Sumaira Farrakh , Meinan Wang Xianming Chen

1 Department of Plant Pathology, Washington State University, Pullman, WA 99164-6430, USA

2 Department of Biosciences, COMSATS Institute of Information Technology, Park Road, Islamabad, Pakistan

3 USDA-ARS, Wheat Health, Genetics, and Quality Research Unit, Pullman, WA 99164-6430, USA

Abstract Interactions of the stripe rust pathogen (Puccinia striiformis f. sp. tritici) with wheat plants activate a wide range of host responses. Among various genes involved in the plant-pathogen interactions, the expressions of particular pathogenesisrelated (PR) protein genes determine different defense responses. Different types of resistance have been recognized and utilized for developing wheat cultivars for resistance to stripe rust. All-stage resistance can be detected in seedling stage and remains at high levels throughout the plant growth stages. This type of resistance is race-specific and not durable. In contrast, plants with only high-temperature adult-plant (HTAP) resistance are susceptible in seedling stage, but become resistant when plants grow older and the weather becomes warmer. HTAP resistance controlled by a single gene is partial,but usually non-race specific and durable. The objective of this study was to analyze the expression of PR protein genes involved in different types of wheat resistance to stripe rust. The expression levels of 8 PR protein genes (PR1, PR1.2,PR2, PR3, PR4, PR5, PR9 and PR10) were quantitatively evaluated at 0, 1, 2, 7 and 14 days after inoculation in single resistance gene lines of wheat with all-stage resistance genes YrTr1, Yr76, YrSP and YrExp2 and lines carrying HTAP resistance genes Yr52, Yr59, Yr62 and Yr7B. Races PSTv-4 and PSTv-37 for compatible and incompatible interactions were used in evaluation of PR protein gene expression in wheat lines carrying all-stage resistance genes in the seedlingstage experiment while PSTv-37 was used in the HTAP experiment. Analysis of quantitative real-time polymerase chain reaction (qRT-PCR) revealed that all of the PR protein genes were involved in the different types of resistance controlled by different Yr genes. However, these genes were upregulated at different time points and at different levels during the infection process among the wheat lines with different Yr genes for either all-stage resistance or HTAP resistance. Some of the genes were also induced in compatible interactions, but the levels were almost always higher in the incompatible interaction than in the compatible interaction at the same time point for each Yr gene. These results indicate that both salicylic acid and jasmonate signaling pathways are involved in both race-specific all-stage resistance and non-race specific HTAP resistance. Although expressing at different stages of infection and at different levels, these PR protein genes work in concert for contribution to different types of resistance controlled by different Yr genes.

Keywords: wheat, stripe rust, plant resistance, pathogen-related protein, qRT-PCR

1. lntroduction

Stripe rust (yellow rust), caused by Puccinia striiformis Westend. f. sp. tritici Erikss. (Pst), is one of the most damaging diseases of wheat worldwide (Chen 2005;Wellings 2011). Depending upon the severity of epidemic and resistance or susceptibility of wheat cultivars, yield losses caused by stripe rust range from 0 to more than 90% (Chen 2005, 2014). The disease can be controlled by growing resistant cultivars or by timely applying effective fungicides. These different control strategies have their advantages and disadvantages. Use of the fungicides adds extra cost to wheat production; is harmful to humans,animals and the environment; and may be prohibited in the areas that are protected for special habitats or for organic farming. Growing resistant cultivars is the most effective,economic, easy-to-use and environmentally friendly approach for control of the disease (Chen 2005, 2013;Dodds and Rathjen 2010).

When a Pst urediniospore adheres to a leaf surface and starts to infect the plant under the conditions of dew formation and temperature range for germination and penetration, the plant exhibits diverse physiological and biochemical changes. Because of these changes,the plant will either show a resistant (incompatible) or susceptible (compatible) response. The onset of these responses involves the recognition of host plants to the attacking pathogen. Wheat plants respond to the infection of Pst through various defense compounds induced by the pathogen. The plant response can be completely resistant,completely susceptible, or any level in between. Resistance to stripe rust can be classified into different types based on various ways of categorization, but commonly into allstage resistance vs. high-temperature adult-plant (HTAP)resistance (Line 2002; Chen 2013). All-stage resistance,also called seedling resistance, can be detected at seedling stage, but expresses during all stages of plant growth.This type of resistance usually provides a high level of protection against Pst infection but is easily overcome by virulent races of the pathogen, and therefore, not durable.In contrast, HTAP resistance, which starts expressing when plants reach late jointing stage (Zadoks 32) and the weather becomes warm, is non-race specific and durable, but often does not provide complete protection, especially when the disease starts early in the growth season and the weather is favorable for the disease development (Chen 2005, 2013).Although both types of resistance can be controlled by a master gene, referred as a Yr gene for yellow rust resistance,numerous defense-related genes that are regulated by master genes are involved in either type of resistance(Coram et al. 2008a, b, 2010; Chen et al. 2013). Through transcriptomics analysis, different genes were found to be involved in all-stage and/or HTAP resistance (Coram et al. 2008a, b, 2010). However, molecular mechanisms underlying the different types of resistance are still not well understood.

In a plant-pathogen interaction, a complex signaling cascade is involved, including the production of reactive oxygen species (ROS), referred as oxidative burst. ROS are not only antimicrobial compounds, but also inducers of hypersensitive response (HR) that is related to the host cell necrosis in the infected cells (Tenhaken et al. 1995;Lamb and Dixon 1997). Host cell death, or programmed cell death, is beneficial for plants as a defense response against biotrophic pathogens, because it is detrimental for biotrophic pathogens as they need living host cells to grow and reproduce (Williams and Dickman 2008).

Pathogen-related (PR) proteins occur widely in the plant kingdom and comprise a large group of 17 protein families.These proteins include PR-1 (unknown function), β-1,3-glucanase (PR-2), endochitinases (PR-3, PR-4, PR-8 and PR-11), thaumatin-like proteins (PR-5), proteinase inhibitors(PR-6), proteinase (PR-7), peroxidase (PR-9), ribonucleaselike proteins (PR-10), defensins (PR-12), thionins (PR-13),lipid transfer proteins (PR-14), oxalate oxidase (PR-15),oxalate oxidase-like proteins (PR-16) and PR-17 (unknown function). These PR proteins have been reported to play important roles in plant defense against various biotic and abiotic stresses (Van Loon and Van Strien 1999; Van Loon et al. 2006a, b), but the roles of most of them in wheat-Pst interactions have not been studied.

At the molecular level, little is known about how different types of stripe rust resistance affect the development of the pathogen, or the response of the plant to pathogen invasion.One approach for understanding the mechanisms of disease resistance is to profile the expression of different genes.Studying the Yr5 resistance, a typical race-specific all-stage resistance, and the Yr39 resistance, a typical non-race specific HTAP resistance, in comparison with a susceptible reaction using the commercial Affymetrix wheat microarray genechip or a custom genechip revealed expression of few PR protein genes (Coram et al. 2008a, b). The attempts to expend gene expression to include some other Yr genes for either all-stage or HTAP resistance were also limited by the numbers of genes in the custom genechip used in the previous studies (Coram et al. 2010; Chen et al. 2013). It is not known whether other PR protein genes are involved in the different types of resistance, and if involved, whether PR protein genes express differently in the different types of resistance controlled by different Yr genes. Therefore, the objective of this study was to elucidate the roles of different PR protein genes in different types of resistance and their involvement in resistance controlled by different Yr genes.The results may be helpful for improving the understanding of race specificity and durability of stripe rust resistance.

2. Materials and methods

2.1. Plant materials

Spring wheat (Triticum aestivum L.) near-isogenic lines (NILs)AvSYrTr1NIL (YrTr1), AvSYrSPNIL (YrSP) and AvSYr76NIL(Yr76) and recombinant inbred line (RIL) AvS/Exp 1/1-1 line 7(YrExp2), representing race-specific all-stage resistance,and RILs Yr52, Yr59, Yr62 and YrPI182103-7BL (Yr7B),representing non-race specific HTAP resistance, were used in this study. The NILs for YrTr1 and YrSP were developed by Wellings et al. (2004), and the NIL for Yr76 and the RIL for YrExp2 were developed by Xiang et al. (2016) and Lin and Chen (2008), respectively in our program. The RILs for Yr52, Yr59, Yr62 and Yr7B were also developed in our program (Ren et al. 2012; Feng et al. 2013; Lu et al. 2014;Zhou et al. 2014). All of these lines have the spring wheat‘Avocet Susceptible’ (AvS) background.

2.2. Pathogen races

Races PSTv-4 (virulent to Yr1, Yr6, Yr9, Yr17, Yr27, YrSP and Yr76 but avirulent to Yr5, Yr7, Yr8, Yr10, Yr15, Yr24,Yr32, Yr43, Yr44, YrTr1 and YrExp2) and PSTv-37 (virulent to Yr6, Yr7, Yr8, Yr9, Yr17, Yr27, Yr43, Yr44, YrTr1 and YrExp2, but avirulent to Yr1, Yr5, Yr10, Yr15, Yr24, Yr32,YrSP and Yr76) (Wan and Chen 2014) were used to study the expression of PR protein genes in compatible and incompatible interactions (bold highlighted Yr genes are the ones studied in this paper). PSTv-37 was used to study PR protein gene expression in the HTAP resistance experiments, in which RILs Yr52, Yr59, Yr62 and Yr7B were resistant, but they were susceptible to this race and other races in seedlings (Ren et al. 2012; Feng et al. 2013; Lu et al. 2014; Zhou et al. 2014).

2.3. Plant inoculation

The experiments for studying race-specific all-stage resistance and non-race specific HTAP resistance were conducted separately. All experiments were conducted using a randomized complete block design with three biological replications. For each biological replication in the race-specific all-stage experiment, seeds of individual genotypes were planted in separate 25 cm×42.5 cm flats using a potting mix (6 peat moss:4 vermiculite with lime:3 sand:3 commercial potting mix:2 perlite:1.7 g L-1lime:3.3 g L-1Osmocote:2.2 g L-1ammonium nitrate). Each flat contained 5 rows of 10 seedlings each, with rows randomly assigned one of the five time points for sampling, 0, 1, 2,7 and 14 days after inoculation (dai). Seedlings from one of the rows were used to monitor the expected disease responses to Pst inoculation. One row of AvS was also planted in each treatment to check inoculation conditions and efficiency. The seedlings were grown to the two-leaf stage (Zadoks 10, ~12 days after planting) in a greenhouse with a diurnal temperature cycle of 10 to 25°C and a 16-h light/8-h dark cycle. Inoculation was performed by misting the plants with sterile water and applying a 1:20 urediniospore/talcum mixture to leaves with a sterile brush.Talcum was used to aid in the spread and adhesion of spores over leaf surfaces. Control flats were treated in the same way except for the absence of urediniospores in talcum.After inoculation, plants were misted lightly with distilled water. Plastic sleeves were placed around the inoculated pots to prevent contamination. Three biological replications(individual plants) were inoculated in each treatment group.Plants were placed in a dew chamber for 24 h in dark at 10°C with 100% relative humidity and then were removed from the dew chamber and placed in a growth chamber(16 h light at 16°C; 8 h dark at 8°C) (Wan and Chen 2014).

For the HTAP experiment, six seeds were planted in a pot of 15-cm in diameter and 15-cm in height filled with potting soil mixture as described above. Plants were grown in a greenhouse at 18-25°C with a 16-h photoperiod. When the plants reach the boot stage (Zadoks 45), inoculation of flag leaves was performed by misting the plants with sterile water and then applies with a 1:20 urediniospore/talcum mixture with a sterile brush. Mock-inoculated plants were treated in the same way with talcum only. Susceptible check AvS was also included in the experiment. Inoculated plants were placed in a dew chamber for 24 h in dark at 10°C with 100%relative humidity, and then placed in a growth chamber set at a diurnal cycle of 12°C for 12 h, rising to 30°C at the rate of 4.5°C h-1, staying at 30°C for 4 h and then decreasing to 12°C with rate 4.5°C h-1with 16 h light covering the high temperature range and 8 h dark in the low temperature range(Chen 2013). Pst- and mock-inoculated seedlings and adult plants for phenotypic observations were grown under the respective temperature conditions as described above, and infection types were recorded 20 dai using the 0-9 scales described by Line and Qayoum (1992).

2.4. RNA extraction

The inoculated leaves were harvested with sterilized scissors at 0, 1, 2, 7 and 14 dai. These time points were selected based on the major points of the infection process: (1) contact between Pst and plants (0 dai, just after inoculation), (2) fungus entered plant tissue (1 dai),(3) formation of haustoria and starting of the parasitic life cycle (2 dai) (Zhang et al. 2012), (4) appearance of chlorosis disease symptom (7 dai) and (5) sporulation (14 dai). The leaves were placed in a labeled aluminum foil wiped with 95% ethanol and then immediately frozen in liquid nitrogen.

Total 200 mg of leaf tissue for each sample was ground in liquid nitrogen in a sterilized mortar with a sterilized pestle to fine powder. Total 1 mL Trizol (Invitrogen, Carlsbad, CA,USA) was added to each sample. The suspension was homogenized by shaking moderately for 7 s, incubated for 2-3 min at 20-22°C, and added with 0.2 mL chloroform. The samples were shaken for 15 s and incubated for additional 2-3 min at 20-22°C. Samples were centrifuged for 15 min at 13 000 r min-1at 4°C. Approximately 550 μL of the aqueous phase was transferred to an Eppendorf tube, added with an equal volume of phenol:chloroform:isoamyl alcohol (25:24:1)and mixed well by vortexing and centrifuged for 10 min at 13 000 r min-1at 4°C. The aqueous phase was transferred to a new Eppendorf tube, added with an equal volume of chloroform and centrifuged at the same conditions. The aqueous phase was transferred again to a new tube,added with 0.25 mL isopropanol, mixed by inverting, kept at 20-22°C for 2-3 min and centrifuged for 10 min at 13 000 r min-1at 4°C. After decanting the supernatant, the pellet was washed with 1 mL filter sterilized 70% ethanol and dried in a chemical hood. The RNA pellet was dissolved in 50 μL of sterile water and kept at -80°C until use.

RNA was analyzed using 2% agarose gel electrophoresis to ascertain the quality of RNA and the absence of DNA.The concentration of each RNA sample was measured with a NanoDrop 1000 Spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). Only the RNA samples with OD260/OD280ratio of 1.9-2.0, indicating protein free, and OD260/OD230ratio greater than 2.0, indicating reagent free,were used for further analysis.

2.5. qRT-PCR assay

As listed in Appendix A, eight PR protein genes of wheat were selected based on their potential roles in plant defense reported in wheat or other plants. The primer sequences for PR1, PR2, PR3, PR5 and PR10 were from Desmond et al. (2006), PR1.2 from Molina et al. (1999), PR4 and PR9 were from Casassola et al. (2015) and β-actin from Yin et al.(2009). All primers were synthesized by Promega (Madison,WI, USA). The iTaq? Universal One-Step RT-qPCR Kit(Life Science Research, Hercules, California, USA) was used in quantitative real-time polymerase chain reaction(qRT-PCR). qRT-PCR amplification was performed in a total volume of 10 μL containing 60 ng μL-1RNA, 1×iTaq universal SYBER Green Reaction Mix, 1×iScript reverse transcriptase, 300 nmol L-1of gene-specific primers and nuclease-free water. PCR was performed in a Bio-Rad iQ5 Real-Time PCR Detection System Instrument (Bio-Rad, Hercules, California, USA) with the following cycling program: reverse transcription of RNA at 50°C for 10 min and qRT-PCR performed at 95°C for 1 min, 35 cycles of 95°C for 10 s, 60°C for 30 s and 75°C for 30 s. All products were subjected to the melting curve analysis between 55 and 95°C to determine the specificity of the PCR reaction.Experiments included a non-template control and non-RT control. Transcript levels of all eight PR protein genes in Pst inoculated wheat leaf tissue were calculated in relative to their mean expression levels in the mock-inoculated (control)plants, at each time point. The expression value of β-actin gene, which is constitutively expressed and not affected by pathogen infection, was used to calibrate the expression values of PR protein genes in each sample. Three biological replicates were analyzed for each time point. Relative fold changes were calculated using the comparative 2-ΔΔCTmethod (Livak and Schmittgen 2001).

2.6. Statistical analysis

Relative gene transcript levels were analyzed using analysis of variance (ANOVA) to determine significant differences between the expression levels using the Statistical Analysis System (SAS) Program (SAS Institute Inc., Cary, NC,USA). The Tukey’s test was used to determine significant differences among the genes at each time point.

3. Results

3.1. Phenotypic observations

In all experiments, resistance to stripe rust on all Pstinoculated and mock-inoculated plants were monitored until 20 dai. For all incompatible interactions, leaves of all genotypes began to show chlorosis at ~7 dai, which developed into necrotic flecks by 14 dai. For the compatible interactions, sporulation was observed starting at ~14 dpi.Infection types (ITs) were recorded at 20 dai. All mockinoculated plants were free of any disease symptoms over the 20-day observation period. Expected reactions, IT 1 or 2 (necrotic blotches on seedling leaves and necrotic stripes on flag leaves without uredinia) for incompatible(resistant) reaction and IT 8 or 9 (abundant uredinia with or without chlorosis) for compatible (susceptible) reaction,were confirmed for all wheat line - Pst race treatments of the wheat Yr single-gene lines with all-stage resistance genes YrTr1, YrExp2, YrSP or Yr76. Similarly, expected IT 2(necrotic stripes without uredinia) was observed on flag leaves of wheat lines with HTAP resistance genes Yr52,Yr59, Yr62 or Yr7B whereas the susceptible check variety AvS had IT 8. These results indicated successful inoculation and right conditions for the plant samples that were used to determine expression of PR protein genes.

3.2. Expression of PR protein genes in race-specific all-stage resistance

YrTr1The YrTr1 line was resistant to PSTv-4 (incompatible interaction) but susceptible to PSTv-37 (compatible interaction). The mean relative expression values of the PR protein genes at different time points are present in Appendix B and their patterns are shown in Fig. 1. Genes PR2 (1.14), PR3 (1.34) and PR9 (1.61) were induced in the incompatible interaction at 0 dai, but their expression was not significant compared with other genes and also not significantly different from those in the compatible interaction.At 1 dai, all genes (4.58-9.53 in the incompatible interaction and 2.79-6.90 for the compatible interaction except for PR2)were highly induced compared to the mock inoculation.In comparison between the incompatible and compatible interactions, all of these genes, except PR1 and PR5, were significantly higher in the incompatible interaction. PR1.2 had the highest expression level (9.53) in the incompatible interaction. At 2 dai, all genes, except PR1.2 and PR9, were induced by Pst inoculation, and the expression levels of these induced genes (PR1, PR2, PR3, PR4, PR5 and PR10) were significantly higher in the incompatible interaction than in the compatible interaction. PR1 had the highest expression level(8.22) in the incompatible interaction. At 7 dai, PR1, PR2 and PR10 were induced in the compatible interaction, whereas all genes, except PR9, were significantly induced in the incompatible interaction compared to the mock inoculation,and their expression levels were significantly higher than those in the compatible interaction. PR4 had the highest expression level (6.58) in the incompatible interaction. At 14 dai, none of the PR protein genes were induced in the compatible interaction and only PR1 (3.21) was significantly induced in the incompatible interaction.

Fig. 1 Quantitative real-time PCR (qRT-PCR) analyses of expression of eight PR protein genes in wheat line of YrTr1 inoculated with races PSTv-4 (incompatible) and PSTv-37 (compatible) of Puccinia striiformis f. sp. tritici. Three independent biological replications were performed. All data were normalized to the β-actin expression level. The relative expression was expressed as fold change relative to the mock inoculated plants. Relative gene expression was calculated using the comparative 2-ΔΔCT method.The ＊ symbol represents significant difference (P≤0.05) according to analysis of variance. Bars represent standard deviation among the biological replicates.

YrExp2Similar to YrTr1, the YrExp2 line was also resistant to PSTv-4 (incompatible interaction) but susceptible to PSTv-37 (compatible interaction). The mean relative expression values of the PR protein genes at different time points are present in Appendix B and their patterns are shown in Fig. 2. At 0 dai, the expression levels of PR4 and PR9 in the incompatible interaction were significantly higher than those in the compatible interaction, but none of the eight PR protein genes had the relative expression values greater than 2.00. At 1 dai, all eight PR protein genes were significantly induced in the incompatible, but only PR5 was significantly induced in the compatible interaction compared to the mock inoculation. All of these genes had higher expression levels in the incompatible interaction than the compatible interaction. PR10 had the highest expression level (8.03) in the incompatible interaction. At 2 dai, the expression levels of all PR genes were significantly higher in the incompatible interactions than those in the compatible interactions. The similar results were observed at 7 dai, however, the expression levels were relatively low compared to 2 dai and PR2 had the highest expression level(6.87). At 14 dai, PR1.2, PR2, PR3, PR5 and PR9 were significantly induced in the incompatible interaction but only PR1.2 and PR2 were significantly induced in the compatible interaction in comparison with the mock inoculation. The expression levels of PR1, PR1.2, PR2, PR3, PR5 and PR9 were significantly higher in the incompatible interaction than the compatible interaction, and PR2 also had the highest expression level (6.12).

YrSPThe YrSP line was susceptible to PSTv-4 (compatible interaction) but resistant to PSTv-37 (incompatible interaction). The mean relative expression values of the PR protein genes at different time points are present in Appendix B and their patterns are shown in Fig. 3. At 0 dai, the expression levels of PR4, PR9 and PR10 were significantly higher in the incompatible interaction than the compatible interaction. At 1 dai, none of the genes,except PR3, were significantly induced in the compatible interactions, whereas all of the genes were significantly induced in the incompatible interaction compared to the mock inoculation. All of the genes had significantly higher expression levels in the incompatible interaction than the compatible interaction, with PR3 had the highest expression level (24.24) in the incompatible interaction. At 2 dai, the expression levels of all of the genes in the incompatible interaction were significantly higher than those in the compatible interaction with PR2 had the highest expression level (25.88). At 7 dai, the expression levels of PR1,PR1.2, PR3, PR4 and PR9 were significantly higher in the incompatible interaction than the compatible interaction with PR9 the highest (10.05). At 14 dai, only PR9 was significantly induced in the incompatible interaction.

Fig. 2 Quantitative real-time PCR (qRT-PCR) analyses of expression of eight PR protein genes in wheat line of YrExp2 inoculated with races PSTv-4 (incompatible) and PSTv-37 (compatible) of Puccinia striiformis f. sp. tritici. Three independent biological replications were performed. All data were normalized to the β-actin expression level. The relative expression was expressed as fold change relative to the mock inoculated plants. Relative gene expression was calculated using the comparative 2-ΔΔCT method.The ＊ symbol represents significant difference (P≤0.05) according to analysis of variance. Bars represent standard deviation among the biological replicates.

Fig. 3 Quantitative real-time PCR (qRT-PCR) analyses of expression of eight PR protein genes in wheat line of YrSP inoculated with races PSTv-4 (compatible) and PSTv-37 (incompatible) of Puccinia striiformis f. sp. tritici. Three independent biological replications were performed. All data were normalized to the β-actin expression level. The relative expression was expressed as fold change relative to the mock inoculated plants. Relative gene expression was calculated using the comparative 2-ΔΔCT method.The ＊ symbol represents significant difference (P≤0.05) according to analysis of variance. Bars represent standard deviation among the biological replicates.

Yr76Similar to YrSP, the Yr76 line was susceptible to PSTv-4 (compatible interaction) but resistant to PSTv-37(incompatible interaction). The mean relative expression values of the PR protein genes at different time points are present in Appendix B and their patterns are shown in Fig. 4.At 0 dai, PR2, PR3, PR4, PR5 and PR9 had significantly higher expression levels in the incompatible than the compatible interaction. PR4 had the highest expression level (13.93) in the incompatible interaction. At 1 dai,the expression levels of PR3, PR4, PR5 and PR10 were significantly higher in the incompatible interaction than the compatible interaction. PR10 had the highest expression level (18.33) in the incompatible interaction. At 2 dai, all of the genes, except PR9, were significantly induced in the incompatible interaction, and PR1, PR1.2 and PR2 were also induced in the compatible interaction. All genes,except PR1.2 and PR9, had significantly higher expression levels in the incompatible interaction than the compatible interaction. PR1 had the highest expression level (7.81).At 7 dai, all genes, except PR2 and PR5, were significantly induced in the incompatible interaction, while only PR3 was significantly induced in the compatible interaction compared to the mock inoculation. PR1, PR1.2, PR3 and PR10 had significantly higher expression levels in the incompatible interaction than in the compatible interaction. PR1.2 had the highest expression level (10.48) in the incompatible interaction. At 14 dai, only PR9 was significantly induced in the compatible interaction, but PR3, PR4 and PR9 were significantly induced in the incompatible interaction and the expression levels of the three genes were significantly higher in the incompatible interaction than in the compatible interaction.

3.3. Expression of PR protein genes in non-race specific HTAP resistance

As there was no a compatible interaction for each of the non-race specific HTAP resistance genes, the expression levels of the PR protein genes were compared with those in the mock inoculation and compared to each other for their relative levels of induction by Pst inoculation. The expression levels are presented in Table 1.

Yr52At 0 dai, PR1.2, PR3, PR5 and PR9 were significantly induced by the Pst inoculation compared to the mock inoculation (Table 1). At 1 dai, all genes, except PR9 and PR10, were significantly induced, and PR5 had the highest expression level (14.03). At 2 dai, all genes,except PR9, were significantly induced, and PR1 had the highest expression level (11.65). At 7 dai, only PR1.2 was significantly induced. In contrast, all genes, except PR1,were significantly induced with PR2 the highest (14.30) at 14 dai.

Fig. 4 Quantitative real-time PCR (qRT-PCR) analyses of expression of eight PR protein genes in wheat line of Yr76 inoculated with races PSTv-4 (compatible) and PSTv-37 (incompatible) of Puccinia striiformis f. sp. tritici. Three independent biological replications were performed. All data were normalized to the β-actin expression level. The relative expression was expressed as fold change relative to the mock inoculated plants. Relative gene expression was calculated using the comparative 2-ΔΔCT method.The ＊ symbol represents significant difference (P≤0.05) according to analysis of variance. Bars represent standard deviation among the biological replicates.

Yr59None of the PR protein genes were significantly induced at 0 and 1 dai (Table 1). At 2 dai, PR2, PR5 and PR9 were significantly induced, and PR5 had the highest expression level (4.93). At 7 dai, five genes, PR1.2, PR3,PR4, PR5 and PR10, were significantly induced, and PR1.2 had the highest expression level (5.00). At 14 dai, PR1,PR2, PR3, PR4 and PR9 were significantly induced, and PR9 had the highest expression level (21.04).

Yr62PR4, PR5 and PR10 were significantly induced(3.79-5.73) at 0 dai (Table 1). At 1 dai, PR3, PR4, PR5,PR9 and PR10 were significantly induced with PR4 having the highest expression (10.38). At 2 dai, all genes were significantly induced at the levels ranging from 2.91(PR10) to 17.97 (PR5). At 7 dai, PR1, PR5 and PR9 were significantly induced with PR9 the highest (8.55). At 14 dai,PR5, PR9 and PR10 were significantly induced, and PR9 had the highest expression level (3.80).

Yr7BAt 0 dai, PR1, PR3 and PR9 were significantly induced with PR9 the highest (4.82) (Table 1). At 1 dai, all of the genes were significantly induced at the levels ranging from 5.95 (PR10) to 40.30 (PR9). Similarly, all of the genes were significantly induced at 2 dai, with the highest expression level (25.72) observed for PR5. At 7 dai, all genes, except PR1.2, were significantly induced, and PR4 had the highest expression level (7.90). At 14 dai, five genes, PR1, PR1.2,PR2, PR3 and PR4 were significantly induced, and among these PR1.2 had the highest expression level (3.69).

The expression patterns of the PR protein genes were quite different among the four HTAP resistance gene lines(Table 1). In the Yr52 lines, highly expressed (＞10.0 fold)PR protein genes occurred at 1 dai (PR1.2, PR2, PR3 and PR5), 2 dai (PR1 and PR4) and 14 dai (PR2 and PR10).In the Yr59 line, the highest expression levels were mainly observed at 14 dai (PR1 and PR9). In contrast, in the Yr62 line, the highest expression occurred at 2 dai for all PR protein genes except PR10. Similar to the Yr52 line, in the Yr7B line, the majority of the PR protein genes highly expressed (＞10.0 fold) at 1 and 2 dai, but none of the genes had expression levels greater than 10 folds at both 7 and 14 dai.

4. Discussion

Pathogen-related proteins play important roles in plant defense against pathogen infection and also elicit resistance.The accumulation of PR protein gene transcripts is one of the best-characterized plant defense responses in many other plant pathosystems. However, PR proteins had not been well characterized for wheat-Pst interactions. In the present study, we determined the involvement of eight PR protein genes in race-specific all-stage resistance controlled by four individual genes (YrTr1, YrExp2, YrSP and Yr76)and in non-race specific HTAP resistance controlled by four genes independently (Yr52, Yr59, Yr62 and Yr7B). Wefound that all of the tested PR protein genes were involved in resistance controlled by different resistance genes, but different PR protein genes were induced at different stages of infection.

Table 1 Expression levels of pathogen-related (PR) protein genes in wheat near-isogenic lines with Yr52, Yr59, Yr62 and YrPI182103-7BL (Yr7B) for non-race specific high-temperature adult-plant resistance at different days after inoculation (dai) with race PSTv-37 of Puccinia striiformis f. sp. tritici in adult-plant stage

4.1. Expression of PR proteins in plant-pathogen interactions

PR1 proteins have been used as markers for enhanced defense conferred by pathogen-induced systemic acquired resistance (SAR) in various plants (Van Loon et al. 2006b;Hou et al. 2012; Gkizi et al. 2016). As expression of this gene is salicylic acid (SA)-responsive, it is used as an indicator for the SA pathway (Gkizi et al. 2016). Li et al.(2016) reported the involvement of the wheat PR1 protein in the Lr35-mediated adult-plant resistance to leaf rust caused by Puccinia triticina. High accumulation of PR1 transcripts was also reported in wheat lines resistant to Fusarium graminerum and Mycosphaerella graminicola (Pritsch et al.2000; Ray et al. 2003). In the present study, the highest expression of PR1 was observed in the incompatible interaction for YrExp2 at 1 dai, YrTr1 and Yr76 at 2 dai,and YrSP at 7 dai, showing that PR1 was involved in racespecific all-stage resistance controlled by different Yr genes.The highest expression of PR1 was occurred for Yr7B at 1 dai, Yr52 and Yr62 at 2 dai and Yr59 at 14 dai. The results showed that PR1 is involved in both race-specific all-stage resistance and non-race specific HTAP resistance.

The PR1.2 protein is homologous to PR1 proteins from other plant species. It represents the first described monocot PR1 gene with constitutive and pathogen-inducible expression in wheat leaves, but does not respond to activators of systemic acquired resistance (Molina et al.1999). Soltanloo et al. (2010) reported that the transcript of PR1.2 was induced significantly at 7 dai in wheat variety Sumai 3 resistant to F. graminearum. In the present study, the highest expression of PR1.2 was detected in the incompatible interaction for races-specific all-stage resistance genes YrTr1 at 1 dai, YrExp2 and YrSP at 2 dai, and Yr76 at 7 dai. For HTAP resistance, the highest expression of PR1.2 was observed for Yr52 and Yr7B at 1 dai, Yr62 at 2 dai andYr59 at 7 dai. The results showed that PR1.2 was involved in both all-stage resistance and HTAP resistance controlled by all tested Yr genes, but its expression pattern differed greatly from those of PR1 and other genes.

PR2 is an β-1,3-glucanase, a group of enzymes playing a major role in plant defense and general stress responses through the regulation of callose deposition(Levy et al. 2007). β-1,3-Glucanases are also capable of hydrolyzing β-1,3-glucans found in the cell wall of fungi.They can degrade the cell wall of the pathogen or disrupt its deposition, which lead to pathogen death (Mauch et al. 1988). The deposition of cell wall releases cell wall fragments that act as elicitors of active host defense response (Yoshikawa et al. 1993). β-1,3-Glucanase gene transcripts were detected on tobacco leaves within 24 to 48 h and increased up to 21-fold by bacterium Pseudomonas syringae (Alonso et al. 1995). High expression of PR2 was reported in wheat adult-plant resistance against P. triticina(Casassola et al. 2015) and in resistant wheat plants infected with F. graminearum (Li et al. 2001) or M. graminicola (Ray et al. 2003). In the present study, the highest expression of PR2 was detected in the incompatible interaction for racespecific all-stage resistance controlled by YrTr1 at 1 dai,YrSP and Yr76 at 2 dai, and YrExp2 at 7 dai. The highest expression of PR2 in HTAP resistance was detected for Yr52 and Yr7B at 1 dai, Yr62 at 2 dai and Yr59 at 14 dai. The results are generally in consistent with the previous reports that β-1,3-glucanase genes express as early as 6 h after inoculation and the highest expression occurred at 1 and 2 dai in all-stage resistance controlled by Yr5 (Coram et al.2008b) and other genes (Coram et al. 2010; Chen et al.2013). They also reported the induction of PR2 expression in HTAP resistance conferred by Yr39 and other genes at 2 dai (Coram et al. 2008a; Chen et al. 2013). The present study and previous studies clearly show that the PR2 protein contributes to both race-specific all-stage resistance and non-race specific HTAP resistance to stripe rust.

PR3 and PR4 are endochitinases that break bonds between the C1 and C4 of two consecutive N-acetylglucosamines of chitin, a main component of the cell wall in fungi (Collinge et al. 1993). PR-4 proteins from wheat were shown to inhibit spore germination and hyphal growth in in vitro assays (Caruso et al. 2001). Chitinase was also reported to be involved in wheat resistance to stripe rust (Coram et al. 2008a, b; Chen et al. 2013). In the present study, the highest expression of PR3 was observed in race-specific all-stage resistance controlled by Yr76 at 0 dai, YrTr1 and YrSP at 1 dai and YrExp2 at 2 dai; in the HTAP resistance controlled by Yr52 at 1 dai, Yr62 and Yr7B at 2 dai and Yr59 at 14 dai. Similarly, the highest expression of PR4 was observed in race-specific resistance controlled by Yr76 at 0 dai, YrSP at 1 dai, YrExp2 at 2 dai and YrTr1 at 2 and 7 dai. The highest expression of PR4 in HTAP resistance was detected for Yr7B at 1 and 2 dai,Yr52 and Yr62 at 2 dai and Yr59 at 7 dai. The results show the involvement of both PR3 and PR4 proteins in both racespecific all-stage resistance and non-race specific HTAP resistance.

PR5, encoding a thaumatin-like protein (TLP), plays an active role in resistance biotic and abiotic stresses in cereals (Rebman et al. 1991; Pritsch et al. 2000). TLP proteins have been reported to be produced in plants under either biotic or abiotic stresses (Walter et al. 2010). Several studies revealed that PR5 proteins interact specifically with (1,3)-β-D-glucans of the fungal cell wall (Osmond et al. 2001). The binding of PR5 proteins to (1,3)-β-D-glucans result in poor fungal cell wall assembly during hyphal extension, resulting in increased plasma membrane permeability which reduces the growth and development of the fungus in the host (Edreva 2005). Overexpression of PR5 was shown to enhance rice resistance to Rhizoctonia solani (Datta et al. 1999). Barley TLP gene was induced by F. graminearum infection (Geddes et al. 2008). Wheat PR5 induced by the leaf rust pathogen was significantly increased from the second-leaf stage through mature stage with a peak at the mature stage (Li et al. 2016). In the present study, the highest expression of PR5 was detected in race-specific all-stage resistance controlled by YrSP at 1 dai and YrTr1, YrExp2 and Yr76 at 2 dai; and in the HTAP resistance controlled by Yr52 at 1 dai and Yr59,Yr62 and Yr7B at 2 dai. The results generally agreed with Wang et al. (2010) that reported high expression of PR5 at 1-2 dai in the incompatible interaction at seedling stage. In our experiments of non-race specific HTAP resistance, this gene was significantly induced at 0 or 1 dai, and remained at significant levels until 14 dai, except for the Yr59 line in which the gene was not significantly induced until 2 dai.The highest expression level in HTAP resistance (4.93-25.72) was generally higher than in all-stage resistance(5.94-16.18). This is in contrast to the previous report of Chen et al. (2013). In the previous study, the significant transcription of PR5 was reported only in all-stage resistance controlled by Yr5, but not in the same type of resistance controlled by other Yr genes tested, and also not in HTAP resistance conferred by Yr18, Yr29 or Yr39. The difference may be due to the different Yr genes tested, and also the limited time points (1 and 2 dai) and microarray technique in the previous studies (Coram et al. 2008a, b, 2010; Chen et al. 2013). Even in the present study, we did not detected significant induction of PR5 in the Yr59 line until 2 dai at a relatively low level (4.93). However, the present results are in agreement with a previous report that the expression level of PR5 was high in adult rice leaves infected by Xanthomonas oryzae pv. oryzae (Hou et al. 2012).

PR9 is a specific type of peroxidase that acts in cell wall reinforcement by catalyzing lignification and enhance resistance against pathogens by producing reactive oxygen species (ROS) (Odjakova and Hadjiivanova 2001). The production of ROS is central to the defense mounted by plants in reaction to challenges from pathogens (Apel and Hirt 2004). Genes participating in oxidative burst are highly induced at 6-8 hours after inoculation, which forms the first step in triggering plant defense (Wang et al. 2007).The first burst occurs within minutes in both susceptible and resistant interactions, whereas the second, sustained burst occurs within a few hours of infection and only in a resistant interaction (Baker and Orlandi 1995). Casassola et al. (2015) reported high expression of PR9 in wheat plants resistant to leaf rust at 2 dai. Dmochowska-Boguta et al.(2015) identified 12 clones representing transcripts involved in oxidation reduction processes in wheat plants inoculated with P. triticina. Through cytological and histological studies of adult-plant resistance of wheat to stripe rust, Zhang et al.(2012) reported that the first phase of ROS occurred at 1 and 2 dai, which is associated with the beginning of haustorium formation and subsequent hyposensitive response (HR)in infected host cells. The second phase of ROS burst occurring at 5 dai was associated with an increasing number of necrotic host cells and the formation of secondary hyphae surrounding the infected plant cells. Microarray and cDNAAFLP studies showed that the transcription levels of wheat peroxidase genes were high at 0.5 and 1 dai in both all-stage resistance and HTAP resistance with Pst (Coram et al.2008a, b, 2010; Wang et al. 2010; Chen et al. 2013). High transcript accumulation of a peroxidase gene was reported for adult plant resistance to stripe rust in wheat cultivar Xingzi 9104 through suppressive subtractive hybridization(Huang et al. 2013) and transcriptome sequencing (Hao et al. 2016). Transcripts of peroxidases significantly increased from 1 to 2 dpi in adult plants of Xingzi 9104. In the present study, PR9 was significantly induced at 0 dai for all-stage resistance gene Yr76 and HTAP resistance genes Yr52 and Yr7B; and at 1 and/or 2 dai for all resistance genes.The highest expression of PR9 was detected for Yr76 at 0 dai, YrTr1 at 1 dai, YrSP at 2 dai and YrExp2 at 7 dai. In HTAP resistance, the highest expression was observed for Yr7B at 1 dai, Yr62 at 2 dai and Yr52 and Yr59 at 14 dai. The highest expression level varied greatly among resistance genes, ranging from 3.02 (YrExp2) to 11.66 (YrSP) for allstage resistance and from 3.64 (Yr52) to 40.30 (Yr7B) for HTAP resistance.

PR10 encodes a ribonuclease (RNase)-like protein(Park et al. 2004). PR-10 proteins have a wide distribution throughout the plant kingdom, but its functions are not clear although some subclasses are allergic to humans(Fernandes et al. 2013). A recent study reported that overexpression of PR10 in rice plants reduced susceptibility to rice blast and enhanced tolerance to salt and drought(Wu et al. 2016). Expression of PR10 was reported to be high in race-specific all-stage resistance controlled by several genes, such as Yr1, Yr5, Yr8 and Yr10 (Coram et al. 2008b, 2010; Chen et al. 2013). In the present study,the highest expression of PR10 was consistently observed in race-specific all-stage resistance controlled by all YrTr1,YrExp2, YrSP and Yr76 at 1 dai. In contrast, the occurrence of the highest expression of PR10 in HTAP resistance varied greatly, at 0 dai for Yr62, 2 dai for Yr7B, 7 dai for Yr59 and 14 dai for Yr52.

The present study showed that all of the tested PR protein genes were involved in both race-specific all-stage resistance and non-race specific HTAP resistance. However,the expression patterns and the levels of expression varied greatly among different resistance genes. The results indicate that the different PR protein genes work in concert leading to the different types of resistance controlled by different Yr genes.

4.2. PR protein transcripts with SA and JA pathways

PR1, PR1.2, PR2 and PR5 can be activated by salicylic acid (SA) application in wheat (Niu et al. 2007), whereas jasmonate (JA) activates PR3, PR4, PR9 and PR10 (Duan et al. 2014; Wu et al. 2016). Thus, these PR proteins can serve as markers for the two different pathways for responding to pathogen interactions. In our study, all of these genes expressed, but at different time points of both all-stage resistance and HTAP resistance after inoculation,indicating that both SA and JA pathways are involved in the different types of resistance. This finding was in agreement with previous reports for resistance to other diseases. Both SA and JA are known to induce wide-spectrum disease resistance in plants, and co-induction of JA and SA signaling pathways has been reported in various plant pathosystems(Pozo et al. 2004; Takahashi et al. 2004). Specifically,Duan et al. (2014) reported that both SA and JA mediated resistance signaling molecules are involved in wheat resistance to powdery mildew.

4.3. Differences of PR protein gene expression between compatible and incompatible interactions

In the present study, expression of PR protein genes was found in both compatible and incompatible interactions,indicating that incompatible and compatible interactions share signaling pathways. However, when the same gene was induced in both compatible and incompatible interactions, the expression of the gene was most likely higher in the incompatible interaction than the compatible interactions. Also, more PR protein genes were induced in the incompatible interactions than compatible interactions.These results agreed with previous studies of wheat response to stripe rust (Coram et al. 2008a, b). Thus, a resistant plant is capable of rapidly deploying a wide variety of defense responses at high levels that prevent or reduce pathogen colonization and growth. In contrast, a susceptible plant exhibits much weaker and slower responses that fail to restrict pathogen colonization and growth.

5. Conclusion

In the present study, we determined expression levels of eight PR protein genes in eight Yr genes controlling either race-specific all-stage resistance or non-race specific HTAP resistance at five major time points during the Pst infection process. We found that all eight PR protein genes of wheat are involved in resistance to the stripe rust pathogen, no matter which type of resistance or which Yr resistance gene controls resistance. The results indicate the involvement of both SA and JA signaling pathways in both race-specific resistance and non-race specific resistance to stripe rust.Thus, both durable resistance and non-durable resistance utilize some common defense mechanisms against the pathogen infection. However, different PR protein genes express at different stages and at different levels during the pathogen infection process. The different signaling pathways may regulate different defense genes, which work in concert to produce final results of different levels of resistance. The expression level of PR protein genes varied significantly at different time points and among different Yr gene lines. This finding indicates that not a single PR protein gene is involved in resistance, however, any type of the resistance or any Yr gene-controlled resistance is contributed by different defenserelated genes that may regulated by different signaling pathways. The results of the present study greatly improve the understanding of the molecular mechanisms underlying wheat resistance to stripe rust.

Acknowledgements

This study was supported by the U.S. Department of Agriculture, Agricultural Research Service (2090-22000-018-00D), the Washington Grain Commission, USA (13C-3061-5665) and the Idaho Wheat Commission, USA (13C-3061-5665; 13C-3061-4232). The Fulbright fellowship to Sumaira Farrakh is highly appreciated. Mention of trade names or commercial products in this publication is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S.Department of Agriculture (USDA). USDA is an equal opportunity provider and employer.

Appendicesassociated with this paper can be available on http://www.ChinaAgriSci.com/V2/En/appendix.htm