Rare Virulences and Great Pathotype Diversity of a Central European Blumeria hordei Population

Barley is an important crop grown on almost 49 Mha worldwide in 2021 and is particularly significant in Europe where powdery mildew is the most frequent disease on susceptible varieties. The most suitable way to protect crops is by exploiting genetic resistance. However, the causal agent Blumeria hordei is an extremely adaptable pathogen. The aims of this research were to increase our knowledge of the rapidly changing pathogen population and detect rare virulences. Random samples of the pathogen were obtained from the air by means of a mobile spore sampler. Spores were collected by driving across the Czech Republic in 2019, 2021 and 2023, and 299 isolates were analyzed on 121 host varieties. No infection occurred on 35 differentials, rare virulence was recorded on 31 varieties and a higher virulence frequency was found on 55 differentials. A core set of differentials along with four additional varieties distinguishes 295 pathotypes (Simple Index = 0.987) and the virulence complexity of isolates varied from 4 to 19 with an average of 10.39. The detection of new virulences, the increasing frequency of previously rare virulences and high pathotype diversity as well as high virulence complexity confirm that using nonspecific durable resistance is crucial for successfully breeding commercial varieties.


Introduction
Virulence of a pathotype (race) reflects the ability of a pathogen to infect a host variety carrying a gene of specific resistance against a respective disease.Blumeria hordei M. Liu & Hambl.(Bh) is an airborne biotrophic fungus causing powdery mildew (PM) predominantly on cultivated barley (Hordeum vulgare L.) and H. murinum L. [1].Barley PM occurs everywhere barley is grown.In southeastern and southwestern China PM is a major airborne foliar disease of barley [2].In the Qinghai-Tibet plateau barley is the most important cereal crop and in recent years yield has been seriously threatened by PM [3].Among 41 barley diseases occurring in Australia, PM causes the second highest yield losses [4] and on varieties infected by PM, more lodging occurs with an associated effect on other traits including grain yield [5].
In Europe, and particularly in the Central and Northwest areas, PM is the most frequent disease of non-resistant barley varieties [6,7] and the crop is routinely protected against the disease.However, the optimum way to combat the pathogen is to grow resistant cultivars.Many effective resistances have been found in landraces [8][9][10] and especially in wild barley (H.v. subsp.spontaneum) [11][12][13] and have been used to breed commercial varieties [14][15][16].
B. hordei has the highest evolutionary potential of most pathogens of crop plants [17].Hybridization within a PM species or forma specialis creates pathotype diversity and can lead to the emergence of new PM forms and an expanding host range [18,19].Transposable elements play a key role in modifying the Blumeria genome and enable rapid adaptation of the pathogens towards overcoming plant immunity [20].The host-specific resistance genes present in cultivated varieties result in the directional selection of corresponding virulences and thus leave traces in the pathogen population [21].
In Europe, 46% of the total world barley area was grown in 2021 [30] and, mainly in the central part of the continent, there is a high concentration of both winter and spring forms.Simultaneous cultivation of spring and winter barley carrying a diversity of different specific resistance genes, the presence of the host in the field through the year and the given favorable conditions for the pathogen development including unlimited migration create a unique foundation for model population studies.
The aims of the current research are first, to keep pace with the characterization of a rapidly adapting population, second, to understand the pathogen itself and third, to obtain essential isolates for resistance gene postulation [31].Special attention was paid to detecting rare virulences to the corresponding resistances of the investigated varieties.

Host Plant Material
About 25 seeds of the susceptible Australian barley variety Stirling [32] were sown in a pot (80 mm diameter) filled with a gardening peat substrate and placed in a PM-proof greenhouse under natural daylight.Healthy primary leaves were excised when the second leaves were emerging, placed in 120 mm glass Petri dishes on water agar (0.8%) containing benzimidazole (40 mg L −1 ), a leaf senescence inhibitor, and inserted in the bottom of a spore sampler (BMF company, Kroměříž, Czech Republic).
One hundred and twenty-one barley differential varieties containing mostly different resistances against PM were sown as above.These were used to determine the corresponding virulences of individual isolates.Of these, 16 near-isogenic Pallas lines [33], 47 commercial varieties, 31 breeding lines and 27 other genotypes, mainly sources of resistance genes, were included.Seeds of single plant progenies [34] of all differentials had previously been produced.

Samples of the Pathogen Population
Random samples of the pathogen population originating from naturally infected spring and winter barley fields were obtained from the air by a universal doubled jet spore sampler [35] mounted on the roof of a car (Figure 1).Spores were collected by driving across the Czech Republic in three years (2019, 2021 and 2023) from mid-May to early June when tillering of spring barley had usually ended and winter barley was at the ear emergence stage.
ble elements play a key role in modifying the Blumeria genome and enable rapid adaptation of the pathogens towards overcoming plant immunity [20].The host-specific resistance genes present in cultivated varieties result in the directional selection of corresponding virulences and thus leave traces in the pathogen population [21].
In Europe, 46% of the total world barley area was grown in 2021 [30] and, mainly in the central part of the continent, there is a high concentration of both winter and spring forms.Simultaneous cultivation of spring and winter barley carrying a diversity of different specific resistance genes, the presence of the host in the field through the year and the given favorable conditions for the pathogen development including unlimited migration create a unique foundation for model population studies.
The aims of the current research are first, to keep pace with the characterization of a rapidly adapting population, second, to understand the pathogen itself and third, to obtain essential isolates for resistance gene postulation [31].Special attention was paid to detecting rare virulences to the corresponding resistances of the investigated varieties.

Host Plant Material
About 25 seeds of the susceptible Australian barley variety Stirling [32] were sown in a pot (80 mm diameter) filled with a gardening peat substrate and placed in a PM-proof greenhouse under natural daylight.Healthy primary leaves were excised when the second leaves were emerging, placed in 120 mm glass Petri dishes on water agar (0.8%) containing benzimidazole (40 mg L −1 ), a leaf senescence inhibitor, and inserted in the bottom of a spore sampler (BMF company, Kroměříž, Czech Republic).
One hundred and twenty-one barley differential varieties containing mostly different resistances against PM were sown as above.These were used to determine the corresponding virulences of individual isolates.Of these, 16 near-isogenic Pallas lines [33], 47 commercial varieties, 31 breeding lines and 27 other genotypes, mainly sources of resistance genes, were included.Seeds of single plant progenies [34] of all differentials had previously been produced.

Samples of the Pathogen Population
Random samples of the pathogen population originating from naturally infected spring and winter barley fields were obtained from the air by a universal doubled jet spore sampler [35] mounted on the roof of a car (Figure 1).Spores were collected by driving across the Czech Republic in three years (2019, 2021 and 2023) from mid-May to early June when tillering of spring barley had usually ended and winter barley was at the ear emergence stage.While travelling, the collected spores settled on detached leaves of Stirling.Dishes were replaced for each 50-100 km section of the sampling route (Table 1) totalling annually to more than 1000 km.During sampling, dishes were kept in a car-refrigerator at about 8 • C.After sampling, exposed leaves were transferred to glass Petri dishes of 150 mm diameter with fresh agar.Dishes with detached leaves of the susceptible variety and settled spores were incubated for 11 to 13 days at 18.0 ± 1 • C under artificial light (cool-white fluorescent lamps providing 12 h light at 30 ± 5 µmol•m −2 •s −1 ).

Testing Procedure
Leaf segments of differentials 15 mm long were cut from the central part of healthy primary leaves of each variety.Testing of isolates on differentials was performed in two steps.First, two segments of standard near-isogenic lines were placed adjacently with the adaxial surface upward on agar media in 90 mm plastic Petri dishes.For inoculation, conidia from each single-spore colony (Figure 2a) were sucked into a replaceable tip of an AW 1000 varipipette (Figure 2b) and then blown off the tip into a micro-settling tower using a syringe connected to the tip with a plastic tube (Figure 2c).In this manner, spores were spread and settled on leaf segments of differentials in a Petri dish placed at the bottom of the tower.The inoculum density was usually about 5 conidia•mm −2 .Dishes with inoculated differentials were incubated in the described conditions.After evaluation of the first part of the differential set, a second inoculation was performed using conidiospores from nine-day-old colonies of isolates produced on susceptible varieties.For every isolate, a 150 mm glass Petri dish with a leaf segment of each variety of the second part of the differential set was placed at the bottom of the larger settling After evaluation of the first part of the differential set, a second inoculation was performed using conidiospores from nine-day-old colonies of isolates produced on susceptible varieties.For every isolate, a 150 mm glass Petri dish with a leaf segment of each variety of the second part of the differential set was placed at the bottom of the larger settling tower.Conidia of an isolate from an infected leaf segment were shaken onto a square piece of black paper to visually estimate the amount of inoculum deposited.This was then gently rolled to form a blowpipe and the conidia were blown into the settling tower over the Petri dish.The inoculum density was around 10 conidia•mm −2 .The dishes with inoculated leaf segments were kept under the previously described incubation conditions.

Evaluation
Seven days after inoculation, infection response (IR = phenotype of a differential variety × isolate interaction) was scored on a scale of 0-4 [36], where 0 = no traces of the pathogen, and 4 = strong mycelial growth and sporulation.Isolates showing IRs 3, 3-4 and 4 were considered virulent.During phenotyping, special attention was paid to the boundary IRs 2-3 and 3 which pose the greatest risk of error in distinguishing between virulence and avirulence [37].In doubtful cases regarding the virulence of the isolates, and in all cases when rare virulences were found (usually up to 3% of the frequency), reinoculation was carried out.In accordance with the gene-for-gene concept [38] pathotypes were postulated [39] on the base of virulence/avirulence isolate phenotypes on the set of barley differentials ranked in the given order.Details of materials and methods including other demonstration images have been recently published [31].

Pathotype Classification
To classify isolates a set of differential varieties used in all three years was used.The numerical pathotype designation of isolates was based on their virulence/avirulence pattern on the set of 40 differentials divided into 13 triplets and the last variety on its own.Each of the digits indicates virulence to the three differentials of the respective triplet.If virulence to a corresponding variety was detected, the first differential is given the value 1 (2 0 ), the second differential has the value 2 (2 1 ), and the third differential is 4 (2 2 ).Therefore, each digit can have a value from 0 (no virulence to any of the three differentials) up to 7 (1 + 2 + 4), denoting virulence to each of the three varieties [40].The resulting number (reverse-octal) [41] defines the pathotype classification of isolates.The HaGiS program was used for the transcription of the infection response arrays (IRAs) into the notation [42].

Results
Over a three-year period, population samples of 299 isolates were studied on 121 host differentials.Of these, 37 differentials were used in two and 23 in only one year (Table 2).Hence, virulence frequency (VF) was determined on 95 differentials in the first year, on 92 in 2021 and on 93 host genotypes in the last year.

Virulence of Isolates and Virulence Frequency of the Population
In 2019 there were no virulent isolates recorded on 39 differentials (VF = 0%) and none on 31 differentials in 2021 and 2023 (=non-differentiating varieties).

Pathotype Diversity
The virulence of isolates to resistance of differentials resulted in IRAs.For the numerical pathotype designation, the core set of 40 differentials (nos.15-54) was arranged according to the ascending total VF.Based on their IRAs, 299 isolates were assigned to 291 pathotypes (Table S1) when each of six pathotypes (0001221, 0030571, 0005771, 0074671, 0076771 and 0147771) was represented by two and one pathotypes (2203271) by three isolates (note that in designating these pathotypes the first seven zeros were omitted here).Two pathotypes (0030571 and 0005771) found in 2019 were represented by two isolates (G-2, G-18 and R-4, R-28 respectively).In the same year, 34 additional differentials were used, and the first pair of isolates differed in virulence to A-222 and the second pair in virulence to Prosa.Two isolates (E-1 and R-5) found in 2023 belonged to pathotype 0147771 but they differed in virulence to the resistance of Mirko.Also, two other isolates (N-2 and O-1) out of three represented by pathotype 2203271 were found in 2023 and they differed in virulence to resistance of Gilberta.The third isolate (Y-1) was collected in 2021, and in 2021 and 2023 13 joint additional differentials were used (nos.86-98).However, none of these could differentiate among these three isolates nor three other pairs of isolates found in 2021 and 2023 (C-12/2021, M-1/2023; I-1/2021, O-4/2023 and X-4/2021, F-1/2023).Hence, four pairs of isolates were differentiated using the named additional differentials and four were not.In summary, the given population sample of 299 isolates belonged to 295 pathotypes (Simple Index − SI = 0.987) (SI = the number of pathotypes/number of isolates) where almost every isolate belonged to a different pathotype.

Complexity of Virulences
The virulence complexity of isolates varied in a wide range continuously from 4 up to 19 with the exception of 5 (Table 3).The most numerous were isolates with a complexity of 8 up to 12 virulences and a frequency of 34-47 which included 204 (68.2%) isolates.The least common were isolates with extreme complexities 4 (2 isolates), 5 (0), 17 (2), 18 (1) and 19 (2).The average virulence complexity of isolates was 10.39.

Selection of Isolates for Resistance Gene Postulation
In total, 22 isolates were selected for future use in postulating resistance genes, five in 2019, eight in 2021 and nine in 2023 (Table S1).Isolates were chosen according to their rare virulences or suitable virulence combinations.

Discussion
This contribution completes more than six decades of monitoring of the Bh population conducted in our laboratory, and the research was always closely linked with the identification and use of new host-specific resistances [22,43], designation of new resistance genes [44,45] and resistance/gene postulation in varieties including those newly registered as commercial cultivars [46,47].
The last population study was performed on 50 differentials [21]; 47 of them were also used here, whereas two were excluded after identifying their resistance: NORD12/1122 which contained a gene of nonspecific resistance mlo, SJ048311 with a gene combination Mlp, Mlat and KM14/2010 was replaced with a sister line KM12/2010 for technical reasons.
Winter varieties Psaknon and Venezia were used as differentials for many years and the first virulences were found in 2011 (VVe) and 2012 (Vp) [29].Nevertheless, corresponding VFs remained rare.Despite this, it was recognized that these virulences are associated [48,49].In 2021 a greatly increasing number of both VFs was found and, therefore, three more varieties with Mlp (including Psaknon, which was not included in 2021) were added to the differential set in 2023 and a huge increase of both VFs was confirmed (VVe from 2.9% in 2019 to 26.2% in 2023 and Vp from 5.7% to 77.0%).
Saturn [47], a variety that was first registered in 2012 in the Czech Republic possesses Mlp and has occupied a negligible area.However, SU Ellen with the same gene and registered five years later [50] was the most widely grown winter barley variety in 2021 comprising around 10% of the crop area [51].Therefore, it is likely that the directional selection and the migration of virulent pathotypes from surrounding countries, where varieties with Mlp had already begun to be cultivated, contributed to the dramatic increase in corresponding VF.
No variety carrying resistance of Venezia (MlVe) has been grown in the Czech Republic.Then the fast increase in frequency VVe could occur as a result of an increase in Vp due to hitch-hiking selection [52,53] and migration of the relevant virulence from neighboring countries.
Many differentials contain an unknown resistance, some of them possibly with the same major genes present in other varieties.Isolates selected here will serve as a useful tool for studying and identifying (postulation) their genes.At least 10 differentials should have SI-1 resistance, but the virulence of a few isolates differed in these as well as in some other varieties probably because of the presence/absence of some additional genes as was demonstrated on a set of varieties carrying resistance Lv [50].Since 2012 six SI-1 differentials have been registered in the country but only Bente was grown on 1.7% in 2020 up to 3.9% of the spring barley area in 2023 [51].VF to the resistance in Bente reached 6.6% and in Camilla 7.4% in 2023 and their resistance already has no importance in the field.New virulences were found on SJ123063, SY412-329 and KM12/2010.Some other VFs differed non-significantly from a previous study [21], and it is difficult to explain these differences since many factors can play a role such as different evolutionary forces in the population, the epidemiological situation in different parts of the country, meteorological conditions during spore-trapping, etc.Some VFs showed a tendency to increase during the monitored period e.g., Va1, Va3, Va7 or VIM9, whereas some decreased (Va9, Vat and VRo).
In previous tests, three varieties (Adam, Leenke and LG Nabuco) were resistant to all pathotypes, and their resistance gene(s) remained unknown (first two varieties) or unsure (LG Nabuco) [50].A VF of 0% confirmed the complete effectiveness of their resistance and the occurrence of an IR0(2) enabled the presence of mlo to be deduced.
In 2023 Engledow India possessing a resistance gene designated Mla24 [54] was included as a differential.Its IR and VF were identical to these parameters on P11 carrying Mla13 [33].It appears that both varieties have an identical resistance gene previously designated Mla13 [55].
In 2016 and 2017, 226 isolates collected on an almost identical sampling route were analyzed and assigned to 224 pathotypes [21] (SI = 0.991).In this report, 299 isolates were studied in three years and 295 pathotypes were established (SI = 0.987).Thus, the present results confirmed an extremely high diversity of pathotypes that must be one of the highest among fungal plant pathogens.Two other characteristics are also high but further increasing, namely a wide spectrum of virulences and high virulence complexity of isolates.The causes of this situation have already been discussed [21], and apart from the high evolutionary potential of the pathogen, one must also consider the extensive cultivation of host crops, the continued use of a diverse set of resistances and long-term exploitation of specific resistances and suitable conditions for the pathogen development in the given area.The method of sampling spores (on average one analyzed isolate per >10 km of the sampling route) and the number of the selected host varieties with a known set of specific resistances in the differential set surely contributed to the discovery of the exceptional pathotype diversity.
Selected VFs found here can be compared with VFs in some non-European countries to show the role of directional selection in Central Europe (Table 4).High VF to the resistance gene Ml(Ru2) found in Chinese populations is also the result of directional selection because, in old Chinese barley varieties, this gene was found in 69 out of 147 tested accessions [56].References of virulence to the nonspecific gene mlo are probably incorrect [2,3,59].If this finding is correct, then the news should be widely disseminated to breeders and researchers so that remedial action can be taken.A similar announcement that avirulent isolates to the resistance gene Mla8 had been discovered [59,60] has not been confirmed and there is only one known old pathotype (Race I) [61] available for specific research projects [62].Unfortunately, most of the recent population studies [3,58,59] use only Pallas near-isogenic lines [33] carrying old "archival" resistance genes.
Our laboratory staff have studied Bh populations in all non-polar continents and the greatest differences in VFs and virulence complexity were found between Central European and Australian populations [27,57], and differences were confirmed with molecular characteristics of both numerous sets of isolates [28].However, the results of another study of the Australian population were in many cases fundamentally different [60].
In population studies of plant pathogens, isolates should be properly designated reflecting their virulence combinations.Creating numerous systems has a long history and many of them have been developed, e.g., hexadecimal, based on 16 patterns designated with capital letters [63] that was subsequently used mainly in studies of cereal rusts, and an octal system based on a short, simple and logical mathematical row (see Materials and Methods, Section 2.5) [40], which was used preferably in PMs studies.A consensus of researchers agreed that modified octal notation (reverse octal notation) is the most appropriate for the given purpose [41] and it was recommended for general use [64].Reverse octal notation has subsequently started to be employed for also designating resistance (IRAs) of host genotypes [12,13].However, the hexadecimal system was recently proposed as "new" for characterizing oat PM pathotypes [65].This contradicts the general recommendation and established methodology and is surely a retrograde step in plant pathological research.
Population studies should enable researchers to understand a pathogen and processes operating in its population.However, in many cases, the results are not comprehensible, and the conclusions of some papers are misleading, e.g., to designate resistance genes as "compromised" when corresponding VF is up to 50% [60], or to designate them as "highly resistant" even if the VF is about 10% [3].Such statements can lead to false conclusions, overestimation of specific resistance and lead to the unsuccessful breeding of varieties resistant to the pathogen.
Specific resistance against barley PM can be an invaluable tool for pathologists, especially if based on genes expressing resistance as IR0 that maintains plants free from disease symptoms after inoculation with avirulent pathotypes.However, there are many examples of specific resistances being overcome in a short time (Table 5) and probably cannot maintain a sufficiently durable specific resistance even when more genes are combined (pyramided) in a variety.Therefore, until an effective way to prolong the durability of a specific resistance can be discovered, other possibilities should be explored [16].These can be summarised as follows: (1) the wider exploitation of Mlo resistance especially outside Europe, (2) to determine whether Mlo resistance is a universal "weapon" against PM [66] and can be safely used in both forms of barley (winter and spring) in areas where they are intensively grown together, (3) to explore non-specific quantitative resistance genes, or (4) to exploit non-host resistance in updated meaning [67] derived from species naturally attacked by distantly related pathogens.

Figure 1 .
Figure 1.A car with a universal doubled jet spore sampler.

Figure 2 .
Figure 2. (a) Colonies of Blumeria hordei developed from single spores caught with a jet spore sampler; (b) Sucking a spore progeny conidia into an exchangeable tip of a varipipette; (c) Blowing conidia off the tip into a micro-settling tower.

Figure 2 .
Figure 2. (a) Colonies of Blumeria hordei developed from single spores caught with a jet spore sampler; (b) Sucking a spore progeny conidia into an exchangeable tip of a varipipette; (c) Blowing conidia off the tip into a micro-settling tower.

Table 1 .
Sections of the sampling route for the collection of spores of Blumeria hordei from the air in the Czech Republic in 2019-2023 and the numbers of isolates studied.

Table 2 .
Barley differential varieties, their Ml resistance genes and corresponding virulence frequency found in a Central European Blumeria hordei population.

Table 3 .
Virulence complexity and number of isolates found in a Central European population of Blumeria hordei in 2019-2023.

Table 5 .
[68]kdown of powdery mildew resistances of barley varieties carrying specific resistance gene(s) in Czech registration trials due to adaptation of the pathogen (Blumeria hordei)[68].The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/jof9111045/s1.TableS1: Year of sampling a Central European Blumeria hordei population, isolate and pathotype designation and their virulence complexity.The study was funded by the Ministry of Agriculture of the Czech Republic, institutional support no.MZE-RO1123.
Funding:Informed Consent Statement: Not applicable.