Marker-assisted selection to pyramid Fusarium head blight resistance loci Fhb1 and Fhb2 in the high-quality soft wheat cultivar Yangmai 15

HU Wen-jing ,FU Lu-ping ,GAO De-rong ,Ll Dong-shengLlAO Sen,LU Cheng-bin＊

1 Key Laboratory of Wheat Biology and Genetic Improvement for Low &Middle Yangtze Valley,Ministry of Agriculture and Rural Affairs,Lixiahe Institute of Agricultural Sciences,Yangzhou 225007,P.R.China

2 College of Agronomy &Center for Crop Genome Engineering,Henan Agricultural University,Zhengzhou 45002,P.R.China

3 Jiangsu Key Laboratory of Crop Genetics and Physiology/Jiangsu Key Laboratory of Crop Cultivation and Physiology/Agricultural College,Yangzhou University,Yangzhou 225009,P.R.China

4 Jiangsu Co-Innovation Center for Modern Production Technology of Grain Crops,Yangzhou University,Yangzhou 225009,P.R.China

5 College of Agriculture,Yangtze University,Jingzhou 434023,P.R.China

Abstract Fusarium head blight (FHB) is one of the most detrimental wheat diseases which greatly decreases the yield and grain quality,especially in the middle and lower reaches of the Yangtze River of China.Fhb1 and Fhb2 are two major resistance loci against Fusarium graminearum.Yangmai 15 (YM15) is one of the most popular varieties in the middle and lower reaches of the Yangtze River,and it has good weak gluten characters but poor resistance to FHB.Here we used Fhb1 and Fhb2 to improve the FHB resistance of YM15 by a molecular marker-assisted selection (MAS)backcrossing strategy.The selection of agronomic traits was performed for each generation.We successfully selected seven introgressed lines which carry homozygous Fhb1 and Fhb2 with significantly higher FHB resistance than the recurrent parent YM15.Three of the introgressed lines had agronomic and quality characters that were similar to YM15.This study demonstrates that the pyramiding of Fhb1 and Fhb2 could significantly improve the FHB resistance in wheat using the MAS approach.

Keywords: Fusarium head blight,marker-assisted selection,Fhb1,Fhb2,resistance breeding,wheat

1.lntroduction

Fusarium head blight (FHB) caused byFusarium graminearumis one of the most destructive diseases of wheat worldwide,and it causes a drastic reduction in grain yield (Figueroaet al.2018).FHB-infected kernels contain mycotoxins such as deoxynivalenol,which is a major concern for animal production and human health (Bai and Shaner 1994).In China,FHB affects up to 7 million hectares of wheat and causes 2.5 million tons of grain damage each year (Chenget al.2003;Baiet al.2004).The breeding and development of resistant cultivars is the most economic and effective control method against FHB(McMullenet al.1997;Gilbert and Tekauz 2000).

FHB resistance has been classified into five types,the two most important of which are types I and II (Zhanget al.2021).Type I prevents the initial infection of the fungus within spikes and type II prevents the spread of the fungus within spikes.The evaluation of type II resistance is much more stable compared with that of type I resistance,so type II is usually applied in most studies on genetics and breeding (Bai and Shaner 2004).To date,only a few varieties with resistance to FHB have been identified,such as Sumai 3 (SM3) (Andersonet al.2001),Wangshuibai (WSB) (Zhouet al.2004),Goldfield(Gilsingeret al.2005),and Frontana (Steineret al.2004).However,complete resistance or immunity to FHB has not been achieved yet in wheat.In the past two decades,extensive mapping of quantitative trait loci (QTLs) for FHB resistance has been conducted,and many loci have been mapped on almost all 21 wheat chromosomes (Andersonet al.2001;Buerstmayret al.2002;Zhouet al.2002;Linet al.2004;Steineret al.2004;Yanget al.2005;Cuthbertet al.2006).Fhb1andFhb2from SM3 and WSB,respectively,are two major loci for type II resistance (Baiet al.1999;Waldronet al.1999;Andersonet al.2001;Buerstmayret al.2003;Yanget al.2003;Zhouet al.2003,2004).

FHB resistance is a quantitative trait controlled by polygenes and it is greatly affected by the environment(Zhanget al.2021).Therefore,accurately screening out the FHB resistance varieties by phenotypic selection in wheat breeding is difficult because the phenotype evaluation requires suitable facilities,uniform inoculation methods,assessments of repeated trials,and the investment of considerable labor and time inputs (Zhanget al.2021).SM3 and WSB are two major sources of FHB resistance genes in wheat breeding programs worldwide (Ruddet al.2001;Jiaet al.2018;Zhuet al.2019).Some varieties derived from them have high levels of FHB resistance but are not yet commercially available because of their poor agronomic traits.The introduction and pyramiding of resistance genes when processing the multi-generation selection are difficult in traditional breeding (Luet al.2013).However,markerassisted selection (MAS) is an effective breeding strategy for improving FHB resistance and overcoming the adverse agronomic traits of the source germplasms in conjunction with FHB resistance.

Initially,a major FHB resistance QTLs of type II on chromosome 3B that accounted for up to 60% of the phenotypic variation was identified from SM3 (Baiet al.1999).Later,Zhouet al.(2003) confirmed this QTL using a recombinant inbred line (RIL) population derived from a cross between resistant variety SM3 and susceptible variety Alondra’s.It was designated asFhb1,and the two flanking SSR markersXgwm493andXgwm533were developed for MAS.Jiaet al.(2005)studied the DH populations derived from WSB/Alondra’s and SM3/Alondra’s,and found that linkingXgwm533-3BwithFhb1could be helpful in FHB resistance breeding.A pore-forming toxin-like gene (PFT) was reported as the candidate gene ofFhb1in 2016 (Rawatet al.2016),however,Heet al.(2018) reported that the FHB resistance of some cultivars was not controlled exclusively byPFT.A histidine-rich calcium-binding protein (His) was identified as playing a role in theFhb1resistance in 2019 (Liet al.2019;Suet al.2019).Functional markers were developed based on the critical sequence deletion ofTaHRCin theFhb1region and they could be used as the diagnostic markers in wheat breeding (Suet al.2018).However,the resistance mechanism to FHB ofFhb1has not yet been confirmed.Another type II resistance QTL,Fhb2(on 6BS),which was tightly linked by SSR markersXgwm644andXgwm133(Sourdilleet al.2004) and could reportedly reduce FHB severity by more than 50% (Cuthbertet al.2007),has not been cloned yet.However,the SSR markers flanking the two resistance loci were generally used in the early MAS of FHB resistance,and the genotyping was considered to be stable and reliable (Pumphreyet al.2007;Foxet al.2013;Zhanget al.2016;Zhouet al.2018).China has the largest amounts of wheat production and consumption suffering from severe FHB damage,especially in the middle and lower reaches of the Yangtze River with its warm and humid environment.Yangmai 15 (YM15) is the most popular variety with the best weak gluten quality that is suitable for cookies,pastries and wine in the middle and lower Reaches of the Yangtze River.However,YM15 is susceptible to FHB and its yield and quality would be seriously threatened in case of a scab outbreak (Huet al.2020b).Improving the FHB resistance of YM15 could be the most effective way to provide a new high-quality variety for wheat production.In this study,we introducedFhb1andFhb2from SM3 into YM15 using the MAS backcrossing method and determined the agronomic and quality traits of YM15.

2.Materials and methods

2.1.Plant materials and field trials

Wheat cultivars SM3 and YM15 were used as the donor and recurrent parents,respectively.The donor of theFhb1andFhb2loci was SM3,which was released by Taihu Institute of Agricultural Sciences,Jiangsu,China in 1974.SM3 is about 114.1 cm in height,with spindleshaped spikes and 18-19 spikelets per spike.The recurrent parent YM15 was developed in Yangzhou (YZ)by Lixiahe Institute of Agricultural Sciences,Jiangsu,China in 2004,and it has good agronomic features such as 84.3 cm in height and 21-22 spikelets per spike.YM15 is susceptible to FHB and it does not carryFhb1andFhb2.Its promotion has been limited in the middle and lower reaches of the Yangtze River.Hybridization and backcrossing were conducted from 2013 to 2017 for the development of BC3F1in the Wanfu Experimental Base of Lixiahe Institute of Agricultural Sciences.After self-crossing and foreground marker selection,BC3F2was harvested in the 2017-2018 cropping season.One trial for type II resistance evaluation consisted of two blocks,in which each plot was set up with two rows of 2.0 m in length and 0.3 m row spacing,and about 25 seeds were planted per row in the 2018-2019 (2019YZ) and 2019-2020 (2020YZ) cropping seasons.Another trial for the agronomic trait,yield and quality evaluation had two blocks,and the BC3F3were sown in a 10-row plot in 2.5 m rows,with 50 seedlings per row and 0.3 m row spacing,using a randomized complete block design in the 2020YZ.In 2020-2021 (2021YZ) cropping season,one trial for the evaluation of yield and quality had two blocks using a randomized complete block design,and each plot had five 2.5-m rows spaced at 0.3 m with 50 seedlings per row.The main character testing methods were based on the national regional testing standard and both parents were used as controls.The field trials were conducted in accord with local management practices.

2.2.Phenotyping of FHB resistance

Type II resistance was evaluated for the selected lines in 2019 and 2020 using the spikelet single infusion method (Huet al.2020a).About 10 μL of mixed conidial suspension ofFusarium graminearumcontaining 1 000 spores was injected into a flowering floret in the middle of a spike.Twenty spikes were inoculated in each plot.The humidity was maintained by regular irrigation.The number of diseased spikelets per spike and the total number of spikelets of every tagged spike were recorded three weeks after inoculation.The average percentage of symptomatic spikelets (PSS) was used to indicate FHB severity.

2.3.Agronomic trait evaluation

Heading time (HT) and flowering time (FT) were investigated in 2020 for the BC3F3population.HT was the number of days from sowing until more than half of the plants were heading in the plot.FT was the number of days from sowing until more than half of the plants were flowering in the plot.Plant height (PH),spikelet number per spike (SNS),number of kernels per spike (NKS),number of spikes per plant (NSP) and kernel weight per plant (KWP) were determined for five plants randomly chosen from the middle of each plot at physiological maturity and the plot means were used in the analysis.PH was the total length of the aboveground part excluding the awn.SNS and NKS were counted from the main spikes.Thousand-kernel weight (TKW),NSP and KWP were measured after oven-drying.TKW was determined by averaging the values of three samples of 200 kernels and then multiplying by five.

2.4.Quality trait evaluation and yield measurement

The protein content of wheat whole-grain was measured using a near-infrared (NIR) analyzer (DA7200,Perten Instruments Co.,Ltd.,Sweden) according to the AACC method 39-10 (AACC 2000).The total starch content of wheat flour was measured according to the AACC method 76-13.01 (AACC 2000).Kernel hardness and moisture of the tested samples were determined by the Single Kernel Characterization System using the AACC method 55-31.03 (AACC 2000).Wet gluten content of the wheat flour was tested by referring to the AACC method 38-12 (AACC 2000).The sedimentation volume was tested in the wheat flour by referring to the AACC method 56-61A (Zhanget al.2018).The grains in each plot were mechanically harvested and dried naturally for yield measurement.

2.5.Molecular marker assays

Genomic DNA was extracted from fresh leaves of seedlings using the CTAB method (Ma and Sorrells 1995).The SSR primers used in this study were synthesized according to previously reported sequences (R?deret al.1998).Two SSR markers,Xgwm493andXgwm533,were used to identify the presence of theFhb1resistance gene from SM3,and their primer sequences areXgwm493forward,5′-TTCCCATAACTAAAACCGCG-3′;Xgwm493reverse,5′-GGAACATCATTTCTGGACTTTG-3′ andXgwm533forward,5′-AAGGCGAATCAAACGGAATA-3′;Xgwm533reverse,5′-GTTGCTTTAGGGGAAAAGCC-3′,respectively.Likewise,two SSR markers,Xgwm133andXgwm644,were used to genotypeFhb2,and their primer sequences areXgwm133forward,5′-ATCTAAACAAGACGGCGGTG-3′;Xgwm133reverse,5′-ATCTGTGACAACCGGTGAGA-3′ andXgwm644forward,5′-GTGGGTCAAGGCCAAGG-3′;Xgwm644reverse,5′-GGAGTAGCGTGAGGGGC-3′,respectively.PCR was carried out in a total volume of 25 μL containing 50 ng of genomic DNA,5 μmol L-1of each primer,1.5 mmol L-1MgCl2,50 mmol L-1KCl,10 mmol L-1Tris-HCl,0.2 mmol L-1dNTP,and 1 UTaqpolymerase.The PCR Program was composed of 34 cycles of 30 s at 94°C,45 s at 55°C,40 s at 72°C,and a final extension for 5 min at 72°C.PCR products were analyzed by 8.0%polyacrylamide gel electrophoresis.In 2019,we identified the accessions using the gene functional marker ofFhb1(TaHRC-GSM) (Suet al.2018),in order to verify the detection ofFhb1using SSR markersXgwm493andXgwm533.The PCR Program was used according to Suet al.(2018),and the PCR products were separated in a 1.0% (w/v) agarose gel.

2.6.Statistical analysis

SPSS Software (ver.22.0) was used to perform analyses of variance (ANOVA) and the least square differences(LSD) were used for multiple comparisons of all traits.The level of significance wasP＜0.05 for all data analyses.

3.Results

3.1.Development of the introgressed lines for pyramiding Fhb1 and Fhb2

Recurrent parent YM15 was crossed with donor parent SM3,and the F1hybrids were subsequently backcrossed with YM15 to produce the first backcross generation BC1F1.The segregating BC1F1population was screened using the SSR markers flankingFhb1orFhb2,and the plants carrying both loci in this generation of backcrossing accounted for 17.5%,which was slightly lower than the expected 1:4 ratio.Individual plants that contained bothFhb1andFhb2and had agronomic performance similar to YM15 were chosen for further backcrossing.We selected an average of 90 plants per generation that were more similar to the recipient YM15 for detection with the foreground-selection markers to identify the plants carrying the two target QTLs.Following the above operation,we generated 548 plants of the BC3F2populations and 11 individual plants were homozygous for bothFhb1andFhb2(Table 1;Fig.1).To verify the veracity ofFhb1detected using the SSR markersXgwm493andXgwm533,we also identified the accessions using the gene functional marker ofFhb1(TaHRC-GSM,Suet al.2018),and the detection of TaHRC-GSM was consistent with that of the SSR markersflankingFhb1.Among the 11 plants homozygous atFhb1andFhb2,seven plants with agronomic traits that were more similar to YM15 were selected for further analyses.

Fig.1 Scheme for Fhb1 and Fhb2 pyramiding in the introgressed lines.SM3,Sumai 3;YM15,Yangmai 15.

Table 1 Population size and the number of plants carrying Fhb1 and Fhb2 in the backcross and F2 generations

3.2.Genetic similarity between the introgressed lines and their recurrent parent

A total of 107 pairs of SSR markers from the polymorphism of SM3 and YM15 were used to compare the genetic backgrounds between the seven selected introgressed lines and YM15 (Table 2).The numbers of polymorphic sites between the introgressed lines and their parent YM15 ranged from 5 to 17,and the polymorphic sites were mainly detected on chromosomes 1A,1B,2A,2D,3B,3D,4A,4B,5A,5B,5D,6A,6B,7A and 7B(Table 2).Comparatively,the introgressed lines YM15-14-3,YM15-27-1 and YM15-88-5 have higher levels of genetic similarity (above 90%) to YM15.

3.3.FHB-resistance levels of the introgressed lines

YM15 and the seven introgressed lines,as well as SM3,were inoculated with a mixture of virulent strains at the beginning of flowering according to the spikelet single infusion method in the 2019YZ.The plants most similar to YM15 were selected for further evaluation of FHB resistance and agronomic traits in the 2020YZ.The FHB resistance donor parent SM3 showed the highest level of resistance to FHB (5.2 and 5.1% of PSS in 2019YZ and 2020YZ,respectively),whereas the recurrent parent YM15 was highly susceptible to FHB (49.8 and 51.1%of PSS in 2019YZ and 2020YZ,respectively).The percentages of infected spikelets of the selected lines were in the ranges of 20.6-26.0% and 19.3-25.5% in 2019YZ and 2020YZ,respectively (Figs.2-3;Table 3).These results showed that the resistance to FHB of the introgressed lines was significantly improved compared with YM15.

Table 2 Analysis of genetic similarity between the introgressed lines and the recurrent parent

Fig.2 Fusarium head blight (FHB) symptom illustrations.A,Sumai 3 (SM3).B,Yangmai 15 (YM15).C,the introgressed line with improved FHB resistance.Photos were taken three weeks after single floret inoculation.The red arrows indicate the site of the inoculation.

Fig.3 Fusarium head blight (FHB) severity (PSS) of the introgressed lines and the parents in two cropping seasons.SM3,Sumai 3;YM15,Yangmai 15.Bars mean SE (n=3).Different letters above the bars indicate significant differences among the genotypes(P＜0.05).

3.4.Comparison of agronomic characteristics between the introgressed lines and the recurrent parent

The agronomic traits of the introgressed lines and YM15 were investigated in the 2020YZ.The HT,FT,SNS,and NKS values of the lines were similar to YM15 (Table 4),but YM15-20-4 had a significantly lower SNS than YM15.Likewise,YM15-14-3,YM15-27-1 and YM15-88-5 also had TKW,NSP and KWP values similar to YM15.The plot yields of the above three introgression lines ranged from 2 549.2 to 2 589.0 g,with no significant differences from YM15 (2 602.5 g;Fig.4;Table 4).In 2021YZ,the plot yields of these three introgression lines ranged from 2 507.6 to 2 559.6 g,similar to that of YM15 (2 501.9 g;Fig.4;Table 5).Consequently,they had no significant differences in yield in the 2021YZ.These results suggested that lines YM15-14-3,YM15-27-1 and YM15-88-5 were highly similar to their recurrent parent YM15 with respect to their agronomic traits and yield.

Fig.4 Yield (A) and main quality characters (B-F) of the three introgressed lines and Yangmai 15 (YM15) in two cropping seasons.Bars mean SE (n=5).

3.5.Comparison of grain quality between the introgressed lines and the recurrent parent

Grain quality characters such as protein content,starch content,wet gluten content,kernel hardness and sedimentation volume were measured for YM15-14-3,YM15-27-1 and YM15-88-5,all of which had yields equivalent to YM15.The results showed that all three lines had the same soft gluten quality characteristics as YM15,so they all could be defined as soft wheat according to the GB/T 17320-2013 (2013) (Tables 4 and 5).

Table 3 Comparison of Fusarium head blight (FHB) resistance between the introgressed lines and the parents

Table 4 Agronomic characters and comparison of yield related agronomic traits between the introgressed lines and the parents in the 2019-2020 cropping season

Table 5 Multiple comparisons of main quality characters of the three introgressed lines with Yangmai 15 (YM15) in two cropping seasons

4.Discussion

Molecular markers are independent of environmental variables and can be screened at any growth stage during the life cycle of the plant.Thus,MAS is very helpful for breeding FHB resistant varieties of wheat (Bai and Shaner 1994).Zhouet al.(2018) has transferredFhb1from SM3 to Jimai 22 (JM22) by continuous backcrossing combined with MAS to develop and screen the new germplasms with resistance to FHB.The FHB severity of the resultant lines carryingFhb1was 31.9% lower than JM22.Six BC2populations have also been developed by crossing dwarf-male-sterile (DMS)-Zhoumai 16 with threeFhb1donors (Ningmai 9,Ningmai 13,and Jianyang 8)using marker-assisted backcross breeding based on a diagnostic marker forFhb1.The plants carryingFhb1showed reductions in disease severity by 75.6 and 33.3%compared with the recurrent parents Lunxuan 13 and Lunxuan 136,respectively (Liet al.2019).Fhb1andFhb2had different effects on FHB when introduced into the different recurrent parents,but both of them could significantly improve FHB resistance (Jiaet al.2018;Bueratmayret al.2019;Liet al.2019).Zhanget al.(2021)introducedFhb1,Fhb4,andFhb5into five Chinese wheat cultivars or lines from Henan and Sichuan provinces through marker-assisted backcross in early generations,and developed lines with the disease severity reduced by 95%incomparison with the respective recipient lines.In this study,the plants carrying the two QTLs theoretically account for 25.0% in each backcross generation.However,the results showed that plants carrying both of the QTL accounted for 13.8-17.5% in each backcross generation,which was not in accordance with the expected 1:4 ratio.For example,we examined 87 plants with the foreground-selection markers,but only 12 plants (13.8%) contained both target QTLs.We considered two possible reasons for this inconsistency.First,several plants carrying the two QTLs were excluded before marker detection due to poor agronomic traits.As a result,the proportion of plants harboring the two QTLs in the selected plants for marker detection was lower than the theoretical proportion.These results suggested that pyramiding the two QTLs was probably in conflict with agronomic trait improvement.Second,the flanking markers forFhb1andFhb2used in this study were not the functional and most closely-linked markers,so their usefulness and effectiveness were limited.More finely associated molecular markers for the targeted QTL/gene are more useful for marker-assisted breeding to break up unfavorable drags on linkage (Zhanget al.2021).Fhb1andFhb2have been cloned and mapped to a small interval in recent years (Jiaet al.2018;Liet al.2019;Suet al.2019).Therefore,in the future,we should enlarge the backcross population size,together with the use of theFhb1functional marker (Liet al.2019;Suet al.2019) and the closely-linkedFhb2flanking markers (Jiaet al.2018)for assisted background selection to obtain the expected introgression plants which are homozygous atFhb1andFhb2.Most varieties with good weak gluten quality,such as YM15 and YM13,had high FHB severity (Huet al.2020b).It is very important to retain the good traits of the recurrent parent in MAS.Our objective was to obtain the improved FHB-resistant introgressed lines by pyramiding two resistance loci more efficiently than conventional breeding and single-marker assisted breeding.The three chosen introgression lines (YM15-14-3,YM15-27-1,and YM15-88-5) and YM15 had the same agronomic characteristics and quality because of their 90% genetic similarity as shown above.This indicates that the agronomic traits and quality characters of YM15 were strongly inherited in these introgression lines.

Most breeding programs now attempt to improve FHB resistance by recombining different sources and types of resistance,simultaneously selecting for resistance and desirable agronomic traits (Clarket al.2016).Collecting cultivars with better resistance from different kinds of regions,conducting accurate phenotypic identification of FHB resistance to screen for germplasm with stable resistance,and further conducting the genetic study on resistance germplasm are necessary for FHB resistance breeding.Most YM series cultivars were confirmed as carryingQFhb.yaas-2DL/Qfhb.hbaas-2DL,QFhb.yaas-4DSandQFhb.yaas-3BLdetected in YM16 (Huet al.2020c;Zhuet al.2020).Other thanFhb1andFhb2,SM3 also harbors some minor FHB resistance loci/genes on chromosomes 2D and 5A (Braret al.2019).We found that YM15-14-3,YM15-27-1 and YM15-88-5 did not carryQFhb.yaas-2DLorQFhb.yaas-3BL,except forQFhb.yaas-4DS(unpublished data),and the FHB severity of the selected introgressed lines was higher than SM3.The FHB resistance of introgressed lines could probably be further improved if the major loci/genes and key minor resistance loci/genes were selected simultaneously in the MAS.Pyramiding the major and the minor resistant loci/genes through the MAS approach could greatly accelerate FHB resistance breeding.

Compared with YM15,the FHB resistance of YM15-14-3,YM15-27-1 and YM15-88-5 was significantly improved.However,owing to the deficient and low throughput of markers for background detection compared with the gene quantity of the wheat genome,further confirmation of the effect of the genetic background on agronomic traits in the introgressed lines should be considered,as well as the resistance ability and stability of the outputs.In this study,we measured the yield and quality characters for two years.Next,we should evaluate the yield,quality characters and FHB resistance of the best lines under multiple environments in order to commercialize them.

5.Conclusion

YM15 is an elite cultivar in the middle and lower reaches of the Yangtze River with the best weak gluten quality,but it is susceptible to FHB.After pyramidingFhb1andFhb2in YM15,three introgressed lines carrying bothFhb1andFhb2were successfully developed.These introgressed lines not only show an improved level of FHB resistance,but also have similar yield potential and good weak gluten quality compared with the recurrent parent YM15.Thus,these introgressed lines are considered not only for use as germplasms with FHB resistance,but they also have the potential to replace YM15 as the leading varieties in the middle and lower reaches of the Yangtze River in production.These results suggested that pyramiding major resistance loci in susceptible cultivars could be an efficient approach for significantly improving FHB resistance.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (31901544 and 2071999) and the National Key Research and Development Program of China (2017YFD0100801).

Declaration of competing interest

The authors declare that they have no conflict of interest.