key: cord-0021471-u7dku2w4
authors: Omara, Reda Ibrahim; Nehela, Yasser; Mabrouk, Ola Ibrahim; Elsharkawy, Mohsen Mohamed
title: The Emergence of New Aggressive Leaf Rust Races with the Potential to Supplant the Resistance of Wheat Cultivars
date: 2021-09-16
journal: Biology (Basel)
DOI: 10.3390/biology10090925
sha: 410f8bd2c27a86b553da039d08eaaf4b643642c5
doc_id: 21471
cord_uid: u7dku2w4

SIMPLE SUMMARY: The pathogen that causes wheat leaf rust, Puccinia triticina, possesses numerous aggressive races that can erode the resistant genes in its host. This study presents the recognition of the new physiological races of P. triticina, their distribution, and their resistance genes in wheat cultivars, which are critical for directing and improving wheat breeding programs for resistance to leaf rust. Winds often transport the pathogen’s initial inoculum from one country to another. Our findings trigger an alert to the whole world about developing races capable of supplanting leaf rust resistance. ABSTRACT: Characterization of the genetic structure and the physiological races of Puccinia triticina is a growing necessity to apply host genetic resistance against wheat leaf rust as a successful control strategy. Herein, we collected and identified about 130 isolates of P. triticina from 16 Egyptian commercial wheat cultivars grown at different locations, over two seasons (2019/2020 and 2020/2021). The 130 isolates of P. triticina were segregated into 17 different physiological races. TTTST and TTTKS were the most common virulent races, whereas TTTST and MTTGT were the most frequent races. The races were classified into three groups, based on their distinct DNA band sizes (150 bp, 200 bp, and 300 bp) after RAPD analysis. The new wheat cultivars (Sakha-94, Sakha-95, and Shandweel-1) infected with the most virulent race (TTTST), Gemmeiza-12, and Misr-3 were resistant to all physiological races. The resistance of these cultivars was mostly due to the presence of Lr19- and Lr28-resistant genes. Our results serve as a warning about emerging aggressive races capable of supplanting resistance to leaf rust, and help in the understanding of the pathotype–cultivar–location association and its role in the susceptibility/resistance of new wheat cultivars to P. triticina.

Bread wheat (Triticum aestivum L.) is a globally important cereal crop. It is widely produced, and leads the grain market along with corn and rice [1] . Wheat production in Egypt has increased during the last decade, from 7.2 million metric tons (MMT) in 2010 to 8.9 MMT in 2020 (approximately 23.6% comprehensive increase) [2, 3] . Briefly, the most recent forecast of the USDA-Foreign Agricultural Service (FAS-Cairo) estimated that Egypt's wheat production would reach 9.0 MMT during the 2021 marketing year, which reflects about a 1.12% increase compared to 8.9 MMT in 2020 [4] . However, they also forecast an approximately 1.53% increase in Egypt's wheat imports (from 13.0 MMT in 2020 to 13.2 MMT in 2021) [4] . population [25] . Therefore, identifying, characterizing, and developing new sources of resistance is a growing necessity [26] .

According to the study done by Kolmer and Liu [27] , virulence infection types of isolates to wheat lines with resistance genes varied substantially for a random amplified polymorphic DNA (RAPD) distance, indicating a general connection between virulence and RAPD phenotype. The similarity of P. triticina isolates' genetic backgrounds, as determined using RAPD markers, suggests that the observed variations in pathogenicity may have resulted from selection for particular virulence matching to resistance genes in wheat varieties cultivated in the area. The amount of variation found among isolates varied across pathogen groups, but the main differences were constant [28] .

In the current study, our objectives were to define the physiological races of leaf rust that overcome resistance in Egyptian commercial wheat cultivars and to understand the pathotype-cultivar-location association and its role in susceptibility/resistance of new wheat cultivars to P. triticina. Herein, we studied the geographic distribution of P. triticina populations on 16 Egyptian commercial wheat cultivars, in four main wheat-growing governorates in Egypt, during two successive seasons (2019/2020 and 2020/2021). The identification of current, as well as new, physiological races of P. triticina and their resistance genes in wheat cultivars is critical for directing and improving wheat breeding programs for resistance to leaf rust, as well as understanding field reactions of new cultivars to the most current P. triticina populations.

The experiments were performed in four different governorates, Kafrelsheikh, Beheira, Sharkia, and Alexandria, as well as in the leaf rust greenhouse at the Wheat Diseases Research Department, Plant Pathology Research Institute (PPRI), Agricultural Research Center (ARC), Giza, Egypt, during successive 2019/2020 and 2020/2021 seasons. In addition, molecular analysis was performed at the Molecular Biology Laboratory, Faculty of Agriculture, Cairo University.

An annual survey of leaf rust was performed to collect infected leaves showing typical symptoms. Wheat fields and the Egyptian Wheat Trap Rust Nursery (EWTRN) in the governorates of Kafrelsheikh, Beheira, Sharkia, and Alexandria were utilized for sampling. The Trap Nursery included 130 wheat entries which were planted in 2-m long rows and were 30 cm apart, with a seed rate of 5 g per entry. These entries included 16 wheat cultivars, monogenic lines for stripe, leaf, stem rusts, and highly susceptible varieties for the three rusts.

Samples were collected and kept in paper bags (6 × 20 cm). For each sample taken, all relevant information was recorded, including the date, governorate, cultivar, severity, and the collector's name. Collected samples (infected leaves) were preserved in an icebox, then left in their bags overnight at room temperature to reduce the humidity in the samples. Samples then were preserved in desiccators containing calcium chloride under 4 • C in a refrigerator [29] .

Samples with enough urediniospores were utilized to inoculate seedlings of the highly susceptible wheat variety "Morocco" at 7 days old. The seedlings were treated by spraying with water that contained a few drops of Tween-20 before being inoculated with a spatula. The infected plants were maintained in a growth chamber for 24 h at 100% humidity [29] . The plants were transferred to greenhouse benches to produce rust pustules. To collect sufficient urediniospores for inoculating the differential sets (Lrs), three single pustules were collected separately from each sample and used to inoculate the seedlings of a highly susceptible wheat variety (Morocco).

The procedure for identifying leaf rust races was based on inoculating isogenic lines (Lrs) with urediniospores of P. triticina [30] . The plant response was determined using this method on twenty lines separated into five sets of four lines. The first set included isogenic Lr lines 1, 2a, 2c and 3; the second-9, 16, 24 and 26; the third group-3Ka, 11, 17 and 30; the fourth-10, 18, 21 and 2b, and the fifth-Lr14b, Lr15, Lr36 and Lr42. Plants representing each rust agent isolate were labeled in letters based on a mixture of low infection type (L) and high infection type (H) responses. As a consequence, each pathotype is assigned a code consisting of five consonant letters from the English alphabet, ranging from B to T (Table S1 ).

At the seedling stage, the infection types for all near-isogenic lines were recorded at 12 days after the formation of pustules on near-isogenic lines, using a disease rating scale of 0 to 4 (Table S2 ) [31] . Based on the infection types developed by each line, the virulence patterns on different sets were assessed. Low infection was represented by infection types 0, 1, and 2, while high infection was represented by infection types 3 and 4 [30] .

At the adult stage, disease severity (DS%) was reported for the four governorates on 16 wheat cultivars, based on the percentage of leaf area covered with rust pustules. Field reaction of leaf rust infection types was classified into five categories: highly resistant (0), resistant (R), moderately resistant (MR), moderately susceptible (MS), and susceptible (S) [29] .

Genomic isolation was used to extract DNA from spores (17 races) according to the technique of Dellaporta et al. [32] . In a 25 µL reaction volume, a polymerase chain reaction was carried out using 2.5 µL of 50 ng/µL genomic DNA, 1 µL each primer (10 pmol, F&R), and 8 µL MQ H 2 O [33] . Electrophoresis of the amplification products was done at 100 V/1 h. The gel was stained with ethidium bromide after electrophoresis, and bands were seen with UV light and photographed using a Syngen UV visualizer. The tested primers' nucleotide sequence, and G + C percentage (Bioneer Company, Oakland, CA, USA) were utilized in RAPD experiment (Table S3 ). Table S4 against physiological races was conducted in the greenhouse of Wheat Diseases Dept., PPRI, ARC, Giza. Inoculation, incubation, and disease assessment were carried out as previously mentioned.

To evaluate leaf rust resistance genes, it is better to consider the number of susceptible responses over the total number of responses to the isolates. However, virulence percent was determined by recording the number of resistant responses over the total responses, as described by Green [34] , as follow: Sixteen Egyptian wheat cultivars, i.e., Gemmeiza-7, Gemmeiza-9, Gemmeiza-10, Gemmeiza-11, Gemmeiza-12, Sakha-93, Sakha-94, Sakha-95, Sids-1, Sids-12, Sids-13, Sids-14, Misr-1, Misr-2, Misr-3 and Shandweel-1, as well as two monogenic lines (Lr19 and Lr28) were used for identifying leaf rust resistance gene(s).

DNA was extracted following the procedures of Dellaporta et al. [32] . PCR was carried out with a 2.5 µL reaction volume comprising 2.5 µL of 50 ng/L genomic DNA, 1 µL of forward and reverse primers (10 pmol), and 8 µL of MQ H 2 O [33] . Table S5 shows the unique SSR primers used to confirm the expression of the Lr19 and Lr28 genes in 16 cultivars.

PCR products (20 µL) were electroporated on 1.5% agarose gel in 1× TBE buffer. DNA Ladder Plus (100 bp-3 kbp, Jena Bioscience, M-203) was utilized to measure the size of DNA fragments. Ethidium bromide was used to stain the gel, then photographed with a Syngen™ UV Transilluminator.

2.6. The Principal Component Analysis (PCA) and Two-Way Hierarchical Cluster Analysis (HCA)

Principal component analysis (PCA) was performed using the number of isolates and frequency (%) of P. triticina races and the associated loading plots were generated using the singular value decomposition (SVD). In addition, standardized two-way hierarchical cluster analysis (HCA) was used to differentiate the interactions between the 17 individual physiological races of P. triticina, 16 Egyptian commercial wheat cultivars, and 4 main wheat-growing governorates, based on the number of isolates, frequency (%), and virulence formula of P. triticina races in these populations. Distance and linkage were done using Ward's minimum variance method [35] with 95% confidence between groups. The differences in the isolates number and frequency are also visualized and presented as a heat map, combined with two-way HCA, as described above.

Overall, we sampled 179 wheat samples that show the characteristic symptoms of leaf rust disease from 16 different cultivars grown in four different locations during the 2019/2020 and 2020/2021 growing seasons, which yielded approximately 130 isolates ( Table 1 ). The first season (2019/2020) yielded the greatest number of samples (94 out of 179 samples) and isolates (71 out of 130 isolates) ( Table 1 ). In terms of locations, the Sharkia governorate had the highest number of samples (27 and 25 samples during 2019/2020 and 2020/2021, respectively) and isolates (20 and 17 isolates during 2019/2020 and 2020/2021, respectively), followed by Kafrelsheikh and Beheira, while the lowest numbers of samples and isolates were recorded from Alexandria (Table 1) . Figure 1A ). However, all other cultivars showed different degrees of susceptibility to P. triticina in most studied locations, particularly new cultivars, such as Sakha-95, Sids-14, and Shandweel-1 (Table 2 and Figure 1A ). Gemmeiza-7, Gemmeiza-11, Sakha-93 and Sids-1 were the most susceptible wheat cultivars in the four governorates during the two seasons ( Figure 1A ). In terms of locations, the Sharkia governorate recorded the highest number of susceptible wheat cultivars in both seasons, followed by Alexandria, then Beheira and Kafrelsheikh, which were almost similar, particularly during the 2019/2020 season ( Figure 1B ). 

During the 2019/2020 season, the highest number of isolates was recorded for Gemmeiza-7 and Gemmeiza-11, followed by Sids-1 and Sakha-93, respectively, whereas during the 2020/2021 season, the highest number of isolates was recorded for Gemmeiza-11, followed by Sakha-93, Sids-1, and Gemmeiza-7, respectively ( Figure 1C ). In both seasons, the highest number of isolates was recorded on samples collected from the Sharkia governorate, followed by Kafrelsheikh, Beheira, and Alexandria, respectively ( Figure 1D ).

Generally, 130 isolates were identified from 16 different cultivars grown in 4 different locations during the two growing seasons. However, a higher number of P. triticina isolates was recorded during the 2019/2020 (71 isolates) than the 2020/2021 season (59 isolates) ( Table 3 ). In both seasons, the highest number of isolates was recorded in Sharkia (20 and 17 isolates during 2019/2020 and 2020/2021, respectively), followed by Kafrelsheikh (19 and 15 isolates during 2019/2020 and 2020/2021, respectively) and Beheira (17 and 14 isolates during 2019/2020 and 2020/2021, respectively), whereas the lowest number of isolates was recorded in Alexandria (15 and 13 isolates during 2019/2020 and 2020/2021, respectively) ( Table 3 ).

In terms of cultivars, in the 2019/2020 season, the highest numbers of isolates were identified from Gemmeiza-7 and Gemmeiza-11 (11 isolates that donated approximately 15.5% frequency). While in the 2020/2021 season, Gemmeiza-11 recorded the highest number of isolates (9 isolates that donated approximately 15.3% frequency), followed by Sakha-93 and Sids-1, which recorded eight isolates each (about 13.6 frequency). Interestingly, no disease symptoms (pustules) were recorded for Gemmeiza-12 and Misr-3 at the four studied governorates during both seasons, which support the suggestion that both cultivars are resistant to P. triticina (Table 3 ). 

Principal component analysis (PCA) was performed using the number of isolates of individual cultivars in the four studied locations. The PCA-associated scatter plots and loading plots are shown in Figure 2A ,B. The scatter plot obtained using the 2019/2020 data showed a clear separation among all studied locations/governorates ( Figure 2A ). However, in 2020/2021, both the Sharkia and Alexandria governorates were overlapped each other slightly in the upper-left side of the scatter plot ( Figure 2B ). In the 2019/2020 season, PC1 and PC2 were responsible for 95.47% of the variation, whereas in the 2020/2021 season they were responsible for only 76.12% of the variation. Moreover, the loading plot showed that the isolates collected during 2019/2020 from six cultivars (Misr-2, Gemmeiza-9, Sids-14, Sakha-95, Sakha-94, and Misr-1) were positively correlated with the Sharkia governorate and five cultivars (Sids-1, Gemmeiza-10, Shandweel-1, Gemmeiza-11, and Gemmeiza-7) were correlated with the Kafrelsheikh governorate. Nevertheless, only two cultivars were correlated with the Beheira governorate (Sids-12 and Sakha-93) and Sids-13 was correlated with the Alexandria governorate ( Figure 2A ). On the other hand, in the 2020/2021 season, isolates collected from only four cultivars (Sakha-94, Misr-1, Gemmeiza-9, and Sakha-95) were correlated with the Sharkia and Alexandria governorates and three cultivars (Gemmeiza-7, Sids-12, and Sakha-93) were correlated with the Kafrelsheikh governorate, while only two cultivars were correlated with the Beheira governorate (Gemmeiza-10 and Gemmeiza-11) ( Figure 2B ). Additionally, standardized two-way hierarchical cluster analysis (HCA) was used to differentiate the individual cultivars among the studied location based on the number of isolates and frequency ( Figure 2C ,D). The differences in the isolates' number and frequency are also visualized and presented as a heat map. In both seasons, the total HCA dendrogram among locations/governorates (presented at the bottom of Figure 2C ,D) showed that the cultivar profiles of Kafrelsheikh were closer to the profile of Beheira, whereas that of Sharkia was closer to that of Alexandria.

To evaluate the range of pathogenic variation in a particular area, the race analysis was carried out for all isolates based on the reaction of distinct lines containing 20 monogenic resistance genes at the seedling stage in the greenhouse. Briefly, 17 physiological races were identified during the seasons 2019/2020 and 2020/2021 ( Table 4 ). The pathological pathotype MTTGT was the most abundant race (21 isolates in both seasons, which donated about 16.20% of total frequency), followed by STSJT (19 isolates), TTTKS (17 isolates), and TTTST (16 isolates) as a total of both seasons, while MBGJT was the lowest abundant race (only one isolate during 2019/2020 season) ( Table 4 ). It worth noting that, both pathological pathotypes, MBGJT and PKTTS, appeared in only the first season and disappeared in the second season.

PCA was performed using the number of isolates of individual physiological races of P. triticina in the four studied locations ( Figure 3A ,B). Generally, in both seasons, the scatter plots showed a clear separation among all studied locations/governorates ( Figure 3A ,B). PC1 and PC2 were responsible for 93.15% and 92.15% of the variation during the 2019/2020 season and 2020/2021 season, respectively. Additionally, the loading plot using the 2019/2020 data showed that seven physiological races (PTKKT, PKTTS, PKKSR, STTLT, NRNJT, LTTFT, and TTTST) were positively correlated with the Alexandria governorate and six pathological pathotypes (MBGJT, LTCGT, PTTJT, STSJT, TTTKS, and PBPPP) were correlated with Beheira governorate ( Figure 3A) . Though, only two pathotypes were associated with Sharkia (MTTGT and GTTPT) and Kafrelsheikh governorate (LGDTT and HTTDS) ( Figure 3A ).

On the other hand, during the 2020/2021 season, five physiological races (NRNJT, TTTST, HTTDS, PKKSR, and LTTFT) were correlated with the Alexandria governorate. Likewise, five pathological pathotypes (GTTPT, PTTJT, TTTKS, LTCGT, and STSJT) were correlated with the Kafrelsheikh governorate. However, only three physiological races (PBPPP, LGDTT, and STTLT) were correlated with the Beheira governorate and two pathological pathotypes (MTTGT and PTKKT) were associated with the Sharkia governorate ( Figure 3B ).

Furthermore, the standardized two-way HCA was used to differentiate the interactions between the 17 individual physiological races of P. triticina and various studied governorates, based on the number of isolates, frequency (%), and virulence formula of P. triticina races in these populations ( Figure 3C ,D). The differences in the isolates number and frequency are also visualized and presented as a heat map. In both seasons, the total HCA dendrogram among locations/governorates (presented in the bottom of Figure 3C ,D) showed that the distribution of physiological races of P. triticina in Kafrelsheikh was closer to the profile of Beheira, whereas the distribution of physiological races of P. triticina in Sharkia was closer to that of Alexandria.

During the 2019/2020 season, the total HCA dendrogram among physiological races of P. triticina (presented on the left side of Figure 3C ) showed that the 17 identified races were separated into four distinct clusters. The first cluster (C-I) included four pathotypes (GTTPT, PKTTS, PTKKT, and TTTST), which were dominant in the Sharkia and Alexandria governorates. Likewise, the second cluster (C-II) included four pathotypes (HTTDS,

LGDTT, PBPPP, and PTTJT), which were mainly prominent in the Kafrelsheikh governorate and some in the Beheira governorate ( Figure 3C ). The third cluster (C-III) included six pathotypes (LTCGT, MBGJT, LTTFT, NRNJT, PKKSR, and STTLT). The last cluster (C-IV) included the three most frequent physiological races of P. triticina (MTTGT, STSJT, and TTTKS) that appeared in almost all studied locations. It worth mentioning that the total HCA dendrogram of the physiological races of P. triticina during the 2020/2021 season (presented on the left side of Figure 3D ) was slightly different to the one described above. Briefly, the 17 physiological races were separated into six distinct clusters. C-I included four pathotypes (GTTPT, LTCGT, PTTJT, and STTLT), C-II included five pathotypes (HTTDS, LTTFT, PKKSR, NRNJT, and PTKKT), C-III included only two pathotypes (LGDTT and PBPPP), C-IV included only two pathotypes (MBGJT and PKTTS), C-V included only two pathotypes (STSJT and TTTKS), and C-VI included only two pathotypes (MTTGT and TTTST) as well ( Figure 3D ) which were dominant in Sharkia and Alexandria governorates. Collectively, these findings suggest that the most virulent race (TTTST) in the Sharkia and Alexandria governorates infected the new wheat cultivars; Sakha-94, Sakha-95, and Shandweel-1.

Due to the potential of P. triticina to develop new races that can target resistant cultivars under favorable environmental conditions, the association between physiological races of P. triticina and the Egyptian wheat cultivars was deeply studied and presented in Table 5 . In the Kafrelsheikh governorate, mainly, five main pathotypes (GTTPT, HTTDS, PTTJT, STSJT, and TTTKS) were identified in the six wheat cultivars (Gemmeiza-7, Gemmeiza-11, Sakha-93, Sids-1, Sids-12, and Shandweel-1) during both seasons. In addition, the physiological races LGDTT and PBPPP were identified only during the 2019/2020 season, and not 2020/2021, for the Gemmeiza-10 and Sakha-93 cultivars, respectively. Likewise, the physiological races LTCGT and STTLT were identified only during the 2020/2021 season, but not 2019/2020, on Sids-1 and Sids-14 cultivars, respectively (Table 5 ).

In the Beheira governorate, generally, six main physiological races (GTTPT, LTCGT, PBPPP, PTTJT, STSJT, and TTTKS) were identified on six wheat cultivars (Gemmeiza-7, Gemmeiza-11, Sakha-93, Sids-1, Sids-12, and Shandweel-1) during both seasons. In addition to these pathotypes, MTTGT and MBGJT were identified only during the 2019/2020 season, but not 2020/2021, for Gemmeiza-7 and Gemmeiza-10, respectively. Similarly, the LGDTT and STTLT pathotypes were identified only during the 2020/2021 season, but not in 2019/2020, for Gemmeiza-10, Sids-13 and Sids-14 cultivars, respectively (Table 5 ).

In the Sharkia governorate, five main physiological races (GTTPT, MTTGT, PTKKT, STTLT, and TTTST) were largely identified for all studied wheat cultivars, except Gemmeiza-12 and Misr-3, which appeared to be resistant to all identified physiological races of P. triticina in the four studied governorates during both seasons. In addition, three pathotypes (PKTTS, PKKSR, and TTTKS) were only identified during the 2019/2020 season, and not 2020/2021 (Sids-1, Sids-13, and Sids-14, respectively). Moreover, a further three pathotypes (PTTJT, HTTDS, and NRNJT) were identified only during the 2020/2021 season, and not 2019/2020, for Sakha-93, Sids-1, and Sids-13 cultivars, respectively (Table 5) .

Finally, in the Alexandria governorate, six main pathotypes (LTTFT, MTTGT, NRNJT, PKKSR, STTLT, and TTTST) were primarily identified for ten wheat cultivars (Gemmeiza-7, Gemmeiza-11, Sakha-93, Sakha-94, Sakha-95, Sids-1, Sids-12, Sids-13, Sids-14, and Shandweel-1) during both seasons. Additionally, five physiological races (STSJT, PTKKT, PKTTS, TTTKS, and GTTPT) were identified only during the 2019/2020 season, but not 2020/2021, on Gemmeiza-7, Sakha-93, Sids-1, Sids-14, and Misr-1 cultivars, respectively. Likewise, the physiological race HTTDS was identified only during the 2020/2021 season, but not 2019/2020, on the Gemmeiza-7 cultivar (Table 5 ).

In the 2019/2020 season, the most frequent race, MTTGT (Table 5) , was recorded in the Sharkia, Beheira, and Alexandria governorates, but not Kafrelsheikh. However, this race was reported only in Sharkia and Alexandria, but neither the Beheira nor Kafrelsheikh governorates during the 2020/2021 season (Table 5) . Besides, the second frequent race, STSJT, was recorded in Kafrelsheikh, Beheira, and Alexandria, but not Sharkia during the 2019/2020 season. Nevertheless, during the 2020/2021 season, STSJT was only reported in samples collected from Kafrelsheikh and Beheira, but neither Alexandria nor Sharkia (Table 5) . Moreover, the third most frequent race, TTTKS, was identified from all studied locations during the 2019/2020 season; however, it was reported only from Kafrelsheikh and Beheira, but neither the Alexandria nor Sharkia governorates during the 2020/2021 season (Table 5 ). PCA was performed using the number of isolates of individual physiological races of P. triticina on the 16 studied cultivars ( Figure 4A,B) . Briefly, in both seasons, scatter plots showed an accepted separation among all studied cultivars ( Figure 4A,B) . PC1 and PC2 were responsible for 65.99% and 58.49% of the variation during the 2019/2020 and 2020/2021 seasons, respectively. It worth mentioning that Gemmeiza-9, Gemmeiza-10, Gemmeiza-12, Misr-1, Misr-2, Misr-3, Sids-13, Sakha-94, and Sakha-95 were clustered together in the negative quarter of the scatter plot during the 2019/2020 season ( Figure 4A ). However, only Gemmeiza-12, Misr-1, Misr-2, and Misr-3 were clustered together in the same part of the graph during the 2020/2021 season ( Figure 4B) .

Additionally, the loading plot using 2019/2020 data showed that the most frequent/ abundant physiological races MTTGT and STSJT strongly influenced PC1, whereas the TTTKS pathotype had more effect in PC2. Moreover, the physiological races PTKKT, PBPPP, PTTJT, PKKSR, GTTPT, NRNJT, MBGJT, LTTFT, and LGDTT were clustered together and appeared to have less weight on PC1 and PC2 ( Figure 4A ). On the other hand, during the 2020/2021 season, the pathotypes TTTKS and TTTST strongly affected PC1, while MTTGT, STSJT, and STTLT strongly manipulated PC2 ( Figure 4B) .

Furthermore, standardized two-way HCA was used to differentiate the individual physiological races of P. triticina among the studied cultivars based on the number of isolates and frequency ( Figure 5A,B) . The differences in the isolates' number and frequency are also visualized and are presented as a heat map. In both seasons, Gemmeiza-7, Sids-12, and Sids-1 were clustered together (C-I) which appeared to have a higher frequency of the physiological races MTTGT and STSJT ( Figure 5A,B) . However, Sakha-93 was also clustered with C-I during the 2020/2021 season ( Figure 5B ). Likewise, Gemmeiza-9, Sakha-94, and Gemmeiza-10 were clustered together, within C-II, which appeared to be associated with the pathotype TTTST in both seasons. Nevertheless, Sakha-95 was also clustered with C-II during the 2019/2020 season ( Figure 5B) . Additionally, the resistant cultivars Gemmeiza-12 and Misr-3 were cluster together with C-III during both seasons. 

The races were split into two major clusters (Figures 6 and 7) , the first of which was further divided into two sub-clusters. At 300 bp, the first sub-cluster contained the races GTTPT, MTTGT, STTLT, TTTKS, and TTTST, while the second sub-cluster included PKTTS, PTKKT, PTTJT and STSJT at 150 bp. HTTDS, LGDTT, LTCGT, LTTFT, MBGJT, NRNJT, PBPPP, and PKKSR were part of the second cluster, which was split into one more sub-cluster at 200 bp.

Since it captures all variation identified by each primer, cluster analysis utilizing all three primers certainly has greater power in separating the investigated isolates, giving a more meaningful grouping pattern. 

For each line, the frequency of virulence was calculated as the ratio of virulent cultures to the total number of cultures. The frequency of virulence to Lr genes varied between P. triticina regional populations in Egypt. Data presented in Table 6 show different frequencies of virulence to the tested Lr genes. The least frequencies of virulence were found in Lr19, Lr28, Lr2b, and Lr10 showing 0.00, 0.00, 39.44, and 38.03%, respectively, in the first season and 0.00, 0.00, 23.73, and 44.07%, respectively, in the second season. On the other hand, virulence against Lr1, Lr2a, Lr2c, Lr3, Lr3ka, Lr9, Lr11, Lr14b, Lr15, Lr16, Lr17, Lr18, Lr21, Lr24, Lr26, Lr30, Lr36, and Lr42 showed the highest frequencies over 57.69%. It worth noting that Lr19 and Lr28 exhibited complete resistance to leaf rust in the current study under Egyptian field conditions ( Table 6 ). 

Sixteen wheat cultivars and two resistant lines (Lr genes) Lr19 and Lr28 were selected for the identification of resistance genes using molecular markers. Briefly, two specific primers were used to detect two resistance genes (Figure 8 ). The markers for Lr19 and Lr28 were defined as 150 bp and 300 bp fragments, respectively, in two wheat cultivars (Gemmeiza-12, and Misr-3) during the polymorphic survey, while fourteen of the tested cultivars did not show the presence of Lr19 ( Figure 8A ) and Lr28 ( Figure 8B ). These findings might explain the molecular mechanism(s) behind the previous results that the two cultivars, Gemmeiza-12 and Misr-3, are resistant to all races of leaf rust pathogen. 

Wheat leaf rust was responsible for the extinction of numerous cultivars in Egypt, including Giza-139, Mexipak-69, Super-X, and Chenab-70, due to their susceptibility under field conditions. Moreover, some wheat genotypes were discarded very shortly after their release, such as Giza-139 [36] . The failure of these cultivars was attributed to the broad pathogen populations, as well as their strong evolutionary potential. The pathogen's dynamic nature resulted in the quick production and emergence of several new virulent races with the capacity to overcome resistance genes in wheat cultivars. As a consequence, the effective lifespan of these recently released cultivars is constantly reduced [11, 12, 16] . Therefore, the identification of the dominant physiological races in each area is one of the most important steps in rust-resistance breeding programs. The program will be effective if all physiological isolates of a pathogen are utilized [37] . Generally, new leaf rust races emerge as a result of mutation [38] , heterokaryosis [39] , gene recombination [40, 41] , migration [42] , vegetative or parasexual recombination, hyphal anastomosis [43] and natural selection of the most virulent races in the region [44] .

The alternative host (Thalictrum spp.) was not discovered in Egypt, and the urediniospores are unable to live in Egypt during the summer owing to the high temperatures [14] . The initial inoculum is frequently transported by the prevailing winds from neighboring countries. Therefore, the current research focused on the frequency of various virulence to investigate changes in virulence formula throughout two seasons. During the growing seasons 2019/2020 and 2020/2021 in Egypt, a survey for wheat leaf rust revealed the prevalence of the disease caused by P. triticina in four governorates, i.e., Kafrelsheikh, Beheira, Sharkia, and Alexandria. Most of the infected samples were collected from the commercial fields and the Egyptian wheat trap rust nurseries (EWTRN). Wheat infected samples (around 179 samples) were collected from 16 distinct cultivars growing in four different locations for the two annual surveys (2019/2020 and 2020/2021) yielding roughly 130 iso-lates. During the two seasons under investigation, Sharkia governorate had the largest number of samples, isolates, and susceptible wheat cultivars, followed by Kafrelsheikh and Beheira, while Alexandria had the lowest number of samples and isolates. In addition, the total HCA dendrogram across locations/governorates in both seasons revealed that the cultivars profile of Kafrelsheikh was closer to that of Beheira, whilst the profile of Sharkia was closer to that of Alexandria. This is because wheat is more often grown in these governorates and the environmental conditions are more suitable for the spread of the disease. Thus, these governorates were regarded as leaf rust hotspots [11, 17] .

Gemmeiza-12 and Misr-3 cultivars were resistant in all tested locations over the two seasons, with no disease symptoms (pustules) observed in these cultivars in the four governorates. Additionally, the cultivars Gemmeiza-12 and Misr-3 were clustered together during both seasons. However, the new wheat cultivars, such as Sakha-95, Sids-14, and Shandweel-1, were susceptible to P. triticina in most of the studied locations. This could be because the most common leaf rust pathotypes change from year to year in pathogen populations [18, 45] . Therefore, it can be suggested that wheat production in Egypt is highly correlated with the changes in cultivar distribution to enhance yield and resistance to fungal infections in new cultivars. Wheat cultivation increased, and the distribution of the most commonly planted cultivars varied considerably as well. Generally, more than half of the wheat cultivated area was planted with seven cultivars: Gemmeiza-11, Sakha-94, Sids-12, Sids-14, Misr-1, Misr-2, and Shandweel-1. However, they are now susceptible or moderately susceptible to wheat leaf rust [46] . Hence, it was necessary to identify the physiological races that supplant the resistance in wheat cultivars. We would like to know whether the physiological races reported in the same governorates may be found in the same cultivars in different governorates. This motivated us to investigate the virulence dynamics, diversity, and degree of similarity of P. triticina populations in various areas, as one of the most crucial stages in rust resistance breeding programs [47] .

Race analysis was carried out in the greenhouse by recording the response of different lines containing 20 monogenic resistance genes at the seedling stage. These genes have varied reactions in different race groups, so they are race-specific. Race analysis was important for determining the range of pathogenic variation in a particular area. The screening for resistance in cultivars revealed that the race differences are responsible for host responses, contributing to our understanding of the mechanism of diversity and driving research and breeding programs to avoid future disease outbreaks.

Seventeen physiological races were identified during the seasons 2019/2020 and 2020/2021. The pathotype MTTGT was found in the Sharkia, Beheira, and Alexandria governorates and was the most common race in both seasons. The second race, STSJT, was found in Kafrelsheikh and Beheira for Gemmeiza-7, Sids-1, and Sids-12 cultivars, as well as in the Alexandria governorate on Gemmeiza-7 cultivars. Moreover, TTTST was the most virulent race in the Sharkia and Alexandria governorates and disappeared in other governorates. The standardized two-way HCA dendrogram between locations/governorate revealed that the distribution of physiological races of P. triticina in Kafrelsheikh was closer to the profile of Beheira, while that of the physiological races of P. triticina in Sharkia was closer to the profile of Alexandria. Additionally, only two pathotypes (MTTGT and TTTST) were detected in the same cluster in the Sharkia and Alexandria governorates. These data imply that the most virulent race (TTTST) infected the new wheat cultivars Sakha-94, Sakha-95, and Shandweel-1 in the Sharkia and Alexandria governorates.

Both pathotypes, MBGJT and PKTTS, originally occurred in the first season and then disappeared in the second. The long-distance dispersal of pathogen propagules and the high gene/genotype flow of such pathogens were mainly held responsible for changes in wheat leaf rust populations and the appearance or disappearance of virulence [44] . This process resulted in noticeable changes in the genetic structure of the recipient pathogen populations, particularly during the summer. Foreign sources of leaf rust inoculum enter Egypt every year, and it is transported from one region to another within the same year [15] . Furthermore, previous studies in Egypt confirmed the conclusion that the source of primary inoculum, which is often carried in by northern winds each year from foreign sources, determines the presence of virulent races in leaf rust populations [14] . According to recent research, urediniospores of leaf rust do not persist or cannot survive in Egypt during the summer [12] . The current study was supported by the results of McVey et al. [15] and Negm et al. [16] who suggested that leaf rust populations in Egypt are made up of a wide range of races.

The races prevailed in certain governorates and disappeared in others. This could be due to climatic differences between governorates, such as temperature, wind directions, and rainfall, which are all necessary for disease occurrence and the potential of P. triticina to develop new races that can attack resistant cultivars under favorable environmental conditions, resulting in significant losses [9, 48] . These results are supported by previous reports indicating that changes in the virulence and genetic structure of leaf rust populations are mostly affected by some environmental factors, such as temperature and relative humidity in the wheat-growing regions [12, 47] .

In the current study, the differences between the races of P. triticina obtained from various regions in Egypt were investigated using RAPD analysis to determine their relationship. This research revealed that P. triticina collections vary on a global scale in terms of virulence and molecular background. P. triticina race analysis revealed differences in virulence and RAPD phenotypes. Previous research has shown that arbitrary decameter primers were used to examine virulence polymorphism in 20 near-isogenic wheat differential lines and three randomly amplified polymorphic DNA (RAPD) [44] . The cluster analysis revealed a correlation between virulence and molecular variation in general [44] . The results support the idea that P. triticina has different regional populations in Canada [44] . The most molecular diversity was found across races with varied virulence phenotypes. There were minor molecular differences across races of similar virulence, with the first cluster consisting of virulent races. The molecular polymorphisms were more efficient in discriminating between the main clusters of P. triticina compared to virulence polymorphisms. There was a correlation between the virulence and molecular dissimilarity matrixes. In P. triticina from Canada, cluster analysis revealed a link between virulence and molecular polymorphism [27] . P. triticina was collected from the different locations. All simple uredinial isolates from the collection were evaluated for virulence polymorphism on 22 Thatcher lines according on virulence phenotypes, and chosen isolates were further analyzed for RAPD using 11 primers [27] . There were 105 virulence phenotypes and 82 RAPD phenotypes among the 131 simple uredinial isolates. Differences in isolates across groups were responsible for 36% of the RAPD variation. RAPD analysis was able to display the connection between the races of P. triticina, which vary in terms of virulence and were collected from different locations in Egypt. Seventeen races were divided into three categories based on RAPD analysis: 150, 200, and 300 bps. GTTPT, MTTGT, STTLT, TTTKS, and TTTST were all members of the same cluster of pathogenic races. As a result, it's more probable that these races (members of the same cluster) shared a lot of genetic material and came from the same location. Furthermore, the isolates that were the most closely related seemed to be from different sources. Since it captures all variation identified by individual primers, cluster analysis employing all four primers probably has greater power in distinguishing among examined isolates, and therefore provides a more meaningful grouping pattern. The results of the cluster analysis were consistent, indicating that the findings were reliable, and that RAPD markers may be used to evaluate P. triticina genetic diversity. Furthermore, the present research revealed that P. triticina has a lot of genetic variation.

As a result of the dynamic changes in leaf rust races, plant breeders should incorporate new efficient resistance genes [17, 18, 49] . Therefore, identifying leaf rust resistance genes is critical for adding new effective resistance genes to wheat breeding programs. During the polymorphism survey, the markers for Lr19 and Lr28 were identified as 150 bp and 300 bp fragments in two wheat cultivars (Gemmeiza-12 and Misr-3), while fourteen of the tested cultivars did not show the presence of Lr19 and Lr28. This explains the earlier findings that both Gemmeiza-12 and Misr-3 cultivars are resistant to all races of leaf rust pathogen. Similar results were reported by Vida et al. [50] , who stated that wheat genotypes containing the three-leaf rust resistance genes, Lr9, Lr19, and Lr28 exhibit good and strong levels of leaf rust resistance at the adult time. In Western Siberia and the Urals, Lr9 has been broadly utilized in breeding, but the first Lr9-virulent isolates were discovered in 2007 [51] .

Lr19, which showed complete resistance to leaf rust in the current study under Egyptian field conditions is also effective in most countries in Asia, Australia, and Europe and has been related to desired genes for grain yield increase, making it a good candidate for wheat breeding [52] . Although Lr28 has a high level of resistance to leaf rust in all common pathotypes in India, it was not related to any undesirable genes [40, 53] . Therefore, it is necessary to determine if these genes are present in Egyptian wheat varieties. Therefore, the present research adds to the growing evidence that leaf rust resistance genes (R-genes) can only protect wheat crops for a limited time. On the other hand, the effect of virulent pathotypes on certain cultivars is determined by the overall level of resistance (partial resistance) in those cultivars [13, 54] .

This study is a pioneer to explain the varietal response to infection with leaf rust disease and physiological races of this fungus in Egypt. Seventeen physiological races were identified on 16 wheat cultivars, during 2019/20 and 2020/21 growing seasons by race analysis. Furthermore, using RABD analysis, the races were divided into three groups: 150, 200, and 300 bp. TTTST was the most virulent race in the Sharkia and Alexandria governorates ( Figure 9 ) and disappeared in other governorates. It infected the new wheat cultivars in Alexandria. On the other hand, Gemmeiza-12 and Misr-3 cultivars were resistant to all races of leaf rust due to the role of resistance genes (Lr19 and Lr28) in these cultivars. Finally, as a warning to the rest of the world, it must explain why some of these new races exist in Egypt. These races move to Egypt every year because there is no alternative host, and they have the potential to supplant the resistance. As a consequence, it may create problems in certain countries. Table S1 : Pt-code for the 20 differential hosts of Puccinia triticina in ordered sets of four and additional set five, Table S2 : Wheat leaf rust infection types used in disease assessment for seedling stage according to Johnston and Browder [31] , Table S3 : Code, nucleotide sequence and G+C (%) of primers used in the random amplified polymorphic DNA (RAPD) reactions, Table S4 : The pedigree list of the monogenic lines (Lr genes) used in this study, Table S5 : Primer names, sequences, annealing temperatures and references from Lr genes associated markers used in this study. 

Food and Agriculture Organization of the United Nations

Annual Wheat Production in Egypt from 2010 to 2020. Statista Global Data Platform

Principal Importing Countries of Wheat

Egypt Is Able to Secure a Steady Supply of Grains during the COVID-19 Pandemic; United States Department of Agriculture (USDA)

The global burden of pathogens and pests on major food crops

Fungal wheat diseases: Etiology, breeding, and integrated management

Rust Diseases of Wheat: Concepts and Methods of Disease Management

Global status of wheat leaf rust caused by Puccinia triticina

Effect of leaf rust infection on yield and technical properties in grains of some Egyptian wheat cultivars

Partial resistance to leaf rust in some Egyptian wheat cultivars

geographical distribution of physiologic races of Puccinia triticina and postulation of resistance genes in new wheat cultivars in Egypt

Abd El-Malek, N.I. Frequency of virulence and virulence formula of wheat leaf rust races identified in Egypt during 2004/05-2007/08

Wheat leaf rust caused by Puccinia triticina

The source of inoculum of wheat leaf rust in relation to wind direction

Patterns of virulence diversity in Puccinia triticina on wheat in Egypt and the United States in 1998-2000

Virulence dynamics and diversity in wheat leaf rust (Puccinia triticina) populations in Egypt during

Geographical distribution of Puccinia triticina physiologic races in Egypt during 2012-2014 growing seasons

Diversity and virulence dynamics within Puccinia triticina populations in Egypt

Genetic of pathogenicity of Puccinia recondita f. sp. tritici, the causal agent of leaf rust of wheat

Catalogue of gene symbols for wheat

Characterization of Lr75: A partial, broad-spectrum leaf rust resistance gene in wheat

Genome-wide association study of leaf rust resistance in Russian spring wheat varieties

Catalogue of Gene Symbols for Wheat: 2015-2016 Supplement. Komugi Wheat Genetic Resources Database

Analysis of the Lr34/Yr18 rust resistance region in wheat germplasm

Virulence analysis of Puccinia graminis f.sp. tritici populations in Ethiopia with special consideration of Ug99

Postulation of leaf rust resistance genes in Iranian wheat cultivars and breeding lines

Virulence and molecular polymorphism in international collections of the wheat leaf rust fungus Puccinia triticina

Population structure of Puccinia recondita in Western Europe during 1995, as assessed by variability in pathogenicity and molecular markers

Identification of Physiologic Races of Puccinia graminis f.sp. tritici

A North American system of nomenclature for Puccinia recondita f. sp. tritici

Seventh revision of the international register of physiologic races of Puccinia recondita f.sp. tritici

A plant DNA minipreparation: Version II

The genetic maps of wheat and their potential in plant breeding

Stem rust of wheat, barley and rye in Canada in 1964

Hierarchical grouping to optimize an objective function

Partial resistance of wheat (Triticum aestivum) to leaf rust (Puccinia triticina) in Egypt. A. Evaluation of seven Egyptian wheat cultivars for partial resistance against leaf rust, under field conditions

Physiologic Specialization of Puccinia triticina in Canada in 1998

Detection and occurrence of a new pathotype of Puccinia triticina with virulence for Lr24 in Australia

Effect of sexual recombination in two populations of the wheat leaf rust fungus Puccinia recondita

Wheat Rusts: An Atlas of Resistance Genes

Evidence for somatic hybridization in nature in Puccinia recondita f. sp. tritici, the leaf rust pathogen of wheat

Studies on the origin, spread, and evolution of an important group of Puccinia recondita f. sp. tritici pathotypes in Australasia

Fusion body formation, germ tube anastomosis, and nuclear migration during the germination of urediniospores of the wheat leaf rust fungus, Puccinia triticina

Physiologic specialization of Puccinia triticina on wheat in the United States in 2003

Molecular markers identification of leaf rust resistant genes Lr19, Lr21, Lr24, Lr47 and Lr51 in selected Egyptian wheat cultivars

Biochemical, histopathological and genetic analysis associated with leaf rust infection in wheat plants

Physiologic specialization of Puccinia triticina on wheat in the United States in 2009

Occurrence and virulence of wheat leaf rust in Hungary during 1999-2001

Molecular and genetic analysis of leaf rust resistance genes in two new Egyptian wheat cultivars

Application of molecular markers in breeding for leaf rust resistance in wheat

Dynamics of distribution of the wheat leaf rust pathotypes virulent to the cultivars with Lr9 gene in Omsk region

Identification and validation of molecular markers linked to the leaf rust resistance gene Lr19 in wheat

Validation and identification of molecular markers linked to the leaf rust resistance gene Lr28 in wheat

Virulence of leaf rust physiological races in Iran from

The authors declare no conflict of interest.