Leaf rust (Puccinia triticina Eriks) resistance genes in wheat cultivars registered in the Czech Republic

Leaf rust (caused by Puccinia triticina Eriks—Pt) occurs regularly in the Czech Republic and can inflict considerable yield losses on wheat, particularly in the southeastern part of the country in warm and dry years. Knowledge of resistance genes (Lr) is useful in the breeding, particularly when combined with data on virulence in the rust population. Seeds of 56 winter wheat cultivars registered in the Czech Republic were obtained from the Central Institute for Supervising and Testing in Agriculture. All cultivars were tested by molecular markers to postulate the presence of Lr1, Lr10, Lr24, Lr26, Lr28, Lr34, Lr37 and Lr68. In total, 39 cultivars were also tested for resistance to leaf rust in the field trials. The Lr37 gene had the highest proportion, occurring in 24 tested varieties. The most effective was the Lr24 gene, which was carried in various combinations with other Lr genes by cultivars Athlon, Futurum, Sheriff, Hyfi, Asory, Campesino, Gordian. Also the gene Lr28 was mostly highly effective.


Introduction
Leaf rust is an important disease of wheat (Triticum aestivum) that is most economically controlled by breeding for resistance. Due to the variability of the pathogen, the resistance breeding is a continuous process. Knowledge of resistance genes (Lr) is useful in the breeding, particularly when combined with data on virulence in the rust population. However, host resistance conferred by major genes has tendency to be easily overcome by new virulent rust pathotypes (McDonald and Linde 2002). Therefore, pyramiding several Lr genes into a single cultivar is a useful strategy as the combined effects of several genes give the cultivar a wider base of disease resistance, thereby extending the period of effectiveness (Roelfs et al. 1992).
There are about 80 known Lr genes and several of those have been commonly pustulated in Czech wheat cultivars in the past. Those include Lr1, Lr10 and Lr13, which have lost their efficacy over time (Bartoš et al. 2002). Pathan and Park (2006) report that the most commonly found gene in European varieties was Lr13, followed by Lr26, Lr20 and Lr37. Adult plant resistance (APR) was found in many European cultivars, but it is mostly not known what genes are responsible for it (Pathan and Park 2006). Molecular markers have been developed for many of the resistance genes to facilitate and speed up their detection in breeding programs during marker-assisted selection (MAS, reviewed by Miedaner and Korzun 2012). Such markers can also be used to screen a selection of cultivars for the presence of genes of interest. For example, in European context, such screening was applied to assess leaf rust resistance in German cultivars (Serfling et al. 2011).
Our paper presents current data on Lr genes in winter wheat cultivars registered in the Czech Republic as well as information on the virulence of local populations of the pathogen against established Lr genes and resistance assessment of varieties with the largest propagation areas in the Czech Republic.

Material and methods
Seeds of 56 winter wheat cultivars registered in the Czech Republic were obtained from the Central Institute for Supervising and Testing in Agriculture (ÚKZÚZ). All cultivars were tested by molecular markers to postulate the presence of Lr1, Lr10,Lr24,Lr26,Lr28,Lr34,Lr37 and Lr68. These particular genes were selected for their relevance to breeding, their supposed presence in Czech cultivars and the availability of reliable molecular markers.

Field trials
Out of the total of 56 cultivars 39 cultivars were also tested for resistance (disease severity) in the field trials. Rust spreaders (cv. Michigan Amber) regularly distributed in the trial were inoculated with a mixture of leaf rust urediniospores from the previous year field scorings. These spores were collected from leaves from the previous season to simulate the range of races that appear naturally. Artificial infection was performed by injecting water suspension of urediniospores into the stems of the susceptible variety spreader at the beginning of stem elongation phase. Visual estimation of disease severity took place in several terms according to the development of the disease using a disease severity scale of 1-9, where 1 stands for no symptoms observed and 9 stands for heaviest sporulation leading to defoliation. More detailed description of the 1-9 scale was published by Bariana et al. (2007).

Virulence survey in the leaf rust population
Samples of wheat leaf rust were obtained every year from different cultivars, mostly from variety trials located across the country and organized by ÚKZÚZ. The pathogen was propagated on a susceptible cultivar. Single pustules were isolated and tested on Thatcher near-isogenic lines (NILs) with genes Lr1, Lr2a,Lr2b,Lr2c,Lr3a,Lr9,Lr10,Lr11,Lr13,Lr15,Lr17,Lr19,Lr21,Lr23,Lr24,Lr26 and Lr28. Seedlings were inoculated by rubbing leaves with an urediniospore water suspension; then, they were sprayed with water and placed in glass cylinders for 24-48 h. Thereafter plants were kept in the greenhouse at 18-22 °C. Infection types were scored after 10-14 days based on the scale by Stakman et al. (1962). Virulence frequency on individual resistance genes to pathogen population was expressed in % of virulent isolates.

Molecular methods
Genomic DNA was isolated from the first two leaves of 20-30 plants per cultivar. Leaf tissue was frozen in liquid nitrogen, ground to a powder and used for DNA extraction by a commercial kit according to the manufacturer's instructions (Qiagen, Germany). Polymerase chain reaction (PCR) protocols used for markers to postulate genes Lr1, Lr10, Lr19, Lr24, Lr26, Lr28, Lr34, Lr37 and Lr68 are listed in Table 1. PCR was carried out in a thermal LabCycler (Sen-soQuest GmbH, Germany). Amplification products were separated by electrophoresis on 1.6% agarose gels stained with ethidium bromide and visualized under UV light. GeneRulerTM 100 bp DNA Ladder (Fermentas, Lithuania) was used as a molecular weight marker. Thatcher NILs containing the corresponding Lr genes were included as positive controls for most of the genes. In case of Lr68 and Lr34 cv. Frontana was used as a positive control.

Results
The most widespread Lr gene in the set of tested cultivars (Table 2) was Lr37, either single or in combination with Lr1, Lr10, Lr13, Lr24 or Lr26. Another important gene was Lr24 single or in combination with Lr34 or Lr28. Gene Lr28 was detected single and in combination with Lr24 or Lr34 and Lr10. One cultivar possessed Lr10, two cultivars Lr26. Gene Lr68 was detected single or in combination with Lr1, Lr10, Lr24, Lr26, Lr28 or Lr34. None of the tested Lr genes was found in six cultivars.
Surprisingly cultivar LG Mocca, without any of the tested Lr genes, was also evaluated 2 indicating high resistance. Cultivars possessing Lr37 in combination with other Lr genes had a higher resistance than those with Lr37 alone. Disease scoring between cultivars possessing Lr37 fluctuated between 3.2 (cultivar RGT Sacramento) and 7.7 (cultivar Sally). Cultivars carrying Lr10 (Patras), Lr26 (Johnson) had a medium scoring.
Average frequency of virulence of local isolates for Lr1, Lr10 and Lr26 exceeded 85% in the years 2014-2020. To Lr24 it was only 12% to Lr28 it was 23%. The virulence frequency to Lr26 appears to have slightly increased from 75% in 2014 to 95% in 2020, and the virulence of local races to Lr24 has slightly decreased from the average to 8% for races collected in 2020. No significant or abrupt shifts in virulence were recorded in this time period for those 5 resistance genes.

Discussion
Data on seed propagation area suggest that cultivars tested in this study cover a considerable part of wheat-growing area in the Czech Republic.
Our data on leaf rust resistance correspond well with the data published by ÚKZÚZ derived from at least 10 different locations in the Czech Republic. To facilitate the comparison of our datasets, the ÚKZÚZ scale (9-resistant, 1-susceptible) was converted to the scale applied in our field trials (1-resistant, 9-susceptible).
Cultivar Frisky was the most resistant one in the period 2017-2020 . Good agreement between our results and data by ÚKZÚZ was observed, also in less resistant cultivars, e.g., cv. Bernstein, LG Imposanto, Patras. In our tests the scoring of the above-mentioned three cultivars was 4.3, 4.9 and 5.5, whereas in the trials undertaken by ÚKZÚZ the converted scoring was 3, 4.3 and 5.5.
Effectiveness of Lr37 as a gene for APR, relatively good until about 2010, is no more satisfactory. Lr68, another APR gene, was detected in several cultivars by molecular markers; however, its effect is not visible in our field trials. This is not too surprising since single APR genes usually offer only slight improvement in resistance and there are several factors undermining the precision of field trials. A more precise greenhouse experiments will have to be undertaken in future to confirm the efficacy of Lr68 in Czech wheat germplasm, such as the integrated seedling and APR phenotyping method (Riaz et al. 2016).
Genes Lr24 and Lr28 offer a high protection (Table 4, Hanzalová et al. 2021); however, they are endangered by virulence in the leaf rust population as was observed in cultivar Tobak.
In the leaf rust virulence survey virulence frequency to Lr24 fluctuated between 6 and 19% to Lr28 between 11 and 35% in the years 2014-2020 (Table 4). However, cv. Futurum (Lr24) remained resistant (score 3.6) in 2015-2020, as well as cv. Sheriff (Lr24) (score 2.7) in 2014-2019, cv. Hyfi (Lr24) (score 2.5) in 2017-2020, cultivar Gordian (Lr24, Lr28) score 3.5 in 2014-2019. Above-mentioned data suggest that the resistance governed by Lr24 is more durable than resistance governed by Lr28 and that combination of the both genes can enhance resistance and the durability of it. LG Mocca can be considered resistant according to our results; however, no known Lr gene was detected in the cultivar by the used range of molecular markers. Further genetic analysis would be necessary to describe the genetic basis of this resistance. This cultivar is completely susceptible to local races of stem rust (Zelba et al. 2022), so it seems the resistance is leaf rust specific rather than pleiotropic.
Genes determined by us may not represent the complete genetic mechanism conferring leaf rust resistance. Other Lr genes may be detected by additional molecular markers in the tested cultivars. Whole genotype with genes enhancing or suppressing the resistance is responsible for the final disease response, and the expression of disease is also affected by the environment.
This must be kept in mind when data on Lr genes are applied in the breeding for disease resistance.  Lr68