A comparative study of the Barley yellow dwarf virus species PAV and PAS: distribution, accumulation and host resistance

Authors


E-mail: jiban@vurv.cz

Abstract

This study investigated the distribution and characteristics of the Barley yellow dwarf virus (BYDV) species BYDV-PAS, which was recently separated from BYDV-PAV, the most commonly studied BYDV species. Throughout 3 years of experimental monitoring of BYDV incidence, PAS was the most frequently occurring species infecting cereals and grasses in the Czech Republic. Furthermore, Rhopalosiphum maidis and Metopolophium dirhodum were recorded as BYDV-PAS vectors, even though M. dirhodum does not usually transmit BYDV-PAV. In field experiments with barley and wheat, where virus accumulation, symptoms and effect on the yield were tested, BYDV-PAV was more severe than PAS. Infection with the BYDV-PAV isolate resulted in greater expression of symptoms and also in a greater reduction in plant height and grain weight per spike than BYDV-PAS. In a sensitive cultivar of barley (Graciosa), the amount of viral RNA of BYDV-PAV was also significantly higher than that of BYDV-PAS. In a tolerant line (Wbon-123), however, no such differences were found. In conclusion, although BYDV-PAS seems to be dominant in the Czech Republic, BYDV-PAV has the potential to cause more significant crop losses in barley and wheat.

Introduction

Barley yellow dwarf (BYD) is one of the most widespread and damaging viral diseases of grasses and cereal crops worldwide (D’Arcy, 1995). The disease is caused by a group of related single-stranded RNA viruses assigned to the Luteovirus (Barley yellow dwarf virus (BYDV) spp. PAV, PAS, MAV and GAV) or Polerovirus (Cereal yellow dwarf virus-RPV) genera, or those unassigned to a genus (BYDV-SGV, BYDV-RMV and BYDV-GPV) in the family Luteoviridae (Miller & Rasochova, 1997). The virus is phloem-restricted and transmitted in a persistent manner, but with variable effectiveness, by many aphid species (Slykhuis, 1967). The BYDV species vary strongly in symptom manifestation, aphid transmission efficiency, host preferences and in their serological and molecular properties. PAV is the best studied BYDV, especially at the molecular level. It is also assumed to be the most widespread (D’Arcy, 1995) and usually causes the most severe symptoms in most of the cereal crops. High PAV incidence is claimed to be caused by high transmission efficiency and a vector with a broad host range (McElhany et al., 1995; Power & Gray, 1995). Previously, only PAV isolates were identified in the Czech Republic on the basis of serological detection, but the presence of BYDV-PAS and BYDV-MAV species was recently identified (Kundu, 2008, 2009). BYDV-PAS (type isolate PAV-129) was recently separated from PAV (Mayo, 2002). As these two species were indistinguishable by serological methods, molecular diagnostic methods such as RT-PCR-RFLP (Kundu et al., 2009) and sequencing were used to this end. Sequence analysis of BYDV field isolates conducted in the Czech Republic revealed that BYDV-PAS and -MAV were the most frequent isolates, whereas BYDV-PAV seemed to be less frequent in cereal crops (Kundu et al., 2009).

Barley yellow dwarf virus (BYDV) induces yield losses ranging from 5 to 80%, with an average of 30% in affected fields (Perry et al., 2000). BYDV has caused significant yield losses over recent decades, particularly in winter crops, and PAV was regarded as the only species occurring in the Czech Republic (Vacke, 1991). The most effective and sustainable control method is the use of genetic resistance/tolerance to the virus complex (Henry et al., 2002). In barley, the semidominant resistance gene Yd2, which was detected in a land race of Ethiopian origin, provides the most efficient protection against different BYDV isolates (Lister & Ranieri, 1995). This gene has also been transferred to several spring and winter barley cultivars and lines (Burnett et al., 1995; Ovesnáet al., 2000; Šíp et al., 2004). Two resistance sources, Bdv1 (Singh et al., 1993) derived from the Brazilian spring wheat cultivar Forntana and Bdv2 (Banks et al., 1995) from Thinopyrum intermedium, have been described in wheat genotypes (Henry et al., 2002). Evaluation of resistance sources carrying these genes (Bdv1 and Bdv2) in central European conditions (Veškrna et al., 2009) suggested the polygenic nature of BYDV resistance. Similar results were obtained with wheat populations segregating for the Bdv1 gene (Ayala et al., 2002). Most resistance studies to date have used PAV, as it was regarded as the most prevalent and detrimental to yields of the BYDV species (Vacke et al., 1996; Ovesnáet al., 2000; Balaji et al., 2003; Jefferies et al., 2003). However, several reports have suggested that PAS may cause resistance breaking in spring oat (Chay et al., 1996).

The aim of this study was to investigate the two most frequently occurring BYDV species in the Czech Republic, PAS and PAV, in particular (i) the incidence of these BYDV species in cereal and grass hosts and aphid vectors in agroecosystems, (ii) the quantitative analysis of viral titre of these species in susceptible and tolerant hosts, and (iii) the resistance of winter wheat and barley to these species.

Materials and methods

Survey of BYDV in host plants and aphid vectors

Samples of cereals, grass weeds and wild grass species, as well as aphids, were collected from 2008 to 2010 randomly. The monitoring took place during the growth season (April to November) all over the Czech Republic. In all cases, only one mixed sample was collected from each field – either the plants with the most symptoms or the plants with the highest aphid incidence. Aphids were found on plant leaves, stems or in the spikes by visual inspection. For each sample, the following information was collected: (i) origin; (ii) plant species; (iii) aphid species (if applicable); (iv) date of collection; (v) season of crop sowing (autumn or spring); and (vi) volunteer plant (if applicable). Plant leaves were collected and transported on ice to the laboratories, where they were immediately used for diagnostic tests, or were ground and stored at −80°C.

Aphids were collected from cereals and grasses into Petri dishes using a thin paintbrush and were transported to an insect-proof greenhouse, where they were transferred onto healthy barley seedlings (cv. Sebastian). An average sample consisted of approximately 20 aphids collected from one field/balk/meadow. An additional sample from an identical area consisting of about 10 aphids was also collected and conserved in 70% ethanol in a 1·5-mL Eppendorf tube. The aphids were allowed to feed and reproduce on the seedling for a period of 28 days. Leaf samples were then collected and tested for the presence of BYDV by both DAS-ELISA and RT-PCR. In case of discordant results, a second test and a test with the reserve aphids sample were performed. The BYDV-positive samples were further tested by RT-PCR-RFLP according to Kundu et al. (2009) so that the BYDV species could be identified.

BYDV polyclonal antibodies (Sediag) were used in DAS-ELISA (Clark & Adams, 1977) to detect the virus in leaf tissues of cereals and grasses. Samples for ELISA were prepared by grinding 1 g leaf tissue in phosphate buffered saline, pH 7·4, with 2% polyvinylpyrrolidone and 0·2% egg albumin in a ratio of 1:20. Microplates were read using a MR 5000 Dynatech reader at 405 nm.

For RT-PCR, plant samples were processed according to Kundu et al. (2009). Aphid samples were first cleared of the ethanol in which they were stored. Next, they were homogenized with a sterilized minishaker (using pellet pestles and a cordless motor; Sigma) in 10 μL lysis buffer from the MasterPure Complete DNA and RNA Purification Kit (Epicentre) in a 1·5-mL Eppendorf tube. After the homogenization, total nucleic acids were purified from the sample using the extraction kit mentioned above according to the manufacturer’s instructions. RT-PCR and RFLP analyses were then carried out as described for the plant samples.

Quantitative analysis of the BYDV titre by real-time RT-qPCR

A winter barley cultivar, Graciosa, and the line Wbon-123 were sown in early autumn (September) 2009 in a small-scale field experiment. Three square plots (each plot 1 m2) were set up in two replicates, each with 100 plants (50 Graciosa plants and 50 Wbon-123 plants). One of the three plots represented BYDV-PAV infection, another BYDV-PAS infection and the third the healthy control. The plots were covered by insect-proof net isolators (1 m2 × 1 m2). When the plants were at the growth stage where two leaves (beginning of tillering – Feekes 2·0) were unfolded, they were infected with BYDV-PAS (isolate PS-RuJK, accession number EU863652) and BYDV-PAV (isolate Blatno85, accession number FJ645745) species by vector transmission. For each plant, approximately five viruliferous Rhopalosiphum padi aphids were allowed to feed on the plant for 2 days. Afterwards, the plants were treated with insecticide, Perfekthion (Dimethoate) at 0·6 L h−1.

At the time of symptom manifestation (early November), 20 randomly chosen plants were collected from each plot, and the green part of the plants (excluding root) was used for the analysis. The plants were ground in liquid nitrogen and 100 mg of the material used for RNA extraction. The RNA was isolated with a Spectrum™ Plant Total RNA Kit (Sigma-Aldrich) according to the manufacturer’s instructions. The concentration and purity of the isolated RNA was then measured spectrophotometrically.

The viral titre was determined by RT-qPCR analysis according to Jarošová & Kundu (2010). Real-time PCR was performed using a 7300 Real-Time PCR system (Applied Biosystems). For analysis with SYBR Green I, the PCR cycling consisted of three steps: a 2-min incubation at 95°C followed by 40 cycles of 95°C for 15 s and 60°C for 60 s; the conditions of the final dissociation curve were as follows: 95°C for 15 s, 60°C for 60 s and 95°C for 15 s. The PCR mastermix composed of the primers (1 μL primer pair mix of 10 μm primer pair stock), 12·5 μL of 2 × Power Sybr Green Master mix (Applied Biosystems) and sterile nuclease-free water was added to a final volume of 20 μL. Finally, 5 μL cDNA was added to this mixture. Three reference genes (GAPDH, TUBB and 18S rRNA) were used to normalize the quantification as recommended and described in Jarošová & Kundu (2010). Results were analysed using statistica v. 9 (StatSoft) and excel (Microsoft). The values from the 20 plants representing one group (e.g. Graciosa infected with BYDV-PAV) were evaluated as a statistical data set by explorative data analysis (EDA) to test for a normal distribution. The statistical program used for EDA was qc.expert v. 2.5 running graphical tests and tests of normality. When normality was ruled out, a Box–Cox transformation of the data was carried out and a two-way anova was then performed using the program statistica. All statistical analyses were performed at the 0·05 level of probability.

Resistance test of wheat varieties to PAS and PAV under field conditions

A comparative study was performed to determine the level of resistance of these traits to both the PAV and PAS species. Ten barley (Wbon-116, Wbon-123, Doria, Wysor, Sigra, Finesse, Traminer, Perry, Luran and Graciosa) and 10 wheat (PSR 3628, Svitava, Dromos, Sparta, Banquet, Roane, Sisson, Meritto, Vlada and SG-S 2703) cultivars and breeding lines with different levels of resistance ranging from tolerant (containing the Yd2 gene in barley) to the most susceptible were selected. Winter barley cultivars and lines (Wysor, Doria, Wbon-116 and Wbon-123) containing the Yd2 gene were included. The selection of winter wheat varieties was based on previous experiments during 2004–2008 as described in Veškrna et al. (2009). The resistance levels of winter barley and wheat to infection by BYDV-PAV (isolate Blatno85) and BYDV-PAS (isolate PS-RuJK) was studied in field experiments running for over 2 years (2008/2009 and 2009/2010). These experiments had both infected and healthy variants and were carried out according to the methods described by Vacke et al. (1996). The plants were grown on two-row 1-m-long plots, each with two replicates (plant spacing 6 × 22 cm). The plant materials were infected with the PAV and PAS species of BYDV at tiller initiation (Feekes 2·0) using viruliferous R. padi greenhouse-reared aphids. Their inoculation access periods lasted 5–7 days and the aphids were then killed by insecticide, Perferthion (Dimethoate) at 0·6 L h−1. To avoid infection by fungal diseases (e.g. powdery mildew, rusts, leaf spot diseases), fungicide Tango Super (Epoxiconazole, Fenpropiomorph) was applied at 1 L h−1 at the stage of flag leaf emergence GS39 (Zadoks et al., 1974). At the full flowering stage symptoms of BYDV infection were evaluated on a scale of 0–9 (where 0 represents no symptoms) developed by Schaller & Qualset (1980). At harvest, 20 randomly selected plants from close stands in each of the infected and control plots were evaluated for visual symptom score (VSS), plant height reduction (PHR) and grain weight per spike (GWS; to determine the percentage reduction in GWS as a result of infection, GWSR). The unistat v. 5.0 package (UNISTAT Ltd.) was used for statistical analyses of the data. The mean trait values across varieties were compared using the least significant difference (LSD) method at the 0·05 level of probability. Simple correlation coefficients between responses of cultivars to BYDV-PAS and BYDV-PAV in the characters examined were calculated for different environmental conditions.

Results

Survey of BYDV species in hosts and vectors

Of the 285 plant samples collected (Table 1), 151 were BYDV-positive. The most prevalent BYDV species was BYDV-PAS, detected in 66% of the cases (total number of samples, = 99), followed by MAV with 18% (= 27 samples) and PAV with 17% (= 25 samples). The PAS and PAV species were found on all the cultivated species (with the exception of Panicum miliaceum), and they were also present on Bromus spp., Festuca spp., Lolium multiflorum, Arrhenatherum elatius and Avena fatua. MAV showed quite high incidence in oat (33% of all samples), wheat, barley and millet plants. However, MAV was not found in corn or in weed species. The incidence of BYDV did not seem to be influenced by the plant species, region of the Czech Republic, or time of sowing (spring vs winter cereals) (Table 2). However, BYDV incidence was twice as high in volunteer plants (70%) than in cultivated or wild species (36%). The majority of BYDV infection in volunteer plants was associated with BYDV-PAS, which comprised 93% of all BYDV infections in volunteer plants (Table 2).

Table 1. Distribution of Barley yellow dwarf virus (BYDV) species in plants sampled
Plant speciesNumber of tested samplesNumber of BYDV-positive samples
PASPAVMAV
Triticum aestivum 185531014
Hordeum vulgare 812556
Panicum miliaceum 3001
Festuca spp.1100
Elytrigia repens 5210
Echinochloa crus-galli 5500
Lolium multiflorum 3030
Arrhenatherum elatius 1100
Bromus spp.3110
Avena sativa 20936
Avena fatua 3110
Zea mays 12110
Total number of plants285992527
Percentage10065·56·518
Table 2. Percentage distribution of individual Barley yellow dwarf virus (BYDV) species in field samples
SamplingIncidence of BYDV speciesBYDV-negative
PAV (%)PAS (%)MAV (%)
Winter cereals526663
Spring cereals including corn7221358
Permanent grasses1423064
Volunteer plants vs non-volunteer plants7 vs 1893 vs 620 vs 20 

During the 3 years of monitoring, approximately 300 samples of aphids were collected; of these, 285 samples survived and reproduced on barley seedlings. The four aphid species R. padi, Sitobion avenae, Rhopalosiphum maidis and Metopolophium dirhodum were found on the cereals or other Poaceae species (Triticum aestivum, Hordeum vulgare, Avena sativa, Zea mays, Bromus spp., Festuca spp., Elytrigia repens). The most prevalent aphid species were S. avenae and R. padi. In 2008 and 2010, the majority (≥70%) of samples collected were S. avenae, whereas in 2009, R. padi was the most prevalent. R. maidis and M. dirhodum were the least prevalent species during all years surveyed, constituting 5 and 6% of the total on average (Table 3). There was a seasonal pattern in the spread of aphid species. For example, in the summer S. avenae was the main species observed in the fields, whereas in the autumn, especially in late autumn months, R. padi prevailed.

Table 3. Incidence of aphid species sampled and of Barley yellow dwarf virus (BYDV) species among aphid samples
Aphid speciesPercentage abundance (%)BYDV species
MAV (%)PAS (%)PAV (%)
Rhopalosiphum padi 2601920
Sitobion avenae 531006780
Rhopalosiphum maidis 505 
Metopolophium dirhodum 609 
Not determined10   

All colonized barley plants were tested 1 month later by both DAS-ELISA and RT-PCR. A total of 232 aphid colonies were BYDV-free and 53 aphid colonies were BYDV-positive. The infectivity of the aphids collected from the fields was therefore lower than 20%. For individual years, the infectivity of the aphids varied from a maximum of 33% in 2008 to a minimum of 13% in 2010. The most prevalent BYDV species was BYDV-PAS, with 85% of all positive BYDV cases; followed by BYDV-PAV (11%) and BYDV-MAV (4%). Sitobion avenae was the only aphid species that was found as the vector of BYDV-MAV. Rhopalosiphum padi and S. avenae were able to transmit BYDV-PAV. The four aphid species identified were found to be BYDV-PAS vectors (Table 3). No significant correlations were found between aphid species and plant species.

Quantitative analysis of BYDV uptake in winter barley

At the time of symptom manifestation, symptoms appeared clearly in cv. Graciosa, but no symptoms were found in line Wbon-123. The symptoms caused by PAV were noticeably more severe than those caused by PAS (Fig. 1). Viral titres were determined by RT-qPCR and the efficiency ranged between 97% and 101% for the reference genes and target genes (Table 4). Relative values and results are shown in Table 5. Statistical differences were recorded between the two barley cultivars/lines for both species’ titres. In both cases, the viral titre in Graciosa was much higher than in Wbon-123. In the sensitive variety, the viral titres of BYDV-PAV and BYDV-PAS also differed greatly. In contrast, differences in viral titres within the resistant line were not significant. In the sensitive variety, however, the viral titre of BYDV-PAV was approximately twice that of BYDV-PAS.

Figure 1.

 Symptoms of Barley yellow dwarf virus (BYDV)-PAS and -PAV on winter barley cv. Graciosa (sensitive to BYDV) and line Wbon-123 (carrier of the Yd2 resistance gene) in a small-scale outdoor field experiment approximately 40 days after infection.

Table 4. Characteristics of gene-specific real-time RT-PCR
Gene symbolGene nameAccession no.Primer sequence (5′→3′)Amplicon sizePCR efficiency (%)
GAPDHGlyceraldehyde-3-phosphate dehydrogenase LOC548217 Forward: TGTCCATGCCATGACTGCAA105101
Reverse: CCAGTGCTGCTTGGAATGATG
TUBBβ-tubulin U76897 Forward: CAAGGAGGTGGACGAGCAGATG8497
Reverse: GACTTGACGTTGTTGGGGATCCA
18S rRNA18S ribosomal RNA M82356 Forward: GTGACGGGTGACGGAGAATT15199
Reverse: GACACTAATGCGCCCGGTAT
BYDV cpCoat protein of BYDV EF521849 Forward: GTTGAGTTTAAGTCACACGC29499
Reverse: TGTTGAGGAGTCTACCTATTTG
Table 5. Results of a two-way anova analysis of Barley yellow dwarf virus (BYDV) viral titres in sensitive and tolerant barley varieties
BYDV speciesWbon-123 vs GraciosaVarietyBYDV-PAV vs BYDV-PAS
  1. > 0·05 was considered nonsignificant (ns), < 0·01 as significant (*), < 0·001 as very significant (**) and < 0·0001 as extremely significant (***).

BYDV-PAV***Wbon-123ns
BYDV-PAS***Graciosa***

Comparison of BYDV-PAS and BYDV-PAV for resistance traits

Significant differences were found between BYDV species in all tested traits. In fact, using the BYDV-PAV isolate resulted in higher disease severity and also in a greater reduction in plant height and grain weight per spike. In general, high levels of resistance were recorded in winter barley breeding lines (Wbon-116, Wbon-123) and cultivars (Wysor, Doria) containing the Yd2 gene resistant to both PAV and PAS. The winter wheat breeding line PSR 3628, a hybrid of wheat and couch grass, was highly resistant to both tested BYDV species. Cultivars Svitava, Dromos, Sparta and Banquet were regarded as moderately resistant in the test. It was noted that the use of two species, namely the cultivars carrying the Yd2 gene in barley and the line PSR 3628 in wheat, allowed the detection of the sources of resistance for all the tested traits. The detection of susceptible cultivars/lines of barley and wheat was also confirmed with both BYDV species (Table 6). A significant and positive correlation between symptom scores (VSS), reduction in plant height and reduction in grain weight per spike was detected in both years. The correlation between infections with BYDV-PAS and BYDV-PAV was especially tight for symptom scores (r = 0·83, < 0·001 and 0·91, < 0·001, respectively), but also highly significant for reductions of grain weight per spike and plant height (r ranging from 0·54, < 0·01 to 0·87, < 0·001) (Table 7).

Table 6. Comparison of Barley yellow dwarf virus (BYDV)-PAS and BYDV-PAV for resistance traits of wheat and barley varieties
CropCultivar/lineVSSa (%)PHRb (%)GWSRc (%)Range
PASPAVPASPAVPASPAV
  1. aAverage VSS (visual symptom score recorded on a scale of 0–9, where 0 represents no symptoms).

  2. bPHR: plant height reduction.

  3. cGWSR: reduction of grain weight per spike.

  4. Means in columns followed by the same letter are not significantly different from each other at P = 0·05 in a LSD test.

Winter barleyWbon-116 (Yd2 gene)1·50a2·75ab3·9a16·0a12·6a17·2a1·7
Wbon-123 (Yd2 gene)1·38a2·13a5·3a16·4a23·4ab29·8ab2·3
Doria (Yd2 gene)3·38b3·88b0·0a21·1ab17·4a23·8a3·0
Wysor (Yd2 gene)1·63a2·13a9·0ab19·6a25·4ab17·8a3·2
Sigra5·13c5·88c18·8bc35·3bc23·9ab31·3ab5·3
Finesse4·88c6·25cd25·6c43·1c28·3abc44·2b6·8
Traminer4·75c7·13de18·8bc40·4c36·3bcd65·1c6·8
Perry5·88cd6·13cd21·5c43·2c48·1de62·7c7·7
Luran6·75de7·88ef17·8bc49·4c54·8e73·2cd8·5
Graciosa7·38e8·63f26·8c68·6d44·4cde84·0d9·7
Winter wheatPSR 36280·75a1·17a4·3a1·8a3·0a8·0a1·0
Svitava4·13b5·63bcd6·5a11·9abc35·4bcd48·2b3·8
Dromos3·88b4·90b7·7a22·3bcd40·0cd43·5b4·0
Sparta4·00b4·70b12·5ab11·0ab28·6bc51·7b4·0
Banquet4·13b6·25d10·6a22·0bcd24·0b46·0b4·3
Roane4·38b6·17cd7·4a27·8d37·1bcd57·0b6·3
Sisson5·00bc6·58d11·4ab19·9bcd38·4bcd49·2b6·5
Meritto4·38b5·20bc19·3b28·6d45·3d56·9b7·5
Vlada6·38cd6·30d11·3ab26·1cd41·1d55·8b7·5
SG-S 27037·25d7·92e31·8c47·8e64·1e76·9c10·0
Average4·44a5·40b13·5a27·6b33·4a46·7b 
Table 7. Coefficients of correlation between examined resistance traits and Barley yellow dwarf virus (BYDV) species PAS and PAV across 20 plant genotypes in 2009 (above diagonal) and in 2010 (below)
 VSS-PASVSS-PAVPHR-PASPHR-PAVGWSR-PASGWSR-PAV
  1. VSS: average visual symptom score recorded on a scale of 0–9, where 0 represents no symptoms; PHR: plant height reduction; GWSR: reduction of grain weight per spike.

  2. *, **, *** significant at P = 0·05, P = 0·01 and P = 0·001, respectively.

VSS-PAS 0·914***0·590**0·496*0·763***0·736***
VSS-PAV0·834*** 0·724***0·626**0·805***0·832***
PHR-PAS0·628**0·584** 0·853***0·612**0·685***
PHR-PAV0·898***0·841***0·586** 0·614**0·795***
GWSR-PAS0·608**0·699***0·665***0·541** 0·874***
GWSR-PAV0·669***0·851***0·622**0·647**0·774*** 

Discussion

BYDV-PAS (type isolate PAV-129) was recently separated from PAV (Mayo, 2002), but found to have a worldwide distribution – the USA (Chay et al., 1996; Robertson & French, 2007), France (Mastari et al., 1998), Morocco (Bencharki et al., 1999), New Zealand (Delmiglio, 2008) and the Czech Republic (Kundu, 2008). This study confirms the previous findings that BYDV-PAS is indeed the most prevalent BYDV species found in the Czech Republic (Kundu et al., 2009). BYDV-PAS was the most prevalent species during the 3 years monitored, but was especially prevalent in 2010, when the incidence of BYDV was the lowest of the 3 years. BYDV-PAV is generally accepted as being the most prevalent worldwide (Clement et al., 1986; Miller et al., 1988; Webby et al., 1993; Gourmet et al., 1996; Anderson et al., 1998, 2003; Ayala et al., 2001; Royer et al., 2005) because the most commonly employed method used for detection (DAS-ELISA) is not able to discriminate between BYDV-PAV and BYDV-PAS. Therefore, all the PAS isolates characterized could have been incorrectly considered to belong to BYDV-PAV. Mastari et al. (1998) reported an interesting distribution of the BYDV-PAV and -PAS groups between the host species they examined. Interestingly, 98% of the French isolates found in ryegrass belonged to the PAV group, whereas 83% of the French isolates present in barley contained the PAS virus. Most BYDV monitoring studies usually focus on cereals (rather than grasses generally) as they are economically more important and because plants with symptoms are easier to identify. In the present study, the majority of collected samples were also from cereals and only about 8% were from grasses. However, the three samples of ryegrass included in the study were infected with BYDV-PAV. Interestingly, the distribution of BYDV species in barley samples did not seem to be influenced by the plant species because it matched the general BYDV species distribution – 69% of BYDV-positive samples belonged to BYDV-PAS, 17% were BYDV-MAV and 14% were BYDV-PAV. Furthermore, when all the grasses were taken as one group, the majority of BYDV-positive samples (63%) were infected with BYDV-PAS and only 38% were infected with BYDV-PAV. However, the correlations between host species and virus species cannot be confirmed from this study nor the study of Mastari et al. (1998).

The high incidence of BYDV-PAS could be partly explained by the wide vector spectrum. Although the only aphid species in this study infectious to BYDV-MAV was S. avenae and BYDV-PAV was transmitted by R. padi and S. avenae, BYDV-PAS was transmitted by all four aphid species found on cereals. Peiffer et al. (1997) reported that R. padi, S. graminum, S. avenae and one biotype of M. dirhodum were able to transmit BYDV-PAV when injected with the virus. However, R. maidis did not transmit PAV, even when very high concentrations (200 μg mL−1, 4 ng PAV per aphid) were injected into aphids. The experiment was conducted with the NY-PAV isolate, which is now known to be a true PAV member (Mastari et al., 1998). In the present study, 5% of PAS isolates were transmitted by R. maidis and 9% by M. dirhodum. No PAV isolates transmitted by M. dirhodum were found. Therefore, the transmission of BYDV-PAS seems to be less vector-specific than the transmission of BYDV-PAV. However, the discrepancy in the incidence of the two species is orders of magnitude apart and cannot be explained only by a wider spectrum range of aphid vectors. In biological comparisons of PAV and PAS, Chay et al. (1996) claimed that the efficiency of transmission by R. padi and S. avenae was not significantly different. Therefore, there may be an additional causal factor for the widespread incidence of BYDV-PAS.

Differences in pathogenicity of the BYDV species could also explain the high prevalence of BYDV-PAS. Because the less aggressive species may cause fewer plant deaths, the reservoir of the virus may exponentially increase with time. In the studies presented here, viral titre analysis, together with symptom manifestation, revealed that PAV seemed more harmful than PAS to the sensitive plant variety. Furthermore, PAV caused more severe symptoms and a greater reduction in plant height and grain weight per spike in different genotypes in the field infection trials. These findings are in contrast to the findings of Chay et al. (1996) and Bencharki et al. (1999), who both reported that the PAS caused more severe symptoms when inoculated to a range of oat and barley genotypes. Both of these studies used spring varieties in greenhouses and growth chambers. In contrast, the tests in the present study were carried out on winter varieties in fields, where the effect of wintering also played a role.

There was a significant difference between PAV and PAS in terms of resistance traits. Inoculation with the BYDV-PAV isolate resulted in higher disease severity and also in a greater reduction in plant height and grain weight per spike than the BYDV-PAS isolate. With both species of BYDV (cultivars carrying the Yd2 gene in barley and line PSR 3628 in wheat), the sources of resistance were recognized in all the tested traits. The high level of resistance to BYDV-PAV in PSR 3628 was described recently by Veškrna et al. (2009). The use of a more aggressive BYDV-PAV isolate led to better differentiation between the tested genotypes, especially for visual symptom score and reduction in plant height. The results obtained also suggested that differentiation between genotypes on the basis of their resistance to BYDV could be facilitated by using BYDV-PAV species because of its ability to cause greater disease severity than BYDV-PAS.

From the results of this study, it is concluded that PAS is the most frequently occurring BYDV in this region. A significant difference in virus RNA accumulation between the PAS and PAV species was found. In susceptible plants, the viral titre of PAS was greater than that of PAV, whereas no significant differences were recorded in tolerant plants carrying the Yd2 resistance gene. PAV resulted in more severe symptoms and caused a greater reduction in plant height and yield than PAS. Finally, the level of resistance traits in winter barley and winter wheat was comparable for both PAS and PAV.

Acknowledgements

The authors would like to express their gratitude to Jindra Štolcová, Šárka Bártová, Zuzana Červená, Eva Svobodová, Jana Drabešová and Blanka Jaňourová for their excellent technical assistance. This work was supported by grants from the Ministry of Agriculture, Czech Republic from projects QH81269 (50%) and QH91158 (50%).

Ancillary