• hierarchical ascendant classification;
  • Nicotiana tabacum;
  • pathotype;
  • principal component analysis;
  • Potyvirus;
  • serotype


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Improved tobacco cultivars introgressed with alleles of the recessive resistance va gene have been widely deployed in France to limit agronomical consequences associated with Potato virus Y (PVY) infections. Unfortunately, necrotic symptoms associated with PVY have been reported on these cultivars suggesting that PVY is able to overcome the resistance. A field survey was performed in France in 2007 to (i) estimate the prevalence of PVY in tobacco plants showing symptoms and (ii) characterize PVY isolates present in susceptible and va-derived tobacco cultivars. A serological typing procedure, applied to 556 leaves collected from different French tobacco growing areas, was performed using polyclonal antisera raised against different viral species including PVY. Viral species were detected in 80·8% of leaves and PVY was present in 83·5% of infected samples. However, statistical analysis confirmed that the probability of a tobacco plant being infected with PVY is reduced in va hosts. Eighty-six PVY isolates were mechanically inoculated on one susceptible and three va-derived tobacco cultivars used as indicator hosts to define virulence of these isolates against alleles 0, 1 and 2 of the va gene. Both qualitative and quantitative analyses showed that 55 PVY isolates were able to overcome the three va alleles. Moreover, the monitored biological diversity of PVY isolates was higher in the susceptible tobacco hosts than in the va-derived ones. This study helps to understand consequences of the deployment of the va gene in tobacco on diversity and virulence of PVY isolates.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Viruses are obligate parasites which exploit the cellular machinery of hosts to complete their infection cycle. RNA viruses undergo clonal replication without proofreading mechanisms. Thus, errors introduced by viral polymerases in newly synthesized RNA molecules lead to the diversification of genomic information (Escriu et al., 2007). The molecular polymorphism of the viral genome generated through mutations (Drake et al., 1998) and recombination (Aaziz & Tepfer, 1999) can result in the production of genomic variants with new biological properties (e.g. modification of host range, ability to overcome resistance genes or increased aggressiveness). However, the emergence of such variants in natural populations depends on numerous biotic and abiotic factors linked to virus/host/environment interactions (Garcia-Arenal et al., 2003).

Potato virus Y (PVY, Potyvirus genus) is one of the most variable plant RNA viral species. The filamentous and flexuous PVY particle contains a viral genome which consists of a single stranded positive-sense RNA molecule of about 10 kb in length (Shukla et al., 1994). A VPg protein is covalently attached at the 5’ end of the RNA molecule and a polyadenylated tail at the 3’ end. The viral genome includes one large open reading frame (ORF), which encodes a polyprotein cleaved into nine products by three viral proteases, and a second short ORF (PIPO, (Chung et al., 2008)) embedded within the previously described large ORF. PVY, transmitted in a non-persistent manner by more than 40 aphid species, has a wide host range including cultivated (e.g. potato, tomato, tobacco and pepper) and wild species in the Solanaceae (Singh et al., 2008). PVY is both one of the most economically important plant viruses and one of the most damaging viruses affecting tobacco and potato crops (Blancard, 1998; Valkonen, 2007). Indeed, PVY infections can induce necrotic symptoms on infected tissues (e.g. tobacco leaves) and organs (e.g. potato tubers) that, in addition to yield reduction due to the infected status of the hosts, impact on the quality of the product (Verrier et al., 2001).

The diversity of biological, serological and molecular characteristics of PVY has led to a complex classification of this viral species. The first level of classification groups isolates into strains according to the host they were collected from (Kerlan & Moury, 2008). Potato strain isolates (Blanchard et al., 2008; Rolland et al., 2008) are distributed in three main groups (PVYO, PVYN and PVYC) and in two PVYN subgroups (PVYNTN and PVYN-W) on the basis of (i) their capacity to induce necrotic symptoms on Nicotiana tabacum and to overcome selected resistance genes introgressed in some Solanum tuberosum cultivars, and (ii) their serological and molecular characteristics. In contrast to the potato strain isolates, and despite the economical and agronomical impacts of PVY on tobacco production, studies of the tobacco/PVY pathosystem are scarce. According to the comparison of nucleotide and amino acid identity of CP sequences, PVY isolates collected in Chinese tobacco fields were classified in three groups, together with reference PVYN, PVYNTN, or PVYO/PVYN-W potato isolates (Li et al., 2006). Phylogenetic analyses performed using either CP or full length genome sequences of PVY isolates collected on different solanaceous hosts (Blanco-Urgoiti et al., 1998; Lorenzen et al., 2006; Feki & Bouslama, 2008; Moury, 2010) clusters potato PVY isolates in three groups named O, N and C2, corresponding to PVYO, PVYN and PVYC isolates, whereas non-potato isolates, including isolates collected on tobacco and pepper, constitute a fourth group, C1. Nevertheless, the scientific literature on the tobacco/PVY pathosystem is mainly focused on the interaction between tobacco strain PVY isolates and hosts (Blancard et al., 1995; Verrier & Doroszewska, 2004). Thus, tobacco strain PVY isolates have been classified, on the basis of mosaic (M) and necrotic (N) symptoms induced on tobacco hosts known to be susceptible (S) or resistant (R) to the root-knot nematode Meloidogyne incognita (Gooding & Tolin, 1973), in three groups corresponding to MSMR, MSNR and NSNR. Indeed, the necrotic response to PVY infection and the resistance to M. incognita have been found to be pleiotropic effects of the Rk gene (Rufty et al., 1983). However, this Rk-based indirect viral typing is inappropriate to describe the diversity of biological properties of PVY populations in the current agronomical context.

Due to the lack of efficient methods to directly protect tobacco plants from PVY infections or to control aphid-mediated non-persistent viral transmissions, genetic resistance against PVY has been used to limit impacts of PVY in tobacco fields. Thus, tobacco cultivars with the recessive resistance va gene, which represents the most reported genetic resistance source against PVY in this host species, have been intensively deployed in tobacco fields in France for three decades (Blancard, 1998). According to the amount of necrotic symptoms induced by a range of PVY isolates on different N. tabacum cultivars, including VAM, Perevi, VD, Paraguay 48, Havana 307, Havana IIC and Polalta, three levels of resistance have been described, from high (0), moderate (1) to low (2). F1 hybrids resulting from the cross between these different cultivars are all resistant and allelic tests have shown that the resistance in these cultivars is controlled by a single recessive gene at the same locus (Yamamoto, 1992; Ano et al., 1995; Blancard et al., 1995). In France, the allelic form 0 present in VAM has been introgressed in burley types whereas the allelic form 2 of the va gene, initially present in VD and Paraguay 48 genotypes, was used in flue cured and dark air cured breeding programmes (J-L. Verrier, Imperial Tobacco Group, France, personal communication). However, the va gene does not stop viral infection of plants, but limits cell to cell movement of viral particles during the infection of the host (Acosta-Leal & Xiong, 2008) and reduces the development of vein necrosis symptoms induced by PVY on tobacco leaves (Verrier & Doroszewska, 2004). The use of genetic resistance could induce selection pressures on the pathogen which can lead to the emergence of variants putatively more virulent and/or aggressive than the parental viral entities (Pelham et al., 1970; Fargette et al., 2002; Chain et al., 2007). Virulence refers here to the genetic ability of a pathogen to overcome a genetically determined host resistance and cause a compatible interaction (Shaner et al., 1992). Viral adaptations to the PVY resistance va gene have been already published (Latorre & Flores,1985; Piccirillo & Piro, 1986). However, information about the impact of alleles of the va gene already deployed on PVY epidemiology is not yet available.

The objective of this study was to describe the biological properties of PVY isolates present in tobacco fields at a national scale in order to analyse the serological and biological diversities of PVY present in susceptible and va-derived resistant tobacco cultivars. This will help to test the impact of va resistance gene on the virulence of tobacco strain PVY isolates.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Sampling of tobacco plants

During a 4 week period (from 16 July to 9 August 2007), 556 leaf samples were collected from individual susceptible and PVY-resistant (va gene) tobacco plants. The collected plants were distributed in 108 fields from the three main French tobacco growing areas defined as north-west (resistant cultivars; 30 fields; 128 leaves), north-east (resistant cultivars; 28 fields; 166 leaves), and south (resistant cultivars; 24 fields; 127 leaves and susceptible cultivars; 26 fields; 135 leaves). Tobacco plants sampled during this field survey belong to flue-cured, burley and dark air-cured types. In addition to leaves with necrotic and/or mosaic symptoms close to or corresponding to those known to be induced by PVY isolates on tobacco, both symptomless plants and plants with unconventional symptoms were collected. Sampled leaves were individually stored at −20°C until used.

Serological identification of viral pathogens in tobacco leaves

The presence of PVY, Cucumber mosaic virus (CMV), Alfalfa mosaic virus (AMV) and Tobacco mosaic virus (TMV) in sampled tobacco leaves was determined using enzyme-linked immunosorbent assays (ELISA) (Clark & Adams, 1977).

Polyclonal antisera raised against PVY (INRA Rennes/FNPPPT), CMV (INRA Avignon) and AMV (LCA Bordeaux), and monoclonal antibodies anti-PVYO/C (Neogen) and anti-PVYN (INRA Rennes/FNPPPT) isolates were used in double antibody sandwich ELISA procedures. Each serological reagent was diluted in carbonate buffer (15 mm Na2CO3, 35 mm NaHCO3, pH 9·6) according to the recommendations of the providers. Wells of microtitre plates (NUNC, Maxisorp) were coated for 4 h at 37°C with 100 μL of diluted polyclonal antiserum. Plates were washed three times with PBS buffer (137 mm NaCl, 8 mm Na2HPO4.12H2O, 2·7 mm KCl, 1·5 mm KH2PO4, pH 7·4) supplemented with 0·05% (v/v) Tween 20 (PBS-T buffer) between each step of the ELISA protocol. Leaf samples were ground in the presence of PBS-T buffer supplemented with 2% (w/v) polyvinylpyrrolidone 40T (grinding buffer). Ninety five microlitres of the plant sap were then incubated in the coated wells overnight at 4°C. According to the expected specificity of the detection assay, polyclonal or monoclonal antisera conjugated to alkaline phosphatase were used. Thus, 90 μL of appropriate antiserum, diluted according to manufacturers’ recommendations in grinding buffer supplemented with 0·2% (w/v) ovalbumine (conjugate buffer), were added in wells.

The detection of TMV was carried out using a triple antibody sandwich ELISA protocol which includes the use of primary and secondary antibodies (antiserum provided by CNRS, Strasbourg) and two additional steps. After the coating step, the plate wells were saturated with 95 μL of gelatin 1% (w/v) diluted in PBS buffer. After 1 h incubation at 37°C with 85 μL of the secondary antibody anti-TMV, wells were filled with 80 μL of an alkaline phosphatase conjugated anti-mouse antibody (Sigma) diluted in conjugate buffer, according to manufacturer’s recommendations.

After 4 h incubation at 37°C in the presence of the alkaline phosphatase conjugated antiserum, plate wells were washed and filled with 85 μL of p-nitrophenyl-phosphate (1 mg mL−1) diluted in substrate buffer (10 mm, diethanolamine, pH 9·8). Following incubation at room temperature in the dark for 1–2 h, the absorbance at 405 nm was recorded for each well using a micro-plate reader (Titertek Multiscan [MCC]). A positive detection was considered when the OD405 value of the tested sample was twofold greater than the OD405 value of the healthy control.

Tobacco cultivars used in the experiments

The PVY-susceptible N. tabacum cv. Xanthi was used to produce viral inoculum for biological characterization experiments. Healthy and infected plants were maintained in separate thermo-regulated insect proof greenhouses (18/25°C night/day). Biological characterization of PVY isolates was performed using four N. tabacum cultivars. MN944 was used as PVY-susceptible control whereas cvs VAM, Wislica and PBD6 correspond to the PVY-resistant hosts with allele 0, 1 and 2 of the recessive resistance va gene, respectively (Ano et al., 1995; Julio et al., 2006). VAM (flue-cured) was obtained from mutagenesis with X-ray irradiation of Virgin A cultivar (Koelle, 1958). Wislica (flue-cured) resulted from varietal selection performed using an oriental tobacco cultivar (Verrier & Doroszewska, 2004). PBD6 (dark air-cured) was obtained from the cross between the P48 and the Bel 61-10 tobacco lines (Schiltz, 1967). Healthy and inoculated plants were grown and maintained under controlled conditions in a growth chamber (16 h day, 20°C, 60–70% humidity; 8 h night, 18°C, 85–98% humidity).

Biological characterization of PVY isolates

To produce the inoculum required for biological characterization of PVY isolates, N. tabacum cv. Xanthi plants were individually inoculated with PVY-infected leaves from field surveys. Each collected tobacco leaf serologically identified as being infected by PVY was ground in a mortar in the presence of 3 mL of inoculation buffer (50 mm Na2HPO4.12H2O, 50 mm KH2PO4, 40 mm sodium diethyldithiocarbamate, pH 7·2). The produced plant sap was mixed with 1·5 g activated charcoal and 2·5 g carborundum. The mixture was rubbed onto two newly expanded leaves of healthy N. tabacum cv. Xanthi plants. Non-inoculated leaves from infected plants were sampled 15 days post-inoculation, ground in a mortar in the presence of liquid nitrogen. Ground material was then aliquoted in fractions and stored at −80°C until used. Each fraction was used to inoculate four plants of each of the N. tabacum MN944, VAM, Wislica and PBD6 cultivars. Fractions were individually mixed with 4 mL of inoculation buffer. For each plant inoculated, 100 μL of the prepared sap was rubbed on each of two newly expanded leaves previously dusted with 0·1 g active charcoal and 0·22 g carborundum. Systemic PVY infection of MN944, VAM, Wislica and PBD6 plants was tested 30 days post-inoculation using non-inoculated leaves with the PVY-specific ELISA protocol described above.

Statistical analysis

Statistical analysis was performed using the software r (R Development Core Team, 2005) by means of Generalized Linear Models (McCullagh & Nelder, 1989) assuming a binomial distribution and a logit link function. Pairwise comparisons (contrasts analysis) were carried out with the function ‘esticon’ in the ‘doby’ package (Author: Søren Højsgaard). The impact of the allelic forms of the va gene on the virulence of PVY isolates was analysed with a principal component analysis (PCA) using the Pearson correlation coefficient (Joliffe & Morgan, 1992) and a hierarchical ascendant classification (HAC) using Euclidian distance as a measure of dissimilarity and Ward aggregative method (Arabie et al., 1998). Finally, chi-squared, binomial tests and z tests were performed with xlstat software (2009, Addinsoft).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Characterization of collected tobacco leaves

During field surveys performed in France in 2007, 556 leaf samples were collected on flue cured, burley and dark air cured tobacco plants. The number of collected leaves for each area/host combination ranged from 0 to 150. The sampling was performed according to the heterogeneous distribution of the cultivated tobacco types in France and reflects the absence in the north of the country of both PVY-susceptible tobacco cultivars and dark air cured tobacco. Most of the collected leaves expressed necrosis, mosaic and/or unconventional symptoms (N = 471) at the time they were collected. However, 85 symptomless plants were also sampled to complete the survey. Each collected sample was tested by ELISA for the specific detection of AMV, CMV, TMV and PVY. The percentage of infected leaves for sampling areas (north-west, north-east and south), susceptible (S) and PVY-resistant (R) tobacco plants, and absence (Sym−) and presence (Sym+) of symptoms on tested leaves are presented in Table 1. None of the tested viruses was detected in 19·2% (107/556) of tested leaves whereas one or a combination of several viruses was detected in 80·8% (449/556) of samples. The percentage of infected samples ranged from 71·9 to 83·7% in the north and from 78 to 88·1% in the south. Most of the infected and healthy (according to the list of viruses tested by the ELISA procedures) samples were collected on plants with (90%, 424/471) and without (70·6%, 60/85) symptoms, respectively. However, infected samples were also detected in 29·4% (25/85) of symptomless plants and 10% (47/471) of leaves showing symptoms were considered ‘healthy’ under these experimental conditions.

Table 1.   Percentages of infected Nicotiana tabacum leaves with (Sym+) or without (Sym−) symptoms according to the geographical and host origins of samples
Origin of samplesInfected samples
  1. aR: resistant tobacco cultivars; S: susceptible tobacco cultivars.

  2. bNumber of infected plants/number of tested plants.

North-WestR29·0 (9/31)b85·6 (83/97)71·9 (92/128)
North-EastR17·4 (4/23)94·4 (135/143)83·7 (139/166)
SouthR47·7 (8/17)82·7 (91/110)78·0 (99/127)
S28·6 (4/14)95·0 (115/121)88·1 (119/135)
Total 29·4 (25/85)90·0 (424/471)80·8 (449/556)

Infected samples were assigned to two groups corresponding to (i) samples associated with positive detection for at least one virus except PVY (‘other viruses’ group, 16·5%, 74/449), and (ii) samples infected by PVY (‘PVY’ group, 83·5%, 375/449). PVY was detected in 77·7% (366/471) and 10·6% (9/85) of plants with and without symptoms, respectively, whereas at least one of the other viruses tested was present in 12·3% (58/471) and 18·8% (16/85) of plants with and without symptoms, respectively. Moreover, the proportions of samples infected with PVY (PVY group) or with AMV, CMV and TMV (other viruses group) differed according to the region and the susceptibility of the hosts (Table 2). In the PVY group, 326 samples (86·9%) were diagnosed by the procedures used as singly infected by PVY whereas 49 samples (13·1%) corresponded to mixed infections with PVY and at least one of the other three tested viruses. Mixed infection between PVY and either AMV, CMV or TMV were detected in 4% (18/449), 2·9% (13/449) and 3·6% (16/449) of infected samples, respectively (Table 3). Most of the PVY mixed infection samples with CMV or TMV were detected in resistant va2 flue cured tobacco in the south (31·8%, 7/22) and in va0 burley cultivars in the north-east (28·6%, 4/14), respectively.

Table 2.   Percentages of Nicotiana tabacum leaves infected by Potato virus Y (PVY), Alfalfa mosaic virus(AMV), Cucumber mosaic virus (CMV) and Tobacco mosaic virus (TMV) according to the geographical and host origins of samples
Origin of samplesTargeted viral species
  1. aR: resistant tobacco cultivars; S: susceptible tobacco cultivars.

  2. bDue to mixed infected samples, sum of percentages for each sampled area is above 100%.

Table 3.   Percentages of Nicotiana tabacum leaves infected by Potato virus Y (PVY) in single and/or mixed infection with other targeted viral species according to the geographical and host origins of samples
Origin of samplesType of infection
  1. aB va0, F va2 and D va2 : resistant burley (B), flue cured (F) and dark air cured (D) tobacco cultivars possessing allele 0 or 2 of the va gene; S: susceptible tobacco cultivars.

North-WestB va078·60·00·00·00·00·078·6
F va282·10·00·01·32·60·086·0
North-EastB va050·00·00·00·028·60·078·6
F va280·06·40·80·08·00·095·2
SouthB va035·38·80·02·90·00·047·0
F va245·50·00·031·80·00·077·3
D va233·311·10·011·10·011·166·6

Finally, samples belonging to the PVY group were analysed with PVY monoclonal antibodies raised against PVYN (YN) and PVYO/C (YO/C) serotypes (Table 4). Resulting data assigned 73·6 and 10·7% of PVY isolates in PVYN and PVYO/C serogroups, respectively. In addition, 2·7 and 13·1% of tested samples were efficiently (YON group) or not (YU group) detected by the two specific PVYN and PVYO/C antibodies. The YN isolates members were prevalent in the PVY infected samples from resistant cultivars in the north (82 and 80·8% of YN group members in north-west and north-east, respectively). Distribution in serogroups of isolates collected in the south was different, as 43·7, 23·6 and 32·7% of PVY isolates detected in leaves from resistant hosts sampled in the south were assigned to YN, YO/C and YU serogroups, respectively (Table 4).

Table 4.   Percentages of Potato virus Y (PVY) samples of the different PVY serogroups
Origin of samplesPVY serogroupb
  1. aR: resistant tobacco cultivars; S: susceptible tobacco cultivars.

  2. bPVY serogroups were defined according to detection assays performed using monoclonal antibodies raised against PVYN (INRA Rennes/FNPPPT) and PVYO/C (Neogen). YO/C, YN, YON and YU serogroups were defined according to specific detection of isolates by YO/C, YN, both YO/C and YN, and neither of these two monoclonal antibodies, respectively.


Analysis of infected samples

The generalized linear model was used to test the impact of parameters associated with the samples such as (i) area (north-west, north-east and south), (ii) type of tobacco (flue cured, burley and dark air cured) and (iii) PVY-resistance (va gene) on both the sanitary status of tested samples and the assignment of infected ones in the previously defined ‘other viruses’, PVY, YO/C, YN and YU groups. As the YON serogroup members represented only 2·7% (10/375) of PVY-infected samples, they were not included in the statistical analysis. Among all tested variables, the non-infected samples were significantly more frequent in resistant va tobacco hosts than in susceptible ones (= 0·02). For infected samples, the assignment of isolates to ‘other viruses’ and to PVY groups was affected by both the area (= 0·01 and < 0·01, respectively) and the type of tobacco (= 0·02 and < 0·01, respectively). Pairwise comparisons (contrasts analysis) were performed between modalities of each of these two factors. The proportion of samples assigned in the ‘other viruses’ group was not statistically different between north-west and north-east, flue cured and dark air cured, and between burley and dark air-cured but was higher in the south compared both to the north-west (= 0·02) and to the north-east (< 0·01). This variable was also higher in burley when compared with flue cured (< 0·01), and in resistant tobacco lines when compared with susceptible ones (< 0·01). Similar analyses performed for the proportion of isolates assigned to the PVY group showed significantly higher values i) in the north-east than in both the north-west (= 0·015) and south (< 0·01), ii) in dark air cured than in burley (< 0·01), (iii) in flue cured than in burley (< 0·01) and (iv) in PVY-susceptible tobacco lines than in resistant ones (< 0·01). For PVY-infected samples, the probability of being associated with serotype-N and -O was significantly affected by the area (= 0·01 and < 0·01, respectively). The percentage of YO/C isolates was significantly higher in the south than in the north-east (= 0·013) and north-west (< 0·01), whereas the percentage of YN isolates was higher in the north-east (< 0·01) and north-west (= 0·019) when compared to the south (Table 4). No significant difference was obtained for the percentages of YO/C and YN isolates observed between north-east and north-west. Finally, the percentage of YU isolates was affected by the va gene, as the presence of this type of PVY isolate was more frequent in resistant tobacco plants than in susceptible ones (< 0·01).

Biological characterization of PVY isolates

From the 375 PVY infected samples, 86 isolates, not co-infected with one of the other tested viral species (AMV, CMV and TMV), were selected according to their geographical (19 from north-east, 18 from north-west, 49 from south) and host (43 from PVY-susceptible and 43 from PVY-resistant plants) origins. The virulence of these selected PVY isolates was analysed by mechanical inoculation of four tobacco cultivars (four plants per cultivar; two replicates) bearing none or one of the alleles (0, 1 or 2) of the va gene. According to the ELISA results obtained from inoculated plants of each tobacco cultivar, eight pathotypes can be defined. Pathotype VA describes PVY isolates with an infection pattern restricted to the susceptible cultivar. Pathotypes 0, 1, 2, 0-1, 0-2, 1-2 and 0-1-2 define isolates able to infect, in addition to the susceptible MN944 cultivar used as infectivity control in the experiment, one, two or three of the resistant VAM (va allele 0), Wislica (va allele 1) and PBD6 (va allele 2) hosts. The 86 PVY isolates tested were distributed in the eight pathotypes (Table 5). Most isolates were assigned to pathotypes 1-2 (N = 16) and 0-1-2 (N = 55). Isolates described as members of pathotypes VA, 2, 1, 0-2 and 0-1 were collected only on susceptible tobacco hosts. PVY isolates assigned to pathotypes 0, 1-2 and 0-1-2 were collected on both va2 and va0 resistant (N = 1, N = 1, and N = 41, respectively) and susceptible (N = 3, N = 15, and N = 14, respectively) tobacco plants. Finally, analysis of the serological characteristics of isolates assigned to the pathotype 0-1-2 showed that the proportions of YO/C (12/19) and YN (42/66) isolates were not statistically different (= 0·97).

Table 5.   Characteristics and biological properties of 86 Potato virus Y (PVY) isolates collected on tobacco plants in different
IsolateOriginSerogroupbTobacco genotypescPathotypeHAC group
  1. aB va0, F va2 and D va2 : resistant burley (B), flue cured (F) and dark air cured (D) tobacco cultivars possessing allele 0 or 2 of the va gene; S: susceptible tobacco cultivars.

  2. bPVY serogroups were defined according to detection assays performed using monoclonal antibodies raised against PVYN (INRA Rennes/FNPPPT) and PVYO/C (Neogen). YO/C, YN, YON and YU serogroups were defined according to specific detection of isolates by YO/C, YN, both YO/C and YN, and neither of these two monoclonal antibodies, respectively.

  3. cMN944 is a susceptible host; VAM, Wislica and PBD6 present a PVY-resistant host phenotype due to the presence of allele 0, 1 and 2, respectively, in their genotype.

239 Ob3North-EastF va2O71780–1–22
240 501Ob1North-EastB va0N88880–1–23
241 221Gr1North-EastB va0N88880–1–23
241 667Gr1North-EastF va2N88880–1–23
242 Pf1North-EastF va2N88880–1–23
244 Nie2North-EastF va2N74440–1–22
245 Sgv5North-EastF va2N88880–1–23
246 501Mam1North-EastB va0N87670–1–22
248 Iss7North-EastF va2N88880–1–23
250 620Loi2North-EastF va2N76670–1–22
252 661Mh3North-EastF va2O51320–1–22
255 30804Lap2North-EastF va2O88880–1–23
255 30804Lap4North-EastF va2N88880–1–23
256 623Nlc3North-EastF va2N44440–1–22
256 623Nlc5North-EastF va2O74880–1–23
259 30804MLe4North-EastF va2O88880–1–23
259 661MLe1North-EastF va2O83660–1–22
259 683MLe2North-EastF va2N86870–1–23
259 683MLe3North-EastF va2O74870–1–23
260 Pal6North-WestF va2N78780–1–23
268 Ite1North-WestF va2N88680–1–23
269 Csh2North-WestF va2N88880–1–23
270 Bru5North-WestB va0N88870–1–23
270 Bru7North-WestB va0N68870–1–23
271 Cha2North-WestF va2N88880–1–23
272 Ter2North-WestF va2N81760–1–22
273 273Ter2North-WestF va2N88880–1–23
278 Yze1North-WestF va2N88880–1–23
281 Val4North-WestF va2N87880–1–23
283 623Jal1North-WestF va2u84460–1–22
283 623Jal3North-WestF va2N88880–1–23
285 667Bri5North-WestF va2N88880–1–23
285 Bri3North-WestB va0N88880–1–23
287 221Ssl1North-WestB va0N78780–1–23
287 Ssl3North-WestF va2N88880–1–23
288 Cug9North-WestF va2N68870–1–23
289 Bro1North-WestF va2N88880–1–23
209 TH8SouthB va0N88680–1–23
210 TH1SouthB va0N88780–1–23
211 1000SA2FBSouthD va2N88780–1–23
228 PDA6SouthB va0O82330–1–22
212 La8SouthSO82340–1–22
213 MaCsa7SouthSN82350–1–22
214 31612Csa8SouthSO83680–1–22
215 MaLf4SouthSN82740–1–22
221 MaSan1SouthSN81330–1–22
222 Sav7SouthSN81540–1–22
224 Mi1SouthSN82450–1–22
224 Mi2SouthSO81440–1–22
224 Mi3SouthSN81420–1–22
226 K171Fa3SouthSN82330–1–22
229 Da1SouthSO85650–1–22
233 31612Beau3SouthSN81220–1–22
233 31612Beau5SouthSN82510–1–22
236 Bar1SouthSN81220–1–22
201 6SouthSO81100–11
212 La4SouthSO82200–11
213 MaCsa1SouthSN82100–11
214 31612Csa6SouthSN81400–11
225 Fa2SouthSN81300–11
236 Bar2SouthSN81400–11
212 La7SouthSN81020–21
228 PDA2SouthBva0N80311–21
206 BR206-4SouthSN80231–21
212 La6SouthSO80441–22
213 BB16ACsa3SouthSN80441–22
214 31612Csa1SouthSN80451–22
214 31612Csa2SouthSN80421–22
214 31612Csa5SouthSN80211–21
214 31612Csa7SouthSN80341–22
215 MaLf7SouthSN80551–22
217 MaSan1SouthSN80321–21
222 Sav8SouthSO80761–22
223 An5SouthSN80221–21
232 Mo2SouthSN80431–22
233 31612Beau4SouthSN80221–21
237 BB16ASan1SouthSN80241–22
238 BB16APS2SouthSN80231–21
233 623Beau1SouthFva2N610001
198 SN11SouthSN810001
226 K171Fa4SouthSN810001
232 Mo1SouthSN810001
221 MaSan4SouthSN801011
230 Vi5SouthSO800121
198 SN3SouthSO6000VA1
213 MaCsa5SouthSO6000VA1

Analysis of the virulence of PVY isolates

To test the possible impact of the allelic forms of the recessive resistance va gene on the virulence of PVY isolates, a principal component analysis was performed on arcsine-root transformed proportions of infected plants per tobacco cultivar for each of the 86 tested PVY isolates. The number of infected plants for each resistant cultivar/isolate combination ranged from 0 to 8 (Table 5). The assignment of an isolate to a pathotype was made according to its capacity to infect the different resistant cultivars. However, the number of infected plants per cultivar differed between isolates belonging to a specific pathotype. Thus, the PCA analysis was performed on the number of infected plants for each isolate (observations) and alleles of the va gene present in VAM, Wislica and PBD6 cultivars (initial variables). The results of the PCA procedure showed that axis F1, F2 and F3 explained 88·3, 7·7 and 4% of the variability contained in the initial data set, respectively. Thus, 96% of the observed variability was illustrated along the F1 and F2 axis (Fig. 1a). Each of the va0, va1 and va2 variables was associated with the F1 axis (cos2 = 0·842, cos2 = 0·903 and cos2 = 0·904, respectively). Moreover, the absolute contribution of va0, va1 and va2 to the F1 axis was 31·8, 34·1 and 34·1%, respectively. A hierarchical ascendant classification was carried out on the coordinates of isolates along the PCA F1 axis to group PVY isolates sharing similar biological properties. Data associated with isolates can be illustrated on the F1/F2 axes with the observations distributed along the F1 axis (Fig. 1b). The HAC analysis allowed the identification of three groups (Fig. 1b, Table 5). For each HAC group, the variables associated with the geographical (north/south) and biological (susceptible/resistance) origins of PVY isolates were compared using both binomial and chi-square tests (Table 6). Binomial tests showed that isolates were mainly associated with south/susceptible tobacco hosts in HAC groups 1 and 2 (< 0·01) and with north/resistant tobacco hosts (< 0·01) in HAC group 3. Taking into account the number of PVY isolates from south and north, chi-squared tests indicated that HAC groups are different from each other (groups 1–2, = 0·02; groups 1–3, < 0·01; groups 2–3, < 0·01). Finally, chi-squared tests also revealed that HAC groups 1 and 3 (< 0·01), and HAC groups 2 and 3 (< 0·01) were significantly different concerning the number of PVY isolates collected on susceptible and on resistant plants, whereas HAC groups 1 and 2 cannot be considered different for this parameter (= 0·13).


Figure 1.  Schematic representation of principal component analysis (PCA) and hierarchical ascendant classification (HAC) data. (a) Coordinates of initial variables (alleles va0, va1 and va2) and (b) associated observations (number of infected plants for each PVY isolate). The data are presented on a two axes graph constituted by principal components F1 and F2 which illustrate 88·3 and 7·7% of data set variability, respectively. The groups of PVY isolates (1, 2 and 3, graph b) proposed by the hierarchical ascendant classification analysis are presented.

Download figure to PowerPoint

Table 6.   Geographical and host origins of Potato virus Y (PVY) isolates assigned to hierarchical ascendant classification (HAC) groups 1 to 3
Origin of samplesHAC groupTotal
  1. aR: resistant tobacco tobacco cultivars; S: susceptible cultivars.



  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Consequences of the deployment of va tobacco genotypes on the virulence of PVY isolates are not well documented. The objectives of the presented field survey were (i) to estimate the prevalence of PVY on resistant cultivars showing symptoms, and (ii) to characterize the serological and biological diversity of PVY isolates present in susceptible and resistant tobacco crops in France. The collected leaves were tested for the presence of the four most important tobacco viruses reported in France (PVY, AMV, CMV and TMV). The resulting data set revealed that PVY can be considered as the most prevalent virus. However, the detection frequencies do not represent the global epidemiological situation for viruses in French tobacco fields. To estimate more accurately the epidemiology of viruses in tobacco, which was not the aim of this work, a random sampling of leaves would have to be performed. However, characteristics of the performed survey allowed the analysis of the impact of several variables on the proportion of viral species detected in tobacco samples. Statistical analyses showed that the presence of the va gene in a tobacco cultivar was associated with a reduced probability of being infected by PVY and an increased probability of being infected by at least one of the other viruses. These results are consistent with the role of the va gene in the limitation of PVY spread but could also suggest that va hosts are more susceptible to other virus infections than VA hosts. However, statistics were mainly supported by the fact that PVY was responsible for almost all observed symptoms in north and in susceptible/south plants, whereas in south/resistant, samples with symptoms were infected by PVY, AMV and CMV. The main hosts of CMV and AMV, cucurbits, tomatoes and peppers are widely cultivated in the south of France and could constitute efficient viral sources for tobacco infections (Jaspars & Bos, 1980; Morroni et al., 2008). Moreover, the va gene has been introgressed in tobacco to reduce PVY necrotic symptoms. Thus, the observation of necrosis in resistant cultivars reflects the presence of PVY isolates able to overcome the resistance, or of other necrotic pathogens. In consequence, the contrasted results obtained in the south for resistant and susceptible tobacco strongly suggest that the va gene reduces the frequency of PVY necrotic symptoms in resistant cultivars making possible the diagnosis, in addition to PVY, of other pathogens.

To describe factors involved in the prevalence of PVY isolates in tobacco plants, tobacco types and sampled areas were used as variables for statistical analyses. Percentages of PVY-infected samples were significantly higher in resistant va2 flue cured and dark air cured than in va0 burley. This is consistent with the fact that va0 burley cultivars were previously reported to exhibit necrosis less frequently than other types of va tobacco cultivars (Verrier & Doroszewska, 2004). Moreover, the prevalence of the PVYN isolates observed in tobacco (this study) is also found in potato crops, mainly grown in the north/north-west of France (Blanchard et al., 2008). Nevertheless, when values associated with serogroups were taken separately, statistical analysis showed that YO/C and YN isolates were more prevalent in samples collected in the south and north, respectively. As tomato and pepper, known to be efficient hosts for YO/C isolates (Soto et al., 1994; Kerlan & Moury, 2008), are mainly cultivated in the south, these two plant species could constitute efficient reservoirs for PVYO/C isolates. In addition, as susceptible tobacco genotypes are cultivated only in the south, the geographical distribution of non necrotic YO/C (south) and necrotic YN (north) PVY isolates could also be underlined by a susceptible and resistant host selection pressure, respectively. Indeed, a recently published work (Rolland et al., 2009) has shown that the necrotic property of PVY is associated with a fitness cost in Nicotiana hosts which responds with necrotic symptoms to PVY infection. Even if the link between serological and biological properties of PVY isolates is known to be weak (Chrzanowska, 1991; Lorenzen et al., 2008), most of the PVYN serogroup members are able to induce necrosis on tobacco. Thus resistant cultivars, that reduce necrotic symptom expression, can be involved in the process that has led to the prevalence of necrotic PVY isolates in Nicotiana hosts. To strengthen this conclusion, the characteristics of PVY isolates in an environment free from resistant tobacco cultivars needs to be described. Finally, statistical analysis revealed that the probability of being infected with YU isolates was higher in va resistant lines than in susceptible ones. It would be interesting to characterize further these original isolates to know if they effectively correspond to isolates with new properties and to determine the impact of the va resistant line in the emergence and selection of YU isolates.

Biological properties of 86 PVY isolates collected on susceptible or resistant cultivars were assessed using four Nicotiana hosts, MN944, VAM, Wislica and PBD6, with either none or one allele (0, 1 or 2) of the va gene, respectively. Indeed, this ‘va-based’ description allowed assessment of the link between the diversity of biological characteristics of PVY isolates and the presence of the va gene in tobacco host. The biological typing made it possible to assign isolates to VA, 0, 1, 2, 0-1, 0-2, 1-2 and 0-1-2 pathotypes. PVY isolates able to systemically infect N. tabacum VAM hosts have already been described in the literature (Latorre & Flores, 1985; Piccirillo & Piro, 1986) and named VAM-B (VAM-breaking). They correspond to pathotype 0 isolates described in this work. Raw data showed that the eight expected pathotypes were observed. Most of the isolates, mainly collected on va0 and va2 resistant cultivars, were assigned to pathotype 0-1-2. In fact, the biological diversity of PVY isolates observed in resistant tobacco was limited to pathotypes 0-1-2, 0 and 1-2, whereas the eight pathotypes were described in susceptible hosts. The ‘va environment’ seems to impose selection pressure(s) that resulted in the reduction of the diversity of PVY in favour of pathotype 0-1-2 members. Such virulence selection phenomenon has been described since the 1970s for different models in plant pathology (Pelham et al., 1970; Pferdmenges et al., 2009). Moreover, the proportion of isolates in pathotypes 0, 1 and 2, and 0-1, 0-2, 1-2 and 0-1-2 suggests that the acquisition of virulence against one of the va alleles could occur independently and with equivalent probabilities ( 0·17) but seems to be increased for isolates already virulent against one or two alleles. This cooperative phenomenon seems to be more important for interactions between alleles 1 and 2 as denoted by the number of pathotype 1–2 isolates. The genetic variation required for modification of the pathogenicity of PVY in the va pathosystem is not yet known but certainly impacts on the emergence of corresponding virulent isolates and needs to be determined in order to further the analysis of PVY/va gene interactions.

The qualitative analyses of the results were completed by a quantitative approach using the number of infected plants for each host cultivar/isolate combination. Isolates collected from field sampling must be considered as populations of sequences resulting from multiple inoculations of the host and/or mutations introduced in the inoculated sequence(s) during the infection process (Domingo, 2002). Depending on the frequency of sequences with genetic determinant(s) of virulence in each field-sampled population and on the characteristics of mechanical inoculation, isolates may not be able to infect all inoculated PVY resistant plants. The analyses of quantitative variables has improved knowledge on the involvement of alleles of the va gene on the selection of virulent isolates. Both PCA and HAC procedures allowed isolates to be assigned in three groups from low to high virulence along the F1 PCA axis, which was equivalently explained by the three va alleles used as initial variables. This HAC grouping is dependent of environmental parameters since low virulent populations were collected in susceptible tobacco in the south and high virulent ones were from resistant tobacco sampled in the north. This contrasted distribution of PVY isolates in HAC groups suggests that the emergence of virulent PVY isolates is associated with a va resistant host selection pressure.

In conclusion, data collected during a field survey performed in tobacco crops in France in 2007 showed that PVY was prevalent in plants with symptoms but that the probability of being infected with this virus was reduced in va hosts. However, both qualitative and quantitative analysis of infection values associated with resistant cultivars indicated that the presence of va alleles in tobacco cultivars was associated with (i) a reduction in the biological diversity of PVY isolates and (ii) a high frequency of virulent PVY isolate members of the 0-1-2 pathotype. Thus, the deployment of the va gene in French tobacco crops seems to play a role in the prevalence of virulent PVY isolates in the field.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We are grateful to Drs Yannick Outreman and Luc Madec for statistical analysis, to Dr Thomas Baldwin for critical reading of the manuscript, and to Michel Tribodet for technical support. The work presented has been supported by the Institut National de la Recherche Agronomique (France), Imperial Tobacco Group and the Association for Research for Nicotianae.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Aaziz R, Tepfer M, 1999. Recombination between genomic RNAs of two cucumoviruses under conditions of minimal selection pressure. Virology 263, 2829.
  • Acosta-Leal R, Xiong ZG, 2008. Complementary functions of two recessive R-genes determine resistance durability of tobacco ‘Virgin A Mutant’ (VAM) to Potato virus Y. Virology 379, 27583.
  • Ano G, Blancard D, Cailleteau B, 1995. Mise au point sur la résistance récessive aux souches nécrotiques du virus Y de la pomme de terre (PVY) présente chez Nicotiana tabacum. Annales du Tabac 27, 3542.
  • Arabie P, Hubert LJ, De Soete G, 1998. Clustering and classification. Journal of Classification 15, 1513.
  • Blancard D, 1998. Maladies du tabac. Observer, identifier, lutter. Paris, France: INRA Editions.
  • Blancard D, Ano G, Cailleteau B, 1995. Etude du pouvoir pathogène d’isolats de PVY sur tabac : proposition d’une classification intégrant la résistance à la nécrose. Annales du Tabac 27, 4350.
  • Blanchard A, Rolland M, Lacroix C, Kerlan C, Jacquot E, 2008. Potato virus Y: a century of evolution. Current Topics in Virology 7, 2132.
  • Blanco-Urgoiti B, Sanchez F, Pérez de San Román C, Dopazo J, Ponz F, 1998. Potato virus Y group C isolates are a homogeneous pathotype but two different genetic strains. Journal of General Virology 79, 203742.
  • Chain F, Riault G, Trottet M, Jacquot E, 2007. Evaluation of the durability of the Barley yellow dwarf virus-resistant Zhong ZH and TC14 wheat lines. European Journal of Plant Pathology 117, 3543.
  • Chrzanowska M, 1991. New isolates of the necrotic strain of potato virus Y (PVYN) found recently in Poland. Potato Research 34, 17982.
  • Chung BYW, Miller WA, Atkins JF, Firth AE, 2008. An overlapping essential gene in the Potyviridae. Proceedings of the National Academy of Sciences, USA 105, 5897902.
  • Clark MF, Adams AN, 1977. Characteristics of the microplate method of enzyme-linked immunosorbent assay for the detection of plant viruses. Journal of General Virology 34, 47583.
  • Domingo E, 2002. Quasispecies theory in virology. Journal of Virology 76, 4635.
  • Drake JW, Charlesworth B, Charlesworth D, Crow JF, 1998. Rates of spontaneous mutation. Genetics 148, 166786.
  • Escriu F, Fraile A, Garcia-Arenal F, 2007. Constraints to genetic exchange support gene coadaptation in a tripartite RNA virus. PLoS Pathogens 3, 6774.
  • Fargette D, Pinel A, Traore O, Ghesquiere A, Konate G, 2002. Emergence of resistance-breaking isolates of Rice yellow mottle virus during serial inoculations. European Journal of Plant Pathology 108, 58591.
  • Feki S, Bouslama L, 2008. Molecular phylogeny and genetic variability of the Potato virus Y (PVY) strains on the CP-encoding region. Annals of Microbiology 58, 4338.
  • Garcia-Arenal F, Fraile A, Malpica JM, 2003. Variation and evolution of plant virus populations. International Microbiology 6, 22532.
  • Gooding GV, Tolin SA, 1973. Strains of Potato Virus Y affecting flue-cured tobacco in southeastern United States. Plant Disease Reporter 57, 2004.
  • Jaspars EMJ, Bos L, 1980. Alfalfa mosaic virus. CMI/AAB Descriptions of Plant Viruses No. 229. Wellesbourne, UK: Commonwealth Mycological Institute/Association of Applied Biologists.
  • Joliffe IT, Morgan BJ, 1992. Principal component analysis and exploratory factor analysis. Statistical Methods in Medical Research 1, 6995.
  • Julio E, Verrier JL, Dorhlac De Borne F, 2006. Development of SCAR markers linked to three disease resistances based on AFLP within Nicotiana tabacum L. Theoretical and Applied Genetics 112, 33546.
  • Kerlan C, Moury B, 2008. Potato virus Y. In: MahyBWJ, Van RegenmortelMHV, eds. Encyclopedia of Virology, 3rd edn. Oxford, UK: Elsevier, 28796.
  • Koelle G, 1958. Versuche zur vererbung der kankheitsresistenz bei tabak; 2 mitt. eine rippenbräune-resistente Virgin A Mutante nach anwendung künstlicher mutations aulösung durch röntgenstrahlen. Tabak-Forschung 24, 834.
  • Latorre BA, Flores V, 1985. Strain identification and cross-protection of potato virus Y affecting tobacco in Chile. Plant Disease 69, 9302.
  • Li N, Wang X, Zhou G, Dong J, 2006. Molecular variability of the coat protein gene of Potato virus Y from tobacco in China. Acta Virologica 50, 10713.
  • Lorenzen JH, Meacham T, Berger PH, Shiel PJ, Crosslin JM, Hamm PB, 2006. Whole genome characterization of Potato virus Y isolates collected in the western USA and their comparison to isolates from Europe and Canada. Archives of Virolology 151, 105574.
  • Lorenzen JH, Nolte P, Martin D, Pasche JS, Gudmestad NC, 2008. NE-11 represents a new strain variant class of Potato virus Y. Archives of Virology 153, 51725.
  • McCullagh P, Nelder J, 1989. Generalized Linear Models. New York, USA: Chapman and Hall.
  • Morroni M, Thompson JR, Tepfer M, 2008. Twenty years of transgenic plants resistant to Cucumber mosaic virus. Molecular Plant-Microbe Interactions 21, 67584.
  • Moury B, 2010. A new lineage sheds light on the evolutionary history of Potato virus Y. Molecular Plant Pathology 11, 1618.
  • Pelham J, Fletcher JT, Hawkins JH, 1970. The establishment of a new strain of Tobacco mosaic virus resulting from the use of resistant varieties of tomato-D. Annals of Applied Biology 65, 2927.
  • Pferdmenges F, Korf H, Varrelmann M, 2009. Identification of rhizomania-infected soil in Europe able to overcome Rz1 resistance in sugar beet and comparison with other resistance-breaking soils from different geographic origins. European Journal of Plant Pathology 124, 3143.
  • Piccirillo P, Piro F, 1986. PVY strains on Burley tobacco in southern Italy. Proceedings of the Coresta Symposium . Paris, France: CORESTA.
  • R Development Core Team, 2005. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computings.
  • Rolland M, Lacroix C, Blanchard A, Baldwin T, Kerlan C, Jacquot E, 2008. Potato virus Y (PVY): from its discovery to the latest outbreaks. Virologie 12, 26173.
  • Rolland M, Kerlan C, Jacquot E, 2009. The acquisition of molecular determinants involved in potato virus Y necrosis capacity leads to fitness reduction in tobacco plants. Journal of General Virology 90, 24452.
  • Rufty RC, Powell NT, Gooding GV, 1983. Relationship between resistance to Meloidogyne incognita and a necrotic response to infection by a strain of Potato virus Y in tobacco. Phytopathology 73, 141823.
  • Schiltz P, 1967. Création de N. tabacum Résistants à P. tabacina. Analyse Histologique et Biologique de la Résistance. Bordeaux, France: University of Bordeaux, PhD thesis.
  • Shaner G, Stromberg EL, Lacy GH, Barker KR, Pirone TP, 1992. Nomenclature and concepts of pathogenicity and virulence. Annual Review of Phytopathology 30, 4766.
  • Shukla DD, Ward CW, Burnt AA, 1994. Genome structure, variation and function. The Potyviridae. Wallingford, UK: CAB International, 82112.
  • Singh RP, Valkonen JPT, Gray SM et al., 2008. Discussion paper: the naming of Potato virus Y strains infecting potato. Archives of Virology 153, 113.
  • Soto MJ, Arteaga ML, Fereres A, Ponz F, 1994. Limited degree of serological variability in pepper strains of potato virus Y as revealed by analysis with monoclonal antibodies. Annals of Applied Biology 124, 3743.
  • Valkonen J, 2007. Viruses: economical losses and biotechnological potential. In: VreugdenhilD, BradshawJ, GebhardtC et al., eds. Potato Biology and Biotechnology: Advances and Perspectives. Oxford, UK: Elsevier, 61941.
  • Verrier JL, Doroszewska T, 2004. The “va” resistance to PVYN in Nicotiana tabacum: an assessment of the frequency of “va” breaking PVYN strains based on seven years of field survey on a worldwide basis. Proceedings of the 12th European Association for Potato Research (Virology Section) Meeting 2004, 13-19 June 2004. Rennes, France. 86 (Abstract).
  • Verrier JL, Marchand V, Cailleteau B, Delon R, 2001. Chemical change and cigarette smoke mutagenicity increase associated with CMV-DTL and PVY-N infection in burley tobacco. Proceedings of the CORESTA Meeting Agro-Phyto Groups, 2001.Cape Town, South Africa, Paris, France: CORESTA, 29.
  • Yamamoto Y, 1992. Studies on breeding of tobacco varieties resistant to veinal necrosis by potato virus Y strain T. Bulletin of the Leaf Tobacco Research Laboratory 2, 185.