Evaluation of host resistance to Barley yellow mosaic virus infection at the cellular and whole-plant levels

Authors

  • Y. You,

    1. Graduate School of Agricultural and Life Sciences, University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657
    Search for more papers by this author
  • Y. Shirako

    Corresponding author
    1. Graduate School of Agricultural and Life Sciences, University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657
    2. Asian Natural Environmental Science Center, University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan
      E-mail: yshirako@anesc.u-tokyo.ac.jp
    Search for more papers by this author

E-mail: yshirako@anesc.u-tokyo.ac.jp

Abstract

This study analysed the resistance and susceptibility of several barley cultivars harbouring one of six resistance genes (rym1–6) to Japanese Barley yellow mosaic virus (BaYMV) isolate JK05 using infectious RNA transcripts. At the cellular level, the virus replicated efficiently in mesophyll protoplasts from rym3 cultivars, less efficiently in protoplasts from a rym2 cultivar and poorly or not at all in protoplasts from other rym cultivars. None of the rym cultivars was susceptible at the whole-plant level. These results indicate that rym2 and rym3 cultivars are resistant to this virus isolate at the movement level, whereas the other rym cultivars are resistant at the cellular level. In addition, among five barley cultivars without known resistance genes, it was found that the cv. Haruna Nijo was resistant to this virus isolate at the cellular level. Based on the host responses, it was inferred that the BaYMV isolate JK05 is in the Japanese pathotype II group. A possible mechanism for how resistance-breaking variants continually emerge in fields is discussed.

Introduction

Barley yellow mosaic virus (BaYMV), the type species of the genus Bymovirus in the family Potyviridae, is a pathogen that causes a yellow mosaic disease in winter barley in east Asia and Europe (Berger et al., 2005). BaYMV infects barley plants in fields by itself or in combination with Barley mild mosaic virus (BaMMV), another Bymovirus species. The severity of yellow mosaic and stunting symptoms resulting from these diseases depends on the virus isolate, the presence of coinfecting viruses and the barley cultivar (Kühne, 2009). BaYMV is transmitted in the soil by zoospores of Polymyxa graminis, a plasmodiophoraceous protist, and can be retained inside its thick-walled resting spores for decades (Adams et al., 1988; Kanyuka et al., 2003). BaYMV has a bipartite plus-sense RNA genome. RNA1 encodes eight proteins, including the RNA-dependent RNA polymerase, a genome-linked protein (VPg), the capsid protein (CP) and a serine proteinase (NIa-Pro), and it replicates autonomously in cells. RNA2 encodes a cysteine proteinase (P1) and a putative vector-transmission factor (P2), both of which are required for systemic infection (Urcuqui-Inchima et al., 2001; Berger et al., 2005; You & Shirako, 2010).

Breeding barley cultivars that harbour one or more resistance gene(s) is currently the only approach to control the yellow mosaic disease caused by BaYMV and/or BaMMV infection because of the manner of natural transmission by P. graminis in the field (Kanyuka et al., 2003). Thus far, 15 recessive genes termed rym (resistance to yellow mosaic) and three dominant genes termed Rym have been identified in the germplasms of Hordeum vulgare or H. bulbosum genotypes resistant to different isolates of BaYMV and BaMMV (Werner et al., 2003; Diaz-Pendon et al., 2004; Ordon et al., 2004; Truniger & Aranda, 2009; Kai et al., 2012). Although the rym1–6 genes have been used for breeding BaYMV-resistant cultivars, none of these genes has been shown to confer complete resistance to any BaYMV isolate on their own, and new virulent BaYMV variants are continually emerging (Ordon et al., 2005; Kühne, 2009). The mechanisms of the rym-mediated resistance and of the resistance breaking by new virus isolates remain poorly understood (Kanyuka et al., 2003). Genes rym4 and rym5 are located on chromosome 3HL and have been shown to encode barley eukaryotic translation initiation factor 4E (eIF4E). An interaction between eIF4E and the virus genome-linked protein (VPg) has been implicated in breaking rym4-mediated resistance (Kanyuka et al., 2005; Stein et al., 2005), similar to the interaction between viruses in the genus Potyvirus and their hosts (Ruffel et al., 2002; Nicaise et al., 2003; Gao et al., 2004; Moury et al., 2004; Charron et al., 2008). The resistance functions of rym1, rym2, rym3 and rym6, which map to the chromosomes 4HL 7HL, 5HS and 3HL, respectively, have yet to be analysed.

Recently, a reverse genetics system was developed for BaYMV isolate JK05 (BaYMV-JK05) as the first case in the genus Bymovirus and the functions of the RNA2-encoded proteins, P1 and P2, during systemic infection were studied (You & Shirako, 2010). P1 is essential in systemic infection whereas P2 facilitates systemic infection. In the study presented here, infectious in vitro transcripts were used to analyse host resistance and susceptibility to BaYMV infection at the cellular and whole-plant levels using barley cultivars with one of genes rym1–6 and five cultivars without known resistance genes.

Materials and methods

Barley cultivars

The 16 barley cultivars used in this study are listed in Table 1. Seeds of New Golden, Haruna Nijo, KoA, Hadaka 1, Kashimamugi, Mihori Hadaka 3 (rym2), Haganemugi (rym3), Ishuku Shirazu (rym3), Franka (rym4), Mikamo Golden (rym5), Misato Golden (rym5), Miho Golden (rym6), Amagi Nijo (rym6) and Mokusekko 3 (rym1 + rym5) were provided by the Institute of Plant Science and Resources at Okayama University with support from the National Bio-Resource Project of the Ministry of Education, Culture, Sports, Science and Technology, Japan. Nittakei 68 (rym1) seeds were provided by the Bioresources Research and Development Department of Sapporo Breweries Ltd, Japan. Cultivar Ryofu is one of the common barley cultivars grown in Japan and the seeds were obtained at a local JA (Japan Agriculture Cooperative) branch.

Table 1. Barley cultivar responses to Barley yellow mosaic virus (BaYMV) isolate JK05 at the cellular and whole-plant levels
CultivarResistance geneaIn protoplastsbMechanical inoculationResponsed
SymptomSystemic infectionc
  1. aInformation on resistance genes is from Takahashi et al. (1970); Graner & Bauer, (1993); Konishi et al. (1997); Iida et al. (1999); Saeki et al. (1999); Konishi & Kaiser-Alexnat, (2000); Okada et al. (2004); Kanyuka et al. (2005); Stein et al. (2005).

  2. bLevels of capsid protein (CP) expression in protoplasts from different barley cultivars transfected with BaYMV-JK05 RNA1 and RNA2 transcripts and analysed by western blotting using anti-CP serum. The number of ‘+’ symbols represents the amount of CP detected by western blotting analysis; −, CP was not detected by western blotting analysis nor was GFP fluorescence detected with fluorescence microscopy.

  3. cAll of the inoculated plants were analysed via western blotting to quantify the level of CP in the upper leaves. The number of CP-positive plants/the number of plants inoculated is indicated. Results from separate experiments are shown in parentheses.

  4. dS, susceptible; R, resistant.

Ryofu+++Yes15/28 (3/6, 5/8, 3/8, 4/6)S
KoA+++Yes2/5 S
New Golden+++Yes6/9 (3/5, 3/4)S
Hadaka 1+++Yes5/11 (2/6, 3/5)S
Kashimamugi++No3/6 (1/2, 2/4)Ssymptom-free
Haruna Nijo+No0/11 (0/6, 0/5)Rcellular
Nittakei 68 rym1 +No0/18 (0/6, 0/6, 0/6)Rcellular
Mihori Hadaka 3 rym2 ++No0/15 (0/5, 0/4, 0/6)Rmovement
Ishuku Shirazu rym3 ++No0/11 (0/4, 0/4, 0/3)Rmovement
Haganemugi rym3 +++No0/19 (0/5, 0/6, 0/8)Rmovement
Franka rym4 No0/8Rcellular
Mikamo Golden rym5 No0/11 (0/6, 0/5)Rcellular
Misato Golden rym5 No0/6Rcellular
Miho Golden rym6 +No0/8 (0/3, 0/5)Rcellular
Amagi Nijo rym6 No0/11 (0/6, 0/5)Rcellular
Mokusekko 3 rym1 + 5 No0/11 (0/5, 0/6)Rcellular

Infectious cDNA clones of BaYMV RNA1 and RNA2

Full-length cDNA clones of a BaYMV isolate collected from a field in Kurashiki, Japan, in 2005 (JK05) were prepared in the previous study and were designated as pBY1 for RNA1 and pBY2 for RNA2 (You & Shirako, 2010). pBY2.GFP harbours a green-fluorescent-protein gene from pQBI25 (Takara Shuzo) cloned in the place of the P12-coding region in RNA2 (You & Shirako, 2010). Plasmid DNA was linearized immediately downstream of the inserted genomes using the restriction enzymes XhoI (pBY1) or SpeI (pBY2 or pBY2.GFP), and this served as the template for the generation of in vitro transcripts of wildtype (WT) RNA1 and WT RNA2 or GFP RNA2. The transcription reaction was performed using T7 RNA polymerase (Takara Bio) in the presence of a cap analogue (New England Biolabs) and ribonucleotide substrates at 37°C for 1 h.

Barley mesophyll protoplast preparation and RNA transfection

Barley seedlings were grown in a mixture of vermiculite and peat moss (1:1) with a half-strength Hoagland solution at 22°C for 6 days under 16 h light with 20 000 lux/24 h in a growth cabinet (Type MLR-350; SANYO Electric Co.). The primary leaves were harvested and mesophyll protoplasts were isolated as previously described (Ohsato et al., 2003). Approximately 0·5 × 105 cells were transfected with WT RNA1 transcripts, in combination with WT RNA2 or GFP RNA2 transcripts, by a polyethylene glycol method, then incubated in one well of a 24-well plate at 15°C for 60 h in the dark in a Peltier device-controlled incubator (Type CN-25C; Mitsubishi Electric Engineering Co.) as previously described (You & Shirako, 2010).

Barley plant inoculation

Barley seedlings were grown in horticultural soil at 22°C in a growth cabinet under 16 h of light with 20 000 lux/24 h for 2 weeks until the 3- to 4-leaf stage. WT RNA1 and WT RNA2 transcripts were diluted in an RNA inoculation buffer (50 mm glycine, 30 mm K2HPO4, pH 9·2, containing 1% [w/v] bentonite) and inoculated on leaves dusted with carborundum (600 mesh) by rubbing with a cotton ball. Inoculated plants were grown at 15°C for up to 12 weeks under 16 h light with 20 000 lux/24 h in a growth cabinet.

Western blot analysis

Protoplast samples

After incubation, protoplasts from one well of a 24-well plate with approximately 0·5 × 105 cells were collected by centrifugation at 100 g for 5 min and suspended in 50 μL of 1 × sample buffer (SB) (5% SDS, 20 mm DTT, 50 mm Tris–HCl, pH 9·0).

Plant tissue samples

Fifty milligrams of plant tissue were ground in 500 μL of 1 × SB. After heating samples at 95°C for 3 min, the samples were separated on a 12·5% SDS-polyacrylamide gel. The samples were normalized based on the intensity of the large subunit of ribulose-1,5-bisphosphate carboxylase oxygenase stained with Coomassie brilliant blue G-250. After electrophoresis, the proteins were transferred from the gel onto a nitrocellulose membrane (Protoran BA85; Whatman). The BaYMV capsid protein (CP) was detected with anti-GST:BaYMV-CP rabbit polyclonal antiserum (You & Shirako, 2010) as the primary antibody and a goat-anti-rabbit IgG conjugated with alkaline phosphatase (Jackson Immunotechnology) as the secondary antibody, and using 5-bromo-4-chloro-3-indolyl phosphate and nitroblue tetrazolium as the substrates. To quantify the levels of CP accumulation in the protoplasts from different barley cultivars, the CP bands on multiple blots were scanned using NIH ImageJ software (v. 1.42q) and converted from digital images to numerical data. Samples from Hadaka 1 were included as the reference in most of the blots and the scores were regarded as 10 to calibrate the values from different cultivars on different blots.

Fluorescent light microscopy

Protoplasts transfected with the WT RNA1 and GFP RNA2 transcripts were incubated at 15°C for 60 h in one well of a 24-well plate. Subsequently, the levels of GFP expression from the GFP RNA2 were examined using a fluorescence light microscope (IX70; Olympus) with a GFP filter.

Results

Analysis of gene expression from BaYMV RNA transcripts in protoplasts

Mesophyll protoplasts from each barley cultivar were prepared and transfected with WT RNA1 and WT RNA2 transcripts. After incubation at 15°C for 60 h, the RNA replication efficiency, as indicated by the levels of CP expression from RNA1 in the transfected protoplasts, was examined by western blotting using a primary antiserum against CP. It should be mentioned that expression of P1 protein from WT RNA2 is required for efficient replication of RNA1 and expression of CP, as was previously reported (You & Shirako, 2010). Representative blots are depicted in Figure 1. CP was strongly detected in the transfected protoplasts from KoA (Fig. 1a, lane 2), New Golden (Fig. 1a, lane 3), Hadaka 1 (Fig. 1a, lane 8 and 1b, lane 5) and Haganemugi (rym3) (Fig. 1b, lane 4). Less CP was detected in the transfected protoplasts from Kashimamugi (Fig. 1a, lane 4) and Ishuku Shirazu (rym3) (Fig. 1a, lane 7). CP was faint but detectable in the transfected protoplasts from Nittakei 68 (rym1) (Fig. 1a, lane 5 and 1b, lane 2), Mihori Hadaka 3 (rym2) (Fig. 1a, lane 6 and 1b, lane 3), Miho Golden (rym6) (Fig. 1b, lane 7) and Haruna Nijo (Fig. 1b, lane 9). However, CP was not detectable in the transfected protoplasts from Mikamo Golden (rym5) (Fig. 1a lane 9), Mokusekko 3 (rym1 + rym5) (Fig. 1a, lane 10 and 1b, lane 8), Amagi Nijo (rym6) (Fig. 1a, lane 11), Misato Golden (rym5) (Fig. 1b, lane 6) and Franka (rym4) (Fig. 1c, lane 2). Figure 1d summarizes the levels of CP accumulation in different barley cultivars from five different blots, which contained approximately 10 samples including the negative controls. The blots were scanned using ImageJ software and the CP band images were converted to numerical data. The Hadaka 1 data from the different blots, with a score of 10, were regarded as the reference and were used to calibrate the values from the different cultivars on different blots. The bar heights are averages from at least three blots for New Golden, Hadaka 1, Nittakei 68 (rym1), Mihori Hadaka 3 (rym2), Haganemugi (rym3) and Mikamo Golden (rym5), and from two blots for Haruna Nijo and Miho Golden (rym6). The data for Franka (rym4) are from a single blot.

Figure 1.

 (a–c) Capsid protein (CP) expression in barley protoplasts from different cultivars transfected with Barley yellow mosaic virus (BaYMV) isolate JK05 RNA1 and RNA2 transcripts, incubated at 15°C for 60 h. The accumulated CP was analysed via western blotting using anti-CP serum. (d) Relative CP accumulation levels in protoplasts from different barley cultivars. The CP bands on western blots were scanned using ImageJ software. Hadaka 1 bands were used as the controls for blots and the relative accumulation scores for other cultivars were calibrated against those of Hadaka 1, which were set to 10.

Protoplasts from different barley cultivars were also transfected with a combination of WT RNA1 and GFP RNA2 transcripts, the latter of which contains a GFP gene in place of the P12 polyprotein gene. After incubation at 15°C for 60 h, the levels of GFP expression from the GFP RNA2 were observed with fluorescent microscopy. The experiments were performed in duplicate. The responses of individual cultivars after RNA transfection were characterized both by the strength of the GFP fluorescence in single cells and by the transfection efficiency. GFP fluorescence was observed at four different levels: (i) approximately 40–50% of the protoplasts of New Golden, KoA, Hadaka 1 and Ryofu fluoresced brightly; (ii) approximately 5–30% of the protoplasts from Kashimamugi, Haganemugi (rym3) and Ishuku Shirazu (rym3) fluoresced less strongly than the protoplasts from (i); (iii) a few protoplasts from Haruna Nijo, Nittakei 68 (rym1), Mihori Hadaka 3 (rym2) and Miho Golden (rym6) fluoresced faintly; and (iv) the Franka (rym4), Mikamo Golden (rym5), Misato Golden (rym5), Amagi Nijo (rym6) and Mokusekko 3 (rym1 + rym5) protoplasts did not fluoresce. Examples of these four levels are illustrated in Figure 2. Thus, the results obtained by detecting CP from the protoplasts transfected with WT RNA1 and WT RNA2 transcripts by western blotting mimicked those obtained by the detection of GFP fluorescence from the protoplasts transfected with WT RNA1 and GFP RNA2 transcripts using fluorescence microscopy.

Figure 2.

 GFP-fluorescence in barley protoplasts from different cultivars transfected with Barley yellow mosaic virus (BaYMV) isolate JK05 RNA1 and GFP RNA2 transcripts, as observed using a fluorescent light microscope (Olympus IX-70) under UV light with a GFP filter (a–d) and under visible light (e). The transfected protoplasts were incubated at 15°C for 60 h. (a, e) Hadaka 1 with approximately 50% brightly fluorescing cells among round-shaped viable cells; (b) Kashimamugi with approximately 25% fluorescent cells; (c) Mihori Hadaka 3 (rym2) with few fluorescent cells; (d) Misato Golden (rym5) with no fluorescent cells. Scale bar = 100 nm.

Infectivity assay of the in vitro transcripts at the whole-plant level

To evaluate the host resistance and susceptibility of the entire plant, different cultivars of barley plants were mechanically inoculated at the 3- to 4-leaf stage (2 weeks after sowing and grown at 22°C under 16 h light with 20 000 lux/24 h in a growth cabinet) with WT RNA1 and WT RNA2 transcripts. The inoculated plants were kept at 15°C for up to 12 weeks under 16 h light with 20 000 lux/24 h in a growth cabinet. Twelve weeks after inoculation (14 weeks after sowing seeds), the barley ears were fully formed in all of the cultivars except for Franka, which is known to flower late. At 3–4 weeks after inoculation, approximately half of the inoculated Ryofu, New Golden, KoA and Hadaka 1 plants exhibited severe symptoms; yellow patches, streaks and mosaic patterns on the uninoculated upper leaves were apparent. Only the infected KoA plants became severely stunted and died by the fifth week. Strong CP was detected by western blotting in extracts of the uninoculated upper leaves of the plants with symptoms. Plants without symptoms were negative for CP, including Mihori Hadaka 3 (rym2), Haganemugi (rym3) and Ishuku Shirazu (rym3), in protoplasts from which BaYMV-JK05 RNA replicated (Fig. 1a, lanes 6 and 7; Fig. 1b, lanes 3 and 4; and Fig. 1d). Thus, development of conspicuous symptoms on upper leaves corresponded with detection of CP by western blot analysis. Kashimamugi was the only cultivar that did not develop any recognizable symptoms until 12 weeks after inoculation, but CP was detected in symptomless leaves and roots at levels similar to those detected in other systemically infected plants (data not shown).

Discussion

In this study, the infectious in vitro transcripts were used to analyse BaYMV–host compatibility precisely and identify resistant cultivars both at the cellular and whole-plant levels. The responses of 16 barley cultivars after inoculation with infectious in vitro transcripts of BaYMV-JK05 are summarized in Table 1. Ryofu served as a positive control (You & Shirako, 2010). New Golden, KoA and Hadaka 1 were susceptible to BaYMV-JK05, whereas Kashimamugi was symptom-free, i.e. tolerant, in response to this virus isolate. Conversely, Mihori Hadaka 3 (rym2) and two rym3 cultivars (Ishuku Shirazu and Haganemugi) supported viral replication at a single-cell level but were not systemically infected, indicating that they are resistant at the whole-plant level through the blockage of either cell-to-cell or long-distance viral movement. Because the level of BaYMV gene expression in Mihori Hadaka 3 (rym2) was limited compared with the other susceptible cultivars, it is anticipated that Mihori Hadaka 3 (rym2) should also be partially resistant to BaYMV-JK05 at the cellular level. Virus replication in root cells of rym2 and rym3 cultivars also needs to be examined to confirm whether the resistance to BaYMV-JK05 at the whole-plant level was determined by the deficiency of virus movements or virus replication in root cells. Nittakei 68 (rym1) and Miho Golden (rym6) were resistant to BaYMV-JK05 at the cellular level, whereas the rym4 and rym5 cultivars and Amagi Nijo (rym6) were completely immune to this virus isolate.

Among the cultivars without known resistance genes examined in this study, Haruna Nijo exhibited resistance at the cellular level, indicating that Haruna Nijo harbours an unidentified resistance gene against BaYMV-JK05. Haruna Nijo (Sapporo Breweries Ltd) was bred for grain yield and quality for brewing, not for BaYMV resistance; thus the sources of BaYMV resistance were not identified. Based on the breeding history described in one study (Seko et al., 1985), identifying the sources of BaYMV resistance by examining individual parental cultivars may be beneficial.

Previously, the responses of barley genotypes to different bymovirus isolates could be analysed only at the level of systemic infection after leaves were mechanically inoculated with infected leaf sap or after roots were inoculated with viruliferous P. graminis (Kashiwazaki et al., 1989; McGrann & Adams, 2004). Based on the symptomatology of the infected barley cultivars and serological assays, seven strains in four pathological groups of BaYMV have been identified in Japan and named pathotypes I-1, I-2, I-3, II-1, II-2, III and IV (Kashiwazaki et al., 1989; Konishi et al., 1997; Okada et al., 2004; Nishigawa et al., 2008; Sotome et al., 2010). Two European BaYMV strains, BaYMV-1 and BaYMV-2, can be distinguished by their responses to barley cultivars carrying the rym4 resistance gene (Kühne et al., 2003). The present study employed the Japanese system to classify isolate JK05 (Table 2). Based on its virulence in the tested barley cultivars, BaYMV-JK05 was most closely related to Japanese pathotype II. Japanese BaYMV pathotypes I, III and IV differ from pathotype II in overcoming resistance mediated by different rym genes. Pathotype I breaks rym6-mediated resistance; pathotype III breaks rym1-, rym5- and rym6-mediated resistance; and pathotype IV breaks rym3-mediated resistance; whereas pathotype II cannot overcome rym1-, rym3-, rym5- or rym6-mediated resistance. There are two strains, II-1 and II-2, within pathotype II; II-1 does not infect Kashimamugi and cannot break rym2-mediated resistance, whereas II-2 infects Kashimamugi systemically and breaks rym2-mediated resistance. BaYMV-JK05 infects Kashimamugi systemically without developing symptoms but cannot break rym2-mediated resistance, so it differs from both of the pathotype II strains. In addition, BaYMV-JK05 is phylogenetically distinct from the II-1 isolate at the whole genome scale as well (You & Shirako, 2010). Therefore, it is concluded that BaYMV-JK05 is a new member of the pathotype II group.

Table 2. Barley yellow mosaic virus (BaYMV) pathotypes in Japan and Europe
Barley cultivarResistance geneJapanese strainEU strain
IIIIIIIV12
I-1I-2I-3II-1II-2JK05
  1. Responses of barley cultivars to BaYMV-JK05 were evaluated in this study; responses to other BaYMV strains were published in references Kashiwazaki et al. (1989); Konishi et al. (1997); Okada et al. (2004); Ordon et al. (2005); Nishigawa et al. (2008); Sotome et al. (2010).

  2. S, susceptibility; R, resistance; –, no data.

New GoldenSSSSSSSS
Haruna NijoSSSRRRS
KashimamugiRSSRSSS
Nittakei 68 rym1 RRRRRRSRR
Mihori Hadaka 3 rym2 RRRRSRRRRR
Ishuku Shirazu/Haganemugi rym3 RRRRRRRSRR
Franka rym4 RRRRRS
Misato Golden/Mikamo Golden rym5 RRRRRRSRRR
Amagi Nijo rym6 SSSRRRSRSS

The use of a combination of infectious in vitro transcripts and barley protoplasts enabled a distinction to be made between host resistance against BaYMV infection at the cellular level and at the virus-movement level. Although they exhibited resistance at the whole-plant level, Mihori Hadaka 3 (rym2) and two rym3 cultivars (Ishuku Shirazu and Haganemugi) supported BaYMV-JK05 RNA replication in single cells fairly efficiently; Nittakei 68 (rym1), Miho Golden (rym6) and Haruna Nijo also exhibited signs of RNA replication at the cellular level. When a virus is not completely inviable in a cell, the possibility remains that a variant virus in a progeny virus population produced in a single infected cell may emerge with the ability to replicate more efficiently and move from cell to cell and over long distances. Previously, it was observed that a BaYMV RNA2 mutant that was defective in processing the P1/P2 cleavage site in the polyprotein P12 could not replicate efficiently in transfected protoplasts nor move systemically in inoculated plants; however, this mutant caused systemic infection in one inoculated barley plant after incurring a spontaneous mutation that restored the wildtype amino acid sequence at the P1/P2 cleavage site (You & Shirako, 2010). This phenomenon may explain how Bymovirus variants that overcome resistance continue to emerge in barley fields (Adams, 1991; Nomura et al., 1996; Hariri et al., 2003; Kanyuka et al., 2004, 2005; Habekuss et al., 2008; Nishigawa et al., 2008; Okada et al., 2008; Sotome et al., 2010). When cultivated barley plants in a field are systemically infected with an isolate of BaYMV, the field would become infested by P. graminis carrying the specific BaYMV strain that is composed of a population of variant viruses. After introduction of a new barley cultivar harbouring a resistance gene, a portion of existing variant viruses in the field may still be able to replicate in the root cells of the resistant cultivar, but the level of virus replication could be very low and may not move from cell to cell nor cause systemic infection. However, once one or more mutations in a variant virus genome allow sufficient replication in the root cells and support the movement of the variant virus, the entire plant will be systemically infected with the new variant virus. This situation increases the titre of the new variant virus acquired by P. graminis feeding on the plant root cells, resulting in an increase of the resting spores carrying the variant virus in the root debris and in soil in a limited area in the field. The viruliferous P. graminis could then be spread by agricultural implements through the entire field in the following years. It may still be years before a wide area of the field becomes infested by P. graminis carrying a resistance-breaking variant virus, i.e. before farmers and researchers recognize that a new pathogenic isolate of BaYMV has emerged. In this scenario, the continuous cultivation of a barley cultivar harbouring identified rym recessive resistance genes in a barley field infested with a strain of BaYMV would not prevent the emergence of new resistance-breaking variant viruses.

To analyse Bymovirus pathogenicity and host resistance mechanisms further, the viral factors involved in overcoming rym-mediated resistance should be identified, by constructing and analysing chimeric infectious cDNA clones containing cDNA fragments from Japanese pathotypes I, III or IV, as well as from the European strain 2 in the pBY1 and pBY2 backgrounds.

Acknowledgements

YY was supported by the University of Tokyo Fellowship for International Students. This work was supported by a research grant to YS from the University of Tokyo.

Ancillary