In the previous study (Zhao et al., 2019), the urediospores of the Pst physiological race CYR32 were irradiated with UV-B radiation under a radiation dose (the radiation intensity was 250 μw/cm², and the radiation time was 95 min) for which the relative lethal rate of urediospores was approximately 90% (i.e., the relative germination rate of urediospores was approximately 10%); then the irradiated urediospores were inoculated on the seedlings of Guinong 22 to screen virulence-mutant strains. Finally, two virulence-mutant strains named CYR32-5 and CYR32-61, with stable infection types on Guinong 22 throughout four successive generations, were obtained. In this study, the original strain of the Chinese physiological race CYR32 of Pst and the two virulence-mutant strains (CYR32-5 and CYR32-61) were used. The wheat cultivar Mingxian 169, which is susceptible to all known Chinese physiological races of Pst, was used for Pst multiplication. Multiplication of the urediospores of each Pst strain was conducted in an artificial climate chamber (environmental parameters: light time, 12 h/d; light intensity, 10,000 lux; temperature, 11–13°C; and relative humidity, 60–70%) using the method described by Cheng et al. (2014).

Fresh urediospores of each of the three strains CYR32, CYR32-5, and CYR32-61 were collected from the diseased leaves of Mingxian 169 seedlings. For each strain, a spore suspension with a concentration of 2.5 mg/mL was prepared with 0.2% Tween-80 solution. Then, 4 mL of the spore suspension was transferred into a Petri dish (16 cm in diameter) containing 1% water agar medium. The urediospores (10 mg) contained in the suspension were evenly scattered in the Petri dish. When the suspension was almost dry, the Petri dish was sealed with plastic wrap and incubated for 10 h at 9°C in a dark environment. After incubation, a small agar block was picked up and placed on a glass slide for microscopic observation. At least 300 urediospores were checked, and the germination rate of the urediospores was recorded. A urediospore with a germ tube longer than half of the diameter of the urediospore was regarded as germinated. If the germination rate of the urediospores was more than 90%, the urediospores and germ tubes on the surface of the water agar medium were gently collected with a cover glass. For each Pst strain, the urediospores and germ tubes collected from five Petri dishes (approximately 50 mg in total) were treated as a sample.

In total, two samples of CYR32, three samples of CYR32-5, and three samples of CYR32-61 were acquired for protein extraction. Grinding tools were pre-cooled with liquid nitrogen, and then each of the prepared samples was pulverized with liquid nitrogen in a mortar and pestle. The pulverized sample was suspended in a trichloroacetic acid (TCA) and acetone solution (TCA:acetone = 1:10, w/v; pre-cooled at −20°C) and then precipitated for 2 h at −20°C. After centrifugation at 4°C under 20,000 × g for 30 min, the supernatant was discarded, and the precipitate was suspended in pre-cooled pure acetone, and then precipitated for 30 min at −20°C prior to centrifugation at 4°C under 20,000 × g for another 30 min. This process was repeated several times until the precipitate was substantially white. The precipitate was resuspended in a lysis buffer (8 M urea, 30 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, 1 mM phenylmethanesulfonyl fluoride, 2 mM ethylenediaminetetraacetic acid, and 10 mM dithiothreitol) and sonicated for 5 min (pulse on 2 s, pulse off 3 s, power 180 W), followed by centrifugation at 20,000 × g for 30 min. The supernatant was collected and added to dithiothreitol to a final concentration of 10 mM. After incubation in a water bath at 56°C for 1 h, iodoacetamide was quickly added, to a final concentration of 55 mM. After incubation for 1 h in a dark environment, the supernatant solution was supplemented with a fourfold volume of pre-chilled acetone, and then precipitated at −20°C for more than 3 h. After centrifugation at 4°C under 20,000 × g for 30 min, the precipitate was dissolved in 400 μL of digestion buffer to a final concentration of 50% triethylammonium bicarbonate (TEAB) and 0.1% sodium dodecyl sulfate (SDS). Subsequently, sonication was performed for 3 min with a pulse on time of 2 s and a pulse off time of 3 s at an ultrasonic power of 180 W, and centrifugation was performed for 30 min at 4°C and 20,000 × g. Finally, the supernatant was obtained, and the protein concentration was quantified using the Bradford assay (Supplementary Table 1).

From each sample solution, 100 μg of protein was transferred to a new clean centrifuge tube. The TEAB solution (0.1% SDS) was added to make the protein solution of each sample in the new tube up to the same volume. After adding 3.3 μL of trypsin (1 μg/μL) to the tube for protein digestion, the solution was incubated in a water bath at 37°C for 24 h. Subsequently, the tube was supplemented with 1 μL of trypsin (1 μg/μL), and the mixture in the tube was incubated in a water bath at 37°C for 12 h. After lyophilization of the mixture, 30 μL of TEAB (ddH₂O:TEAB = 1:1, v/v) was added to the tube to dissolve the peptides. Using an iTRAQ^® Reagent-8Plex Multiplex Kit (Applied Biosystems, Foster City, CA, United States), peptides from the eight samples were labeled with the iTRAQ tags as follows: CYR32 (tags 113 and 114), CYR32-5 (tags 115, 116, and 117), and CYR32-61 (tags 118, 119, and 121), respectively (Supplementary Table 1).

The solution containing labeled peptides of each sample was diluted tenfold with buffer A (25% acetonitrile (ACN), 10 mM KH₂PO₄, pH 3.0), and the pH was adjusted to 3.0 with phosphoric acid. After centrifugation for 15 min at 15,000 × g, the labeled peptides in the supernatant were fractionated using a Phenomenex Luna SCX column (250 mm × 4.60 mm, 100Å) with a high performance liquid chromatography (HPLC) system (Thermo Fisher Scientific, Waltham, MA, United States) at a flow rate of 1 mL/min. The system was equilibrated for 10–20 min with buffer A at a flow rate of 1 mL/min, prior to strong cation exchange fractionation. The HPLC gradient was as follows: 0–45 min, 100% buffer A; 45–46 min, 0–5% buffer B (25% ACN, 2 M KCl, 10 mM of KH₂PO₄, pH 3.0); 46–66 min, 5–30% buffer B; 66–71 min, 30–50% buffer B; 71–76 min, 50% buffer B; 76–81 min, 50–100% buffer B; 81–91 min, 100% buffer B. Absorbance was recorded at 214 nm. The eluted peptides were desalted with a Strata-X C18 column (Phenomenex, Torrance, CA, United States), and then dried by vacuum centrifugation at a low temperature.

The desalted, dried peptides were resuspended with solvent A (0.1% formic acid (FA) in H₂O), transferred to an Acclaim PePmap C18-reversed phase column (75 μm × 2 cm, 3 μm, 100 Å, Thermo Scientific), and then separated using a Dionex Ultimate 3000 Nano LC system with a C18 reversed phase column (75 μm × 10 cm, 5 μm, 300 Å, Agela Technologies) at a flow rate of 400 nL/min. The mobile phases were solvent A and solvent B (0.1% FA in ACN). The elution gradient was as follows: 0–10 min, 5% solvent B; 10–40 min, 5–30% solvent B; 40–45 min, 30–60% solvent B; 45–48 min, 60–80% solvent B; 48–55 min, 80% solvent B; 55–58 min, 80–5% solvent B; 58–65 min, 5% solvent B. Subsequently, 16 pre-isolated and purified components of peptides were detected using a Q-Exactive mass spectrometer (Thermo Fisher Scientific, Waltham, MA, United States) with set parameters as follows: polarity, positive ion mode; MS scan range, 350–2000 m/z; MS/MS scan resolution, 17,500; capillary temperature, 320°C; ion source voltage, 1,800 V; MS/MS acquisition modes, higher collision energy dissociation; normalized collision energy, 28.

The acquired raw mass spectrometry data were processed using the PD (Proteome Discoverer 1.3; Thermo Fisher Scientific, San Jose, CA, United States) software with the following parameters: mass range of parent ion, 350–6,000 Da; minimum number of peaks in MS/MS spectrum, 10; signal-to-noise ratio (S/N) threshold, 1.5. The spectra extracted using the PD software were searched against the Basidiomycota_UniProt database using Mascot 2.3.0 (Matrix Science, London, United Kingdom) with the following identification parameters: fixed modification, carbamidomethyl (C); variable modification, oxidation (M), Gln→Pyro-Glu (N-term Q), iTRAQ 8 plex (K), iTRAQ 8 plex (Y), iTRAQ 8 plex (N-term); peptide tolerance, 15 ppm; MS/MS tolerance, 20 mmu; max missed cleavages, 1; enzyme, trypsin. Based on the Mascot search results and the extracted spectra, protein quantitative analysis was performed using the PD software with the following parameters: protein ratio type, median; minimum peptides, 1; normalization method, median; P-value, < 0.05; ratio, ≥ 1.2. In this study, proteins with at least two unique peptides, scores more than 70, P-values less than 0.05, and fold changes more than 1.2 or less than 0.83 were identified as DEPs between the different Pst strains. The fold changes of the up-regulated DEPs were more than 1.2, and those of the down-regulated DEPs were less than 0.83. For the convenience of expression, CYR32-5 vs. CYR32, CYR32-61 vs. CYR32, or CYR32-61 vs. CYR32-5 was used to represent the combination in which the proteome of the former was compared with that of the latter to screen DEPs.

Functional annotations of the proteins were carried out using the Blast2Go program¹ (Conesa et al., 2005). The COG annotations of proteins were conducted using WebMGA² (Wu S. T. et al., 2011). The annotations of the identified DEPs at the biological pathway level were performed using the KEGG database³. The KOBAS 2.0 software was used to annotate pathways by comparing similar protein sequences (Xie et al., 2011). Signal peptides were predicted and analyzed using the SignalP 5.0 program⁴ (Armenteros et al., 2019).

The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium⁵ via the iProX partner repository (Ma et al., 2019) with the dataset identifier PXD018136.

Using the method described above, 20 mg of the germinated urediospores and germ tubes of each of the three strains CYR32, CYR32-5, and CYR32-61 were collected and treated as a sample. Total RNA from each sample was extracted using the RNeasy kit (Omega) according to the manufacturer’s instructions, and then genomic DNA contaminants were removed by DNase I (RNase-free) treatment. The synthesis of the first-strand cDNA was performed by reverse transcription using the PrimeScript^TM RT Master Mix kit (Takara, Dalian, China). Specific primers were designed for the selected DEPs according to their corresponding coding sequences using NCBI Primer-BLAST⁶ (Table 1). Pst β-tubulin gene TUBB (GenBank accession No. EG374306) was used as the reference gene, and the corresponding primer reported by Huang et al. (2012) (as shown in Table 1) was used in this study. SYBR Green real-time fluorescence quantitative PCR assays were performed using an Applied Biosystems ABI Model 7500 Real Time PCR system. The qRT-PCR reactions were carried out in a total 20 μL volume of reaction mixture containing 10 μL 2 × SYBR Premix DimerEraser, 0.6 μL forward primer (10 μM), 0.6 μL reverse primer (10 μM), 2.0 μL template DNA, 6.4 μL sterile purified water, and 0.4 μL 50 × ROX Reference Dye or Dye II. The amplification procedure was as follows: pre-denaturation at 95°C for 30 s, followed by 40 cycles of 95°C for 5 s, 55°C for 30 s, and 72°C for 30 s. The relative gene expression levels were calculated using the 2^−ΔΔC_T method described by Livak and Schmittgen (2001). The experiment was carried out with three biological replicates.

A total of 337,731 spectra were acquired from the eight samples of the three Pst strains, and 52,290 spectra were matched by searching against the Basidiomycota_UniProt database using Mascot 2.3.0 (false discovery rate < 1%). In total, 8,882 peptides and 2,271 proteins were obtained (Supplementary Table 2).

The results of screening the DEPs in the urediospores and germ tubes after germination in three combinations, CYR32-5 vs. CYR32, CYR32-61 vs. CYR32, and CYR32-61 vs. CYR32-5, are shown in Figure 1. As shown in Figure 1, 59 DEPs were obtained in the combination CYR32-5 vs. CYR32, of which 35 proteins were up-regulated and 24 proteins were down-regulated, and the DEPs E3JUB4 and Q9C1C1 (3.39%) were predicted to contain signal peptides. Seventy-four DEPs were obtained in the combination CYR32-61 vs. CYR32, of which 43 proteins were up-regulated and 31 proteins were down-regulated, and the DEPs Q9C1C1, R7SW21, E3JQF7, E3JUB4, E3JSA3, E3KXP2, and E3K3X8 (9.46%) were predicted to contain signal peptides. Sixty-four DEPs were obtained in the combination CYR32-61 vs. CYR32-5, of which 40 were up-regulated and 24 were down-regulated, and the DEPs E3K302, E3KDD8, E3JSA3, and E3JQF7 (6.25%) were predicted to contain signal peptides. In total, 144 DEPs were obtained in the three comparison combinations, of which four were common in CYR32-5 vs. CYR32, CYR32-61 vs. CYR32, and CYR32-61 vs. CYR32-5. The DEPs that were predicted to contain signal peptides may be secretary proteins or membrane proteins.

All obtained DEPs were subjected to COG functional annotation, and then assigned to 25 categories that were represented by A–Z (Figure 2). Among all the DEPs, seven were annotated with multiple functions, and 22 were not annotated and had no annotation information. The four main functional categories were as follows: [J] translation, ribosomal structure and biogenesis (31, 21.68%), [O] posttranslational modification, protein turnover, chaperones (24, 16.78%), [C] energy production and conversion (17, 11.89%), and [U] intracellular trafficking, secretion, and vesicular transport (11, 7.69%). The detailed functional annotations of the up-regulated DEPs and the down-regulated DEPs are shown in Tables 2, 3, respectively. The acquired DEPs were classified into four major functional categories, including “information storage and processing,” “cellular processes and signaling,” “metabolism,” and “poorly characterized.” Of the up-regulated DEPs, 14, 43, 27, and 4 were involved in “information storage and processing,” “cellular processes and signaling,” “metabolism,” and “poorly characterized,” respectively, and 18 had no annotation information. Of the down-regulated DEPs, 23, 15, 18, and 5 were involved in “information storage and processing,” “cellular processes and signaling,” “metabolism,” and “poorly characterized,” respectively, and 7 had no annotation information.

The biological pathways that the DEPs were involved in were investigated using the KEGG database, and 5, 7, and 7 significant enriched KEGG pathways (P < 0.05) were acquired for the DEPs of CYR32-5 vs. CYR32, CYR32-61 vs. CYR32, and CYR32-61 vs. CYR32-5, respectively (Figure 3 and Supplementary Table 3). The DEPs of CYR32-5 vs. CYR32 were involved in pathways including metabolic pathways, biosynthesis of secondary metabolites, carbon metabolism, ribosome, and pyruvate metabolism. The DEPs of CYR32-61 vs. CYR32 were mainly involved in the carbon metabolism pathway, calcium signaling pathway, and melanogenesis pathway. The DEPs of CYR32-61 vs. CYR32-5 were mainly involved in the ribosome pathway, melanogenesis pathway, and calcium signaling pathway.

To confirm the differences in protein abundance in each of the three combinations including CYR32-5 vs. CYR32, CYR32-61 vs. CYR32, and CYR32-61 vs. CYR32-5, after germination of the urediospores, qRT-PCR was used to investigate the relative expression levels of the genes encoding the DEPs identified by iTRAQ analysis. The relative expression levels of the genes encoding six proteins (G9B235, F4S8X5, V2XWY1, Q9C1C1, F4S0Z3, and F4RWN9) in CYR32-5 vs. CYR32, seven proteins (G9B235, F4S8X5, V2XWY1, Q9C1C1, E3L0W8, D4QFJ2, and J6EXB0) in CYR32-61 vs. CYR32, and one protein (F4S8X5) in CYR32-61 vs. CYR32-5 were determined, and the results are shown in Figure 4. In CYR32-5 vs. CYR32, the relative expression levels of the genes encoding Q9C1C1, G9B235, F4S8X5, F4RWN9, and F4S0Z3 were not consistent with the corresponding protein abundance based on iTRAQ data, but the relative expression of the gene encoding V2XWY1 was consistent with the corresponding protein abundance. In CYR32-61 vs. CYR32, the relative expression levels of the genes encoding Q9C1C1 and J6EXB0 were inconsistent with the corresponding protein levels, and the genes encoding G9B235, F4S8X5, V2XWY1, E3L0W8, and D4QFJ2 were consistent with the corresponding protein levels. In CYR32-61 vs. CYR32-5, the relative expression level of the gene encoding F4S8X5 at the transcriptional level was consistent with the corresponding protein abundance obtained via iTRAQ analysis.