cases, the primary determinant of seed transmission was found to be in the virus genomic RNA1, which contained genes involved in virus replication. Typically, different tobraviruses have RNA1 of similar sizes and similar coding capacity. In contrast, the RNA2 varies in both respects for different tobraviruses. For example, the RNA2 of the PEBV TPA56 isolate is 3.37 kb long and contains three genes coding for nonstructural proteins, whereas RNA2 of pepper ringspot virus (PRV) is 1.8 kb long and only encodes the viral coat protein (MacFarlane and Brown, 1995; Bergh et al., 1985). However, PRV is 15–30% seed transmissible in tomato (Costa and Kitajima, 1968), suggesting that the tobravirus nonstructural proteins from RNA2 may not be essential for viral seed transmission. Recently, the construction of chimeric viruses from isolates differing in seed transmission has been used to map seed transmission determinants in barley stripe mosaic virus (BSMV) and pea seed-borne mosaic virus (PSbMV) (Edwards, 1995; Johansen et al., 1996). In both studies, the major determinant of seed transmission was the viral nonstructural gene coding for a protein containing a cysteine-rich domain, the gb gene for BSMV and the HC-Pro gene for PSbMV.
Pea early browning virus (PEBV) has a bipartite, positive sense, single-stranded RNA genome. The RNA1 of the PEBV isolate SP5 encodes four polypeptides of 201, 141, 30, and 12 kDa (MacFarlane et al., 1989). The 141kDa (141K) and the 201-kDa (201K) polypeptides are proteins associated with the viral replicase, the 201K being produced by read-through of a leaky termination codon; the 30-kDa (30K) polypeptide probably serves as the viral movement protein. The 12-kDa (12K) polypeptide contains a cysteine-rich putative zinc finger structure but has had no function assigned to it. PEBV RNA2 (Goulden et al., 1990; MacFarlane and Brown, 1995) encodes four polypeptides: the coat protein and the three nonstructural proteins of 9, 29, and 23 kDa, respectively. PEBV is transmitted naturally by soil-inhabiting nematodes and through seeds from infected plants. By comparing nematode transmission of different mutants, it was found that all the proteins encoded by RNA2 are involved in transmission (MacFarlane and Brown, 1995; MacFarlane et al., 1996). In contrast, the viral genes affecting PEBV seed transmission have not been investigated. The existence of viral determinants of seed transmission was originally deduced from the observation that different isolates of a seed-borne virus may have different rates of seed transmission in the same host (Bennett,
METHODS
1
Current address: National Laboratory for Plant Cell and Chromosome Engineering, Institute of Genetics, Chinese Academy of Sciences, Beijing 100101, China. 2 To whom correspondence and reprint requests should be addressed. Fax: 44-1603-456844. E-mail. [email protected]. 0042-6822/97 $25.00
112
Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
AID
VY 8637
/
6a3d$$$201
Virus and host plants Two isolates of PEBV were used in this work. RNA1 and RNA2 of PEBV isolate SP5 have been fully character-
06-25-97 11:24:24
vira
AP: VY
VIRAL DETERMINANTS OF PEBV SEED TRANSMISSION IN PEA
ized and cloned full-length viral cDNAs produced (MacFarlane et al., 1991). A full-length cDNA clone of RNA2 of isolate TPA56 was also available (MacFarlane and Brown, 1995). Infections with PEBV (MacFarlane and Brown, 1995) were established initially in Nicotiana benthamiana using in vitro-generated transcripts derived from cloned viral cDNAs as described previously (MacFarlane et al., 1991, 1996). Homogenates of infected tissue in 0.1 M sodium phosphate, pH 7.5, were manually inoculated to seedlings of Pisum sativum cv. Rondo, and the plants grown to maturity at 15–207 with supplementary lighting to give a 14-hr photoperiod. To assess the incidence of seed transmission, mature seeds (50–100) were harvested from infected pea plants and germinated and grown under the same conditions. Virus mutants Recombinant DNA techniques were carried out using standard protocols (Sambrook et al., 1989). PEBV RNA2 mutants DBE, 29D, 23D, 23FS, and 9FS have been described previously (MacFarlane et al., 1996; Fig. 1). Two mutants in the 12K gene of PEBV RNA1 were made (Fig 1). The 12K deletion (D12K) was made by introducing new NheI sites three nucleotides (nt) after the 12K translational start codon and 3 nt before the translational stop and removing the 12K coding sequence as a NheI fragment. The NheI sites were introduced into a 3* terminal BamHI (nt 5110) to SmaI (nt 7073) M13mp19 subclone of the RNA1 cDNA (pT7FL3; MacFarlane et al., 1991) using mutagenic oligonucleotides and the dut0 ung0 mutagenesis method (Kunkel et al., 1987). The mutagenized sequence was confirmed by DNA sequencing. After excision of the NheI fragment, the deleted cDNA was digested with MluI (nt 5235) and SmaI (nt 7073) and inserted into similarly digested pT7FL3. The 12K frameshift mutant (12FS) was made from a clone with only a single NheI site introduced three nucleotides after the translational start codon. The DNA was cut with NheI, the site filled-in with T4 DNA polymerase, and the DNA religated to introduce a /1 frameshift. The modified sequence was reinserted into pT7FL3 as above. The frameshift results in expression of a nine amino acid peptide unrelated in sequence to the 12K protein. The stability of the RNA2 mutants after replication in pea was assessed as described previously (MacFarlane et al., 1996). Stability of the D12K mutant was assessed using RT-PCR amplification (using a primer representing nt 5608–5631 of PEBV RNA1 and a primer complementary to the terminal 19 nt of all tobravirus RNAs) and restriction enzyme digestion of the PCR product with NheI. None of the mutants showed reversion after replication and systemic infection of pea.
113
(Ding et al., 1992) with a PEBV-specific rabbit polyclonal antiserum (MacFarlane et al., 1989). The accumulation of wild-type PEBV and the various mutants was assessed using ELISA and Northern analysis. For the latter, 20 mg of total RNA, extracted from the maternal pod tissues, was denatured according to Gru¨ndemann and Koepsell (1994), separated on a 1.25% agarose gel, transferred to Hybond N membrane (Amersham), and hybridized (Brisson et al., 1994) using a a radiolabeled DNA probe derived from pT7FL3 (MacFarlane et al., 1996). To assess the distribution of wild-type PEBV and the D12K mutant, unpollinated flowers from infected plants were dissected and the anthers and carpels separately fixed, wax-embedded, and processed for immunohistochemical analysis. Sections were probed with rabbit polyclonal anti-PEBV serum and specific reactions were visualized following incubation with alkaline phosphatase (AP)-conjugated secondary antibody (Sigma) and Fast Red/Naphthol AS-MX (Sigma) AP substrate, as described before (Wang and Maule, 1997). The sections were counterstained by brief immersion (15 sec) in an aqueous solution (0.25% w/v) of Alcian Blue (Aldrich Chemical Co.), which binds to cell wall material. Slides were mounted in Entellan (Merck) before microscopic examination and photography. RESULTS AND DISCUSSION
Infection of the mother plants and transmission of PEBV to progeny seedlings were confirmed using ELISA
At the start of this work, the fully characterized PEBV SP5 isolate and a pseudorecombinant composed of SP5 RNA1 and TPA56 RNA2 were assessed for seed transmissibility in pea cultivar ‘Rondo’. Although the levels of seed transmission in infected plants showed a large variation between experiments, the pseudorecombinant showed consistently higher levels, with an average of about 40% as opposed to approximately 10% for the pure SP5 isolate. This result, and the availability of mutants within TPA56 RNA2 (MacFarlane et al., 1996) led to the use of the pseudorecombinant for all the subsequent analyses. From here this pseudorecombinant is referred to as wild-type virus. To assess the contribution of the nonstructural genes within PEBV RNA 2 to PEBV seed transmission, infections of pea were established for some deletion and frameshift mutants that had been used previously in an investigation of nematode transmissibility of PEBV (Fig. 1; MacFarlane et al., 1996). Mutant DBE deleted the complete coding region of the 29- and 23-kDa and a part of the coding region of the 9-kDa proteins. Mutants 29D and 23D, and 23FS and 9FS, were deletion and frameshift mutants, respectively, of the individual nonstructural genes. Symptoms induced by either the wild-type virus or the mutants on pea plants were very mild and, based on ELISA assays, all the viruses accumulated to a similar extent in infected leaf tissue (data not shown). When progeny seedlings were assessed for PEBV infection it was found
AID
vira
Virus detection
VY 8637
/
6a3d$$$201
06-25-97 11:24:24
AP: VY
114
WANG, MACFARLANE, AND MAULE TABLE 1 Seed Transmission of the PEBV Wild Type and Viruses Mutated in RNA2 Experiment 1 (Nov. 1994–Feb. 1995)
Experiment 2 (Jan. 1995–April 1995)
Virus
Number of seedlings tested
Seed transmission (%)
Number of seedlings tested
Seed transmission (%)
WT 9FS DBE 29D 23FS 23D
96 44 61 44 30 50
29 41 34 50 13 44
45 48 69 115 74 79
75 31 22 63 31 56
that all the mutants were seed transmitted (Table 1). The stability of these mutants after transmission into the progeny seedlings was confirmed by RT-PCR analysis. There was a large variation between experiments in the extent of seed transmission of the mutant and wild-type viruses. This might be due to environmental influences as these two experiments were done at different times of the year. Environmental factors such as temperature have commonly been found to affect the efficiency of plant virus seed transmission (reviewed by Maule and Wang, 1996). Despite considerable variation in the levels of seed transmission between the two experiments, two conclusions were clear. Neither the removal of the sequence coding for all the nonstructural genes (mutant DBE) nor disruption of individual nonstructural genes abolished PEBV seed transmission. While the removal of the nonstructural gene products did not prevent seed transmission, the comparative data for the pseudorecombinant and the pure SP5 isolate suggests that RNA2 may have a minor role in this process. There are 11 nucleotide differences between the RNA2s of SP5 and TPA56, 3 of which result in altered protein coding in the 29K (2 amino acids) and coat protein (1 amino acid) genes (MacFarlane and Brown, 1995). Unfortunately, the variability in the seed transmission data did not allow the important nucleotide changes to be identified. Considerably more stable environmental conditions would have to be established to achieve this. An involvement of the larger genomic RNA (RNA1) in seed transmission might result from specific RNA sequences or from the RNA1-encoded proteins. Mutations that abolish the expression of the 141K or 201K, or the 30K genes would prevent virus replication, or virus movement, respectively, and would de facto affect transmission of the virus into the seed. To assess the possible involvement of the 12K cysteine-rich protein in seed transmission, two different mutants were made, a frame shift mutation at the start of the 12K coding region and a complete deletion of the 12K coding region (Fig. 1).
AID
VY 8637
/
6a3d$$$201
06-25-97 11:24:24
Previously, premature termination or deletion of the homologous gene (16K gene) from tobacco rattle tobravirus was shown not to destroy infectivity (Guilford et al., 1991). Infections with PEBV RNA1 D12K or 12FS, and RNA2, were established on N. benthamiana and tissue extracts used to infect pea. PEBV D12K infections resulted in the formation of necrotic symptoms on pea leaves and pods (Figs. 2A and 2B). No symptoms were seen following inoculation with 12FS. When inoculated plants were grown to maturity and progeny seedlings tested for seed transmission, none was recorded for 12FS, and none or õ1% in two experiments for D12K (Table 2). The wildtype isolate achieved high levels of seed transmission in the same experiments (Table 2). Unfortunately, we could not maintain the virus from the single seed showing transmission of the D12K mutant and it could not be analyzed further. To investigate the mechanism responsible for the reduced seed transmission, the inoculated plants were assessed for virus accumulation. After inoculation of peas,
FIG. 1. Mutations in PEBV RNA1 and 2 used to define the viral determinant of seed transmission. (A) Five RNA2 mutations (9FS, DBE, 29D, 23FS, and 23D; MacFarlane et al., 1996), shown relative to the organization of wild-type (WT) RNA2, were tested. (B) Two RNA1 mutants (D12K and 12FS), shown relative to the organization of WT RNA 1, were made and tested. Relevant restriction sites (Bst, BstBI; Bgl, BglII; Eco, EcoR1; Afl, AflIII; Pfl, PflMI) are shown. Destroyed restriction sites (Bst*; Afl*) are identified. Filled boxes show deleted region.
vira
AP: VY
VIRAL DETERMINANTS OF PEBV SEED TRANSMISSION IN PEA
115
FIG. 2. PEBV accumulation in the leaf and reproductive tissues of infected peas. The symptomatic effects of wild-type (WT) and D12K mutant PEBV on leaves (A) and pods (B) are shown; healthy (H) tissues are included for comparison. The symptoms are very mild for WT PEBV and characterized by small necrotic lesions on leaves and larger necrotic areas on pods for D12K PEBV. Sections of anthers (C and D) and carpels (E and F) sampled from D12K-infected (C and E) or WT-infected (D and F) plants were subjected to immunohistochemistry to detect PEBV coat protein (red color; illustrated by arrows). The wide distribution of WT PEBV throughout the anther, including the tapetum (t), and the carpel (c) tissues results in the infection of pollen (ip; red color) and the ovules (o). In contrast, the limited spread of the D12K mutant leaves the pollen (p) and ovules (o) free of infection. c, carpel; ip, infected pollen; o, ovule; p, uninfected pollen; t, tapetum. Bars: C and D, 500 mm; E and F, 1 mm.
12FS was not detected in the inoculated or younger leaves by ELISA (Fig. 3A). In contrast, both ELISA and Northern analysis showed that D12K accumulated at 1–
2 times the level of wild-type virus in leaf tissues and mature pod tissues (Fig. 3). These results demonstrated that the decrease in seed transmission of the D12K mu-
AID
vira
VY 8637
/
6a3d$$$201
06-25-97 11:24:24
AP: VY
116
WANG, MACFARLANE, AND MAULE
tant was neither due to a reduced viral accumulation in the infected plants nor due to an altered proportion of viral RNA to virus particles that, potentially, could affect the efficiency of virus movement. Although our inability to detect the 12FS mutant in pea did not allow us to distinguish between effects associated with the 12K protein and the 12K RNA sequence, it is interesting that, like the cysteine-rich proteins from BSMV and PSbMV, which act as determinants of seed transmission, the PEBV 12K is also a cysteine-rich protein. In our previous study on the plant structural basis of PEBV seed transmission, we showed that PEBV accumulated extensively in the male and female reproductive tissues and cells, i.e., anthers, pollen grains, carpels, ovules and egg cells (Wang and Maule, 1997). To analyze further the basis of the reduction in seed transmission in D12K-infected plants, the distribution of the mutant in the floral tissues was investigated (Figs. 2C and 2E). The wild-type virus accumulated extensively in the anther tissues, including the tapetum and in pollen grains (Fig. 2D), although not all of the pollen grains were infected. In contrast, the mutant virus was detected in the central vascular tissues of the anther (Fig. 2C, arrow) , but the pollen grains and other tissues of the anther remained uninfected. Sections of carpels revealed fully developed ovules. The wild-type isolate again accumulated abundantly in the carpel and the ovules (Fig. 2F), while the D12K mutant was found only sparsely distributed in the vascular tissues in the carpel (Fig. 2E, arrow). No virus was found in the ovule (Fig. 2E) or in the egg cells (data not shown). Because pollen grain and/or egg cell infection is the route for PEBV to achieve seed transmission, it is not surprising that seed transmission in D12K-infected plants was almost completely abolished when infection of the gametes was absent. Apparently the 12K protein (or its coding sequence) is involved in the infection of the gametes. The lack of D12K accumulation in the male and female gametes contrasted with the elevated level of this mutant in the vegetative leaf tissue and the mature pod tissue. Clearly, the removal of the 12K gene can have
FIG. 3. ELISA and Northern analysis of PEBV accumulation in infected peas. (A) ELISA assay of PSbMV in leaf samples from the upper parts (systemic leaves) of plants inoculated with wild-type (WT), D12K, and 12FS PEBV. Control, uninfected material (H) was processed in parallel; data are expressed as A405 { SE. (B) Northern blot of RNA extracted from healthy plants (H) and plants infected with WT, 9FS, and D12K PEBV. The positions of PEBV RNA 1 and RNA 2 are marked; RNA 2 could be detected by the probe prepared to RNA 1 sequences through the conserved sequence in the 3* untranslated region (MacFarlane et al., 1991).
different effects on virus/host interactions in different organs and in the tissues at different developmental stages. ACKNOWLEDGMENTS We thank Margaret Boulton, Jeff Davies, and George Lomonossoff for comments on the manuscript prior to publication. D.W. was in receipt of a Plant Molecular Biology II grant from the Biotechnology and Biological Research Council (BBSRC). The John Innes Centre is grant-aided from the BBSRC. The Scottish Crop Research Institute is grant-aided from the Scottish Office Agriculture, Environment and Fisheries Department (SOEAFD). The work was carried out under Ministry of Agriculture Fisheries and Food Licence No. PHF 1419C/1669 and SOEAFD Licence No GM/43/1993.
REFERENCES TABLE 2 Seed Transmission of PEBV Wild-Type Isolate and the Mutant D12K Experiment 1 (Feb. 1995–May 1995)
Experiment 2 (March 1995–June 1995)
Virusa
Number of seedlings tested
Seed transmission (%)
Number of seedlings tested
Seed transmission (%)
WT D12K
36 59
40 0
64 75
25 1
a
PEBV12FS not included as no infection of pea was recorded.
AID
VY 8637
/
6a3d$$$201
06-25-97 11:24:24
Bennet, C. W. (1969). Seed transmission of plant viruses. Adv. Virus Res. 14, 221–261. Bergh, S. T., Koziel, M. G., Huang, S-C., Thomas, R. A., Gilley, D. P., and Siegel, A. (1985). The nucleotide sequence of tobacco rattle virus RNA2 (CAM strain). Nucleic Acids Res. 13, 8507–8518. Brisson, L. F., Tenhaken, R., and Lamb, C. (1994). Function of oxidative cross-linking of cell wall structural proteins in plant disease resistance. Plant Cell 6, 1703–1712. Costa, A. S., and Kitajima, E. E. (1968). TransmissaV o do võB rus do anel do pimentaV o atrave´s da semente de tomate. Rev. Soc. Bras. Fitopatol. 2, 25–28. Ding, X. S., Cockbain, A. J., and Govier, D. A. (1992). Improvements in the detection of pea seed-borne mosaic virus by ELISA. Ann. Appl. Biol. 121, 75–83. Edwards, M. C. (1995). Mapping of the seed transmission determinants
vira
AP: VY
VIRAL DETERMINANTS OF PEBV SEED TRANSMISSION IN PEA
117
of barley stripe mosaic virus. Mol. Plant-Microbe Interact. 8, 906– 915. Goulden, M. G., Lomonossoff, G. P., Davies, J. W., and 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 subgenomic promoters. Nucleic Acids Res. 18, 4507–4512. Gru¨ndemann, D., and Koepsell, H. (1994). Ethidium bromide staining during denaturation with glyoxal for sensitive detection of RNA in agarose gel electrophoresis. Anal. Biochem. 216, 459–461. Guilford, P. J., Ziegler-Graff, V., and Baulcombe, D. C. (1991). Mutation and replacement of the 16-kDa protein gene in RNA-1 of tobacco rattle virus. Virology 182, 607–614. Hampton, R. O., and Francki, R. I. B. (1992). RNA-1 dependent seed transmissibility of cucumber mosaic virus in Phaseolus vulgaris. Phytopathology 82, 127–130. Hanada, K., and Harrison, B. D. (1977). Effects of virus genotype and temperature on seed transmission of nepoviruses. Ann. Appl. Biol. 85, 79–92. Johansen, I. E., Dougherty, W. G., Keller, K. E., Wang, D., and Hampton, R. O. (1996). Multiple determinants affect seed transmission of pea seedborne mosaic virus in Pisum sativum. J. Gen. Virol. 77, 3149– 3154. Kunkel, T. A., Roberts, J. D., and Zakour, R. A. (1987). Rapid and efficient site-specific mutagenesis without phenotypic selection. Methods Enzymol. 154, 367–382.
MacFarlane, S. A., Taylor, S. C., King, D. I., Hughes, G., and Davies, J. W. (1989). Pea early browning virus RNA1 encodes four polypeptides including a putative zinc-finger protein. Nucleic Acids Res. 17, 2245– 2260. MacFarlane, S. A., Wallis, C. V., Taylor, S. C., Goulden, M. G., Wood, K. R., and Davies, J. W. (1991). Construction and analysis of infectious transcripts synthesized from full-length cDNA clones of both genomic RNAs of pea early browning virus. Virology 182, 124–129. MacFarlane, S. A., and Brown, D. J. F. (1995). Sequence comparison of RNA2 of nematode-transmissible and nematode non-transmissible isolates of pea early-browning virus suggests that the gene encoding the 29kDa protein may be involved in nematode transmission. J. Gen. Virol. 76, 1299–1304. MacFarlane, S. A., Brown, D. J. F., and Bol, J. F. (1995). The transmission by nematodes of tobraviruses is not determined exclusively by the virus coat protein. Eur. J. Plant Pathol. 101, 535–539. MacFarlane, S. A., Wallis, C. V., and Brown, D. J. F. (1996). Multiple virus genes involved in the nematode transmission of pea early browning virus. Virology 219, 417–422. Maule, A. J., and Wang, D. (1996). Seed transmission of plant viruses: A lesson in biological complexity. Trends Microbiol. 4, 153–158. Shepherd, R. J. (1972). Transmission of viruses through seed and pollen. In ‘‘Principles and Techniques in Plant Virology’’ (C. I. Kado and H. O. Agrawal, Eds.), pp. 267–292. Van Nostrand Reinhold, New York. Wang, D., and Maule, A. J. (1997). Contrasting patterns in the spread of two seed-borne viruses in pea embryos. Plant J. 11, 1333–1340.
AID
vira
VY 8637
/
6a3d$$$201
06-25-97 11:24:24
AP: VY
234, 112–117 (1997) VY978637
ARTICLE NO.
Viral Determinants of Pea Early Browning Virus Seed Transmission in Pea Daowen Wang,*,1 Stuart A. MacFarlane,† and Andrew J. Maule*,2 *Department of Virus Research, John Innes Centre, Norwich Research Park, Colney, Norwich NR4 7UH, United Kingdom; and †Department of Virology, Scottish Crop Research Institute, Invergowrie, Dundee DD2 5DA, United Kingdom Received February 14, 1997; returned to author for revision March 10, 1997; accepted May 16, 1997 Pea early browning virus (PEBV) is a member of the genus, Tobravirus. It is transmitted by soil-inhabiting trichodorid nematodes and through seeds from diseased plants. By introducing mutations into the PEBV genome, we have studied the viral determinants of seed transmission in pea. Neither deleting a portion of the genome containing the three nonstructural genes in RNA2 nor the interuption of any of the three genes individually prevented PEBV seed transmission. However, a comparison of two PEBV isolates indicated a minor role for RNA2 or its products. In contrast, the removal of the coding sequence of the 12K gene in RNA1 almost completely abolished viral seed transmission. The virus lacking the 12K gene caused more severe symptoms on leaves and pods, and accumulated to a higher level than the wild-type virus in both types of tissues. However, the 12K deletion mutant accumulated poorly in anthers and carpels, and could not be detected in pollen grains and ovules. These results suggest that the 12K gene is involved in the infection of the gametic cells and hence the seed transmission of PEBV in pea. q 1997 Academic Press
INTRODUCTION
1969; Shepherd, 1972). This was confirmed later for nepoviruses (Hanada and Harrison, 1977) and cucumoviruses (Hampton and Francki, 1992) when highly and weakly seed-transmissible isolates were compared. In these cases, the primary determinant of seed transmission was found to be in the virus genomic RNA1, which contained genes involved in virus replication. Typically, different tobraviruses have RNA1 of similar sizes and similar coding capacity. In contrast, the RNA2 varies in both respects for different tobraviruses. For example, the RNA2 of the PEBV TPA56 isolate is 3.37 kb long and contains three genes coding for nonstructural proteins, whereas RNA2 of pepper ringspot virus (PRV) is 1.8 kb long and only encodes the viral coat protein (MacFarlane and Brown, 1995; Bergh et al., 1985). However, PRV is 15–30% seed transmissible in tomato (Costa and Kitajima, 1968), suggesting that the tobravirus nonstructural proteins from RNA2 may not be essential for viral seed transmission. Recently, the construction of chimeric viruses from isolates differing in seed transmission has been used to map seed transmission determinants in barley stripe mosaic virus (BSMV) and pea seed-borne mosaic virus (PSbMV) (Edwards, 1995; Johansen et al., 1996). In both studies, the major determinant of seed transmission was the viral nonstructural gene coding for a protein containing a cysteine-rich domain, the gb gene for BSMV and the HC-Pro gene for PSbMV.
Pea early browning virus (PEBV) has a bipartite, positive sense, single-stranded RNA genome. The RNA1 of the PEBV isolate SP5 encodes four polypeptides of 201, 141, 30, and 12 kDa (MacFarlane et al., 1989). The 141kDa (141K) and the 201-kDa (201K) polypeptides are proteins associated with the viral replicase, the 201K being produced by read-through of a leaky termination codon; the 30-kDa (30K) polypeptide probably serves as the viral movement protein. The 12-kDa (12K) polypeptide contains a cysteine-rich putative zinc finger structure but has had no function assigned to it. PEBV RNA2 (Goulden et al., 1990; MacFarlane and Brown, 1995) encodes four polypeptides: the coat protein and the three nonstructural proteins of 9, 29, and 23 kDa, respectively. PEBV is transmitted naturally by soil-inhabiting nematodes and through seeds from infected plants. By comparing nematode transmission of different mutants, it was found that all the proteins encoded by RNA2 are involved in transmission (MacFarlane and Brown, 1995; MacFarlane et al., 1996). In contrast, the viral genes affecting PEBV seed transmission have not been investigated. The existence of viral determinants of seed transmission was originally deduced from the observation that different isolates of a seed-borne virus may have different rates of seed transmission in the same host (Bennett,
METHODS
1
Current address: National Laboratory for Plant Cell and Chromosome Engineering, Institute of Genetics, Chinese Academy of Sciences, Beijing 100101, China. 2 To whom correspondence and reprint requests should be addressed. Fax: 44-1603-456844. E-mail. [email protected]. 0042-6822/97 $25.00
112
Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.
AID
VY 8637
/
6a3d$$$201
Virus and host plants Two isolates of PEBV were used in this work. RNA1 and RNA2 of PEBV isolate SP5 have been fully character-
06-25-97 11:24:24
vira
AP: VY
VIRAL DETERMINANTS OF PEBV SEED TRANSMISSION IN PEA
ized and cloned full-length viral cDNAs produced (MacFarlane et al., 1991). A full-length cDNA clone of RNA2 of isolate TPA56 was also available (MacFarlane and Brown, 1995). Infections with PEBV (MacFarlane and Brown, 1995) were established initially in Nicotiana benthamiana using in vitro-generated transcripts derived from cloned viral cDNAs as described previously (MacFarlane et al., 1991, 1996). Homogenates of infected tissue in 0.1 M sodium phosphate, pH 7.5, were manually inoculated to seedlings of Pisum sativum cv. Rondo, and the plants grown to maturity at 15–207 with supplementary lighting to give a 14-hr photoperiod. To assess the incidence of seed transmission, mature seeds (50–100) were harvested from infected pea plants and germinated and grown under the same conditions. Virus mutants Recombinant DNA techniques were carried out using standard protocols (Sambrook et al., 1989). PEBV RNA2 mutants DBE, 29D, 23D, 23FS, and 9FS have been described previously (MacFarlane et al., 1996; Fig. 1). Two mutants in the 12K gene of PEBV RNA1 were made (Fig 1). The 12K deletion (D12K) was made by introducing new NheI sites three nucleotides (nt) after the 12K translational start codon and 3 nt before the translational stop and removing the 12K coding sequence as a NheI fragment. The NheI sites were introduced into a 3* terminal BamHI (nt 5110) to SmaI (nt 7073) M13mp19 subclone of the RNA1 cDNA (pT7FL3; MacFarlane et al., 1991) using mutagenic oligonucleotides and the dut0 ung0 mutagenesis method (Kunkel et al., 1987). The mutagenized sequence was confirmed by DNA sequencing. After excision of the NheI fragment, the deleted cDNA was digested with MluI (nt 5235) and SmaI (nt 7073) and inserted into similarly digested pT7FL3. The 12K frameshift mutant (12FS) was made from a clone with only a single NheI site introduced three nucleotides after the translational start codon. The DNA was cut with NheI, the site filled-in with T4 DNA polymerase, and the DNA religated to introduce a /1 frameshift. The modified sequence was reinserted into pT7FL3 as above. The frameshift results in expression of a nine amino acid peptide unrelated in sequence to the 12K protein. The stability of the RNA2 mutants after replication in pea was assessed as described previously (MacFarlane et al., 1996). Stability of the D12K mutant was assessed using RT-PCR amplification (using a primer representing nt 5608–5631 of PEBV RNA1 and a primer complementary to the terminal 19 nt of all tobravirus RNAs) and restriction enzyme digestion of the PCR product with NheI. None of the mutants showed reversion after replication and systemic infection of pea.
113
(Ding et al., 1992) with a PEBV-specific rabbit polyclonal antiserum (MacFarlane et al., 1989). The accumulation of wild-type PEBV and the various mutants was assessed using ELISA and Northern analysis. For the latter, 20 mg of total RNA, extracted from the maternal pod tissues, was denatured according to Gru¨ndemann and Koepsell (1994), separated on a 1.25% agarose gel, transferred to Hybond N membrane (Amersham), and hybridized (Brisson et al., 1994) using a a radiolabeled DNA probe derived from pT7FL3 (MacFarlane et al., 1996). To assess the distribution of wild-type PEBV and the D12K mutant, unpollinated flowers from infected plants were dissected and the anthers and carpels separately fixed, wax-embedded, and processed for immunohistochemical analysis. Sections were probed with rabbit polyclonal anti-PEBV serum and specific reactions were visualized following incubation with alkaline phosphatase (AP)-conjugated secondary antibody (Sigma) and Fast Red/Naphthol AS-MX (Sigma) AP substrate, as described before (Wang and Maule, 1997). The sections were counterstained by brief immersion (15 sec) in an aqueous solution (0.25% w/v) of Alcian Blue (Aldrich Chemical Co.), which binds to cell wall material. Slides were mounted in Entellan (Merck) before microscopic examination and photography. RESULTS AND DISCUSSION
Infection of the mother plants and transmission of PEBV to progeny seedlings were confirmed using ELISA
At the start of this work, the fully characterized PEBV SP5 isolate and a pseudorecombinant composed of SP5 RNA1 and TPA56 RNA2 were assessed for seed transmissibility in pea cultivar ‘Rondo’. Although the levels of seed transmission in infected plants showed a large variation between experiments, the pseudorecombinant showed consistently higher levels, with an average of about 40% as opposed to approximately 10% for the pure SP5 isolate. This result, and the availability of mutants within TPA56 RNA2 (MacFarlane et al., 1996) led to the use of the pseudorecombinant for all the subsequent analyses. From here this pseudorecombinant is referred to as wild-type virus. To assess the contribution of the nonstructural genes within PEBV RNA 2 to PEBV seed transmission, infections of pea were established for some deletion and frameshift mutants that had been used previously in an investigation of nematode transmissibility of PEBV (Fig. 1; MacFarlane et al., 1996). Mutant DBE deleted the complete coding region of the 29- and 23-kDa and a part of the coding region of the 9-kDa proteins. Mutants 29D and 23D, and 23FS and 9FS, were deletion and frameshift mutants, respectively, of the individual nonstructural genes. Symptoms induced by either the wild-type virus or the mutants on pea plants were very mild and, based on ELISA assays, all the viruses accumulated to a similar extent in infected leaf tissue (data not shown). When progeny seedlings were assessed for PEBV infection it was found
AID
vira
Virus detection
VY 8637
/
6a3d$$$201
06-25-97 11:24:24
AP: VY
114
WANG, MACFARLANE, AND MAULE TABLE 1 Seed Transmission of the PEBV Wild Type and Viruses Mutated in RNA2 Experiment 1 (Nov. 1994–Feb. 1995)
Experiment 2 (Jan. 1995–April 1995)
Virus
Number of seedlings tested
Seed transmission (%)
Number of seedlings tested
Seed transmission (%)
WT 9FS DBE 29D 23FS 23D
96 44 61 44 30 50
29 41 34 50 13 44
45 48 69 115 74 79
75 31 22 63 31 56
that all the mutants were seed transmitted (Table 1). The stability of these mutants after transmission into the progeny seedlings was confirmed by RT-PCR analysis. There was a large variation between experiments in the extent of seed transmission of the mutant and wild-type viruses. This might be due to environmental influences as these two experiments were done at different times of the year. Environmental factors such as temperature have commonly been found to affect the efficiency of plant virus seed transmission (reviewed by Maule and Wang, 1996). Despite considerable variation in the levels of seed transmission between the two experiments, two conclusions were clear. Neither the removal of the sequence coding for all the nonstructural genes (mutant DBE) nor disruption of individual nonstructural genes abolished PEBV seed transmission. While the removal of the nonstructural gene products did not prevent seed transmission, the comparative data for the pseudorecombinant and the pure SP5 isolate suggests that RNA2 may have a minor role in this process. There are 11 nucleotide differences between the RNA2s of SP5 and TPA56, 3 of which result in altered protein coding in the 29K (2 amino acids) and coat protein (1 amino acid) genes (MacFarlane and Brown, 1995). Unfortunately, the variability in the seed transmission data did not allow the important nucleotide changes to be identified. Considerably more stable environmental conditions would have to be established to achieve this. An involvement of the larger genomic RNA (RNA1) in seed transmission might result from specific RNA sequences or from the RNA1-encoded proteins. Mutations that abolish the expression of the 141K or 201K, or the 30K genes would prevent virus replication, or virus movement, respectively, and would de facto affect transmission of the virus into the seed. To assess the possible involvement of the 12K cysteine-rich protein in seed transmission, two different mutants were made, a frame shift mutation at the start of the 12K coding region and a complete deletion of the 12K coding region (Fig. 1).
AID
VY 8637
/
6a3d$$$201
06-25-97 11:24:24
Previously, premature termination or deletion of the homologous gene (16K gene) from tobacco rattle tobravirus was shown not to destroy infectivity (Guilford et al., 1991). Infections with PEBV RNA1 D12K or 12FS, and RNA2, were established on N. benthamiana and tissue extracts used to infect pea. PEBV D12K infections resulted in the formation of necrotic symptoms on pea leaves and pods (Figs. 2A and 2B). No symptoms were seen following inoculation with 12FS. When inoculated plants were grown to maturity and progeny seedlings tested for seed transmission, none was recorded for 12FS, and none or õ1% in two experiments for D12K (Table 2). The wildtype isolate achieved high levels of seed transmission in the same experiments (Table 2). Unfortunately, we could not maintain the virus from the single seed showing transmission of the D12K mutant and it could not be analyzed further. To investigate the mechanism responsible for the reduced seed transmission, the inoculated plants were assessed for virus accumulation. After inoculation of peas,
FIG. 1. Mutations in PEBV RNA1 and 2 used to define the viral determinant of seed transmission. (A) Five RNA2 mutations (9FS, DBE, 29D, 23FS, and 23D; MacFarlane et al., 1996), shown relative to the organization of wild-type (WT) RNA2, were tested. (B) Two RNA1 mutants (D12K and 12FS), shown relative to the organization of WT RNA 1, were made and tested. Relevant restriction sites (Bst, BstBI; Bgl, BglII; Eco, EcoR1; Afl, AflIII; Pfl, PflMI) are shown. Destroyed restriction sites (Bst*; Afl*) are identified. Filled boxes show deleted region.
vira
AP: VY
VIRAL DETERMINANTS OF PEBV SEED TRANSMISSION IN PEA
115
FIG. 2. PEBV accumulation in the leaf and reproductive tissues of infected peas. The symptomatic effects of wild-type (WT) and D12K mutant PEBV on leaves (A) and pods (B) are shown; healthy (H) tissues are included for comparison. The symptoms are very mild for WT PEBV and characterized by small necrotic lesions on leaves and larger necrotic areas on pods for D12K PEBV. Sections of anthers (C and D) and carpels (E and F) sampled from D12K-infected (C and E) or WT-infected (D and F) plants were subjected to immunohistochemistry to detect PEBV coat protein (red color; illustrated by arrows). The wide distribution of WT PEBV throughout the anther, including the tapetum (t), and the carpel (c) tissues results in the infection of pollen (ip; red color) and the ovules (o). In contrast, the limited spread of the D12K mutant leaves the pollen (p) and ovules (o) free of infection. c, carpel; ip, infected pollen; o, ovule; p, uninfected pollen; t, tapetum. Bars: C and D, 500 mm; E and F, 1 mm.
12FS was not detected in the inoculated or younger leaves by ELISA (Fig. 3A). In contrast, both ELISA and Northern analysis showed that D12K accumulated at 1–
2 times the level of wild-type virus in leaf tissues and mature pod tissues (Fig. 3). These results demonstrated that the decrease in seed transmission of the D12K mu-
AID
vira
VY 8637
/
6a3d$$$201
06-25-97 11:24:24
AP: VY
116
WANG, MACFARLANE, AND MAULE
tant was neither due to a reduced viral accumulation in the infected plants nor due to an altered proportion of viral RNA to virus particles that, potentially, could affect the efficiency of virus movement. Although our inability to detect the 12FS mutant in pea did not allow us to distinguish between effects associated with the 12K protein and the 12K RNA sequence, it is interesting that, like the cysteine-rich proteins from BSMV and PSbMV, which act as determinants of seed transmission, the PEBV 12K is also a cysteine-rich protein. In our previous study on the plant structural basis of PEBV seed transmission, we showed that PEBV accumulated extensively in the male and female reproductive tissues and cells, i.e., anthers, pollen grains, carpels, ovules and egg cells (Wang and Maule, 1997). To analyze further the basis of the reduction in seed transmission in D12K-infected plants, the distribution of the mutant in the floral tissues was investigated (Figs. 2C and 2E). The wild-type virus accumulated extensively in the anther tissues, including the tapetum and in pollen grains (Fig. 2D), although not all of the pollen grains were infected. In contrast, the mutant virus was detected in the central vascular tissues of the anther (Fig. 2C, arrow) , but the pollen grains and other tissues of the anther remained uninfected. Sections of carpels revealed fully developed ovules. The wild-type isolate again accumulated abundantly in the carpel and the ovules (Fig. 2F), while the D12K mutant was found only sparsely distributed in the vascular tissues in the carpel (Fig. 2E, arrow). No virus was found in the ovule (Fig. 2E) or in the egg cells (data not shown). Because pollen grain and/or egg cell infection is the route for PEBV to achieve seed transmission, it is not surprising that seed transmission in D12K-infected plants was almost completely abolished when infection of the gametes was absent. Apparently the 12K protein (or its coding sequence) is involved in the infection of the gametes. The lack of D12K accumulation in the male and female gametes contrasted with the elevated level of this mutant in the vegetative leaf tissue and the mature pod tissue. Clearly, the removal of the 12K gene can have
FIG. 3. ELISA and Northern analysis of PEBV accumulation in infected peas. (A) ELISA assay of PSbMV in leaf samples from the upper parts (systemic leaves) of plants inoculated with wild-type (WT), D12K, and 12FS PEBV. Control, uninfected material (H) was processed in parallel; data are expressed as A405 { SE. (B) Northern blot of RNA extracted from healthy plants (H) and plants infected with WT, 9FS, and D12K PEBV. The positions of PEBV RNA 1 and RNA 2 are marked; RNA 2 could be detected by the probe prepared to RNA 1 sequences through the conserved sequence in the 3* untranslated region (MacFarlane et al., 1991).
different effects on virus/host interactions in different organs and in the tissues at different developmental stages. ACKNOWLEDGMENTS We thank Margaret Boulton, Jeff Davies, and George Lomonossoff for comments on the manuscript prior to publication. D.W. was in receipt of a Plant Molecular Biology II grant from the Biotechnology and Biological Research Council (BBSRC). The John Innes Centre is grant-aided from the BBSRC. The Scottish Crop Research Institute is grant-aided from the Scottish Office Agriculture, Environment and Fisheries Department (SOEAFD). The work was carried out under Ministry of Agriculture Fisheries and Food Licence No. PHF 1419C/1669 and SOEAFD Licence No GM/43/1993.
REFERENCES TABLE 2 Seed Transmission of PEBV Wild-Type Isolate and the Mutant D12K Experiment 1 (Feb. 1995–May 1995)
Experiment 2 (March 1995–June 1995)
Virusa
Number of seedlings tested
Seed transmission (%)
Number of seedlings tested
Seed transmission (%)
WT D12K
36 59
40 0
64 75
25 1
a
PEBV12FS not included as no infection of pea was recorded.
AID
VY 8637
/
6a3d$$$201
06-25-97 11:24:24
Bennet, C. W. (1969). Seed transmission of plant viruses. Adv. Virus Res. 14, 221–261. Bergh, S. T., Koziel, M. G., Huang, S-C., Thomas, R. A., Gilley, D. P., and Siegel, A. (1985). The nucleotide sequence of tobacco rattle virus RNA2 (CAM strain). Nucleic Acids Res. 13, 8507–8518. Brisson, L. F., Tenhaken, R., and Lamb, C. (1994). Function of oxidative cross-linking of cell wall structural proteins in plant disease resistance. Plant Cell 6, 1703–1712. Costa, A. S., and Kitajima, E. E. (1968). TransmissaV o do võB rus do anel do pimentaV o atrave´s da semente de tomate. Rev. Soc. Bras. Fitopatol. 2, 25–28. Ding, X. S., Cockbain, A. J., and Govier, D. A. (1992). Improvements in the detection of pea seed-borne mosaic virus by ELISA. Ann. Appl. Biol. 121, 75–83. Edwards, M. C. (1995). Mapping of the seed transmission determinants
vira
AP: VY
VIRAL DETERMINANTS OF PEBV SEED TRANSMISSION IN PEA
117
of barley stripe mosaic virus. Mol. Plant-Microbe Interact. 8, 906– 915. Goulden, M. G., Lomonossoff, G. P., Davies, J. W., and 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 subgenomic promoters. Nucleic Acids Res. 18, 4507–4512. Gru¨ndemann, D., and Koepsell, H. (1994). Ethidium bromide staining during denaturation with glyoxal for sensitive detection of RNA in agarose gel electrophoresis. Anal. Biochem. 216, 459–461. Guilford, P. J., Ziegler-Graff, V., and Baulcombe, D. C. (1991). Mutation and replacement of the 16-kDa protein gene in RNA-1 of tobacco rattle virus. Virology 182, 607–614. Hampton, R. O., and Francki, R. I. B. (1992). RNA-1 dependent seed transmissibility of cucumber mosaic virus in Phaseolus vulgaris. Phytopathology 82, 127–130. Hanada, K., and Harrison, B. D. (1977). Effects of virus genotype and temperature on seed transmission of nepoviruses. Ann. Appl. Biol. 85, 79–92. Johansen, I. E., Dougherty, W. G., Keller, K. E., Wang, D., and Hampton, R. O. (1996). Multiple determinants affect seed transmission of pea seedborne mosaic virus in Pisum sativum. J. Gen. Virol. 77, 3149– 3154. Kunkel, T. A., Roberts, J. D., and Zakour, R. A. (1987). Rapid and efficient site-specific mutagenesis without phenotypic selection. Methods Enzymol. 154, 367–382.
MacFarlane, S. A., Taylor, S. C., King, D. I., Hughes, G., and Davies, J. W. (1989). Pea early browning virus RNA1 encodes four polypeptides including a putative zinc-finger protein. Nucleic Acids Res. 17, 2245– 2260. MacFarlane, S. A., Wallis, C. V., Taylor, S. C., Goulden, M. G., Wood, K. R., and Davies, J. W. (1991). Construction and analysis of infectious transcripts synthesized from full-length cDNA clones of both genomic RNAs of pea early browning virus. Virology 182, 124–129. MacFarlane, S. A., and Brown, D. J. F. (1995). Sequence comparison of RNA2 of nematode-transmissible and nematode non-transmissible isolates of pea early-browning virus suggests that the gene encoding the 29kDa protein may be involved in nematode transmission. J. Gen. Virol. 76, 1299–1304. MacFarlane, S. A., Brown, D. J. F., and Bol, J. F. (1995). The transmission by nematodes of tobraviruses is not determined exclusively by the virus coat protein. Eur. J. Plant Pathol. 101, 535–539. MacFarlane, S. A., Wallis, C. V., and Brown, D. J. F. (1996). Multiple virus genes involved in the nematode transmission of pea early browning virus. Virology 219, 417–422. Maule, A. J., and Wang, D. (1996). Seed transmission of plant viruses: A lesson in biological complexity. Trends Microbiol. 4, 153–158. Shepherd, R. J. (1972). Transmission of viruses through seed and pollen. In ‘‘Principles and Techniques in Plant Virology’’ (C. I. Kado and H. O. Agrawal, Eds.), pp. 267–292. Van Nostrand Reinhold, New York. Wang, D., and Maule, A. J. (1997). Contrasting patterns in the spread of two seed-borne viruses in pea embryos. Plant J. 11, 1333–1340.
AID
vira
VY 8637
/
6a3d$$$201
06-25-97 11:24:24
AP: VY