Non-2NS blast resistant wheat genotypes evaluated in the Brazilian Cerrado

ABSTRACT Searching for novel sources of resistance to head blast is essential to strengthen wheat production in the Cerrado´s biome. The objective of this work was to evaluate disease intensity measures and yield for 2NS and non-2NS carriers wheat genotypes with varying heading times in Minas Gerais, Brazil. A total of fourteen wheat genotypes, two susceptible and twelve resistant to head blast, were sown in 2014, 2015, and 2017 at the Sertãozinho Experimental Station of Empresa de Pesquisa Agropecuária de Minas Gerais (Epamig) during three sowing dates without fungicide applications. The experimental design was a randomized complete block with three replicates. Blast incidence, severity, disease index and grain yield were influenced by the cycle of wheat genotypes and the sowing date, with the highest disease intensities and the lowest yields in the earliest sowing date. Blast incidence and disease index correlated negatively with grain yield and positively with percent yield losses. The group of wheat genotypes with higher grain yield (between 2,104.7 and 2,917.8 kg ha-1) and lower yield losses (between 44.3 and 54.8%) includes BR 18 as well as other five that do not carry the 2NS/2AS translocation: BRS Angico, PF 909, BRS 229, Embrapa 27, and CPAC 07340.


INTRODUCTION
_______________________________________________ Submitted on April 28 th , 2022 and accepted on October 24 th , 2022. 1 Universidade de Passo Fundo, Passo Fundo, Rio Grande do Sul, Brazil.natywebber@hotmail.com;carolinadeuner@gmail.com 2 Empresa de Pesquisa Agropecuária de Minas Gerais, Campo Experimental de Sertãozinho, Patos de Minas, Minas Gerais, Brazil.macoelho62@gmail.com 3 Embrapa Trigo, Passo Fundo, Rio Grande do Sul, Brazil.gisele.torres@embrapa.br;luciano.consoli@embrapa.br 4 Universidade Federal de Viçosa, Viçosa, Minas Gerais, Brazil.cecon@ufv.br*Corresponding author: gisele.torres@embrapa.brIn Brazil, wheat (Triticum aestivum) is mainly cultivated in the South region, with Paraná and Rio Grande do Sul states accounting for 87% of national production in 2021 (Conab, 2022).Despite the 23.2% increase of national production compared to 2020, it is estimated that the 7.7 million tons produced will be sufficient to supply 61% of domestic consumption.The Central Brazil region, in the Cerrado biome, is the current agricultural frontier for the expansion of wheat cultivation (Farias et al., 2016).In this region with a hot and dry climate, wheat is produced in both rainfed (higher altitude areas) and irrigated conditions (Cunha et al., 2011).Considering areas of high altitudes, the potential of this region for wheat cultivation is two to three million hectares (Albrecht et al., 2007).However, the availability of wheat cultivars with resistance to wheat head blast is indispensable for strengthen such production.
Magnaporthe oryzae (syn: Pyricularia oryzae) is a fungal species that infects more than 50 species of grasses and causes the disease called blast.Different pathotypes are designated depending on the host plant.In wheat, Magnaporthe oryzae Triticum (MoT) pathotype causes disease symptoms on all above-ground parts of wheat plants but spike infection is more destructive.The penetration of the pathogen in the rachis prevents the translocation of assim-Natália Forchezato Webber et al.
ilates for grain filling.Grain development is impaired and the spikes become bleached.Disease symptoms are more severe the earlier the infection resulting in small and shriveled grains, with low test weight (Goulart et al., 2007).Losses of up to 100% are reported depending on the susceptibility of wheat cultivars (Kohli et al., 2011).
Blast is present in all wheat producing regions of Brazil and for 30 years it was restricted to countries of South America (Bolivia, Paraguay and Argentina) (Kohli et al., 2011).In 2016, wheat blast was reported in Bangladesh with wheat yield reductions ranging from 5-51% in the affected fields (Islam et al., 2016).Bangladesh is close to Asian countries that rank among the world's top ten wheat producers (China, India, and Pakistan).During the 2017-2018 rainy season, wheat blast symptoms were observed in Zambia, Africa (Tembo et al., 2020).The intercontinental spread of wheat blast poses serious threats to food security since wheat is the staple crop for 40% of world population (Giraldo et al., 2019).
The use of resistant cultivars is the most economic and environmental friendly strategy to control wheat blast.
The search for resistant cultivars has been the focus of research on wheat blast and thus far five genes of specific resistance to MoT isolates have been described (Ferreira et al., 2020).In addition to these genes, it was found that the presence of a chromosomal translocation (2NS/2AS) from Aegilops ventricosa in wheat confers resistance to the pathogen (Cruz et al., 2016).Resistance sources carrying this translocation have been used in South America and Bangladesh crops to control the disease (Cruppe et al., 2020).However, the high genetic variability of the pathogen and the strong isolate-cultivar interaction are widely known (Maciel et al., 2014).Even the resistance conferred by the 2NS translocation is not so effective faced to new isolates of the pathogen (Cruz & Valent, 2017).Therefore, searching for novel sources of wheat resistance to blast is imperative.

Annually, the Brazilian Wheat and Triticale Research
Commission publishes data about the reaction to blast of wheat cultivars indicated for cultivation in the country (RCBPTT, 2020).One hundred seventeen wheat cultivars were currently registered for use in Brazil and for 41 (35%) of them some degree of blast resistance is reported.
Nevertheless, no data concerning wheat yield performance under conducive disease conditions is presented.
In the context of tropical and subtropical regions of wheat cultivation in Brazil, more than characterize spike reaction it is essential to evaluate crop losses under field conditions.A strong correlation between blast incidence on the spikes and yield loss was found for early cycle genotypes evaluated in a hotspot site for wheat blast (Dianese et al., 2021).In this work, the objective was to assess disease intensity and yield losses for 2NS and non-2NS carriers wheat genotypes with different heading times in the Cerrado of Minas Gerais under conditions of wheat cultivation.1).These genotypes were selected due to their lower blast incidences than Trigo BR 18-Terena, hereinafter called BR 18.For comparison of disease and yield data, one resistant (BR 18) and two susceptible checks (Anahuac 75 and BRS 209) were also sown.Among the 12 resistant genotypes, two of them (CBFusarium ENT014 and CPAC 07434) possess the 2NS/2AS translocation.Three classes of cycle were represented by wheat genotypes with different heading times (in days): early (from 54 to 62 days), medium (from 66 to 73 days) and late (from 82 to 86 days).Each experimental unit consisted of five 4-m row, which was replicated three times per genotype.The distance between rows was 0.2 m.A useful 1.8m 2 area per plot was marked and included the three central rows, with the elimination of 0.5 m at each end line.The fertilization was made with 350 kg ha -1 NPK (8-28-16) before planting, and 80 kg ha -1 N as urea were applied at 25 days after sowing.Plots were weekly irrigated at a 10-mm depth, except when rainfall occurred.Experiments were harvested from 77 until 140 days after sowing for date 1, from 90 until 143 days after sowing for date 2, and from 94 until 142 days after sowing for date 3.

MATERIAL AND METHODS
Non-2NS blast resistant wheat genotypes evaluated in the Brazilian Cerrado The experimental design was a randomized complete block with 14 genotypes and three replicates.Blast incidence, severity, disease index and grain yield were evaluated.For each wheat genotype, 100 spikes were randomly collected from the useful area per plot.Blast incidence was calculated as the proportion of spikes with bleaching symptoms and blast severity was calculated as the average percentage of bleached spikelets, including spikes with zero severity.Thereafter spikes with no symptoms were also accounted to the final severity of the plot.The incidence and severity values were then used to estimate the blast index, which is the product of incidence and severity (Juliana et al., 2020).Spikes were threshed, grains were weighed, and plot yield was estimated in g m -2 and then converted to kg ha

RESULTS AND DISCUSSION
All variables -blast incidence, severity, disease index and grain yield were affected by the length of the cycle and sowing date, as well as their interaction.Despite the interaction observed, the results show the occurrence of disease with the same magnitude regardless of the cultivar cycle.In late sowing dates, reduction of disease and yield losses were observed.For all classes of crop cycle, blast incidence was highest at the earliest sowing date (Table 2).Blast incidence observed in sowing date 2 was significantly higher than that of sowing date 3 for early genotypes, but not for medium and late ones.Considering blast severity values, there was no difference between sowing date 1 and 2 for genotypes with early heading time.For medium and late cycle genotypes, sig-nificant differences were observed among blast severity of sowing dates 1, 2 and 3. Blast disease index varied similarly to observed for blast incidence with higher values in sowing date 1.For a different group of wheat genotypes, higher disease intensities were also observed upon earlier sowing (Coelho et al., 2016).Grain yield was higher at sowing date 3 for genotypes with early and medium cycles.For the latter, there was no difference between dates 2 and 3.For late genotypes, productivity at date 2 was higher but not significantly different to date 3, and means of date 3 and 1 did not differ statistically.
Differences in grain yield upon different sowing dates are the result of disease intensity influence and of adaptability of wheat genotypes analyzed to this region of cultivation in Brazil.In the sowing date 1, blast incidence of wheat genotypes with medium cycle was significantly lower than that of early and late ones (Table 2).The establishment of the disease depends on temperature and wetting period (Cardoso et al., 2008).It is possible that weather conditions less favorable for the disease occurred during the period comprised between 66 and 73 days after sowing date 1 (the heading time for genotypes with medium cycle).On the other hand, for sowing dates 2 and 3, blast incidences are similar between genotypes with different heading classes.
During sowing dates 1 and 3, the highest blast severities were observed for wheat genotypes with late heading.This could be explained by the longer exposure time of these genotypes to field conditions, allowing greater progress Non-2NS blast resistant wheat genotypes evaluated in the Brazilian Cerrado of the spike bleaching and also new infections on non-infected heads.Considering blast disease index, wheat genotypes with different cycles could be clearly discriminated only in sowing date 1; significantly greater values were observed for genotypes with late heading > early heading > medium heading.For all three sowing dates, grain yield data of early and medium heading wheat genotypes were significantly higher than that observed for late heading genotypes.Based on these results, it can be recommended for this region the use of wheat genotypes with early or medium heading that have globally lower blast intensity and higher grain yields.
Data from sowing date 1 of the three years of evaluation (2014, 2015 and 2017) were considered for joint ANOVA.
For all variables analyzed, there was a significant effect of the wheat genotype (p ≤ 0.01).The comparison of means of genotypes by the Scott-Knott test at 5% significance is presented in Table 3. Anahuac 75 presented the highest blast incidence (63.5%), blast severity (65.8%) and disease index (43.9).This wheat cultivar is widely known to be susceptible to a large number of isolates of the pathogen (Arruda et al., 2005)  Natália Forchezato Webber et al.
Greater differences among genotypes were observed for blast incidence rather than for blast severity to which two groups of means were identified (Table 3).Ten genotypes had higher values of severity ranging from 47.8 to 65.8% (group a).Both susceptible (Anahuac 75) and resistant (BR 18) checks were in this same group.Therefore, we postulate that despite the reduced incidence BR 18 presents lower resistance to spread of the pathogen in the spike tissues.After a successful infection, further colonization (and probably progress of blast severity) can continue even in conditions above or below the optimum for the pathogen (Cardoso et al., 2008).In contrast to our observation, Rios showed significantly lower blast severity than BRS 209 (Ferreira et al., 2020).Upon inoculation, the differential reaction of wheat genotypes cannot be evaluated through incidence because every inoculated spike showed bleaching symptoms.On the other hand, severity levels reveal which genotypes present mechanisms to restrain pathogen colonization (Ferreira et al., 2020).Dianese et al. (2021) infer that upon field prone conditions to wheat blast, once a spike is infected by MoT blast severity on that spike will generally be very high.Anahuac 75 showed the highest blast disease index (43.9)which was significantly different from all other genotypes.The lowest disease indexes were observed for eleven genotypes: Safira (4.1), BRS 229 (5.0), CBFusarium ENT014 (5.7), PF 909 (7.4), Embrapa 27 (7.7),BRS Angico (8.1), CPAC 07340 (9.8), BR 18 (10.7),CPAC 07434 (12.9),Trigo Chapéu (14.5) and Huanca (19.0) (Table 3).93.8% of the lines without the 2NS translocation had mean blast indices greater than 30 (Juliana et al., 2020).
Crop losses may be expressed in absolute terms (kg ha -1 ) or in relative terms (loss in %).The FAO definition of yield loss is the difference between the attainable yield (determined by the genotype in a specific environment) and the actual yield, which is actually harvested (Savary et al., 2012).In the present study, yield losses were estimated in terms of percent yield reduction relative to sowing date with the highest yield, for each wheat genotype.Although few results are described comparing yield losses of varieties with different resistance levels, this trait is important to identify tolerance mechanisms (Shankar et al., 2021).
The lowest yield losses varied between 33.3 and 54.8% and were observed for BR 18 and other nine genotypes (group b, Table 3).In Mato Grosso do Sul damages to yield oscillated from 10.5 to 13% under natural infection (Goulart et al., 2007).However, considering a highly susceptible wheat cultivar in the same region, yield reduction attained 51% (Goulart & Paiva, 2000).We can conclude that environmental conditions in Patos de Minas were more favorable for the development of the disease than they were in Mato Grosso do Sul.
The highest yield losses varied between 62.9 and 79.2% (group a, mean of 72.2%, Table 3) and are similar to that ranging from 62.8 and 80.1% found for a different set of wheat genotypes evaluated (Coelho et al., 2016).For localities where blast is not endemic, the occurrence of the disease depends on inoculations (Rios et al., 2016).The analysis of four wheat genotypes with different levels of resistance, upon inoculation, revealed yield reductions from 19 to 42% (Gomes et al., 2017) infer obtaining even higher yields with the use of fungicide applications to control the disease on these resistant wheat cultivars.The authors reported both strategies combined reducing the negative effects of the disease on wheat phys-Non-2NS blast resistant wheat genotypes evaluated in the Brazilian Cerrado iology and increasing photosynthetic performance (Rios et al., 2017).
Correlation between blast incidence and severity was moderate but significant (0.55) (Table 4).The calculation of a blast disease index search to estimate the effect of both parameters of quantification of the disease, incidence and severity.Blast index showed moderate to high correlations with severity (0.66) and with incidence (0.97).Grain yield was negatively correlated to incidence (-0.81) and disease index (-0.73).In their turn, yield losses were positively correlated to blast incidence (0.63) and disease index (0.59) but negatively to grain yield (-0.70).In the same municipality, similar results were obtained by Coelho et al. (2016).Under blast hotspot conditions, a positive correlation between yield losses and blast incidence was also observed for early cycle genotypes (Dianese et al., 2021).
Actually, yield impacts due to the disease depend on the resistance of the host plant, on the environmental conditions and even on the pathogen populations, which tend to be diverse in different regions (Maciel et al., 2014).Mean disease index values varied with similar ranges in Brazil and Bolivia (Cruz et al., 2019).Nonetheless, under high pressure of disease, penalties to yields are higher in Brazil (Cruz et al., 2019).The significance of these correlations varied with different wheat genotypes (Gomes et al., 2017) but they are proven to be significant for wheat genotypes with different heading timing under the conditions of the present work (Table 4).In field trials, besides being laborious the estimation of blast severity is dispensable since as incidence as disease index showed significant negative correlations to yield losses.Therefore, blast incidence can be used as the parameter for evaluation of wheat reaction to the disease.The 2NS/2AS translocation-based lines are currently used as sources of blast resistance in South America countries and in Bangladesh (Cruppe et al., 2020).Besides the variability of resistance levels conferred by this translocation, their solely use can risk to select more aggressive races of the pathogen.Though finding novel sources of blast resistance is imperative (Cruz & Valent, 2017;Juliana et al., 2020;Dianese et al., 2021;Juliana et al., 2022).Cruppe et al. (2020) evaluated over 780 accessions searching for identifying non-2NS sources of resistance to wheat blast, and only 1% of them was characterized as resistant or moderately resistant to the pathogen.However, no data about grain yield was obtained (Cruppe et al., 2020).The present work identified wheat genotypes with reduced blast intensity and also reduced yield losses upon high blast pressure in Minas Gerais State (located at Cerrados region) where wheat blast is endemic.Interestingly, most of the wheat genotypes with lower losses had the highest grain yields under these conducive conditions for the disease (

CONCLUSIONS
Regardless of the heading time of the wheat genotype, blast intensity is higher and grain yield is lower in the first sowing date.
Blast incidence and disease index showed a significant

Juliana
et al. (2020) considered highly resistant wheat lines having mean blast indices less than 10.This is the case for seven of the cited eleven genotypes.Interestingly, six (Safira, BRS 229, PF 909, Embrapa 27, BRS Angico and CPAC 07340) out of these 11 genotypes (54.5%) are non-2NS carriers while in field trials in Bolivia and Bangladesh, negative correlation with grain yield, and a positive correlation with yield losses.Associated with blast incidence, yield losses is an important trait to identify mechanisms of tolerance to wheat blast.Six wheat genotypes without the 2NS translocation (BR 18, BRS Angico, PF 909, BRS 229, Embrapa 27 and CPAC 07340) showed the highest grain yields and the lowest yield losses similar to those of the two genotypes (CBFusarium ENT014 and CPAC 07434) carrying the 2NS translocation, known as the major source of wheat blast resistance.

Table 1 :
Wheat (Triticum aestivum)genotypes analyzed in this study

Table 2 :
Mean blast incidence, blast severity, disease index, and grain yield for wheat (Triticum aestivum) genotypes with different cycles in three sowing dates in 2014 and 2015Means followed by equal capital letters in the rows and equal lowercase letters in the columns, for each trait, do not differ by the Tukey´s test, at 5% significance.
. Its cultivation in Brazil was aban-

Table 3 :
Mean blast incidence, blast severity, disease index, grain yield and yield losses for fourteen wheat (Triticum aestivum) genotypes in the first sowing date in three years of cultivation(2014, 2015 and 2017) Means followed by equal letters, in the columns, belong to the same group by the Scott-Knott test, at 5% significance.

Table 4 :
Coefficients of Pearson´s correlation between blast incidence, blast severity, disease index, grain yield and yield losses for fourteen wheat (Triticum aestivum) genotypes in the first sowing date in three years of cultivation(2014, 2015 and 2017) *, **, *** Significant by the t test, at 5%, 1% and 1 o /oo probability.

Table 3 )
. These data indicate that Safira did not show any tolerance to the disease.Unlike this result, Dianese et al.
Table 3).Apart from CBFusarium ENT014 and CPAC 07434, that have the 2NS translocation, the group Natália Forchezato Webber et al. with higher grain yield and lower yield losses includes BR 18 as other five genotypes that do not carry the 2NS/2AS translocation: BRS Angico, PF 909, BRS 229, Embrapa 27, and CPAC 07340.All these wheat genotypes are good options for breeding programs to improve both reaction to the disease and grain yield performance.