SEARCH

SEARCH BY CITATION

Keywords:

  • gene expression;
  • microarray;
  • PVYN;
  • PVYNTN;
  • quantitative real-time PCR;
  • Solanum tuberosum;
  • symptomatology

Abstract

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

Differences in the early responses of two potato cultivars, Igor and Nadine, to two isolates of Potato virus Y (PVY), the aggressive PVYNTN and the mild PVYN, were monitored. Microarray and quantitative real-time PCR analyses were carried out to identify differentially expressed genes after inoculation with each virus isolate. Additionally, symptom severity and development was observed and the amount of virus isolate accumulated in systemically infected leaves was evaluated, where a significantly higher amount of PVYNTN was detected. Microarray analysis revealed 572, 1288 and 1706 differentially expressed genes at 0·5, 12 and 48 h post-inoculation, respectively in cv. Igor, with a similar pattern observed in cv. Nadine. Microarray and quantitative real-time PCR results implied an earlier accumulation of sugars and lower photosynthesis in leaves inoculated with the aggressive isolate than in leaves inoculated with the mild isolate. The PVYNTN isolate did not activate early differential expression of the Fe-superoxide dismutase and pectin methylesterase inhibitor (PMEI) genes, indicating a delay in plant response relative to that following PVYN inoculation. Differences in the expression of the β-glucanase-I gene were also observed in early plant responses to inoculation with each virus isolate.


Introduction

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

Plants respond to virus infection with a number of physiological alterations which lead to disease development. The process of disease development in plants is still not fully understood, but two models have been proposed: the competitive model, where the viruses compete with the host for resources; and the interaction model, where the disease develops as a consequence of viral components interfering with host processes (Culver & Padmanabhan, 2007). The latter can be related to changes in levels of different metabolites in infected leaves, namely carbohydrates (Herbers et al., 2000) and antioxidants (Li & Burritt, 2003; Diaz-Vivancos et al., 2008), and changes in the expression of stress-associated genes (Xu et al., 2003). With the development of powerful transcriptomic tools, more genes and, consequently, more pathways have been described in plant–microbe interactions (Yang et al., 2007; Babu et al., 2008; Baebler et al., 2009; Garcia-Marcos et al., 2009). Comparisons of plant responses to two related or unrelated virus isolates were previously performed on different plant–virus interactions (Xu et al., 2003; Love et al., 2005; Sajnani et al., 2007; Garcia-Marcos et al., 2009), but early responses to two closely related virus isolates have not yet been described. The timing of the response is critical for a successful defence, as was shown by proteome and transcriptome levels in sugar beet systemically infected with Beet necrotic yellow vein virus (Larson et al., 2008).

Potato virus Y (PVY), a member of the Potyviridae family, is economically the most important viral potato pathogen and is distributed worldwide. The three most common PVY isolate groups which have been described are PVYN, PVYO and PVYC. PVYN isolates induce necrosis in tobacco plants and very mild mottling in almost all potato cultivars, but no symptoms in potato cultivars bearing resistance genes Nc or Ny (Singh et al., 2008). The PVYN isolate group includes the extremely aggressive PVYNTN isolates (Singh et al., 2008), possessing common immunogenic characteristics with PVYN isolates (Barker et al., 2009). In sensitive potato cultivars, PVYNTN elicits the development of potato tuber necrotic ringspot disease (PTNRD), causing a decrease in the quality and quantity of potato production. On PVYNTN-inoculated leaves, necrotic and chlorotic spots appear a few days after the infection, whilst on non-inoculated leaves, wrinkles and mosaic chlorosis appear later as the virus spreads, leading to a palm-tree appearance (leaf drop). There are only a few potato cultivars which demonstrate extreme resistance, exhibiting no symptoms following PVYNTN infection. It is not yet understood which viral sequence or domain is actually responsible for the development of the PTNRD phenotype in sensitive potato cultivars (Singh et al., 2008). However, comparative studies showed that it is unlikely to be present within the coat-protein (CP) region (Glais et al., 2002) or within the HC-Pro region (Schubert et al., 2007). Recently, it was proposed that determinants responsible for symptom development may originate from different amino acid changes in several regions, but that none of them alone are essential for disease development (Barker et al., 2009).

Previous research on the susceptible interaction of potato and PVYNTN showed changes in the cytokinin level in inoculated leaves (Dermastia et al., 1995). In inoculated leaves with symptoms modifications in chloroplast structure and size (Pompe-Novak et al., 2001) and lower chlorophyll levels were observed (Milavec et al., 2001). Various activities of soluble and ionically and covalently bound peroxidases were detected in a range of potato cultivars with different reactions to PVYNTN infection (Milavec et al., 2008). In addition, photosynthesis-related genes and genes involved in perception, signalling and defence response were shown to be involved in the early response to virus inoculation (Baebler et al., 2009).

To identify genes and pathways differentially expressed after inoculation with two closely related virus isolates, a comparative study of potato plants inoculated with PVYNTN and PVYN was performed. cDNA microarray results obtained from two sensitive potato cultivars with similar reactions to PVY infection, i.e. Igor and Nadine (Pompe-Novak et al., 2006; Singh et al., 2008), allowed the differences in early plant responses to the aggressive and mild virus isolates to be interpreted. Early changes in plant responses revealed the first signals which triggered different levels of disease severity. The results were further verified with quantitative real-time PCR and symptom observation. Additionally, the responses of the plants to each virus isolate were compared against healthy plants, with a view to better understanding of plant responses to aggressive and mild virus isolates.

Materials and methods

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

Plant and virus material

To observe and compare the responses of potato plants to inoculation with the two virus isolates, PVYNTN and PVYN, two sensitive potato cultivars, Igor and Nadine, were selected. Plants grown from node tissue culture were planted in soil and kept at 21 ± 2°C in growth chambers, with a photoperiod of 16 h and relative humidity of 70%. After 4 weeks, 3–4 basal leaves of each plant were mechanically inoculated with the sap of either PVYNTN-infected plants (isolate NIBNTN, AJ585342), PVYN infected plants (isolate N-RB, AJ390285) or the sap of healthy plants (mock-inoculation). At 0·5, 12 and 48 h post-inoculation (h.p.i.), inoculated leaves from six to eight plants were harvested, immediately frozen and ground in liquid nitrogen. Plant material was stored at −80°C prior to further analysis. A separate group of 6–8 inoculated plants was left for symptom observation, testing for virus spread to upper leaves and tuber production. The same inoculation procedure and collection of samples were repeated three times; each series was grown independently at different times of the year.

RNA preparation, microarray hybridization and data analysis

RNA preparation, microarray hybridization and data analysis were carried out as recently described (Baebler et al., 2009) with minor modifications. In brief, total RNA was isolated using the RNeasy Plant Mini kit (Qiagen) from 300 mg plant material. In order to remove the remaining genomic DNA, digestion by DNase I (deoxyribonuclease I, amplification grade; Invitrogen) was carried out. The resulting RNA was further purified and concentrated using an RNeasy MinElute Cleanup kit (Qiagen). RNA quality and quantity were assessed using agarose gel electrophoresis, an RNA 6000 Nano Lab Chip kit in combination with Bioanalyzer (Agilent Technologies) and NanoDrop1000 (NanoDrop Technologies).

A 3DNA Array 900 kit (Genisphere) based on dendrimer labelling technology was used for cDNA synthesis and tagging as well as for microarray hybridization. For cDNA synthesis Superscript II Reverse Transcriptase (Invitrogen) was used and luciferase control RNA was added to the reverse transcription reaction as an external control. Samples of PVYNTN- and PVYN-inoculated plants were labelled with Cy-3 and Cy-5 and dye swap was undertaken to minimize variation associated with probe labelling and detection. The microarrays used in all experiments were cDNA microarrays (TIGR 10K v3 or v4 potato arrays, http://www.jcvi.org/potato/sol_ma_microarrays.shtml). The TIFF images were quantified and analysed using arraypro®analyzer 4·5 software (Media Cybernetics). Quality control and image analysis were performed as described by Baebler et al. (2009).

The microarray data were deposited in the NCBI Gene Expression Omnibus (GEO, http://www.ncbi.nlm.nih.gov/geo/) and are accessible through GEO series accession number GSE18196.

Data analysis was carried out in an r computing environment, limma package, as previously described (Baebler et al., 2009). Two normalizations, ‘vsn’ and ‘loess’, were applied separately, thus producing two datasets. Statistical analysis using linear models was carried out for each dataset, giving lists of differentially expressed genes in PVYNTN- vs. PVYN-inoculated plants for a given cultivar at a given time point. The intersection of genes which were found to be differentially expressed after two separate preprocessing procedures was used in the final list of differentially expressed genes. The differential expression was visualized in the context of biological processes using mapman (Rotter et al., 2007). For the identification of the processes, pathways or gene family members whose expression was significantly altered, the Wilcoxon rank sum test available in mapman and a ‘gene set enrichment analysis’ (GSEA, Subramanian et al., 2005) were applied.

Quantitative real-time PCR experiments

Quantitative real-time PCR (Q-PCR) was used to analyse changes in the expression of genes encoding RuBisCO activase (RA), chlorophyll a-b binding protein 4 (CAB4), granule-bound starch synthase I (GBSSI), sucrose synthase (SuSy), β-glucanases II and III (β-glu-II, β-glu-III), pectin methylesterase inhibitor (PMEI), Fe-superoxide dismutase (FeSOD), photosystem II 22-kDa protein (Baebler et al., 2009) and β-glucanase I (β-glu-I) (Oufir et al., 2008) (Table 1). For assay design, TA sequences (TIGR TA database, http://plantta.jcvi.org/index.shtml) presenting selected microarray clones were aligned using vectornti (InforMax) software and regions of homology were determined. Primers were designed using primerexpress software (ABI) or by Eurofins MWG Operon (Table 1). The concentrations of primers were optimized and the lowest concentration that gave the highest reporter fluorescence (ΔRn) and the lowest Ct were chosen. Dissociation curve analysis was carried out and primer-dimer formation examined. For normalization, the expression of COX and EF-1 genes was tested (Baebler et al., 2009).

Table 1.   Target sequences (GenBank accession nos) and forward (FP) and reverse primer (RP) sequences (FP and RP) of primers for the genes encoding RuBisCO activase (RA), chlorophyll a-b binding protein 4 (CAB4), granule-bound starch synthase I (GBSSI), sucrose synthase (SuSy), β-glucanases II and III (β-glu-II, β-glu-III), pectin methylesterase inhibitor (PMEI) and Fe-superoxide dismutase (Fe-SOD) representing selected gene groups; CHO, carbohydrates
Gene groupPrimerTarget sequencePrimer sequence (5′–3′)
PhotosynthesisRABQ120698FPAGAGGCAGCACTCGGAGATGC
RPGGGAAAAAAGCGCGTCCTGTTGC
CAB4BQ113443FPGGGCCTGCCAAGTTGTTCTTATGG
RPGATGATCCCGAGGCTTTTGCTGAATTG
Major CHO metabolismGBSS1BQ112290FPCCAAGAAATGGGAGACATTGCTATTGGG
RPTGATCTTATTGTTGAAACAAGGATAACCAAAGCTC
SuSyBQ119048FPCATCCCCCGTATTCTTATTGTTACTCGTCTAC
RPCGATACCCTTCTCAGTCCTAAAAGGGACC
Virus spreadβ-glu-IIM60462FPGATGCCCTTKTGGATTCWATGTA
RPGTATCKGAAAGTGGYTGGCCTT
β-glu-IIIX54456FPCCCTGGAGTTGTTGTAAATGATAAYG
RPATGCYACATACTCRGCCCTTGAGA
PMEIBQ504752FPAGGCGCCATGCGTGACT
RPAAAATCTTTAGGGGAAATGAAGGAAT
Antioxidant metabolismFe-SODBQ118260FPCTCCWCCTTATCCWATGRATGCT
RPCATCACAGGGCVTATGTCGAC

cDNA for Q-PCR was prepared from 1 μg DNase-treated total RNA, previously used in microarray experiments, using a High Capacity cDNA Reverse Transcription kit (ABI) according to the manufacturer’s recommendations with minor modifications.

Universal Master Mix (TaqMan or SybrGreen, ABI) together with primers and probes of appropriate concentrations was used to analyse 2 μL cDNA. The final concentration of all the probes in the reaction mix was 250 nm, whilst primer concentration was 300 nm, except with PMEI, COX and EF-1, where it was 900 nm. Two different dilutions of cDNA were tested in two parallel reactions in order to check the inhibition of amplification. All Q-PCRs were carried out in 384-well reaction plates on the ABI 7900HT Sequence Detection System using real-time data collection. For amplification, standard cycling conditions were used with an added dissociation-curve stage for SybrGreen reactions.

Data was first analysed using the sds 2·3 software (ABI), whilst relative quantification was calculated as previously described by Pfaffl (2001) and Vandesompele et al. (2002). Expression in individual samples was normalized to the expression of two endogenous control genes, EF-1 and COX (Baebler et al., 2009). The results were expressed as log2 of the ratio between gene expressions in two samples. The calculations for difference in expression between plants inoculated with PVYNTN or PVYN and between virus- and mock-inoculated plants were made. Average expression values of three biological replications were calculated. MeV (TIGR) was used for the visualization of the results.

Single-step RT Q-PCR specific for each virus isolate was used for testing the infection of systemically infected leaves of 4–6 potato cv. Igor plants inoculated with PVYNTN or PVYN. From each plant a leaf (100 mg) showing systemic symptoms was harvested, and total RNA was extracted. Total RNA (1 μL) was analysed in a duplicate of two dilutions to check amplification efficiency. The amount of viral RNA was estimated from a standard curve (Kogovšek et al., 2008). For total RNA load control, a COX assay was performed.

Results

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

Symptom development and systemic virus spread

Both cultivars reacted to infection with PVYNTN and PVYN isolates as expected. Chlorotic and/or necrotic ringspot lesions developed on PVYNTN-inoculated leaves 5–7 days post-inoculation (d.p.i.) (Fig. 1; Table 2). By 12 d.p.i. most of the inoculated leaves had fallen off and systemic mosaic was observed on upper non-inoculated leaves. PVYN inoculation triggered very mild chlorotic ringspot development, which became visible later and on few leaves (Fig. 1; Table 2). In comparison to cv. Igor, cv. Nadine developed slightly milder symptoms after PVYNTN inoculation, with smaller numbers of chlorotic and/or necrotic ringspot lesions per leaf (data not shown).

image

Figure 1.  Symptoms on leaves of potato cv. Igor plant 9 days post-inoculation with Potato virus Y isolates PVYN (left) and PVYNTN (right). PVYN induced mild chlorotic ringspot lesions, whilst PVYNTN induced severe chlorotic and necrotic ringspot lesions.

Download figure to PowerPoint

Table 2.   Time of appearance (days post-inoculation; d.p.i.) of the first symptoms on leaves of potato cvs Igor and Nadine following the inoculation with Potato virus Y isolates PVYNTN and PVYN. The percentage of leaves with symptoms present at 9 d.p.i. was determined on 6–8 inoculated plants (n = 18–24 leaves)
 Appearance of symptoms (d.p.i.)Percentage of leaves with symptoms (9 d.p.i.)
Cultivar/virus isolatePVYNTNPVYNPVYNTNPVYN
Igor598314
Nadine788313

At 21 d.p.i., plants of both cultivars infected with either virus isolate were showing a typical palm-tree effect. Viral infection and systemic spread were confirmed using single-step RT Q-PCR, which enabled the differentiation and detection of low amounts of virus. The efficiency of amplification was above 0·99 in all reactions, indicating only a minor effect of plant material on amplification (data not shown). When comparing the amounts of each virus isolate in systemically infected leaves in cv. Igor, higher levels of PVYNTN than PVYN were detected (Table 3). Even though there is no simple direct link between virus titre and symptom severity (Love et al., 2005), the difference in PVYNTN and PVYN virus titres could at least in part be related to the disease severity observed in inoculated potato plants. Additionally, the virus was confirmed in all tested tubers and PTNRD symptoms developed in PVYNTN-infected tubers (data not shown).

Table 3.   Systemic spread of Potato virus Y isolates PVYNTN and PVYN 21 days after inoculation; 4–6 inoculated potato cv. Igor plants were tested by single-step RT Q-PCR. The amount of viral RNA detected in 100 mg plant material was estimated from Ct values for each virus isolate. Average Ct values for COX were taken into account to avoid misinterpretation of the results regarding different total RNA loading into a single-step RT Q-PCR
Virus isolateAmount of viral RNA detected (pg mg−1 plant material)Virus Ct ±SDCOX Ct ±SD
PVYNTN1000–1000019·7 ± 3·622·6 ± 1·3
PVYN0·1–129·0 ± 2·721·9 ± 0·8

Microarray analysis of PVYNTN- vs. PVYN-inoculated plants

Using the setup of cDNA microarray experiments, the effects of two closely related viral isolates on the expression of genes in inoculated potato leaves were directly analysed. The results were therefore interpreted as differences in gene expression between plants inoculated with PVYNTN or PVYN, rather than differences between virus- and mock-inoculated plants. Differential expression of genes was monitored at three different time points: 0·5, 12 and 48 h.p.i. Microarray analysis resulted in lists of differentially expressed genes (Fig. 2). The number of differentially expressed genes was similar in both cultivars and increased over time. The number of up- and down-regulated genes was evenly distributed at all time points in both cultivars.

image

Figure 2.  Venn diagram showing numbers of differentially expressed (DE) genes (total, down-regulated and up-regulated) in plants of potato cvs Igor (a) and Nadine (b) inoculated with Potato virus Y isolates PVYNTN and PVYN. Each sector shows a number of unique DE genes for each time point after inoculation and the intersections of the sectors show shared DE genes. Supporting information can be found at http://www.nib.si/index.php/o-institutu/telefonski-imenik.html?view=details&id=93.

Download figure to PowerPoint

Pathway analysis of microarray data

For identification of the relevant processes involved in the studied system, microarray data were analysed with two algorithms: GSEAnalysis and the Wilcoxon rank sum test using the mapman gene ontology system. Both algorithms gave the same results for larger BINs (pathways or processes including more than 15 genes), but for smaller gene groups only the Wilcoxon rank sum test was efficient. Therefore, the results of the latter analysis are presented in Table 4. Genes related to photosynthesis, antioxidant metabolism and the invertase/pectin methylesterase inhibitor family were significantly differentially expressed in both potato cultivars (< 0·05). In cv. Igor, abiotic stress-associated genes (= 6 × 10−6), and in cv. Nadine, cell-wall genes ( 3 × 10−4), were significantly differentially expressed after inoculation with PVYNTN or PVYN.

Table 4.   Processes or protein families showing significantly altered expression in potato plants of cvs Igor and Nadine inoculated with Potato virus Y isolates PVYNTN and PVYN, 0·5, 12 and 48 h post-inoculation (h.p.i.). The results are part of the mapman gene ontology. The Wilcoxon rank sum test was carried out to predict BINs exhibiting expression profiles differing from those of other BINs (Usadel et al., 2005)
 IgorNadine
Process/protein familyExpressionaP-valueProcess/protein familyExpressionaP-value
  1. CHO: carbohydrates; PR: pathogenesis related.

  2. aExpression of genes is described as higher or lower in PVYNTN- than in PVYN-inoculated plants.

0·5 h.p.i.Invertase/pectin methylesterase inhibitor familyLower1 × 10−5Invertase/pectin methylesterase inhibitor familyLower4 × 10−4
PhotosynthesisHigher2 × 10−5Development – storage proteinsHigher5 × 10−4
Tetrapyrrole synthesisHigher9 × 10−4Tetrapyrrole synthesisHigher9 × 10−3
Metal handlingHigher6 × 10−3   
12 h.p.i.PhotosynthesisLower2 × 10−10Cell wallHigher3 × 10−4
SignallingLower3 × 10−4RedoxLower2 × 10−2
Amino acid metabolismHigher3 × 10−3PhotosynthesisLower2 × 10−2
Major CHO metabolismLower2 × 10−2Beta 1,3 glucan hydrolasesHigher3 × 10−2
   Major CHO metabolism – starch synthesisHigher3 × 10−2
48 h.p.i.Abiotic stressHigher6 × 10−6Cell wallHigher2 × 10−5
Transport – major intrinsic proteinsHigher4 × 10−4Biotic stress – PR proteinase inhibitorsHigher2 × 10−3
Glutathione S-transferasesHigher2 × 10−3PeroxidasesHigher2 × 10−2
Redox – dismutases and catalasesHigher3 × 10−2Glutathione S-transferasesLower2 × 10−2

Analysis of selected pathways

Based on previous knowledge of plant–virus interactions and the results of the two independent algorithms used, it was decided to focus the analysis on four metabolic processes: photosynthesis, tetrapyrrole synthesis, major carbohydrate (CHO) metabolism and antioxidant metabolism; and enzymes associated with virus movement, namely β-glucanases. PMEI was also assigned to the latter group. The percentage of differentially expressed genes out of all genes in the corresponding mapman BIN was similar in both cultivars (Fig. 3 for cv. Igor). In the gene groups for CHO metabolism, β-glucanases and antioxidant metabolism, the number of differentially expressed genes mainly increased with time, especially during the initial 12-h period. The number of differentially expressed PMEI genes was decreasing, whilst the number of differentially expressed photosynthesis related genes had a transient pattern, meaning that the highest number of differentially expressed genes was detected at 12 h.p.i.

image

Figure 3.  Percentage of differentially expressed clones in potato cv. Igor plants inoculated with Potato virus Y isolates PVYNTN and PVYN for selected metabolic pathways and gene groups. A very similar response was observed in cv. Nadine. Percentage was calculated from the number of differentially expressed clones out of all clones in corresponding mapman BINs. CHO metabolism (bsl00001), β-glucanases (inline image), PMEI (bsl00066), antioxidant metabolism (x), photosynthesis (•).

Download figure to PowerPoint

Only when comparing the number of differentially expressed photosynthesis-related genes was a considerable difference observed between cultivars. In cv. Igor, the highest number of differentially expressed photosynthesis-related genes (41) was detected at 12 h.p.i., whilst at the same time point, the lowest number of differentially expressed genes (14) was detected in cv. Nadine (Fig. 4). At the other two time points, the two cultivars had more similar numbers of differentially expressed photosynthesis-related genes.

image

Figure 4.  Numbers of differentially expressed (DE) genes related to photosynthesis in potato cvs Igor (white) and Nadine (grey) after inoculation with Potato virus Y isolates PVYNTN and PVYN.

Download figure to PowerPoint

Visualization of selected pathways

mapman visualization of the microarray results offered a schematic overview of the changes in gene expression within the metabolic pathways. The differential expression of selected gene groups at different time points in cvs Igor and Nadine is shown in Figure 5, where the difference in gene expression between plants inoculated with the two virus isolates is presented. In both cultivars, genes associated with photosynthesis light reactions, photorespiration and the Calvin cycle were more expressed in PVYNTN- than in PVYN-inoculated plants at 0·5 h.p.i. However, at later time points the expression of photosynthesis-related genes in PVYNTN-inoculated plants was mainly lower than in PVYN inoculated plants. The pattern of differential expression of tetrapyrrole synthesis-related genes was in parallel with the expression of photosynthesis-related genes (Fig. 5). A very similar pattern of expression of photosynthesis- and tetrapyrrole synthesis-related genes was observed in a previous microarray experiment on cv. Igor when comparing PVYNTN-inoculated and healthy plants (Baebler et al., 2009).

image

Figure 5.  Presentation of differentially expressed genes from selected gene groups in potato cvs Igor and Nadine inoculated with Potato virus Y isolates PVYNTN and PVYN. MapMan BINs: Photosynthesis – light – 1·1: Light reactions, respiration – 1·2: photorespiration, Calvin – 1·3: Calvin cycle, tetrapyrrole – 19: tetrapyrrole synthesis; major CHO – Syn. – 2·1·1: Sucrose synthesis and 2·1·2: Starch synthesis, Deg. – 2·2·1: Sucrose degradation and 2·2·2: Starch degradation; antioxidant – THIO – 21·1: thioredoxin peroxidase, 21·2: ASC – ascorbate and glutathione peroxidase, glutathione reductase, SOD - 21·6: dismutases and catalases; GST – 26·09: glutathion S-transferase, POX – 26·12: peroxidases; 26·4: β-glucanases; 26·18: PMEI. Each square presents log2 ratio of the expression of one clone (red, up-regulated; green, down-regulated).

Download figure to PowerPoint

Immediately after inoculation, expression of genes involved in major carbon metabolism showed only slight differences in response to the two virus isolates in either cultivar (Fig. 5). Later, in cv. Igor, genes involved in the synthesis of sucrose and starch were less expressed in PVYNTN- than in PVYN-inoculated plants, whilst in cv. Nadine, the opposite effect was observed. At the same time point, genes encoding sucrose-degradation enzymes were more highly expressed in PVYNTN-inoculated plants in both cultivars and in cv. Igor they remained more expressed at 48 h.p.i. However, in cv. Nadine the expression was lower in PVYNTN- than in PVYN-inoculated plants at 48 h.p.i.

Immediately after inoculation with the virus isolates a differential expression of antioxidant metabolism-associated genes was observed in both cultivars. They remained differentially expressed through all time points tested, although differences in expression between cultivars were observed. Antioxidant metabolism-related genes, namely ascorbate and glutathione peroxidase, glutathione reductase and glutathione S-transferase (GST), were predominantly more expressed in PVYNTN- than in PVYN-inoculated plants of cv. Igor, whilst in cv. Nadine, genes were mainly less expressed in PVYNTN- than in PVYN-inoculated plants.

In cv. Igor a differential expression of PMEI genes was observed only at 0·5 h.p.i., where in PVYNTN-inoculated plants expression was lower than in PVYN-inoculated plants (Fig. 5). Similarly, in cv. Nadine a lower expression of PMEI genes was observed in PVYNTN-inoculated plants than in PVYN-inoculated plants at 0·5 h.p.i. In cv. Igor a difference in expression of β-glucanase genes was observed at 12 h.p.i. and in cv. Nadine it was observed immediately after inoculation, when lower expression was detected in plants inoculated with PVYNTN than in plants inoculated with PVYN. The expression of β-glu-I and β-glu-II genes was mainly lower in PVYNTN-inoculated cv. Igor plants, with the exception of a higher expression of β-glu-I at 12 h.p.i. in PVYNTN- than in PVYN-inoculated plants.

Confirmation of a differential response by Q-PCR

Q-PCR analysis was used to verify microarray results and to analyse individual gene expression within selected gene groups in cv. Igor. Expression of genes encoding RA, CAB4 and PSII from the photosynthesis group, GBSSI and SuSy from the major carbon metabolism group, β-glu-I, β-glu-II and β-glu-III and PMEI from the virus movement group and Fe-SOD from the antioxidant metabolism group were analysed. A high correlation (R = 0·85) was observed between the results of Q-PCR and microarray results (Fig. 6).

image

Figure 6.  Q-PCR verification of DNA microarray results. Correlation was calculated between log2 ratios indicating a change in expression after inoculation of potato cv. Igor with Potato virus Y isolate PVYNTN and PVYN as measured by microarrays and Q-PCR. A high correlation factor was determined (R = 0·85). Supporting information can be found at http://www.nib.si/index.php/o-institutu/telefonski-imenik.html?view=details&id=93.

Download figure to PowerPoint

Q-PCR analysis of PVY- vs. mock-inoculated plants

To better understand the early response to viral infection, the response to each virus isolate was compared to that in healthy, mock-inoculated, plants by Q-PCR analysis. Based on a similarity in the response between cultivars observed by microarrays, only samples of cv. Igor were tested. The results are presented as expression in PVYNTN- and in PVYN-inoculated plants relative to healthy (mock-inoculated) plants. The overall changes in expression of genes encoding GBSSI and Fe-SOD were lower in PVYNTN- than in PVYN-inoculated plants. In the case of Fe-SOD, no differences in expression were observed between PVYNTN-inoculated and healthy plants, whilst the expression of GBSSI was mildly up-regulated in PVYNTN-inoculated plants compared with healthy plants (Fig. 7). However, in plants inoculated with PVYN, both genes were down-regulated at 0·5 h.p.i. and up-regulated 12 h later, relative to healthy plants. At 48 h.p.i., the expression of Fe-SOD in PVYN-inoculated plants decreased. In plants inoculated with either virus isolate, genes encoding SuSy were down-regulated, with stronger down-regulation relative to healthy plants detected at 12 h.p.i. in PVYNTN-inoculated plants and at 48 h.p.i. in PVYN-inoculated plants. Potato plants responded similarly to both virus isolates in terms of β-glu-I and -III and PMEI, the expression of which was mainly down-regulated. Nevertheless, lower expression was detected in plants inoculated with PVYNTN (Fig. 7). The gene encoding isoform β-glu-I was up-regulated only at 12 h.p.i. in PVYNTN-inoculated plants, and a strong up-regulation of the gene encoding PMEI was detected in PVYN-inoculated plants at 0·5 h.p.i. The gene encoding isoform β-glu-II was up-regulated at 0·5 and 48 h.p.i., with intermediary down-regulation at 12 h.p.i., after inoculation with either virus isolate.

image

Figure 7.  Heatmap showing differential expression of seven selected genes in potato cv. Igor plants inoculated with Potato virus Y isolate PVYNTN (NTN) or PVYNTN (N) vs. mock-inoculated (healthy) plants. Q-PCR was used to determine average log2 ratios of expression at three time points after inoculation. Green signifies down-regulation, red signifies up-regulation relative to mock-inoculated plants. Grey, not tested.

Download figure to PowerPoint

Discussion

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

To investigate the early response to two closely related PVY isolates, the plants of sensitive potato cvs Igor and Nadine were inoculated with an aggressive PVYNTN and a mild PVYN isolate. Microarray hybridization and Q-PCR were used to identify differentially expressed metabolic pathways or gene groups and specific genes.

Accumulation of soluble sugars [sucrose (Shalitin & Wolf, 2000), glucose and fructose (Herbers et al., 1997)] and starch (Arias et al., 2003; Love et al., 2005; Handford & Carr, 2007) is often observed in virus-infected plants. Sugars are known to play a major role in the regulation of various functional gene groups (Graham, 1996). Sucrose was shown to induce the expression of defence genes in rice (Gomez-Ariza et al., 2007). The exact role of changes in sugar and starch levels in the development of the disease is not understood. However, catabolism of starch was shown not to be required in meeting the carbon demand for the biosynthesis of virus-specific gene products (Handford & Carr, 2007). Immediately after inoculation (0·5 h.p.i.) a higher expression of the GBSSI gene was detected in PVYNTN- than in PVYN-inoculated potato cv. Igor plants, whilst at 12 h.p.i. the opposite effect was observed (Figs 5 and 7). In parallel, the expression of the sucrose degradation enzyme (SuSy) mRNA was down-regulated in plants inoculated with either isolate, with PVYNTN inoculation causing a faster response than PVYN inoculation, compared with healthy plants (Fig. 7). A similar response, although on an enzymatic level, was observed in tobacco plants inoculated with PVYN, where a decrease in SuSy activity was detected 48 h.p.i., associated with an accumulation of soluble sugars (Sindelarova et al., 1999). A higher expression of GBSSI, possibly leading to starch accumulation, and down-regulation of SuSy, implying accumulation of soluble sugars, were observed in response to both virus isolates, although in potato cv. Igor the response was faster in plants inoculated with the milder isolate. When comparing the cultivars, higher accumulation of starch could be predicted in potato cv. Nadine plants inoculated with PVYNTN than in potato cv. Igor, since the expression of starch synthesis genes remained higher for longer in the former (Fig. 5). It was concluded that, in general, cvs Nadine and Igor have similar responses to virus inoculation, although differences in expression of the specific genes involved in carbohydrate metabolism were observed.

The differential expression of the genes involved in major carbon metabolism observed in the two cultivars implied an accumulation of free sugar and starch, affecting photosynthetic capacity. A lower expression of photosynthesis-related genes was observed in both cultivars at 12 and 48 h.p.i. in PVYNTN- than PVYN-inoculated plants (Fig. 5). Immediately after inoculation, the genes involved in photosynthesis were more highly expressed in both cultivars in plants inoculated with PVYNTN than in those inoculated with PVYN. The effect of virus infection on photosynthesis rate, chlorophyll content and photosynthesis-related gene expression has been described in different plant–virus interactions (Herbers et al., 2000; Milavec et al., 2001; Pompe-Novak et al., 2006; Synkova et al., 2006; Dardick, 2007; Babu et al., 2008; Baebler et al., 2009). Down-regulation of photosynthesis-related genes was described in two independent studies on potato cv. Igor infected with PVYNTN compared with healthy plants at 14 d.p.i. (Pompe-Novak et al., 2006) and 12 h.p.i. (Baebler et al., 2009), whilst at 0·5 h.p.i. up-regulation of photosynthetic genes was observed (Baebler et al., 2009). Similarly, at the time of symptom appearance in Nicotiana benthamiana plants infected with PVY or Potato virus X, or co-infected with both viruses the genes related to photosynthesis were down-regulated in all three types of interaction, and the number of differentially expressed genes and their magnitude of repression was correlated with symptom severity (Garcia-Marcos et al., 2009). An identical pattern of expression to that of photosynthesis-related genes was observed for tetrapyrrole synthesis-related genes (Fig. 5), which were previously reported to be connected with changes in photosynthesis as a response to virus inoculation (Dardick, 2007; Baebler et al., 2009; Garcia-Marcos et al., 2009).

To prevent the overproduction of reactive oxygen species as a result of energy-production impairment and defence response, plants activate antioxidant mechanisms, involving catalases, superoxide dismutase (SOD), peroxidases (ascorbate, glutathione), glutathione reductase, GST and others (Kiraly et al., 2002; Li & Burritt, 2003; Diaz-Vivancos et al., 2008). The differential activation of three groups of peroxidases in a range of potato cultivars infected with PVYNTN was described previously (Milavec et al., 2008). The present results showed an immediate response to virus inoculation at the level of genes connected to antioxidant metabolism in both potato cultivars (Fig. 5). Antioxidant-associated genes were expressed more in PVYNTN- than in PVYN-inoculated cv. Igor plants, whilst in cv. Nadine plants the expression of those genes was lower in PVYNTN- than in PVYN-inoculated plants. The most pronounced difference between cultivars was in the expression of GST genes at 48 h.p.i. (Fig. 5). Involvement of GST in response to infection has been observed in different plant–virus interactions, where in the early stages of infection the amount of GST transcripts (Whitham et al., 2003) or enzyme activity (Kiraly et al., 2002) decreased, but later the transcript (Whitham et al., 2003) and enzyme (Diaz-Vivancos et al., 2008) levels and enzyme activity increased (Kiraly et al., 2002). The differential expression of GST genes in virus-inoculated plants of cvs Igor and Nadine could be explained by a time shift in response, or by the activation of different antioxidant mechanisms, which can be partially deduced from the differential expression of other antioxidant-associated genes observed in both potato cultivars (Fig. 5). Moreover, the unresponsiveness of Fe-SOD mRNA expression to PVYNTN inoculation observed in cv. Igor (Fig. 7) could also indicate an activation of different antioxidant mechanisms. The expression of Fe-SOD mRNA in PVYN-inoculated plants was transient, whilst no difference in expression was observed in PVYNTN-inoculated plants when compared to healthy plants of cv. Igor (Fig. 7). Similarly, no changes in activity of SOD were detected in the early response to local lesion-producing Tobacco mosaic virus (TMV) in tobacco (Kiraly et al., 2002). However, in symptomless virus–plant interactions, the activity of SOD declined (Clarke et al., 2002).

Reactive oxygen species, especially hydrogen peroxide, and callose have similar patterns of induction in plants inoculated with mild and aggressive virus isolates (Xu et al., 2003). Callose deposition was shown to play a role in the movement of plant viruses (Iglesias & Meins, 2000) and the expression of β-glucanase mRNA, encoding the key enzymes degrading callose, was shown to be differentially regulated after virus inoculation (Ward et al., 1991; Pompe-Novak et al., 2006). Three structural classes of β-glucanase have so far been identified in the Solanaceae family. Class I isoforms are basic proteins constantly accumulated in the cell vacuole, whilst the class II and III isoforms are inducible acidic proteins secreted into the extracellular space and were described as PR-2 proteins (Ward et al., 1991). The results of the present study showed a general down-regulation of β-glu-I and β-glu-III in plants of cv. Igor inoculated with either virus isolate when compared to healthy plants (Fig. 7). An interesting and perhaps important exception was in the expression of the β-glu-I gene at 12 h.p.i., where up-regulation was detected in plants inoculated with PVYNTN and down-regulation in plants inoculated with PVYN. The expression of the β-glu-II gene in both virus-inoculated groups of plants indicates a transient pattern, signifying up-regulation with intermediary down-regulation at 12 h.p.i. (Fig. 7), but when comparing the response in PVYNTN- to PVYN-inoculated plants, a lower expression of β-glu-II was observed in the former (Fig. 5). A lower expression of β-glu-I mRNA in PVYNTN- than in PVYN-inoculated plants was also detected in cv. Nadine at 0·5 h.p.i. (Fig. 5). The results showed uniformity in expression of β-glucanase genes in response to both virus isolates, but some differences in expression of specific gene isoforms were detected.

On the basis of the literature and the microarray results, PMEI was assigned to the group of genes connected to virus spread. Pectin methylesterase (PME) was shown to be involved in the cell-to-cell movement of TMV (Dorokhov et al., 1999; Chen et al., 2000) and reduced levels of PME in plant vascular tissue limited virus spread, resulting in a significant delay in TMV systemic infection (Chen & Citovsky, 2003). However, inverse correlation between PME gene activity and TMV lesion size was shown in tobacco plants expressing elevated levels of the PME gene (Gasanova et al., 2008). The inhibitor of PME (PMEI) was shown to have antifungal (Lionetti et al., 2007) and antimicrobial (An et al., 2008) properties. PMEI was also speculated to be involved in limiting virus movement through the formation of a PME–PMEI complex (Giovane et al., 2004). The present study showed an immediate increase in PMEI gene expression in PVYN-inoculated plants of cv. Igor (Figs 5 and 7) and a similar response in cv. Nadine (Fig. 5). At later stages of infection (21 d.p.i.) higher amounts of PVYNTN than the milder PVYN isolate were detected in systemically infected leaves. Even though changes in β-glucanase and PMEI genes expression observed in the early stages of infection and the lower PVYN titre could not be directly connected, it would be interesting to further investigate this.

Although both potato cultivars are sensitive to PVY infection and develop typical PTNRD symptoms, variations in symptom severity and the timing of their appearance were observed. Gene expression analysis showed slight differences in the early expression of the gene groups presented here between cultivars, although very similar pathways or gene groups were expressed after inoculation with the two closely related virus isolates.

The mechanism of early plant response to virus inoculation is still poorly investigated and not fully understood. Genetic variation within the virus and within the host, the developmental stage of the host and environmental factors all contribute to the complexity of the host response to a virus infection (Love et al., 2005; Handford & Carr, 2007). The present study has shown the uniqueness of the early plant response to the closely related aggressive and mild virus isolates PVYNTN and PVYN, respectively. Virus isolates interfere with plant metabolism, inducing different changes within inoculated plants. As the results show, viruses trigger changes on two levels, relating to energy production and defence response. It can be concluded that when comparing plant responses to PVYN and PVYNTN, the latter employs a better strategy to overcome the first signal induction, delaying plant defence responses. In addition, PVYNTN has the ability to alter the defence response, leading to unsuccessful protection.

Acknowledgements

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

We are grateful to Mrs Lidija Matičič for her excellent technical support in plant material preparation and inoculation. The work was partially conducted at the Center for Functional Genomics and Bio-Chips, Ljubljana, Slovenia. Lisa Gow was in receipt of a Defra Seedcorn fellowship awarded jointly to FERA and the University of Bristol. The project was supported by the Slovenian Research Agency (grant numbers P4-0165 and 3311-04-831046).

References

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