LIU Yuan,Takele Weldu GEBREWAHID,,ZHANG Pei-pei,LI Zai-feng,LIU Da-qun

1 College of Plant Protection, Hebei Agricultural University, Baoding 071001, P.R.China

2 College of Agriculture, Aksum University, Shire-Indaslassie 314, Ethiopia

Abstract Wheat leaf rust,triggered by Puccinia triticina Eriks (Pt),is among the most important diseases of wheat worldwide.Deploying resistant varieties against leaf rust is the most effective,environmentally-friendly and economic way to control the disease.In the present study,66 wheat varieties form China and foreign countries were tested with 17 Pt races for gene postulation during the seedling stage in the greenhouse.All the varieties were also planted to identify slow rusting responses to leaf rust at the adult plant stage in Baoding and Zhoukou field trials during the 2016/2017 to 2017/2018 cropping seasons.Moreover,12 closely linked molecular markers to known leaf rust resistance (Lr) genes were used for assessing all the varieties.The results of both gene postulation and molecular marker identification showed that a total of eight Lr genes,Lr1, Lr10,Lr17, Lr20,Lr26, Lr34, Lr37 and Lr46,either singly or in combination were detected in 32 varieties.Known Lr genes were not identified in the remaining 34 varieties.Seventeen varieties were found to have slow rusting resistance.The resistance sources identified in this study can be used as resources for resistance against leaf rust in wheat breeding programs in China and the respective foreign countries.

Keywords:wheat,gene postulation,molecular marker,Lr gene

1.Introduction

Wheat (Triticum aestivumL.),one of the most important food crops grown throughout the world,is also extensively planted in China.Wheat leaf rust caused byPucciniatriticinaEriks (Pt),is a fungal aerially-spread disease which is mainly damaging to wheat leaves reducing the plant’s photosynthetic area.When a large number ofPtspore piles break through the wheat leaf surface,it causes a serious loss of water in the wheat plant,thus affecting its growth and grain filling and resulting large production losses (Gaoet al.2010).Wheat leaf rust usually causes production loss as high as 40% in susceptible varieties (Khanet al.2013),and remains as one of the most damaging diseases of wheat worldwide.Wheat leaf rust is also widespread in China,mainly in Hebei,Henan,Shandong,Jiangsu,Shanxi,Guizhou,Sichuan,Yunnan,Heilongjiang,and Jilin provinces.Growing wheat varieties resistant to leaf rust is the most effective,environmentally-safe and economical way to control the disease.

Resistance to wheat leaf rust is mainly divided into two types;vertical resistance and horizontal resistance (Yuan 1998).Vertical resistance,also known as race-specific resistance,is often controlled by a single major gene and expressed throughout all the growth stages of the plant.This type of resistance is not durable,and varieties carrying this type of resistance tend to induce species variation ofPtraces after deployment in a large-scale causing loss of resistance of the varieties (Wanget al.2015).On the other hand,horizontal resistance,which is also known as race non-specific resistance,is controlled by minor genes which are mainly expressed at an adult plant stage.This type of resistance has a long incubation period,low and small sporulation size,and a low severity and infection rate.Race non-specific resistance is relatively stable and can remain effective for several decades (Daset al.1992;Wanget al.2015).

Thus far,a total of 79Lrgenes have been reported and mapped on the wheat chromosomes(Qureshiet al.2018).However,most of theLrgenes have lost their resistance in China,and those that still show good resistance includeLr9,Lr19,Lr24,Lr38,Lr47,Lr51,andLr53(Liet al.2010;Gaoet al.2019).Reports have indicated there are about 12 adult-plant resistance (APR) genes to leaf rust,namelyLr12,Lr13,Lr22a,Lr22b,Lr34,Lr35,Lr37,Lr46,Lr48,Lr49,Lr67andLr68.Among these APR genes,only four (Lr34,Lr46,Lr68andLr67) are known to be genes for slow-rusting resistance to leaf rust (Liet al.2014).

Gene postulation and molecular marker detection are suitable for screening resistant sources and identifying important genes in wheat varieties.Gene postulation is based on the gene-to-gene hypothesis proposed by Flor(1956) and involves postulating the disease resistance genes possibly present in tested genotypes.According to the results of Renet al.(2015),12Lrgenes (Lr1,Lr3bg,Lr3,Lr3ka,Lr11,Lr13,Lr26,Lr27,Lr14a,Lr16,Lr30andLr31)were identified from 84 winter wheat varieties.TwelveLrgenes (Lr1,Lr2b,Lr3ka,Lr10,Lr13,Lr26,Lr11,Lr21,Lr34,Lr37,Lr44andLr46) were reported in 83 wheat varieties from China using 18Ptraces (Gebrewahidet al.2017).LrgenesLr1,Lr26,Lr13andLr46were detected in 30 wheat varieties from China using 18Ptraces (Yanet al.2017).SixLrgenes (Lr1,Lr2c,Lr3a,Lr10,Lr16andLr26) were identified in 28 wheat varieties in China using 19Ptraces(Chen and Hu 1993).SevenLrgenes,Lr1,Lr2b,Lr14a,Lr20,Lr21,Lr34andLr36were reported by Zhanget al.(2019) in 46 Chinese wheat landraces using 16Ptraces.SeveralLrgenes (Lr1,Lr2c,Lr3,Lr3ka,Lr9,Lr10,Lr14a,Lr14b,Lr13,Lr16,Lr18,Lr20,Lr21,Lr23,Lr27+31,Lr30,Lr37andLr44) were detected in 36 wheat cultivars from Ethiopia and Germany using 31Ptraces (Mebrateet al.2008).The combination of gene postulation,pedigree analysis and molecular marker detection can effectively improve the accuracy ofLrgene identification.

In China,there is limited information on theLrgenes present in the released wheat varieties,which has made it difficult to utilize resistant varieties for controlling leaf rust.The objective of the present study was to identifyLrgenes in 66 wheat varieties from China and foreign countries.

2.Materials and methods

2.1.Plant materials

A total of 66 wheat varieties from China,North America,Russia,Mexico,Spain,France,Italy,Portugal and some other countries (Table 1),were used to test their seedling responses using 17Ptraces.Thirty-six differential lines,mostly near isogenic lines (NILs) in the background of Thatcher with knownLrgenes (Table 2),were provided by International Maize and Wheat Improvement Center(CIMMYT) and included in the seedling testing.All the 66 wheat varieties,including the slow rusting resistance check,Saar and the susceptible check Zhengzhou 5389(Zhanget al.2009),were tested in the fields in Zhoukou in 2016/2017 and Baoding in 2017/2018 cropping seasons.

2.2.Pt races

SeventeenPtraces with different virulence levels and collected from different wheat growing regions of China were used for seedling test in the greenhouse.ThePtraces were named according to the three-letter code naming system given by Longet al.(1989) with the addition of a fourth letter from the reactions to the fourth quartet of the differentials later.THTS and PHTS were used as mixedPtraces for inoculation in the field trials based on their known virulence to most of the varieties in the seedling test.All thePtraces were provided by the Hebei Biological Control Center,China.

2.3.Seedling test

On average,about seven seeds of each of the 36 differential lines,66 wheat varieties and the susceptible check,Zhengzhou 5389 were planted in a plastic growth chamber in the greenhouse for the seedling test.Inoculation in the greenhouse was performed when the flag leaf was fully expanded using 17Ptraces (FGJQ,FHJQ①,FHJT,FHJR,THTS,FHJL,PHTS,FHJQ②,FHSQ,FHGS①,FHGS②,FHDQ,KHJS①,FHNS,KHJS②,FGSQ and FHTQ),according to the method described by Liet al.(2010).Inoculated seedlings were then incubated at 18°C and 100% relative humidity (RH) for about 24 h.They were then transferred to a growth chamber maintained under 12 h darkness/12 h light at 18 to 20°C and 70% RH.Infectiontypes (ITs) were collected based on a 0 to 4 range based on the standard modified by Roelfset al.(1992).According to the method given by Dubinet al.(1989),the gene postulation was carried out in the greenhouse to identify theLrgenes or gene combinations that may be present in the tested varieties.

(Continued on next page)

Table 1 (Continued from preceding page)

2.4.Field test

All the 66 varieties,the slow rusting check,Saar and the susceptible check,Zhengzhou 5389 were planted for testing adult plant responses to leaf rust in Zhoukou,Henan Province and in Baoding,Hebei Province in the 2016/2017 and 2017/2018 cropping seasons,respectively.Field trials were conducted in a randomized complete block with two replicates at each location.About 50 seeds of each variety were planted in a row of 1.5 m length,with 25 cm between rows.Zhengzhou 5389 was planted perpendicularly and adjacent to the test rows as a spreader row.Field inoculation was carried out by applying a mixture of equal amounts of uredinospores from THTS and PHTSPtraces suspended in 0.03% Tween 20 onto the spreader rows at the tillering stage on April 10 at Zhoukou and Baoding.The final disease severity (FDS) was recorded when the susceptible check,Zhengzhou 5389 was fully infected.The FDS data collection was conducted according to the method described by Liet al.(2010).

2.5.Molecular marker detection

The genomic DNA of the wheat varieties was extracted according to the improved CTAB method (Sharpet al.1988),and the concentration and purity of DNA samples were detected by a spectrophotometer.Twelve molecular markers linked to 10 knownLrgenes (Lr1,Lr9,Lr10,Lr19,Lr20,Lr24,Lr26,Lr34,Lr37andLr46) were used to assess all varieties.The sequences and annealing temperatures of all the primers are provided in Table 2.Zhengzhou 5389 was used as a susceptible control,whereas the differential lines with their respectiveLrgenes were used as a resistance control during molecular marker detection.A total of 20 µL PCR reaction mixture contained 2 µL (40 ngµL–1) DNA template,2 µL (4 mol µL–1) primer,10 µL 2×TaqPCR master Mix and 6 µL ddH2O.The amplification products of CAPS marker,csLV46G22specific toLr46were detected by 12% non-denatured polyacrylamide gel electrophoresis,while the remaining amplification products were detected by the agarose gel electrophoresis with a concentration of 1.5%.

2.6.Statistical analysis

Statistical Analysis System (version 9.1,SAS Institute Inc.,Cary,NC) with the Generalized Linear Model procedure(PROC GLMMIX) was used for the analysis of variance(ANOVA) and to calculate the least significant difference(LSD) between varieties atP=0.05.Environments,varieties and the variety×environment interaction were treated as fixed effects,and replications nested in environments were treated as random.Varieties with an FDS value obviously less than or not significantly greater than the slow rusting check,Saar were considered to have slow rusting resistance.

3.Results

3.1.Lr-genes from seedling reactions and molecular marker detection

The response patterns generated by the interaction of 36 differential lines and 66 tested varieties using 17Ptraces with different virulence levels are shown in Tables 2 and 3,respectively.Nine differential lines withLr9,Lr24,Lr19,Lr28,Lr29,Lr42,Lr47,Lr51andLr53had low ITs to all testedPtraces,while ten differential lines withLr2c,Lr3,Lr16,LrB,Lr3bg,Lr13,Lr14b,Lr23,Lr33andLr39were susceptible to all testedPtraces.Therefore,it was not possible to postulate these genes.However,the remaining 17Lrgenes,Lr1,Lr2a,Lr26,Lr3ka,Lr11,Lr17,Lr30,Lr10,Lr14a,Lr18,Lr2b,Lr15,Lr20,Lr21,Lr36,Lr44andLr45showed different ITs to thePtraces used and can be postulated for absence or presence in the wheat varieties.Based on the gene postulation and molecular marker detection,eightLrgenes,Lr1,Lr10,Lr17,Lr20,Lr26,Lr34,Lr37andLr46either singly or in combination were identified in 32 varieties,while the remaining 34 varieties likely contained unknownLrgene(s)or lacked detectableLrgenes.

Lr1was susceptible to only twoPtraces,THTS and PHTS and resistant to the remaining 15Ptraces (Table 2).Twelve varieties,XYZ319,XYZ320,XYZ321,AC Reed,Ф2a,DANPHE #1,IANTER,PRIASPA,STEPNO,MAIORCA,EUREKA and ALMONSOR containedLr1based on their reaction patterns (Table 3).Four varieties,AC Reed,DANPHE #1,STEPNO and ALMONSOR only containedLr1.XYZ 319 and MAIORCA containedLr1in combination withLr26andLr20,respectively.Four varieties,XYZ 321,Ф2а,IANTER and PRIASPA containedLr1andLr34in combination.Lr1in combination withLr37was detected in XYZ 320 and EUREKA.All varieties found to be carryingLr1by gene postulation were also confirmed by molecular marker detection specific forLr1.

Lr26had low ITs to three avirulentPtraces,FGJQ,KHJS①and FGSQ,and high ITs to the otherPtraces(Table 2).Four varieties were found withLr26based on their ITs responses to thePtraces (Table 3).Livia containedLr26gene andLr34in combination,whereas XYZ322 had onlyLr26.XYZ 319 and PBW343 hadLr26in combination withLr1andLr10,respectively.All varieties found to be carryingLr26by gene postulation were also confirmed by the molecular marker specific forLr26.

Lr10was identified in GABO 54,RUIVO and PBW343 because these varieties had low ITs to oneLr10avirulentPtrace,FHJL and were susceptible to 16Lr10virulentPtraces.The presence ofLr10in these three varieties was also confirmed by the molecular marker specific forLr10.Two varieties,Copt-1 and FLAMINIO,were postulated to containLr17because of their intermediate reactions to twoLr17avirulentPtraces,FHGS①and FHGS② and high ITs to the remaining 15Ptraces.

Twelve SCAR or STS molecular markers closely linked toLr1,Lr10,Lr9,Lr19,Lr24,Lr20,Lr26,Lr34,Lr37andLr46genes were used to assess all the wheat varieties(Table 4).The presence ofLr1,Lr10,Lr20,andLr26was confirmed by both molecular marker detection and gene postulation.All varieties lackedLr9,Lr19andLr24based on the seedling reaction and molecular marker genotyping.SinceLr34,Lr46andLr37are APR genes,it was not possible to postulate them in the seedling tests.Hence,the closely linked molecular markers to these genes were used to assess all the varieties.The results indicatedLr34,Lr46andLr37were detected in 14,5 and 3 varieties,respectively (Table 3).

3.2.Slow rusting resistance

The analysis of variance results for FDS (atP=0.05) of the tested varieties showed significant differences among the varieties,environments,and variety×environment interactions in the fields,indicating that the phenotypes were affected by varieties and environments.No significant difference was obtained between replications or variety×replication interactions (Table 5).

In the 2016/2017 and 2017/2018 cropping seasons,the FDS values of the susceptible check Zhengzhou 5389 and slow-rusting resistance check,Saar were 85 and 2%,respectively,indicating an adequate development of leaf rust.A total of 17 varieties which showed high ITs to mixedPtraces in the seedling stage and low FDS values in the field were supposed to have slow rusting resistance (Table 6).

4.Discussion

In this study,66 wheat cultivars from China and foreign countries were tested for identifyingLrgenes by combining gene postulation,molecular marker detection and pedigree analysis.A total of eight knownLrgenes,includingLr1,Lr10,Lr17,Lr20,Lr26,Lr34,Lr37andLr46were identified in 32 wheat varieties.Three varieties,XYZ 341,Mv Tamara and ELIA showed resistant responses to allPtraces at the seedling stage in the greenhouse and were resistant to the mixedPtraces in the field.However,no knownLrgene was detected in XYZ 341 or ELIA indicating unknown majorLrgene(s) might be present in these varieties.

Table 4 PCR primers and their conditions for detecting leaf rust resistance (Lr) genes,as well as specific fragment sizes

Table 5 Analysis of variance of final disease severity (FDS,%) in 66 wheat genotypes,including slow rusting cultivar Saar and susceptible Zhengzhou 5389 checks,tested in the 2016/2017 and 2017/2018 growing seasons

Gebrewahidet al.(2017) reported thatLr1,Lr26,Lr34,Lr37andLr46were the predominantLrgenes identified in the tested wheat cultivars from China.According to Liet al.(2010) and Singhet al.(1999),Lr1,Lr26andLr34were the most frequently detectedLrgenes in the tested Chinese wheat cultivars.Consistent with their findings,Lr1,Lr26,Lr34,Lr37andLr46were the most commonly identifiedLrgenes in the present study.It seems reasonable to say that these are the predominantLrgenes present in wheat cultivars from China.

Lr1in variety Malakoff (Mainset al.1926) was mapped on chromosome 5DL by Cloutieret al.(2007) for the first time.Lr1is reported widely distributed in wheat varieties from China (Singhet al.1999;Gebrewahidet al.2017;Gaoet al.2019).In this study,Lr1was the most prevalent(in 12 varieties)Lrgene identified.A report by Boroujeniet al.(2011) also showed thatLr1was the most commonly identified gene (in 25 varieties) among 36 tested wheat varieties in Iran.According to the pedigree analysis,Zhengmai 366,one of the parents of XYZ319,was obtained from the cross between Yumai 47×PH82-2-2,henceLr1in XYZ319 might be from Yumai 47 (Liet al.2010).

Lr26was derived from rye and exists in bread wheat as 1BL/1RS translocation.This translocation also carriesSr31,Yr9andPm8genes (Rajaramet al.1983) and has been used as a source of resistance genes in wheat breeding programs.In line with the results of the present study,Bramaet al.(2003) and Stepieńet al.(2003) reported thatLr26was detected in PBW343 and Livia,respectively.According to the studies of Gurjaret al.(2014),PBW343 hadLr26andLr10in combination.Lr26was also reported in Livia by Stepieńet al.(2003).The pedigree analysis showed thatLr26in PBW343,Livia and Mv Tamara might be derived from VEERY-5 (Singh and Rajaran 1991),KAVKAZ(Kochumadhavanet al.1988;Rizvi and Buchenau 1994)and FATIMA-2 (Wanget al.2009),respectively.

Table 6 Seedling Infection types (ITs) with mixed Puccinia triticina Eriks (Pt) races in the seedling test and final disease severity(FDS,%) in field experiments in Zhoukou in 2016/2017 and in Baoding in 2017/2018 growing seasons for wheat cultivars with slow rusting resistance to leaf rust

Lr34is the first slow rust gene which has been found and widely distributed around the world.Dyck (1977) had reportedLr34from the very beginning in Canada.Since then,Lr34had been used in the development of new wheat varieties.A gene-specific DNA marker has been developed forLr34(Lagudahet al.2009),providing a powerful tool for MAS in wheat breeding of slow-rusting resistance to leaf rust.In this study,Lr34was detected in 12 varieties.Consistent with the present study,Mishra(1992) and Ciucaet al.(2015) reported that YECORA F 70 and ARIESAN containedLr34,respectively.Based on the pedigree analysis,Lr34in PADEIRA,Livia and Mv Tamara might be derived from MENTANA (Dakouriet al.2013),KAVKAZ (Kochumadhavanet al.1988;Rizviet al.1994)and FATIMA-2 (Wanget al.2009).

Lr46was originally reported in Pavon 76 on chromosome 1BL (Singhet al.1998),and it is a slow rusting gene.Williamet al.(2003) reported thatLr46is widely found in wheat varieties developed from CIMMYT.At present,Lr46still maintains good resistance,so it can greatly improve host resistance when it occurs together with other diseaseresistance genes.RAYON-89,one of the parents of PICAFLOR #1,was reported to containLr46(Kolmeret al.2008),thereforeLr46in PICAFLOR #1 might be derived from RAYON-89.

Molecular marker detection showed that eitherLr34orLr46was found in Ф2а,Ак-Урук,Чуuонuозюахоп,Безостая 2,Carodejkas,Ianter,Taras,Mv Tamara,Leone and Elvas 61P17;however,the FDS responses to leaf rust of these varieties at the field trials were significantly high than the slow rusting resistance control,Saar.This inconsistency might occur due to environmental conditions or changes of thePtrace pathogenicity.Thus,in this study,in order to improve the reliability of the results,we present the varieties with possible slow rusting resistance to leaf rust that were consistent between the molecular marker detection and FDS in the field trials.

5.Conclusion

In the present study,66 wheat varieties form China and foreign countries were used to identify their leaf rust resistance genes.A total of eight Lr genes,includingLr1,Lr10,Lr17,Lr20,Lr26,Lr34,Lr37andLr46either singly or in combination,were identified in 32 varieties based on the gene postulation and molecular marker detection.KnownLrgenes were not identified in the remaining 34 varieties.Seventeen varieties showed slow leaf rusting resistance in two cropping seasons.The resistance sources identified in the study can be used as resources of resistance against leaf rust in wheat breeding programs in China.

Acknowledgements

This study was supported by the National Key Research and Development Program of China (2017YFD0300906-07).

Declaration of competing interest

The authors declare that they have no conflict of interest.