The goal of this study was to determine the expression profile of a late blight resistance gene, Rpi-phu1, after challenge with an incompatible Phytophthora infestans isolate. Relative expression levels were measured by quantitative real-time PCR, before as well as 1, 3 and 5 days after contact with the pathogen. Plants tested included diploid and tetraploid breeding lines with the gene in different genetic backgrounds and plants of a chosen breeding line at different developmental stages. All tested potato breeding lines with the Rpi-phu1 gene were highly resistant to P. infestans and so was the line tested at different developmental stages. No significant changes were detected in Rpi-phu1 transcript level after pathogen challenge in any of the tested tetraploid potato lines. However, Rpi-phu1 transcription in diploid lines was enhanced by contact with P. infestans. Basal expression of the Rpi-phu1 gene was significantly lower in the youngest, 3-week-old plants than in 6-week-old plants. Nevertheless, 1 day after inoculation with the pathogen, transcript levels in these plants were slightly greater than in 6-week-old plants, indicating that plant age can influence transcription of the Rpi-phu1 gene, but that this does not affect late blight resistance.
In response to proteins called effectors which are produced by pathogens to suppress plant defences, plants have evolved NBS-LRR (nucleotide binding site, leucine-rich repeat) genes (Eitas & Dangl, 2010). They are the most common resistance genes (R genes) and are of great interest amongst scientists and breeders of various plant species. R genes have been consciously applied in agriculture at least since the publication of Harold Flor's gene-for-gene hypothesis in 1942 (Flor, 1942). The number of identified and sequenced R genes is rapidly growing in many plants, including model and economically important species. A good example is the potato–late blight pathosystem. The first potato R gene against its most economically important enemy, Phytophthora infestans, which causes late blight, was cloned in 2002 (Ballvora et al., 2002) and since then the sequences of approximately 20 others have been determined (Śliwka & Zimnoch-Guzowska, 2012). Knowledge of gene sequences enables a range of analyses that can address some vital questions about the mechanisms of R gene function or regulation of expression. The latter has been investigated in various plant species and diverse factors potentially affecting R gene expression have been taken into account. To date, three patterns of resistance gene expression after pathogen challenge have emerged from these studies.
In the first one, the NBS-LRR gene is expressed constitutively, at a low level that does not change after contact with a pathogen. An L6 gene encoding flax resistance to Melampsora lini follows this pattern. No significant difference in its expression level was detected in plants before and up to 48 h after inoculation with an incompatible M. lini isolate (Ayliffe et al., 1999). Similarly, the potato R3a gene for resistance to late blight was shown to be expressed at the same level before and until 72 h after pathogen attack (Huang et al., 2005). In a comprehensive study on Arabidopsis thaliana, the majority of the 170 investigated NBS-LRR and related genes were expressed at low levels with a variety of tissue specificities and irrespective of salicylic acid treatment (Tan et al., 2007). In that work, low levels of constitutive expression of R proteins supported their postulated constitutive ability to recognize early stages of infection and induce downstream defence responses.
However, in the same study, 15 other NBS-LRR genes were induced above their low levels of constitutive expression 4 h after salicylic acid treatment (Tan et al., 2007). This resembles an increase of expression of 10 A. thaliana R genes induced by flagellin peptide, flg22 (Zipfel et al., 2004), as well as several other cases from diverse plant species in which expression of R genes has been enhanced after pathogen challenge. One such example is the sugar beet gene Hs1pro1, encoding resistance to the nematode Heterodera schachtii; its low expression level increased fourfold 1 day after infection (Thurau et al., 2003). The tobacco N gene for resistance to Tobacco mosaic virus is another interesting example of an R gene up-regulated upon infection; it was shown that 72 h after contact with the virus, the level of N gene expression relative to controls increased 38-fold in inoculated leaves and 165-fold in younger, uninoculated leaves (Levy et al., 2004). Lastly, in the potato–late blight pathosystem, it was demonstrated that the transcript levels of the RB gene originating from Solanum bulbocastanum increased after infection with P. infestans in transgenic potato lines and, even more dramatically, in the wild donor species (Kramer et al., 2009). These data provide evidence that some R genes are regulated during contact with a pathogen, probably with the result of strengthening the recognition capacity of the plant.
The third pattern of R gene transcription regulation is probably the most energy-efficient. Here, transcription of the R gene is induced by contact with the pathogen and is not detectable prior to that. A rice gene, Xa1, for resistance to Xanthomonas oryzae pv. oryzae, exhibited this expression pattern; its transcripts were not detected until the third day after infection and the level of expression was correlated with the number of bacteria infecting the plant. Wounding was also sufficient to trigger weak transcription of the Xa1 gene, detected 5 days after treatment (Yoshimura et al., 1998). Similarly, expression of the Capsicum resistance gene Bs3 against another Xanthomonas species (X. campestris pv. vesicatoria), was induced by contact with the pathogen. In this case it was demonstrated that pathogen recognition was based on interaction between the pathogen's effector and the Bs3 promoter region, leading to activation of R gene transcription and subsequent defence reactions (Römer et al., 2007).
Expression regulation studies have so far only been published for two R genes involved in interactions between potato and P. infestans, namely R3a and RB, and they have indicated rather diverse regulation patterns, even for the same R gene (Huang et al., 2005; Bradeen et al., 2009; Kramer et al., 2009; Millett et al., 2009; Iorizzo et al., 2011). While R3a transcription was investigated in transgenic plants inoculated with an incompatible P. infestans isolate (Huang et al., 2005), expression of RB has been measured in both transgenic and donor plants but using a compatible late blight isolate (Bradeen et al., 2009; Kramer et al., 2009; Millett et al., 2009). The results obtained still leave many questions open. The goal of the present study was to determine the expression profile of another late blight resistance gene, Rpi-phu1, in non-transgenic plants and after challenge with an incompatible P. infestans isolate.
The Rpi-phu1 gene was introgressed from the interspecific hybrid between Solanum phureja and Solanum stenotomum to a diploid S. tuberosum genetic background by a series of crosses and then mapped to potato chromosome IX (Śliwka et al., 2006). It has been shown to be effective both in potato leaves and tubers, providing very high resistance and to act independently from the length of vegetation period. Rpi-phu1 has been transferred to tetraploid level and applied in conventional potato breeding together with a molecular marker useful for its detection (Śliwka et al., 2010). It is a CC-NBS-LRR (coiled coil NBS-LRR) gene and its sequence is identical to that of the Rpi-vnt1.1 gene from Solanum venturii (Foster et al., 2009). So far, only one P. infestans isolate, originating from Ecuador, has been found able to infect plants with Rpi-phu1/Rpi-vnt1.1 (Foster et al., 2009).
To improve understanding of the resistance provided by the Rpi-phu1 gene, its relative expression levels were measured by quantitative real-time (qRT)-PCR, before as well as 1, 3 and 5 days after contact with the pathogen. An observation that in the field some plants with the Rpi-phu1 gene showed late blight symptoms late in the season and that the P. infestans isolate obtained from them was not able to infect 6-week-old plants of the same genotype in laboratory tests (J. Plich and J. Śliwka, IHAR-PIB Młochów, unpublished data) inspired a hypothesis that in some conditions the expression of the gene is insufficient for plant defence. Diploid and tetraploid breeding lines with the gene in different genetic backgrounds were tested, as well as plants of a chosen breeding line at different developmental stages.
Materials and methods
For the expression survey of the Rpi-phu1 gene in different genetic backgrounds, the diploid potato lines DG 01-180, DG 92-227 and DG 97-813 and the tetraploid lines 04-IX-21, TG 97-403 and TG 97-411 with the Rpi-phu1 gene were used. They were complex breeding lines obtained by IHAR-PIB Młochów, related to each other as shown in Figure 1. The share of wild germplasm from the S. phureja × S. stenotomum hybrid, from which the Rpi-phu1 gene was introgressed, was estimated to range from 3·12% in the 04-IX-21 genome, through 6·25% in the DG 97-813, DG 01-180, TG 97-411 and TG 97-403 genomes, to 12·5% in the DG 92-227 genome.
The experiment on the influence of plant age on the expression profile of the Rpi-phu1 gene after pathogen challenge was performed using the tetraploid potato line 04-IX-21.
In all experiments five plants of each genotype were tested as biological replications and potato cv. Craigs Royal was used as a late-blight-susceptible control.
Inoculation tests and plant growing conditions
Phytophthora infestans isolate MP324 from the IHAR-PIB Młochów collection was used for inoculation tests. This incompatible isolate was used for identification and mapping of the Rpi-phu1 gene (Śliwka et al., 2006). It was collected in 1997 in Poland. It is of the A1 mating type, resistant to metalaxyl and of complex race (22.214.171.124.126.96.36.199.10.11). Plants were grown in controlled conditions with 16 h daylight at 18°C and 8 h dark at 16°C. Inoculation was performed by spraying whole plants with a sporangial suspension of P. infestans (50 000 sporangia mL−1) prepared as described by Zarzycka (2001). Before each experiment the isolate was propagated on susceptible potato tissue at least twice. Plants were inoculated at the beginning of the night period, after which relative humidity was maintained between 80 and 90%.
Gene expression assay
Leaf samples for the analyses were taken at random, excluding the youngest and the oldest leaves, from all replicate plants at four time points: before inoculation and 1, 3 and 5 days post-inoculation. They were immediately frozen in liquid nitrogen, and stored at −80°C. Isolation of total RNA was performed using the Spectrum Plant Total RNA Kit (Sigma) according to the manufacturer's instructions. RNA was then treated with DNaseI (Sigma) (experiment with different potato genotypes) or the On-Column DNase I Digestion Set (Sigma) (experiment on plant age). The quantification of the isolated RNA was performed on a BioPhotometer (Eppendorf) spectrophotometer and approximately 100–400 ng, depending on the yield, was further reverse transcribed with oligo(dT)18 and random hexamer primers in 20-μL reaction volumes using the Maxima First Strand cDNA Synthesis Kit for RT-qPCR (Fermentas). The initial incubation for 10 min at 25°C was followed by 15 min at 50°C and the cDNA synthesis reaction was later terminated by heating to 85°C for 5 min. Concentrations of cDNA were adjusted to 100 ng μL−1 and 2 μL cDNA were used as a template for qRT-PCR performed in a 20-μL reaction volume along with 10 μL LightCycler® 480 SYBR Green I Master (Roche) and primers at 0·5 μm each (Table 1). Specific primers (phu6, Table 1) for Rpi-phu1 gene amplification in qRT-PCR were designed on the basis of GenBank sequence FJ423044.1, which was shown to be identical to that of Rpi-phu1 (Foster et al., 2009). Every sample was analysed in three technical replications in qRT-PCR and the reactions were performed on the LightCycler® 480 II system (Roche). The following programme was applied: denaturation at 95°C for 5 min; 40 amplification cycles of 95°C for 10 s, 65°C for 20 s and 72°C for 30 s. Then, PCR product melting was performed at a temperature range of 65–97°C and the melting curve was analysed to confirm amplification of gene-specific products. On the basis of previous studies (Nicot et al., 2005; Bradeen et al., 2009; Kramer et al., 2009) the following reference gene candidates were chosen: elongation factor 1-α (Ef1-α), urease, cyclophilin and α-tubulin. Using corresponding primers, PCR reactions in amplification conditions suitable for phu6 primers were performed. Good PCR products were obtained for α-tubulin and Ef1-α, which were also described as the best candidates for reference genes in P. infestans–potato pathosystem in a previous study (Nicot et al., 2005). These genes were tested using the cDNA from four biological replicates of 6-week-old plants, in order to define which was suitable for use as the reference gene, with the result that no significant differences in Rpi-phu1 expression levels between α-tubulin and Ef1-α were detected. The α-tubulin gene showed the most consistent crossing point (Cp) values and was chosen for further gene expression surveys. Efficiency testing for both phu6 and α-tubulin primers was performed using a serial dilution of cDNA and similar efficiencies were observed (E = 2), so this value was used for further calculations. The relative expression data were obtained from LightCycler 480 release 1.5.0 software (Roche). Microsoft excel was applied for data analyses and to draw graphs. Statistical analyses using a t-test for ∆∆Ct cycle threshold values were performed (statistica, Stat Soft Inc.).
Table 1. qRT-PCR primers tested and used in the Rpi-phu1 gene expression survey
Plants of the Rpi-phu1-bearing potato line 04-IX-21 and susceptible cv. Craigs Royal were grown in controlled conditions for 14 weeks. Between weeks 6 and 9 they flowered and after week 12 their foliage started to dry out because of senescence. Plants aged 3, 6, 9 or 12 weeks old (five plants per genotype) were inoculated by spraying with P. infestans isolate MP324. Leaf samples collected before and 1, 3 and 5 days after contact with the pathogen were analysed for Rpi-phu1 transcription by qRT-PCR.
All tested potato breeding lines with the Rpi-phu1 gene were highly resistant to P. infestans isolate MP324 and so was the line 04-IX-21 at all examined developmental stages. Within 5 days after inoculation, plants with the resistance gene, irrespective of their genetic background, either showed no symptoms of the disease or produced small necroses characteristic of the hypersensitive reaction, while on plants of the susceptible cv. Craigs Royal severe sporulating lesions were formed (Fig. 2). Phytophthora infestans isolate MP324 was very aggressive in the conditions provided.
Rpi-phu1 gene expression in different potato genotypes
Three diploid and three tetraploid potato lines with the Rpi-phu1 gene, related to each other as shown in Figure 1, were used. Five plants of each of the resistant lines, as well as susceptible cv. Craigs Royal were grown in controlled conditions for 6 weeks. They were then inoculated with a suspension of sporangia and zoospores of P. infestans isolate MP324. Five independent inoculations were performed but only 6-week-old plants of line 04-IX-21 were inoculated twice. The results obtained for this plant genotype/age combination were consistent (Fig. 3, 04-IX-21; Fig. 5, week 6). Before inoculation and also 1, 3 and 5 days after it, leaf samples were collected from all plants and analysed by qRT-PCR for Rpi-phu1 gene transcript abundance relative to α-tubulin transcript level within each plant. No Rpi-phu1 gene transcription was detected in plants of cv. Craigs Royal and its expression was constitutive but low (a fraction ranging from 0·003 to 0·026 of the α-tubulin transcript level before inoculation) in all plants containing the gene. Amounts of transcripts observed during the P. infestans attack were compared with the baseline transcript level on day 0 for each potato genotype separately in order to reveal expression patterns (Fig. 3). No significant changes were detected in Rpi-phu1 transcript level after pathogen challenge in any of the tested tetraploid potato lines (TG 97–403, TG 97–411, 04–IX-21). However, Rpi-phu1 transcription in diploid lines was enhanced by the contact with P. infestans. In two lines, DG 92–227 and DG 97–813, a significant increase of transcription was observed on the first day after inoculation, while in line DG 01–180 it was on the fifth day. The diploid line DG 92–227 showed the strongest, over threefold enhancement of Rpi-phu1 transcription on the first day. Unexpectedly, on the third day of the experiment a decrease was noted in Rpi-phu1 transcript levels in all lines compared with day 1, but only in diploid lines DG 92–227 and DG 97–813 was it significant. In line DG 97–813 this decrease was also significantly below the level for day 0 (Fig. 3).
To assess differences between the six potato lines in Rpi-phu1 expression level, the tetraploid line 04-IX-21, with the lowest expression of the gene, was chosen as a baseline and other genotypes were compared to it at four time points of the experiment (Fig. 4). Tested potato lines differed both in their basal Rpi-phu1 transcript levels and in their response to pathogen attack. Before contact with P. infestans, one tetraploid (TG 97–411) and two diploid (DG 97–813 and DG 92–227) lines exhibited significantly stronger expression of the Rpi-phu1 gene than line 04-IX-21 (Fig. 4, day 0). In the case of diploid lines DG 97–813 and DG 92–227, this difference persisted 1 day post-inoculation (Fig. 4, day 1) and for line DG 92–227 even until the third day (Fig. 4, day 3). The Rpi-phu1 transcript was most abundant in line DG 92–227 – more than three times as abundant as in line 04-IX-21 on day 0, almost seven times as abundant after enhancement on day 1 and still three times as abundant on day 3 (Fig. 4).
Influence of plant age on Rpi-phu1 gene transcription
The leaf samples collected before and 1, 3 and 5 days after pathogen contact from 3-, 6-, 9- and 12-week-old plants of line 04-IX-21 were analysed for Rpi-phu1 transcription by qRT-PCR. The transcription patterns were visualized for plants at different developmental stages by comparison with transcription levels on day 0 (Fig. 5). A significant change was observed in 3-week-old plants, where expression increased more than three times 1 day post-inoculation (Fig. 5, week 3). A weak, significant enhancement of Rpi-phu1 transcription was also seen on day 5 of the experiment in the case of 9- and 12-week-old plants (Fig. 5, weeks 9 and 12).
When basal expression of the Rpi-phu1 gene was analysed in plants of different ages and compared with expression in 6-week-old plants (Fig. 6), it was significantly lower in the youngest, 3-week-old plants (Fig. 6, day 0). However, even 1 day after inoculation, transcript levels in these plants, as well as in 9- and 12-week-old plants, slightly exceeded those in 6-week-old plants (Fig. 6, day 1). On day 3, the expression of the Rpi-phu1 in 3- and 9-week-old plants was significantly lower than in 6-week-old plants (Fig. 6, day 3). Plants at tested ages did not differ in their Rpi-phu1 transcript levels on day 5 after pathogen contact (Fig. 6, day 5).
The present study is apparently the first to describe R gene expression patterns in an incompatible interaction between non-transgenic potato plants and P. infestans. This is in contrast to the study by Kramer et al. (2009), who used only compatible isolates, as well as to the work by Huang et al. (2005), who applied avirulent P. infestans but only to transgenic plants. The present experiment on the influence of plant on Rpi-phu1 expression differed from the previous ones, because, unlike other studies (Millett et al., 2009; Iorizzo et al., 2011), it was concerned not only with R gene transcript levels in uninoculated plants, but also with potential changes in Rpi-phu1 transcription in the plant's response to the pathogen at different developmental stages.
The Rpi-phu1 gene was constitutively expressed at a low level in tetraploid breeding lines, like the R3a gene (Huang et al., 2005), but its transcription was enhanced by pathogen challenge in diploid lines and young plants of the tetraploid line 04-IX-21, similarly to the RB gene (Kramer et al., 2009).
It was shown that diploid and tetraploid breeding lines, bearing the Rpi-phu1 gene, varied in basal transcription levels of the gene as well as in its expression patterns 1, 3 and 5 days after pathogen contact (Figs 3 and 4). Line DG 92–227 (diploid), which exhibited the highest level of Rpi-phu1 expression before pathogen attack but also the greatest transcription enhancement 1 day after inoculation, was at an earlier stage of backcrossing to S. tuberosum than the other lines. The share of wild Rpi-phu1 donor plant germplasm was estimated to be 12·5% in this line, while in line 04-IX-21 (tetraploid), which showed the lowest basal Rpi-phu1 expression level, the proportion of wild genome was the lowest (3·12%). However, among other lines with 6·25% wild germplasm in their genomes, two showed higher basal Rpi-phu1 expression than 04-IX-21, while two did not, and only in diploid lines DG 97-813 and DG 01–180 was an increase in gene transcription observed on day 1 and 5, respectively. Nevertheless, the proportion of wild genome in the tested line, as well as ploidy, could underly the effect of genetic background on Rpi-phu1 transcription that was clearly seen in this study. A larger share of wild germplasm might increase the probability of the presence of alleles of genes which regulate Rpi-phu1 transcription and which have evolved together with the R gene in the donor species. This hypothesis is supported by the findings of Kramer et al. (2009). In a compatible interaction with P. infestans, gene RB showed variable basal levels and patterns of transcription 1, 3 and 5 days after pathogen contact, even in different transgenic potato lines (tetraploid), indicating the importance of gene copy number and location in the genome for its function. More striking was the difference between the transgenic plants and a wild donor of the RB gene, S. bulbocastanum (2×), that expressed the gene seven to 17 times more than transgenic potatoes before contact with the pathogen; 1 day after infection the wild plant showed a dramatic increase in RB transcription, transcript levels reaching 37 times those observed in some transgenic lines (Kramer et al., 2009). Moreover, in another study, RB transcript level was shown to correspond to transgene copy number but also to late blight resistance level (Bradeen et al., 2009).
Another potential factor underlying the differences between tested potato lines could be the cytoplasm. However, cytoplasmic influence on Rpi-phu1 expression patterns can be excluded here because in the pedigree of tested lines, donors of the gene were used as both seed and pollen parents (Fig. 1).
As reviewed by Millett et al. (2009), some studies suggest that both young and senescing potato plants are less resistant to late blight than flowering ones, while others indicate that susceptibility increases from flowering onwards. The latter is consistent with the trend noted in maize and tomato and is explained by a source–sink relationship change, i.e. development of storage organs and shifting of energy away from other metabolic activities. In the case of potato, tuber formation approximately corresponds with flowering. Field observations in the current study were in good accordance with this theory. It was noted that plants with the Rpi-phu1 gene showed some late blight symptoms at the end of growing season, although they were infected with a P. infestans isolate which was incompatible on young, preflowering plants in laboratory tests. Therefore, in order to test the hypothesis that the lack of or insufficient Rpi-phu1 expression in an aging plant is responsible for successful infection, an avirulent isolate, MP324, was used. However, in the experiment on the influence of plant age on Rpi-phu1 expression performed in controlled laboratory conditions, only the youngest, 3-week-old plants showed significantly lower gene transcript levels before inoculation with the pathogen (Fig. 6, day 0). One day post-inoculation, this was compensated by the expression enhancement resulting from pathogen contact, and in the 3-week-old plants the Rpi-phu1 transcript level was even slightly higher than that of 6-week-old plants (Fig. 6, day 1). A similar response to P. infestans was noted in 9- and 12-week-old plants, in which transcript levels were also significantly increased (Fig. 6, day 1). Plants at all developmental stages remained highly resistant to late blight. One possible explanation could be that in laboratory conditions, the senescence of plants growing in pots was too rapid to detect decreases in Rpi-phu1 transcript levels or resistance level preceding it. However, Millett et al. (2009) demonstrated that although late blight resistance levels of plants with the RB gene in a compatible interaction differ depending on plant developmental stage (pre- and post-flowering and senescence), RB transcript levels remain the same and cannot be responsible for this phenomenon. Another study tested flowering plants as well as 2- and 4-week-older ones and also detected no significant changes in RB transcript abundance (Iorizzo et al., 2011). In both studies the R gene transcript levels were tested in uninoculated plants, while the present study also considered potential changes in Rpi-phu1 transcription in the plant's response to the pathogen.
Potato NBS-LRR genes do not often emerge from global expression studies as differentially induced upon late blight infection (Tian et al., 2006; Gyetvai et al., 2012). However, using subtractive hybridization in combination with cDNA array hybridization, Ros et al. (2004) detected two- to fourfold induction of expression of three resistance genes 72 h post-infection in both 2- and 4-week-old plants of two potato cultivars, Indira and Bettina. The induced genes were: R1 from Solanum demissum and homologues of R genes from the Rx1/Gpa2 locus on chromosome XI and the Cf-9 gene cluster on chromosome I (Ros et al., 2004). In another study, sequencing of EST libraries constructed from potato leaves inoculated and non-inoculated with P. infestans revealed four NBS-LRR sequences unique for a challenged incompatible leaf library, indicating their induction by pathogen attack (Ronning et al., 2003). Finally, in a comprehensive transcriptional analysis using cDNA microarrays, a homologue to the R13 resistance gene cluster in soyabean was shown to be strongly (more than 10-fold) induced from 36 to 72 h post-infection. Interestingly, the expression pattern of the R gene observed in that study seemed to be quite complex, possibly oscillating with two induction peaks within this period (Wang et al., 2005). This pattern resembles some noted in the present study (Fig. 3), although no biological explanation for it has yet been proposed.
The aim of this study was to investigate a phenomenon observed in the field. Controlled experiments were performed in the laboratory in which plants were inoculated with P. infestans and conditions supporting disease development were ensured. However, an essential difference between the field and the laboratory that could have influenced the outcome of this research was disease pressure. In the field, plants are not sprayed with inoculum once, but need to fight under permanent pathogen attack, as spores are constantly present in the air, especially late in the growing season. Moreover, in the field potatoes are subjected to infection not only by P. infestans but also by a broad spectrum of other pathogens and pests, including insects, fungi, bacteria and viruses. Various abiotic stresses and changing day length can also affect plant resistance in the field. This may be a potential reason for successful attack of P. infestans isolates that are incompatible in the laboratory tests, although as shown here, Rpi-phu1 gene expression was sufficient for defence of even aging plants.
The authors thank Dr H. Jakuczun and Dr B. Flis for supplying them the plant material. This work was financed by Polish National Centre for Research and Development grant LIDER/06/82/L-1/09/NCBiR/2010.