Molecular mapping of a new recessive wheat leaf rust resistance gene originating from Triticum spelta.

TSD276-2, a wheat genetic stock derived from the cross Agra Local/T. spelta 276 showed broad spectrum resistance against leaf rust pathogen. Genetic analysis was undertaken using F1, F2, F2:3 and BC1F1 generations derived from the cross TSD276-2/Agra Local. The results revealed a single recessive gene for leaf rust resistance, tentatively named as LrTs276-2, in TSD276-2. Molecular mapping of leaf rust resistance gene LrTs276-2 in TSD276-2 was done using SNP-based PCR and SSR markers. For Bulked Segregant Analysis (BSA), two bulks viz. resistant bulk and susceptible bulk, and the parents TSD276-2 and Agra Local were genotyped for SNPs using AFFYMETRIX 35K Wheat Breeders' AXIOM array. T. spelta 276 was also genotyped and used as a check. BSA indicated that the gene for leaf rust resistance in TSD276-2 is located on chromosome arm 1DS. Putatively linked SNPs on chromosome arm 1DS were converted into PCR-based markers. Polymorphic SSR markers on chromosome arm 1DS were also identified. Final linkage map was constructed using one SNP-based PCR and three SSR markers. The rust reaction and chromosomal location suggest that LrTs276-2 is a new leaf rust resistance gene which may be useful in broadening the genetic base of leaf rust resistance in wheat.

Introduction

Leaf rust caused by Puccinia triticina Eriks. is one of the most important and widespread foliar diseases of wheat (Triticum aestivum L.) inflicting significant yield losses in susceptible cultivars[1-4]. Although, rust diseases can be controlled by application of fungicides, genetic resistance remains the most effective, economical and environmentally sustainable method[5,6]. To date, 79 leaf rust resistance genes have been designated and catalogued in wheat and about half of them have their origin in various closely or distantly related species and genera of wheat while remaining resistance genes are native to wheat[7-9]. Many of the leaf rust resistance have become ineffective due to evolution of new virulent pathotypes. This necessitates continuous search for new and effective resistance genes for deployment in wheat cultivars. Spelt wheat (T. spelta) is potentially a good source of rust resistance genes[10]. T. spelta is a hulled wheat and is considered as ancestral to the free-threshing forms of hexaploid wheat[11]. Spelt wheat shares the same genomic structure, 2n = 6x = 42 (BBAADD genome) with common wheat and belongs to primary gene pool of wheat. This facilitates gene transfer from T. spelta by direct hybridization through homologous recombination. Till date, only three leaf rust resistance genes viz., Lr44 on chromosome 1B[10], Lr65 on chromosome 2A[12] and Lr71 on chromosome 1B[13] from spelt wheat have been identified and mapped.

As part of our pre-breeding programme at Indian Agricultural Research Institute, New Delhi, we have been working to identify and map rust resistance genes from primary, secondary and tertiary gene pools[14-19]. The present study reports the inheritance and molecular mapping of leaf rust resistance in T. spelta derived bread wheat line TSD276-2.

Material and methods

Plant materials

Triticum spelta derived bread wheat line TSD276-2 and leaf rust susceptible cultivar Agra Local (AL) were used to study the mode of inheritance and molecular mapping of leaf rust resistance. TSD276-2 is derived from the cross T. spelta accession 276/Agra Local. T. spelta accession 276 (T. spelta276) is a winter wheat requiring either vernalization or a prolonged photoperiod for flowering while AL is spring wheat. TSD276-2 showed spring wheat nature with no vernalization requirement. The F1, F2 and F2:3 population from the cross TSD276-2/AL were used for genetic analysis and molecular mapping of leaf rust resistance. T. spelta 276, the donor of leaf rust resistance to TSD276-2 was also used as a check in this study.

Leaf rust pathotypes

Pure inoculum of Puccinia triticina pathotypes was obtained from ICAR-Indian Institute of Wheat and Barley Research, Regional Station, Flowerdale, Shimla. Pathotypes were multiplied and maintained on susceptible cultivar AL under greenhouse conditions at Division of Genetics, IARI, New Delhi. T. spelta 276 and its derivative TSD276-2 along with susceptible check Agra Local were tested with 17 diverse Puccinia triticina pathotypes during crop season 2017–2018 (Table 1). Pathotype 77-5 (121R63-1), currently one of the most predominant one in India, was used for genetic analysis and molecular mapping.

Table 1

Infection types on Agra Local, T. spelta 276 and TSD276-2 against 17 pathotypes of P. triticina when tested at seedling stage at mean temperature range of 20–28 °C.

S. no	Pathotypes of P. triticina	Agra local	T. spelta 276	TSD276-2
(1)	77	3+	;N	;
(2)	77A	3+	0;	0;
(3)	77A-1	33+	;1	0;
(4)	77-2	33+	;	;
(4)	77-3	33+	;	0;
(5)	77-4	33+	;N	;
(6)	77-5	33+	;1=N	;1−N
(7)	77-6	33+	;N	;1−N
(8)	77-8	3	0;	0;
(9)	77-9	3	;	;1−
(10)	77-10	33+	;	;
(11)	104	3+	;	0;
(12)	104-4	33+	;N	;
(13)	106	3+	0;	0;
(14)	108	3+	;	;1−
(15)	162A	3+	;1=	0;
(16)	12-3	33	;	;
(17)	12-4	3+	;	;

Screening for leaf rust resistance

Screening for leaf rust resistance was done at seedling stage in greenhouse. Seeds were sown in small rectangular metallic trays (28 cm × 10 cm × 7.5 cm). About 10 day old seedlings were inoculated by spraying an aqueous suspension of P. triticina uredospores. The uredospore suspension was mixed with a drop of Tween20. Inoculated seedlings were incubated in a humid glass chamber for 48 h and were subsequently transferred to benches in a greenhouse under ambient condition of light and relative humidity. Disease reaction was recorded 12 days after inoculation as per the method described by Stakman et al.[20].

Molecular marker analysis

Fresh leaf samples collected from 40 to 45 days old plants were crushed in liquid nitrogen with mortar and pestle. DNA isolation was done following CTAB method[21]. DNA was quantified on 0.8% (w/v) agarose gel using Lambda Uncut DNA (THERMO FISHER SCIENTIFIC INC., USA) as standard and confirmed with NanoDrop Lite spectrophotometer (THERMO FISHER SCIENTIFIC INC., USA). DNA was diluted to the working stock concentration of 25 ng/μL and stored at − 20 °C.

For bulked segregant analysis equal amount of DNA from 20 homozygous resistant (HR) and 20 homozygous susceptible (HS) lines from F2:3 population was bulked to constitute two contrasting bulks viz. resistant bulk and susceptible bulk[22]. T. spelta276, TSD276-2, Agra Local and two extreme bulks i.e. RB and SB were genotyped for SNP using AFFYMETRIX 35K Wheat Breeders' AXIOM array[23]. SNPs found to be polymorphic between parents as well as bulks were identified. Chromosomal region found to show maximum polymorphic SNPs between bulks was presumed to carry leaf rust resistance gene. These SNPs were converted to PCR based markers using the software primer 3(v.0.4.0) as described earlier[19]. These SNP-based PCR markers were used for parental polymorphism.

Besides, a total 51 SSR markers spanning across the putatively identified chromosome 1D (on the basis of SNP genotyping) carrying rust resistance gene were tested for polymorphism between TSD276-2 and AL[24,25]. Primer sequences of these markers are available in public domain (https://wheat.pw.usda.gov/cgi-bin/GG3/browse.cgi?class=marker). For studying marker polymorphism between parents, PCR was performed in a reaction volume of 10 μl. Each 10 μl reaction volume included 2 μL of template DNA (50 ng), 1 μL forward primer (5 pm/μl), 1 μL of reverse primer (5 pm/μl), and 3 μL of Taq DNA Polymerase RED 2× master mix (AMPLIQON A/S, Denmark) and 3 μL of nuclease free water (THERMO FISHER SCIENTIFIC INC., USA). The PCR reactions were performed in 96-well PCR plates with thermal seal in an APPLIED BIOSYSTEMS VERITI thermal cycler at specific profile. The PCR conditions for primers used are given in Table 2. PCR amplified products were resolved on 3.5% (w/v) Agarose (LONZA, Rockland, USA) gel stained with ethidium bromide. Gels were visualized with a UV-transilluminator gel documentation system (SYNGENE G-BOX, Cambridge, UK).

Table 2

PCR amplification conditions of molecular markers used in genetic map construction.

Marker	Initial denaturation	Denaturation	Annealing	Extension	Total cycles	Final extension
SNP AX-94393474	94 °C for 4 min	94 °C for 30 s	60 °C for 30 s	72 °C for 30 s	35	72 °C for 10 min
SSR Xcfd15	94 °C for 4 min	94 °C for 30 s	60 °C for 30 s	72 °C for 30 s	35	72 °C for 10 min
SSR Xcfd61	94 °C for 4 min	94 °C for 30 s	60 °C for 30 s	72 °C for 20 s	30	72 °C for 10 min
SSR Xgwm106	94 °C for 4 min	94 °C for 30 s	60 °C for 30 s	72 °C for 20 s	30	72 °C for 10 min

Finally, both polymorphic SNP-based PCR markers and polymorphic SSR markers were used for bulked segregant analysis to confirm the identity of chromosome carrying leaf rust resistance gene[22]. Total 136 F2:3 homozygous lines were genotyped with SNP-based PCR marker as well as SSR markers identified as polymorphic in BSA. Linkage analysis was performed using MAPMAKER v3.0[26] with a minimum LOD score of 3.0 and a maximum genetic distance of 37.2 cM. ‘COMPARE’, ‘TRY’ and ‘RIPPLE’ commands of MAPMAKER v3.0 were used to check the final order of map. The genetic distances (cM) were calculated using the Kosambi mapping function[27]. Chi-square test was conducted to test the goodness-of-fit for segregation of the resistance gene[28]. Putative gene(s) present between flanking marker interval were predicted using wheat sequence (International Wheat Genome Sequencing Consortium, 2018) available at Ensembl Plants (https://plants.ensembl.org/Triticumaestivum/Info/Index) between the two flanking markers utilizing BioMart (https://plants.ensembl.org/biomart/martview/12b2b93c60bfbcedcaf0e4d1e023fee9).

Results

Genetic analysis of leaf rust resistance

TSD276-2 showed high degree of leaf rust resistance with ITs ranging from “0;” to “1−” against different P. triticina pathotypes, whereas the susceptible parent Agra Local showed susceptible reaction with infection type (IT) “3” to “33+” against all the pathotypes used in the study (Fig. 1). The original spelt wheat accession T. spelta 276 also showed high degree of resistance against all the 17 pathotypes (Table 1).

Figure 1

Infection types (ITs) of pathotype 77-5 on (1) Agra Local, (2) T.spelta 276, (3) TSD276-2, (4) F1 (TSD276-2/AL), (5) HR F3, and (6) HS F3 .

For genetic analysis, TSD276-2, AL, F1 (TSD276-2/AL) and 294 F2 plants were screened for leaf rust resistance against P. triticina pathotype 77-5. TSD276-2 showed resistance reaction with ITs “;1−N” whereas Agra Local showed susceptibility (IT 33+). All the 15 F1 plants were susceptible indicating recessive nature of resistance. Out of 294 F2 plants, 75 plants were resistant with ITs ranging from “;” to “1++” while 219 plants showed susceptible reaction (Fig. 1). The F2 segregation showed a good fit to theoretically expected ratio of 1 resistant: 3 susceptible plants (χ2(1:3) = 0.041, p-value = 0.84) for a single recessive gene. The results were further confirmed in F2:3 families. The 284 F2:3 families segregated into 1 resistant: 2 segregating: 1 susceptible with χ2(1:2:1) = 0.254 (p-value = 0.88). BC1 generation also showed expected segregation of IR:1S plants (Table 3).

Table 3

Segregation of leaf rust resistance at seedling stage in F2, BC1F1 and F2:3 populations against pathotype 77-5 at temperatures range of 20–28 °C.

Generation	Total progeny scored	Number of seedlings/families			Expected ratio	χ2(calc)	p -value
		Resistant	Segregating	Susceptible
F2	294	75	–	219	1R:3S	0.041	0.84
BC1F1	245	115	–	130	1R:1S	0.918	0.34
F2:3	284	70	146	68	1HR:2Seg:1HS	0.254	0.88

Mapping of leaf rust resistance

Genotyping data points for 35,143 SNP markers were obtained, which were filtered in a sequential manner. SNPs lacking any chromosome ID and position were removed. Moreover, SNPs showing heterozygous alleles are also filtered out. Further filtering resulted into 2414 SNPs showing polymorphism between parent viz. TSD276-2 and AL. Of these, only 20 SNPs were polymorphic between resistant and susceptible bulks. The 20 polymorphic SNPs were distributed over 10 chromosomes but five SNPs were observed in the short arm of chromosome 1D indicating putative linkage of these SNPs with leaf rust resistance gene in TSD276-2. The five SNPs on chromosome arm 1DS carried the identical alleles in T. spelta 276 and TSD276-2 (Table 4) and were converted into SNP-based PCR markers (Table 5). Among SSR markers on chromosome 1D, twelve were polymorphic between parents TSD276-2 and Agra Local. A combined BSA analysis using five SNP-based PCR markers and twelve SSR markers identified one SNP-based PCR marker (AX-94393474) and three SSR markers (Xcfd15, Xcfd61 and Xgwm106) as polymorphic between resistant and susceptible bulks (Fig. 2). For construction of linkage map, 136 F2:3 families comprising 68 homozygous resistant and 68 homozygous susceptible families were genotyped. Linkage map of leaf rust resistance gene in TSD276-2 was constructed with three SSR and one SNP-based PCR marker covering genetic distance of 18.7 cM on short arm of chromosome 1D (Fig. 3). The leaf rust resistance gene in TSD276-2, hereafter referred as LrTs is flanked by SSR markers Xcfd15 and Xcfd61 spanning 7.8 cM interval on map. SSR marker Xcfd15 mapped closest to the gene at 2.3 cM. The SNP based PCR marker AX-94393474 mapped 7.2 cM distal to the gene LrTs. Xcfd61 mapped 5.5 cM proximal to rust resistance gene. The order of SSR and SNP based marker in the linkage map is consistent with the consensus map of Somers et al. 2004 as well as with CS-IWGSC RefSeq v1 (Table 5).

Table 4

AXIOM array SNP genotyping data showing polymorphic SNPs between parents and bulks on 1DS-chromosome.

SNP Probeset_Id	T. spelta 276	TSD276-2	AL	RB	SB	IWGSC v1.0 position (bp)
AX-95241170	TT	TT	CC	TT	CC	3,965,001
AX-94393474	CC	CC	AA	CC	AA	3,967,540
AX-94772107	GG	GG	TT	GG	TT	3,969,410
AX-94818846	CC	CC	TT	CC	TT	8,727,512
AX-94570332	–	–	TT	–	TT	21,830,064

Table 5

SNP based primers and SSR primers on 1DS-chromosome used in the study.

Markers	Marker type	Designed SNP primer sequence (F&R)	CS-IWGSC RefSeq v1.0 genomic position
AX-95241170	SNP based PCR marker	F 5′ AGAATGAGGATGGCAGCGAT 3′ R 5′ CACCACAAATTCACAGGCCA 3′	3,965,001 bp
AX-94393474	SNP based PCR marker	F 5′ GAGAGAGATCGATATGTTCTGGAC 3′ R 5′ GGCAGCAAACAGAACCTTCA 3′	3,967,540 bp
AX-94772107	SNP based PCR marker	F 5′ GCGTTCGCATGGCGATG 3′ R 5′ ACACCAGTAGCAACCCGTTACCAG 3′	3,969,410 bp
AX-94818846	SNP based PCR marker	F 5′ GGTTGCAGAACTTCCTACCG 3′ R 5′ TGCCAGAAGTTGTGCTTTATTGA 3′	8,727,512 bp
AX-94570332	SNP based PCR marker	F 5′ GCACAAACAGGCTAACAAAACCTTTA 3′ R 5′ GGGCCCTATTTAGGAGATGTGAC 3′	21,830,064 bp
Xcfd15	SSR	F 5′ CTCCCGTATTGAGCAGGAAG 3′ R 5′ GGCAGGTGTGGTGATGATCT 3′	9054 kb
Xcfd61	SSR	F 5′ ATTCAAATGCAACGCAAACA 3′ R 5′ GTTAGCCAAGGACCCCTTTC 3′	15,414 kb
Xgwm106	SSR	F 5′ CTGTTCTTGCGTGGCATTAA 3′ R 3′ AATAAGGACACAATTGGGATGG 3′	18,188 kb

Figure 2

Bulk Segregant Analysis of leaf rust resistance in F2:3 population of TSD276-2/AL cross. Lanes: (L) 100 bp ladder, (1) Donor parent T.spelta 276, (2) Resistant parent TSD276-2, (3) Susceptible parent AL, (4) F1 (TSD276-2/AL), (5) Resistant Bulk, and (6) Susceptible Bulk.

Figure 3

Linkage map of 1DS chromosome in our study based on 136 homozygous F2:3 lines of TSD276-2/AL cross.

The SSR marker Xcfd15, closest to the resistance gene LrTs behaved as a codominant marker amplifying alleles of 180 and 200 bp in TSD276-2 and only one allele i.e. 168 bp in AL (Table 6, Supplementary Fig. S1 online). The 200 bp allele of SSR marker Xcfd15 was found to be linked with the leaf rust resistance in TSD276-2. T. spelta 276, the original source of resistance gene LrTs also amplified alleles identical to TSD276-2. The markers Xcfd61 and Xgwm106 were linked with leaf rust resistance gene in repulsion phase and behaved as dominant markers amplifying alleles of 195 bp and 127 bp in susceptible parent AL, respectively (Fig. 2). Both Xcfd61 and Xgwm106 produced null allele in TSD276-2 and T. spelta 276. The SNP-based PCR marker AX-94393474 was linked with resistance gene in coupling phase and amplified alleles of 211 and 232 bp in TSD276-2 and T. spelta 276 while only single allele i.e. 211 bp was produced in AL showing the dominant nature of marker (see Supplementary Fig. S1 online). Further, the total number of genes between two flanking markers i.e. cfd15 and cfd61 observed in Triticum aestivum (covering 6.36 Mb sequence) and Ae. tauschii (covering 6.21 Mb sequence) were 141 and 84, respectively. Out of these, 45 genes in Triticum aestivum and 27 genes in Ae. tauschii have R gene related domain (see Supplementary Table S1 and Supplementary Table S2 online). Among these, 24 genes are common between Triticum aestivum and Ae. tauschii between two flanking markers (see Supplementary Table S3 online).

Table 6

Polymorphic allele size scored between parents and the bulks and used in mapping population screening.

Markers	Alleles amplified (in bp)			Polymorphic allele used in mapping (TSD276-2/AL)
	T. spelta 276	TSD276-2	Agra local
SNP AX-94393474	211, 232	211, 232	211	232/–
Xcfd15	180, 200	180, 200	168	200/168
Xcfd61	–	–	195	–/195
Xgwm106	–	–	127	–/127

Discussion

The leaf rust resistance gene in TSD276-2 showed a wide spectrum of resistance against Indian P. triticina pathotypes. Genetic analysis showed a single recessive gene conferring leaf rust resistance in TSD276-2. The resistance gene was mapped on short arm of chromosome 1D and was flanked by SSR markers Xcfd15 and Xcfd61. The resistance gene in TSD276-2 is derived from the spelt wheat accession T. spelta276. A large number of rust resistance genes have been transferred into wheat from the species belonging to secondary and tertiary gene pools. These resistance genes often carry some degree of linkage drag and sometimes undesirable genes[17,29,30]. Genetic resources from primary gene pool have the advantage of homologous recombination which can be used to remove the linkage drag. Closely related species of wheat from primary gene pool are rich and diverse source of unique alleles that can be used in wheat improvement[12,31-35]. In the present study, we identified a seedling leaf rust resistance gene tentatively named as LrTs in spelt wheat derived common wheat line TSD276-2. Till date, only three leaf rust resistance genes from spelt wheat viz., Lr44, Lr65 and Lr71 have been identified and mapped. While Lr44 and Lr71 have been located on chromosome 1B[10,13], Lr65 has been mapped on chromosome 2A[12]. The leaf rust resistance gene in the present study has been mapped on short arm of chromosome 1D indicating that LrTs is different from the already characterized leaf rust resistance genes from spelt wheat and potentially a novel leaf rust resistance gene. Moreover, Lr44, Lr65 and Lr71 behaved as dominant genes while the LrTs in TSD276-2 is recessive in nature, differentiating this gene from other spelt wheat genes characterized so far. Lr65 has also been reported to be susceptible to Indian pathotype 77-5[19].

Till date, three leaf rust resistance genes viz., Lr21[36,37], Lr42[36,38-40] and Lr60[41] have been mapped on 1DS chromosome of wheat. Among these genes Lr21 and Lr42 have been transferred in common wheat from diploid progenitor species Aegilops tauschii (2n = 2x = 14, genome DD)[36,37,40,42] while Lr60 is native in bread wheat[41,43]. Although the gene LrTs mapped by us is from T. spelta and is expected to be different, nevertheless it is important to distinguish this gene from other leaf rust resistance genes mapped on chromosome 1DS. This can be done on the basis of differential response to P. triticina pathotypes and by comparing the genetic and physical position on the chromosome. The gene Lr21 is ineffective against several Indian pathotypes of P. triticina[44] whereas in our study both T. spelta 276 and TSD276-2 showed high degree of resistance against all the 17 pathotypes used in the study. The response of Lr60 to Indian pathotypes of P. triticina is not available. However, Lr42 shows resistance to all the P. triticina pathotypes in India[44]. Hiebert et al.[41] analyzed the linkage between Lr21 and Lr60 and observed that Lr60 is about 13.5 cM distal to Lr21 with the SSR marker Xbarc149 co-segregating with Lr21. Lr42 has been reported as race-specific partially dominant resistance gene[36]. However, Czembor et al.[45] mapped Lr42 on chromosome 3D and observed that Lr42 behaved as dominant gene. Sun et al.[38] mapped Lr42 on the distal end of chromosome arm 1DS and marker Xwmc432 was found closest to the gene Lr42 at a distance of 0.8 cM. Liu et al.[39] reported Lr42 as recessive gene and mapped it on 1DS chromosome with flanking markers Xwmc432 and Xgdm33 spanning a genetic distance of 17 cM. Xwmc432 was the closest marker 4 cM proximal to Lr42. Gill et al.[40] narrowed down the Lr42 region to 3.7 cM with marker TC387992 at a distance of 1.7 cM distal to Lr42 and Xwmc432 located at 2 cM proximal to Lr42. The gene LrTs mapped by us in TSD276-2 is flanked by the markers Xcfd15 and Xcfd61. The gene LrTs is 2.3 cM proximal to marker Xcfd15 while Gill et al.[40] reported Lr42 at a distance of 5.4 cM distal to Xcfd15. Thus, Lr42 is located distal to LrTs on chromosome arm 1DS. A comparison of genetic and physical maps unambiguously shows that the locus represented by LrTs is different from other rust resistance loci mapped on chromosome 1DS (Fig. 4). Anchoring of markers linked to Lr21, Lr 42, Lr60 and LrTs on Chinese Spring Reference genome indicates that these genes are located on 1DS chromosome in order of Lr60-Lr21-Lr42 and LrTs. Further, in-silico studies suggested that the flanking markers are more than 6 Mb apart. The predicted number of genes related to disease resistance with R gene domain (using domain reported by Peng et al.[46]) in the species i.e. Triticum aestivum and Aegilops tauschii are very high. Hence, it is essential to narrow down the region for prediction of putative candidate gene. The rust reaction, nature of gene and comparative genomics indicates that LrTs is a novel leaf rust resistance gene that may be useful in resistance breeding programs in wheat.

Figure 4

Comparative analysis of LrTs along with Lr42, Lr60 and Lr21. Map unit is cM in (a) and (c), Map unit is Mbp in (b).

Supplementary Information 1.

Supplementary Information 2.