PLANT GENETICS • ORIGINAL PAPER

Genetic mapping of male sterility and pollen fertility QTLs in triticalewith sterilizing Triticum timopheevii cytoplasm

Marzena Wasiak1 & Agnieszka Niedziela1 & Henryk Woś2 & Mirosław Pojmaj3 & Piotr Tomasz Bednarek1

Received: 20 April 2020 /Revised: 27 October 2020 /Accepted: 5 November 2020# The Author(s) 2020

AbstractCytoplasmic male sterility (CMS) phenomenon is widely exploited in commercial hybrid seed production in economicallyimportant crop species, including rye, wheat, maize, rice, sorghum, cotton, sugar beets, and many vegetables. Although somecommercial successes, little is known about QTLs responsible for the trait in case of triticale with sterilizing Triticum timopheevii(Tt) cytoplasm. Recombinant inbred line (RIL) F6 mapping population encompassing 182 individuals derived from the cross ofindividual plants representing the HT352 line and cv Borwo was employed for genetic map construction using SNP markers andidentification of QTLs conferring pollen sterility in triticale with CMS Tt. The phenotypes of the F1 lines resulting from crossingof the HT352 (Tt) with HT352 (maintainer) × Borwo were determined by assessing the number of the F2 seeds per spike. Agenetic map with 21 linkage groups encompasses 29,737 markers and spanned over the distance of 2549 cM. Composite (CIM)and multiple (MIM) interval mappings delivered comparable results. Single QTLs mapped to the 1A, 1B, 2A, 2R, 3B, 3R, 4B,and 5B chromosomes, whereas the 5R and 6B chromosomes shared 3 and 2 QTLs, respectively. The QTLs with the highest LODscore mapped to the 5R, 3R, 1B, and 4B chromosomes; however, the QRft-5R.3 has the highest explained variance of the trait.

Keywords Triticale . Genetic map . Cytoplasmicmale sterility . QTL

Introduction

Triticale (x TriticosecaleWittmack) is a relatively young synthet-ic species created by hybridization of wheat and rye nearly150 years ago (Wilson 1875), but it took 100 years until the firstvariety was released (Kiss 1971). The evolution of triticale as acommercial crop was slow until the mid-1980s. Since that time,the production in world area is increasing from 91 ha (1980) to4.2 million ha (2018) (FAOSTAT 2018). The species combines

grain quality and productivity typical for wheat with vigour,hardiness, disease resistance and high lysine content specificfor rye (Myer and Barnett 2000). The vigorous root system andtolerance to abiotic stresses arising from rye (Niedziela et al.2014) allow it to grow on sandy soils with low fertility.Triticale is mainly used as animal feed but holds some promisefor human nutrition (Pena 2004). The development of efficientcytoplasmic male sterility system based on, i.e. CMS Tt, andintroducing new triticale hybrid varieties would further stimulatethe interest in breeding the species. However, many questionsconcerning, i.e. the genetic pool of the species, the presence ofheterotic groups that could be exploited in breeding programs,development of the genetic maps dedicated for dissection ofpollen sterility QTLs in triticale with CMS Tt and evaluation ofmarkers useful for breeding programs need to be assessed.

The advances in genotyping provide an opportunity to usegenomic tools in understanding the genetic structure of culti-vated crops, including triticale. There were several attempts todifferentiate triticale materials via DNA-based molecularmarkers in combination with cluster analysis (Tams et al.2005; Kuleung et al. 2006). The analysis of the genetic diver-sity of 232 triticale breeding lines based on DArT markersdemonstrated that the materials divided into three groups.

Key message Identifying QTLs of male sterility and pollen fertility on amap derived from advanced plant materials created especially for thisresearch.

Communicated by: Izabela Pawłowicz

* Piotr Tomasz Bednarekp.bednarek@ihar.edu.pl

1 Plant Breeding and Acclimatization Institute – NRI,05-870 Radzikow, Błonie, Poland

2 Breeding Department Borowo, Plant Breeding Company-Strzelce,Czempin, Poland

3 Breeding Department Laski, Danko Plant BreedingCompany-Choryń, Laski, Poland

https://doi.org/10.1007/s13353-020-00595-z

/ Published online: 23 November 2020

Journal of Applied Genetics (2021) 62:59–71

http://crossmark.crossref.org/dialog/?doi=10.1007/s13353-020-00595-z&domain=pdfhttps://orcid.org/0000-0002-1553-8378mailto:p.bednarek@ihar.edu.pl

The differences between them are not sharp, and most of thevariation (86%) is due to within-population variability(Niedziela et al. 2016), suggesting that there are enough var-iances for the development of new varieties. Recently, ad-vances in genotyping with DArT markers enabled genome-wide characterization of the population structure and linkagedisequilibrium (LD) in a comprehensive set of 161 winter andspring triticale lines (Alheit et al. 2012). Winter and springgrowth habits contributed to the population structure in triti-cale, and a family structure exists in both growth types. Lately,the analyses of 885 various European triticale lines revealed alack of significant population structure but a particular group-ing of materials according to their origin (Losert et al. 2017).The available information suggests that in contrast to rye(Geiger and Miedaner 2009), the distinct triticale heteroticgroups may not exist or are still under development (Geigerand Miedaner 2009). Thus, the development of genetic mapsbased on diverse and well-characterized breeding forms thatmay facilitate the identification of markers towards male ste-rility genes are of importance.

By now, a few genetic maps of triticale are available. Thefirst one (González et al. 2005) exploited 356 markers with anaverage map density of 6.9 cM and evaluated on 73 DH lines.The primary goal of the map was to study the androgenicresponse. The next one based on DArT, SSR, and AFLPmarkers in 90 doubled haploid (DH) lines was published in2011 (Tyrka et al. 2011). Then, Alheit et al. (2011) construct-ed a consensus genetic map evaluation with six DH mappingpopulations using nearly 2000 unique DArT markers span-ning over 2310 cM. The study was designed to deliver a toolfor genomic approaches in triticale. Niedziela et al. (2014)constructed the map of a small fragment of the 7R chromo-some encoding Al-tolerance and using association mappingsuggested markers for MAS (Niedziela et al. 2015). Recently,Tyrka et al. (2015) published a map saturated with SSR,DArT, and SNP. Nevertheless, any of the maps used mappingpopulations useful for hybrid breeding and dissection of QTLsconferring pollen sterility trait in CMS Tt. With the develop-ment of the Next Generation Sequencing technology, geneticmaps based on recombinant inbred lines (RILs) and highlysaturated with molecular markers need to be created, allowinga search for markers tightly linked to putative pollen sterilitygenes or for association mapping and genomic selection pur-poses (Jannink et al. 2010; Meuwissen et al. 2001; Veyrieraset al. 2007).

Despite breeding efforts resulting in commercial triticale va-rieties, i.e. Hyt Prime, Hyt Max, and RGT KEAC (Geiger andMiedaner 2009), little is known on QTLs conferring pollen ste-rility in the CMS Tt system. There are pieces of evidence thatsome of them are present on the 6RL, whereas some are lesseffective on the 4RL chromosome (Curtis and Lukaszewski1993). Some authors (Geiger et al. 1995; Kojima et al. 1997;Ma and Sorrells 1995) documented the importance of the sixth

chromosomes of Triticum aestivum L., as well as the Rf3 locuson the 1BS, in the restoration of pollen fertility. The presence ofrestoration genes was also evidenced on the 2A, 4B, and 6Achromosomes (Ahmed et al. 2001) and 1BS, 5AL, 5D, and6BS using RFLPs (Ma and Sorrells 1995). In parallel to thelimited information on pollen fertility restoration genes in triticalewith CMSTt (Góral 2004), there is no information on howmanyof them participate in male sterility. Finally, their chromosomalassignment, precise chromosomal location, not to mention puta-tive function, is also not known (Geiger et al. 1995). All of thatlimit the evaluation of valuable maintainer lines in triticale. It isworth noting that published molecular studies on pollen sterilitygenes are usually restricted to the F2 generation (Góral 2013) asthe male sterile genotypes cannot be reproduced by self-pollina-tion. However, mapping populations on crosses between main-tainer lines based on non-sterilizing cytoplasm and restorer ones(on CMS Tt) may solve the problem. Such an approach mayallow evaluation of advance recombinant inbred lines (RILs).When crossed to the maternal line with sterilizing Tt cytoplasm(sterile analogue of the maintainer line), the pollen sterility phe-notypes of the RILs could be assessed by analysing the numberof seeds per spike of the F1 progeny (CMS-Tt). Thus, the eval-uation of respective RILs required for genetic mapping purposesas well as the assessment of the phenotypic data for the quanti-tative trait analysis may allow for the identification of molecularmarkers linked to the QTLs and useful for marker-assisted selec-tion (MAS) (Góral and Spiss 1998) purposes.

The aim of the study was to identify QTLs conferring malesterility and pollen fertility in triticale with CMS Tt usingdedicated genetic map evaluated on RILs and saturated withSNP and silicoDArT markers to assess markers linked withthe trait.

Materials and methods

Plant materials

The population of 182 recombinant inbred lines was devel-oped in Plant Breeding Company-Strzelce (BreedingDepartment Borowo, Poland) by single-seed descent methodfrom a cross between a plant fromHT352maintainer line withnon-sterilizing cytoplasm (N) and cv Borwo recognized as agood restorer genotype during earlier breeding experiments.The F1 plant was bag-isolated, and the F2 seeds were groundplanted. Each of the F2 plant was a starting point for thedevelopment of the RILs. The plant materials werereproduced during six consecutive seasons resulting in theF6 generation.

The HT352 maternal line with Tt cytoplasm (Tt), the ana-logue of the HT352 (N), was crossed by common baggingwith spikes of male parents—the individual of RIL F6:(N) × Borwo. The seeds were planted on the field in two rows

60 J Appl Genetics (2021) 62:59–71

(10 plants per row) with a 25-cm spacing during the 2016/17vegetation season. The HT352 (Tt) × RIL F6: (HT352 (N) ×Borwo) F1 plants were bag-isolated, and the F2 seeds werecollected for counting. The male sterility of maternal line wastested before crossing in two locations Borowo and Strzelce(Poland). Phenotyping was performed once in Borowo.

Phenotyping

The number of seeds per each F1 progeny was evaluatedusing main spikes collected from four different plantsrepresenting 154 out of 182 RILs. Only the spikes unin-fected by brown rust were considered; thus, 28 lineswere omitted. The number of spikelets per spike wasnearly constant (15–16 per spike). It was assumed thatthe F1 plants (corresponding to the respective RIL F6line) with the highest average number of seeds per test-crossed spike was 100% fertile. Thus, the fertility ofeach RIL was expressed as a proportion of average seedsnumber per spike to the average number of seeds perspike of the most fertile line. The sterility values werecalculated as the difference: 1 − fertility and transformusing an arcsine square root transformation of sterility(Sokal and Rohlf 1981). The normal distribution of thetrait was tested using Kolmogorov-Smirnov test withLilliefors significance correction implemented in theXlStat software (XlStat 2019).

DNA isolation

The total genomic DNA was extracted from ca. 100 mg offresh leaf tissues of a single plant of each 182 RIL F6: HT352(N) × Borwo encompassing the mapping population accord-ing to the manufacturer instruction using DNeasy Plant MiniKit 250. DNA integrity and purity was tested via electropho-resis on 1% agarose gels stained with EtBr (0.1μg/ml) in TBEbuffer, while its quantity was measured spectrophotometrical-ly on the NanoDrop (ND-1000).

Genotyping

DArTseq genotyping was conducted using PstI and TaqIdigestions, and sequencing was performed with a HiSeq2000 sequencing system (Illumina Inc., San Diego,USA) at Diversi ty Arrays Technology Pty Ltd. ,Australia (Sánchez-Sevilla et al. 2015). The resulting se-quences were filtered for quality, with a cut-off at 90%confidence. The generated single nucleotide polymor-phism (SNP) and silicoDArT markers were encoded tofulfil the MultiPoint Ultra-Dense software requirementsestablished for recombinant inbred line mapping popula-tion (Ronin et al. 2015).

Linkage map

For the construction of the genetic map, all SNP and silicoDArT loci that showed no or limited deviation from the ex-pected segregation of 1:1 (chi-squared ≤ 19.2) wereemployed. Moreover, markers exhibiting more than 15% ofmissing data were removed from the analysis. As markerphases might be assigned incorrectly, both possible phaseswere implemented in the MultiPoint Ultra-Dense software(Ronin et al. 2015).

Mapping consisted of the following steps: (1) Grouping ofmarkers with zero distance and selecting a “delegate” fromeach group (no fewer twins than the predefined threshold wereselected). (2) Except for twins of the candidates, all remainingmarkers were removed to the Heap. (3) Clustering the delegatemarkers and ordering the obtained linkage groups (LG) wereperformed. (4) Filling gaps and extending LG ends usingmarkers from Heap. (5) Removal of markers violating mapstability and monotonic growth of distance from a marker andits subsequent neighbours.

Chromosomal assignment and arm orientation

The wheat LG groups were assigned to triticale chromosomesand oriented in S-L direction based on known wheat genomelocation of SNP and silicoDArT markers (WCM) (DArT-P/LDAT 201 6 ) , p h y s i c a l m a p o f wh e a t (WPM)(www.wheatgenome.org 2018), the genetic map or rye(RM) (Bauer et al. 2017) and triticale (DH-T) (Tyrka et al.2015). The synteny data were applied to verify both the S-Lorientation the LGs representing rye chromosomes and theassignment of the LGs to the rye genome (Martis et al. 2013).

Genetic map comparison

The genetic map of triticale saturated with SNP andsilicoDArT markers (Tyrka et al. 2015) was used for the com-parison. Similarly, the genetic map of rye (Bauer et al. 2017),the wheat consensus map V4.0 (DArT-P/L DAT 2016) andphysical map of wheat (www.wheatgenome.org 2018) wereapplied. The order of common markers was tested by Pearsoncorrelation in the XlStat software (XlStat 2019).

Composite and multiple interval mapping

Composite interval mapping (CIM) and multiple intervalmapping (MIM) were performed in the WinQTLCartographer software (Silva Lda et al. 2012) following sug-gestions presented elsewhere. In the case of CIM, the windowsize was set to 1 cM and walk speed, to 1 cM, while thenumber of control, markers to 25. Backward regression meth-od was applied. The significance of the QTLs was tested using1000 permutation test.

61J Appl Genetics (2021) 62:59–71

Multiple interval mapping (MIM) was also conducted inWin QTL Cartographer (Luciano Da Costa et al. 2012). Thefollowing settings were applied in the MIM model: MIM for-ward search method. MIM forward regression approach wasused with the following MIM selection criteria: Bayesian in-formation criterion (BIC)→ c(n) = ln(n); MIM search walkspeed in cM: 10. The position of the QTLs was optimized,and additional QTLs were searched and epistatic effects test-ed. Heritability, in a narrow sense, was calculated as the addi-tive genetic portion of the phenotypic variance available as theoutput from the WinQTL Cartographer.

Results

Phenotyping

Parental forms of the RIL F6: HT325(N) × Borwo were fullyfertile. The maternal line HT352(Tt) was sterile; however, inthe case of bag-isolated spikes, marginal seed setting (one totwo seeds) could be sporadically present. Phenotyping of theF1 progeny of the HT352 (Tt) × RIL F6: (HT352 (N) ×Borwo) showed that varying numbers of seeds were presentin each of the lines (Supplementary File 1). The highest aver-age number of seeds per spike equalled to 74. The phenotypicdata concerning 28 lines infected and damaged by brown rustwere assigned as missing for analysis.

The Kolmogorov-Smirnov normality tests with Lillieforssignificance correction based on transformed phenotypevalues obtained for seeds of F1 progeny of the HT352(Tt) × RIL F6: (HT352 (N) × Borwo) crosses indicated thatpollen sterility trait showed a normal distribution (D = 0.07,p value (two-tailed) = 0.385 at α = 0.05). Graphical represen-tation of the trait distribution and results of the Kolmogorov-Smirnov test are given in Supplementary File 1.

Genetic map

The RIL F6: HT352 (N) × Borwo mapping populationencompassing 182 individuals genotyped with SNP andsilicoDArT markers resulted in 29,671 SNP and 92,255silicoDArT markers implemented for the construction of agenetic map. The markers with segregation distortion (chi-squared ≤ 19.2) and missing data (≥ 15% of missing data)were removed from the analysis. Assuming all markers (skel-eton, redundant and added: markers with lower reliability re-moved duringmapping steps due to segregation distortion or alarge number of missing data), the map encompasses 29,737markers. In total, 1700 skeleton and 15,568 redundantmarkers were mapped on 21 linkage groups (Table 1,Fig. 1). Most of the markers were silicoDArTs (23,923); theremaining were SNP (5814). The fewest number of skeletonmarkers mapped to 1A while the largest to 3B chromosomes.

The map spanned over the distance of 2549 cM with a skele-ton marker per 1.7 cM on average. The longest LG (6A) cov-ered 183.5 cM, while the shortest (4R) 65.7 cM. Despite thehigh saturation of the map with molecular markers, gaps be-tween skeleton markers were still present (Fig. 1) with thebiggest spanning over 30.6 cM (6A). Eight LGs have gapsover 20 cM in length (Table 1), resulting in a lack of randomdistribution of markers along linkage groups (SupplementaryFig. 1). Analysis of marker density revealed that the 2R, 1A,and 6R chromosomes had the most uneven distribution,whereas the others more or less even distribution (Fig. 1).

The SNP and silicoDArT markers, with the known chro-mosomal location on wheat consensus genetic (WCM)(DArT-P/L DAT 2016) , whea t phys ica l (WPM)(www.wheatgenome.org 2018), and triticale map (DH-T)(Tyrka et al. 2015), allowed the univocal assignment of 14LGs to the respective chromosomes. Knowing the chromo-somal assignment of some markers to either short (S) or long(L) arms of the wheat genome, it was possible to orient linkagegroups in the S to L direction (Supplementary Fig. 2 a, b, c).However, using DH-T (Tyrka et al. 2015), the S-L orientationof the LG representing 5B chromosome of our map was dif-ferent from the one suggested byWCM andWPM. In the caseof the rye genome, the marker pattern reflecting synteny wasrecognized (Martis et al. 2013). For example, the linkagegroup encompassing markers mapped to 5A, 5B, and 5Dfollowed by those from 4A, 4B, and 4D ones were typicalfor 5R chromosome allowing for the proper S-L orientationof the LGs. Similar reasoning allowed the assignment of theother rye LGs. The S-L orientation was estimated based on thecurrent marker density reflecting putative synteny or based onorientation on the other maps, including rye (RM) (Bauer et al.2017) and DH-T (Tyrka et al. 2015) ones (SupplementaryFig. 2a). It should be stressed that the S-L orientation of the21 linkage groups encompassing RIL F6: HT352 (N) ×Borwo genetic map with WCM, WPM, RM, and DH-T (ex-cept but 5B) maps coincided. In all cases, the good collinearityof marker was observed (Supplementary Fig. 2a, b, c). Thecollinearity of the RIL F6: HT352 (N) × Borwo map and RM,DH-T, WCM, and WPM ones revealed no evident transposi-tions or translocations between the respective chromosomes(Supplementary Fig. 2a, b, c). Comparison of the correlationcoefficients of the RIL F6: HT352 (N) × Borwo and WCMshowed that the coefficients varied between 0.790 (5A) and0.997 (5B) (Supplementary Table 1). In the case of the wheatphysical map (WPM), the respective figures ranged from0.686 (1A) to 0.976 (1B) for the wheat genome. A similaranalysis for the rye genome, when RIL F6: HT352 (N) ×Borwo was compared with the rye genetic map (RM) resultedin the range of 0.84(4R)–0.96(6R). The broadest range ofcorrelations (0.57(2A)–1.0 (6A, 7A)) was observed whenour genetic map was aligned with the triticale one. While theRIL F6: HT352 (N) × Borwo linkage groups aligned mostly in

62 J Appl Genetics (2021) 62:59–71

Table1

The

arrangem

ento

fgenetic

mapping

dataevaluatedbasedon

182RIL

F6:H

T352(N

)×Borwomapping

populatio

nandskeleton

markers(SM)

Quantifying

markersandtheir

derivativ

esChrom

osom

esCharacteristics

1A1B

1R2A

2B2R

3A3B

3R4A

4B4R

5A5B

5R6A

6B6R

7A7B

7RΣ

Av/

Chr

Max

Min

SM19

109

6399

110

4779

117

7550

4273

105

102

9770

82109

107

6184

1700

81117

19

TM

112

859

1180

702

1183

645

457

745

1194

248

211

1083

548

805

1224

457

763

1677

950

930

1295

17.268

822.3

1677

112

GM

172

1323

2321

1089

1893

1457

616

1263

2655

350

332

2401

733

1185

2187

732

1117

2826

1499

1219

2367

29.737

1416

2826

172

SM/cM

4.72

1.12

1.16

1.56

1.42

1.92

1.88

1.34

0.93

2.25

2.59

0.9

1.7

1.63

0.8

2.62

1.59

0.94

1.38

2.15

0.99

35.6

1.7

4.7

0.8

TM/cM

0.8

7.02

16.194.54

7.56

7.15

3.08

4.74

17.082.21

1.94

16.483.06

4.85

15.852.49

5.86

16.436.41

7.09

15.56166.4

7.9

17.1

0.8

LGL

89.6

122.472.9

154.7156.490.2

148.6157.169.9

112.4108.665.7

178.8166.177.2

183.5130.3102.1148.1131.283.2

2549

121.4

183.565.7

Gap

20.1

10.8

16.1

20.8

13.1

29.1

24.1

14.7

6.9

22.3

23.9

3.4

17.3

13.7

6.6

30.6

16.8

7.6

13.3

20.9

15.8

347.9

16.6

30.6

3.4

Av/Chr

averagemarkersperchrom

osom

e,SM

skeleton

markers,TM

totalnum

bero

fmappedmarkersincludingskeleton

(S)and

redundantm

arkers,G

Mincludes

skeleton,redundant,and

addedmarkers,

LGLlin

kage

grouplength,SM/cM

skeleton

markerspercM

,TM/cM

totaln

umberof

markerspercM

63J Appl Genetics (2021) 62:59–71

a linear way with WCM, RM, and DH-T maps, nearly in allcases, the “S” shape was observed when compared with theWPM.

Composite and multiple interval mapping of pollensterility

Composite interval mapping showed the presence of at least13 QTLs conferring male sterility and pollen fertility in triti-cale with CMS Tt identified based on the RIL F6: HT352(N) × Borwo mapping population (Fig. 2) and shared amongten chromosomes (Table 2) withmaximum log of odds (LOD)score higher than the permutation test threshold (3.7). SingleQTLs were mapped to the 1A, 1B, 2A, 2R, 3B, 3R, 4B, and5B chromosomes, whereas the 5R and 6B chromosomes have3 and 2 QTLs, respectively. The QTLs mapped on 1A, 1B,3B, 3R, and 4B originated from HT352 genotype, while theothers from Borwo. The LOD score of the QTLs ranges from4.23 (2R) to 16.28 (5R). The QTLs with the highest LOD

score were identified on the 5R, 3R, 1B, and 4B chromo-somes; however, the QRf-5R.3 has the highest explained var-iance of the trait (R2 = 15.1%). Out of the QTLs with thehighest LOD score, four (QRft-1B, QRft-3R QRft-4B, andQRft-5R.2) have a negative additive effect. The QTLs coverfrom 0.5 to 11.2 cM and are bounded by tightly linked mo-lecular markers as indicated by recombination frequency(Rec-L and Rec-R) of the LOD maximum and the nearestmarker from its both sides (Table 2, Fig. 2). Markers4253609, 10597858, 4342600, 3605423, 2359223,4220579, and 5619416 were located exactly at the LOD max-imum of the QRft-1B, QRft-2R, QRft-3R, QRft-5R.1, QRft-5R.2, QRft-5R.3, and QRft-6B.1, respectively. The mosttightly linked markers are marked in red on Fig. 2.

Multiple interval mapping (Table 3) confirmed the pres-ence of ten QTLs (except but QRft-1A, QRft-3B, and QRft-5R.1) with the same or nearly the same chromosomal posi-tions as those detected by CIM. In all cases additive effectswere observed. Two epistatic effects (QRft-3R ×QRft-4B and

4544759

5625561

5605515

4349618

42206584354050361211185161401051745443431063043610|F|0--6:G>C3611621360593085222548531266|F|0--51:A>G119095774369261|F|0--8:A>G36122924214801

A 1-tf

RQ

1A

43631634352841163290508519135421955742151238515793853599685227803604292|F|0--22:G>C851545936161453046426|F|0--16:T>C36193683605205|F|0--55:C>T361933843455443045631|F|0--15:G>C10515114|F|0--10:G>A4340159163295783612761435226343468918521022434495036194324361527|F|0--39:A>G4366235361353036162283615470455687443632333616992|F|0--8:C>A456005543727973041473|F|0--55:A>T4363402421866685189383047799|F|0--55:C>T10516576|F|0--16:G>C42196758519411425998745680454253609425550243611784568731100071792|F|0--20:G>A85336413610642851762685174694346607456281010514094|F|0--28:C>G3612416|F|0--8:C>G434430136119998516652851974836242894357402852091236043823616935|F|0--24:G>A362409736071673043224|F|0--24:T>C3607740421342442023093046904|F|0--51:T>C851472236156213040997|F|0--5:T>C45512673044800|F|0--39:A>G3041221|F|0--39:G>C8534029420711910508858|F|0--64:T>G435338710514529|F|0--10:C>T42147081051668145495563047424|F|0--20:T>C421460614477405|F|0--35:C>A4217648851881345475301632928285357984216877853869243650693608467360885730472894203018105117094368199|F|0--7:G>A36041524339204

QRft-1B

1B

3604706100159541|F|0--47:C>T362089836239874213783851905785191843623477853794336198604349490455386130405584347154455259843639173041513|F|0--36:T>C119117854562598361045810516276420361445471121052099185173583610385|F|0--8:C>T85190098519275436671636094983605998851025643474654349664434211630439984367803|F|0--37:T>C85336911051682343579328533411434729643456844345387360712136153794214186|F|0--58:G>T420926736114364202393|F|0--19:G>C3614709435601010519563|F|0--46:C>G43714904343514104978433605707851468143426613623705360790436103523622018

1R

4562141433946336210093048000|F|0--50:G>T420052615998179852029436125393604044435256643429243041971|F|0--61:C>T360685836162848511798|F|0--10:A>C8512126|F|0--24:G>A4348730436553236229051051669016311593853927042057733043902|F|0--10:C>G104981633047960|F|0--37:T>C8510386|F|0--18:G>T4343918454504110523870|F|0--25:C>A4347112360339785117313609132437343643407594203833421544311908810853591116330062853958236064554364097852058215998401|F|0--15:A>G3040812|F|0--5:T>C3046126|F|0--42:C>G36231263613610163307954208244|F|0--10:C>G3604178100168520|F|0--5:A>C436213942095175041728|F|0--50:G>A4208655|F|0--40:A>G434007536187804354971|F|0--50:C>A36113194371397454997085171824207135|F|0--49:T>G42040114203341|F|0--52:G>T421583643510013609607|F|0--24:T>A36038474343927|F|0--24:A>G435813142136664348063100112184|F|0--23:A>C85336864346225100150521|F|0--10:A>G3040630|F|0--53:T>C435018510524845304482085162723615265100151465|F|0--18:A>G8515433422019142169984544789851706810504543435388645500174352035422043042005644551125

QRft-2A

2A

3608526144702728517354434133485361853041194|F|0--12:T>A36173094369024|F|0--14:A>G8511769|F|0--17:G>A454510985205073624638100125989|F|0--9:C>T8510963434399243479003044849|F|0--16:C>T4350767434287610522547|F|0--66:G>A3616542853789836068233608813853762836065243611095436435685383718511027100185797|F|0--7:T>C4572122851032336127124205208851680910502218851514245702954351203435265636139298511017|F|0--16:T>G853372745586664358669434161142028381446976942212034545889105231353047582435910043464544549235454701610516429|F|0--11:C>G455268336236764358284|F|0--37:T>G4205923434367943548231632919842027058538689|F|0--20:A>G8518802434461715996538159965534571287434657536223998509653|F|0--17:C>T14472111455021485186324353267360854136218534215569105127103041947|F|0--23:G>T43459574340729163292511635388742034844364574853656511909164361034743425573609365163293848538184851927216329107163295488512379455639410499582104996121632996416329994105243378527670456991310513238

2B

105978583057937|F|0--55:C>G4348212437381211912386|F|0--10:T>C10520120|F|0--17:C>G10513333420162214477573851620385196898511631454487242130624220568|F|0--12:G>A4218637|F|0--22:C>G4214009144770071447010843433494360208435292811912951435031743432188510197420252342021158510989853914014471314851437185104294373747851534742058568515253159966081633122345714251191234715996397144769084343367851039642174423614978

QRft-2R

2R

105221374344980851528843562498516170852025542196924556623100151157|F|0--7:A>G3623900|F|0--50:T>G4340843361210042192224218640437259342099924555274434688345517674566428|F|0--23:A>G100182788|F|0--43:G>C10523672|F|0--10:G>C851714236243284341388421528645746694341791435889143499984203675455198410515220|F|0--14:A>G3047102|F|0--5:G>A8516008851746742146773045990|F|0--49:G>A155729263057600|F|0--23:T>A5556820434389785219704219128360602642143403041479|F|0--20:G>A436463442186064364615|F|0--23:C>A45572153041533|F|0--58:T>C3611634100045194|F|0--42:A>T3046239|F|0--9:T>G3617982|F|0--8:A>G3047854|F|0--8:C>T1052429485104424370868|F|0--14:A>G45594741191184236044343623091361502736055804354312360864385217044571427853726236098114373503360588836139621049824343509981049783616329676

3A

455126036239583612120454912630439073615418434458385144434343675104976023604978454917136225643609805|F|0--14:T>C85147134557548851867236139034221542360450185394363623547853936585190818518407361862336220448519188362360536112301052112811912136|F|0--22:T>C851028636201384217475434357110516854|F|0--11:G>A36197163613331105223703623056|F|0--11:G>T85214751051146245625253040605|F|0--44:A>G36144423617867|F|0--49:A>T3044421|F|0--20:C>A435225242209913616787|F|0--8:A>G42035243047168|F|0--13:C>G4354055361615743454944349877360922536189304557708455806236129573621302100129161|F|0--23:T>C42059144565940|F|0--24:T>C36038004358750422094036248163616410362168543446721447128645480994368231|F|0--23:G>A10525046|F|0--38:A>G3617811|F|0--54:T>G36211034370235|F|0--35:A>G3618676|F|0--37:A>G3624438|F|0--32:A>C85259523611405|F|0--62:T>C3603933|F|0--18:C>G456271985259814212900|F|0--25:A>G85179974368879|F|0--30:C>T163297604216210105087754342061104977391052477916358565422049610520180851702642201484216914434753445500334358129421325510513656|F|0--14:T>C43546493046469|F|0--44:C>T855056585210818520936853831336197894205751420172016330059

QRft-3B

3B

105137704575483421992610520151|F|0--33:T>C455709785338001052311736128728534296362379736129323604315853429211911760|F|0--12:A>G8521741851964685216894365220362022243668424201606851241385259633616614362078236062103613970434212042174211632978436049344211193|F|0--9:G>A8533660360415114476208851769914476907434505785181414348915144700414548777435985185156171632903143440423615011852594085259793625000361862685222674320820434260082171118237911|F|0--33:G>C3618167|F|0--18:C>T361234136224283618288|F|0--68:G>C435932285204573606260361615042143954209574|F|0--17:T>C851965736182081052333243727113047890|F|0--32:A>G4340104|F|0--62:G>C3624360433959516329410

QRft-3R

3R

3623584|F|0--22:G>T851531410522166|F|0--25:G>C3605617|F|0--58:C>T3043317|F|0--47:T>C5042775|F|0--27:G>A560580736036734570988436828745569518518953105126984356897435352242195034343921361582085118494219402435562045491383603961|F|0--27:G>A8536081|F|0--13:T>A36035644365923362409642054938531741|F|0--22:G>A8537504852193843540588531377433955810520996|F|0--13:C>G4216235420079785213864369192|F|0--42:G>A4570684144690053618527|F|0--25:G>A45479813617269|F|0--45:T>G360521236064991051600910512198|F|0--6:G>T85155743045781

4A

163564333042631|F|0--39:T>C10514695|F|0--6:C>T3393290233907924853506345465163622839422030842018494206295|F|0--27:C>G852242010510437362286136187144203604434869785176948509694|F|0--13:G>A100174690|F|0--48:C>T8523874434484414475758434081342172794211186|F|0--63:A>G85189294214939|F|0--40:C>T435343642035094202895

36250248519922119079703041031|F|0--39:T>G45456293043991|F|0--68:C>T43418283616394435045285197054218507

QRft-4B

4B

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

105

110

115

120

125

130

135

140

145

150

155

160

165

170

175

180

Fig. 1 The schematic illustration of the RIL F6: HT352 (N) × Borwo genetic map. The QTLs identified by CIM are indicated by red bars (in red are themost tightly linked markers)

64 J Appl Genetics (2021) 62:59–71

QRft-3R ×QRft-5R.3) were identified. Heritability (in a nar-row sense) varied from 0.6 (QRft-5R.2) to 16% (QRft-6B.2).Additive-by-additive effects were identified for the QTLs withepistasis. In total, the heritability of all QTLs, including epi-static effects, was 68.83%.

Based on the origin of markers conferring QTL regions,five QTLs originated from the maternal line, whereas theothers were due to the parental form (Table 2). Four QTLsexplained more than 0.1 of the phenotypic variance of pollenfertility restoration. Male sterility QTLs were assigned towheat genome 1A, 1B, 3B, and 4B chromosomes, except onone localized on 3R. Male sterility QTLs explained less than0.1 of phenotypic variance each.

CIM and MIM analysis identified ten common QTLs,namely, QRft-1B, QRft-2A, QRft-2R, QRft-3R, QRft-4B,

QRft-5B, QRft-5R.2, QRft-5R.3, QRft-6B.1, QRft-6B.2,QRft-3R, and QRft-3R. Both methods indicated exactly thesame chromosome positions for six QTLs located on 2A, 3R,5R, and 6B chromosomes.

Discussion

The RIL F6: HT352 (N) × Borwo map presented in the givenstudy is in concordance with either wheat consensus (DArT-P/L DAT 2016), physical (www.wheatgenome.org 2018), rye(Bauer et al. 2017), and triticale genetic (Tyrka et al. 2015)maps. However, some discrepancies concerning linkagegroup length, the extent of gaps, or number of markers

851672442132601632840736123231051236285143458534054362357743435683041713|F|0--8:G>A853822945599864571573851915636063604340420435469642058624556179159963778510680|F|0--63:C>T3041643|F|0--24:G>T4346447435555245541418535961|F|0--5:C>G8536299852597485228668537339851629036082758520192361434185389653607540456944036167184202997163290183042846|F|0--16:T>C8536471163288843611082105251184352691420986216328830163283813614061104960288521942852191836206623046560851906336122998533903454687285339864555632304461136180444365669853302636240743617304361537536138413616270361471085125268539271

4R

434347543540124349365163288394370117|F|0--31:C>T8510804|F|0--5:G>C3042960|F|0--10:C>G4206021|F|0--21:T>C10501598100114176|F|0--19:T>C3044668|F|0--23:G>T3621122|F|0--40:A>G36120173044451|F|0--13:A>G851804010502930436638736059314217899436803736214203616827|F|0--27:A>G362210785390163616509360627745580724345236454986016329307119083361632977142134684211107|F|0--55:T>C3046401|F|0--19:G>C3042503|F|0--66:C>T3044443|F|0--41:C>G4355282|F|0--52:G>A43410643043629|F|0--53:G>C36121854340941434215285320753042567|F|0--28:C>A4353120163292733615123360515885335993044426|F|0--32:A>C85354128510112|F|0--17:A>G45567368538185144740874344233437189036195038523773360966310518374|F|0--43:A>G45562283616807|F|0--47:A>G100069625|F|0--28:C>A43393001190876585376553610542436101743610043624618105119123612049|F|0--23:G>C4561479360361236131653046132|F|0--48:T>C43610503621301|F|0--12:C>T43412468532550|F|0--25:C>G100144253|F|0--32:C>T8511171|F|0--16:A>G3604347421309085387063043362|F|0--33:G>A3622467|F|0--60:C>T3045867360364685186643042350|F|0--63:T>C36159653046729|F|0--33:T>G3042831|F|0--22:A>T3040872|F|0--53:T>G43678701190985242180383621575|F|0--12:A>G3040778|F|0--33:C>T360975985155824573426

5A

5607554434976845605283619097435153785171944216942105018071049751436130094358789361445143532063043512|F|0--64:T>G10506913|F|0--39:C>G1050160542148954371049|F|0--46:C>T4201021|F|0--10:T>A360464036126533042874|F|0--30:G>A304708043417773610003851016711907905434282485177203046489|F|0--61:C>T4208366|F|0--23:G>A360809211909599435151743635008517461100187163|F|0--32:C>G3608300851540442198323622578|F|0--8:C>G435063936091864550005434587585228848535987435033685235743612909436030736203164366668360659036210224340583435085343537293616234360576636143543613827|F|0--35:C>G16329373104971893616051361495536235211049951836160628536257360949185313894370232|F|0--47:G>A4559137434050945737053617397|F|0--47:G>C36112324571773851624615997669|F|0--25:C>G43489193046334|F|0--49:C>T455441043510073621044144726834358533455894585187873613325|F|0--6:G>C434997614476268|F|0--6:C>T851873185192883040736|F|0--18:C>T3619078361352136144543043986|F|0--9:C>T36083364344137

QRft-5B

5B

85123904215211421852036203474212019852010443731663615275163553654573075|F|0--31:T>C45476124356728455173342139164203329456926330406593616618|F|0--25:C>A360408742051323622240454586085324954205067360767943452271051237236129794217813420514443714191049790816357067163298464357252360529443454578519093455513936118534545472851968743718731050114843540034371607852380210516140361551543525104204462360542311212233|F|0--22:G>C16328495105163193021225|F|0--17:T>A360752223398282359223220207916358117360633385184253619923|F|0--47:G>A85332585605722454659856097774540257422057945615613615679304778043437344363422434224636163864570270421311436112374349658360673685376548510941|F|0--39:C>A8518778853548243504378519636360363285196944218413|F|0--29:G>T42204573041610|F|0--45:G>A1049988843463231051884710510378

QRft-5R

.1

QRft-5R

.2 QRft-5R

.3

5R

11908652851328045593064203530456886745727861191172685192153045111|F|0--8:A>G36038523042555|F|0--31:T>C456221185177813619683434498410302438119128934370626362290236114744360165434961985227751633002436091264201131|F|0--39:A>G10519298|F|0--18:A>G4218227100067834|F|0--16:C>T85395473042739|F|0--26:C>T421326310507955|F|0--66:G>T43450093616380|F|0--12:G>A4350305|F|0--31:T>C36101334367650|F|0--65:T>C10522459|F|0--21:T>A42190263048122|F|0--15:T>C4214806853495643666338538580|F|0--7:C>T36107304344563105127508534067455610243418484219305|F|0--11:A>G3616067853432243618283604877

4217231

8522125

3615427

4557015420328443472333047932163296704216989105176688522421435270945548814212841

6A

10500455105012971632971036089814201151853653411912913|F|0--8:A>G3609402|F|0--15:G>C1632811742149904371576|F|0--56:G>A144764103623999105246994341857421985585189343612019457023836109813614349|F|0--6:A>T4561621119073788535698434045636123291191280542052041051345056194165040595|F|0--6:T>G433957385120878533732851663385391514363559|F|0--42:G>T851678710517268|F|0--5:A>G85193041049796216329004420425842211231050293536163484563065361881085106018512473|F|0--43:C>T4211154|F|0--68:C>T85106303615750|F|0--19:A>G3617898|F|0--14:T>C43594864359033434393936214004339736455251043726183616238|F|0--20:T>C105185934210067|F|0--29:A>G8513916105194253607345456839445662904215078105116504346775853634532033533032339|F|0--8:A>G8525333163290451632837145505038514809104973193609552

QRft-6B

.1QRft-6B

.2

6B10517863421391416328304851483610516248144713563046122|F|0--53:C>T43481421447687243541504202548851866636220361448000143515524344613422033545523674220583159981933611301|F|0--9:G>T4365530105209184361696360448385187544211363|F|0--23:G>C422121545700638536118435663485387358536126119122254572956119122914369146|F|0--42:T>C1599631310522024105012368513286455256385332378537784361289485160284346131853812445624678534965|F|0--12:C>T3610776454610543499368515608105161698514812360691636223508517311853718536202731599636010524944159983543047858104984241635425943394564220372100142111|F|0--8:T>C4221093433987042152094220182851971142182594348251420059243426184208932|F|0--26:T>C4202374436561636113304341243853633243414078519713361491416354292100146558|F|0--61:A>G16329037105216588531336455688436038754210613|F|0--14:A>G85227174204839361455310515227361950885111378538394456921985130568520232422091585396993605287

6R

1635375436107704573491435649836065024202652105205303043275|F|0--11:T>C42167684216463|F|0--11:G>A10525403|F|0--18:A>G85338934349817163561273614798105027784368468|F|0--50:T>G4557804|F|0--40:A>G455184436225974211622|F|0--65:G>A1631191636136301632808942189364365550434010145461674567747|F|0--5:C>T4347752853407614479628105016898519632105123573045048|F|0--63:C>T3041172|F|0--52:T>A435598585109681632918545738641051278085191003042384|F|0--16:C>A85105431632835943566831049791985188753042017|F|0--36:T>C455262943552273045708|F|0--28:G>A8537515435474510497655105020494339246304547943493233047282|F|0--49:T>C85333063619224|F|0--27:G>A43458683612962361434536137508512804|F|0--8:T>C434259036221913610571|F|0--47:T>C36037874354785100176966|F|0--11:T>A436416536037363604487100111956|F|0--14:G>T4342542434383536032714355243852596985317754219961457105436134044370638|F|0--53:C>G4344662|F|0--27:C>T36106193603382360976942154103606617|F|0--28:A>C3047817|F|0--15:A>T3047784|F|0--33:T>C3615524435152443570223043726|F|0--35:G>A42154684566068434007685343241049753343518473604988

7A

85145984344890361583736134773621489|F|0--54:A>C8511215|F|0--6:T>C8537884|F|0--36:T>C104982344347642434313985163734212474|F|0--35:C>G10516838|F|0--30:C>A119083274215020422052545526574219959|F|0--27:G>A4212625|F|0--33:T>C105229298521605421370943584643045476|F|0--40:T>G361103642212443045485|F|0--49:C>T45687963047729|F|0--41:C>A4214000|F|0--56:A>G851786443453364213183455232945728128539510420317445724784212151|F|0--44:G>A42107408510689|F|0--8:C>T43603634206568|F|0--23:A>G436375085365331051272036161311191199236078471635445643569774221773422037785206618522628420567542151528533407436492042136134345950

7B

1632869416353756105114404370127|F|0--60:T>C105019394547174853606436098991191226236076973624646851019636143004347167437377785167843623613362008685337853622594361090843401474555990434380743460174340353119093721190920185121194330103|F|0--38:C>G36237273617502|F|0--64:A>G361950485156491051945514470680455491043478294369413|F|0--68:C>T361604536223163610244|F|0--14:A>G3609883361526743518133617729|F|0--21:C>T3611681163300514364681|F|0--64:C>G45714933607853851491845547444373531|F|0--14:G>A434597843445943622957362297143491544574121362312814475854144712783044680|F|0--13:C>T4353907436374745514473616322361286342136874575477421882885373053606045434420045469074207929851182416328195163524784217354105039851447893416329175

7R

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

75

80

85

90

95

100

105

110

115

120

125

130

135

140

145

150

155

160

165

170

175

180

Fig. 1 continued.

65J Appl Genetics (2021) 62:59–71

encompassing the map even if maps of the same species basedon the same marker types were considered are still present.

Our mapping procedure assumed the selection of markerswith limited segregation distortion (chi-squared ≤ 19.2) andelimination of markers with missing data (more than 15% ofmissing data). The restrictive limitations on markers used forthe construction of our genetic map resulted in exclusion frommapping procedures of nearly 75% of markers forming initialset (initial set. 121,926 markers; selected for mapping,29,737). The successive mapping steps reduced the figure to17,268 (1700 skeleton) markers (ca. 14% of the initial set)what might have reduced the overall map length. The shorterlength of the RIL F6: HT352 (N) × Borwo genetic map(2649 cM) than the one based on DH-T lines (4910 cM)(Tyrka et al. 2015) might be the result of many recombination

events (ca. 80 based on skeleton markers) and the restrictednumber of chromosomal changes not so frequent during gen-erative propagation and rigorous marker selection assay. It isnot without significance that our mapping population is basedon 182 individuals in contrast to 92 used in the case of the DH-T genetic map (Tyrka et al. 2015), which may impact on mapresolution and its size. Alternatively, the reduced size of theRIL F6 genetic map in contrast to the DH-T one (Tyrka et al.2015) might be the result of mapping procedure implementedin the MultiPoint Ultra-Dense software which prefers inser-tion of markers between adjustment markers, leading to thereduction of genetic map distances.

Furthermore, the reduced size of the wheat and rye ge-nomes of the RIL F6: HT352 (N) × Borwo mapping popula-tion was observed, due to the lack of some chromosomal parts

Table 2 The arrangement of the composite interval mapping data ofpollen sterility in triticale with CMS Tt for the RIL F6: HT352 (N) ×Borwo mapping population. Rec-L and Rec-R states for the recombina-tion frequency of themarker nearest to the QTL LOD functionmaximum.

LOD score describes the LOD function maximum value. R2 is the pro-portion of variance explained by a QTL at a test site with the estimatedparameters. Marker position reflects the number of the marker on thegenetic map, whereas “Position” is its position in cM

No. QTLname

QTLorigin

Chromosome Marker code Position maxLOD (cM)

LODscore

Additive R2

(%)Rec-L Rec-R Range between

adjustment markersbounding QTL(from R to L)

1 QRft-1A ♀ 1A 4544759 15 4.2935 − 0.0899 8.8 0.1296 0.0438 11.22 QRft-1B ♀ 1B 4253609 47.4 8.2271 − 0.0778 6.6 0 0.0089 3.33 QRft-2A ♂ 2A 5041728|F|0--50:G >A 94.9 6.5434 − 0.0740 5.5 0.0003 0.0056 1.54 QRft-2R ♂ 2R 10597858 0 4.2268 0.0531 3.2 0 0.0119 1.4

5 QRft-3B ♀ 3B 8550565 148.7 4.9121 0.0976 3.8 0.0003 0.0171 1.2

6 QRft-3R ♀ 3R 4342600 48.2 10.5793 − 0.1439 8.8 0 0.0059 4.47 QRft-4B ♀ 4B 33907924 8.2 7.3656 −0.1572 6.7 0.0003 0.0081 3.68 QRft-5B ♂ 5B 5,607,554 1 5.0853 0.1387 17.8 0.0099 0.0152 0.5

9 QRft-5R.1 ♂ 5R 3605423 32.8 8.3856 0.1502 6.8 0 0.0094 2.9

10 QRft-5R.2 ♂ 5R 2359223 37.9 13.2053 − 0.2164 11.6 0 0.0033 3.811 QRft-5R.3 ♂ 5R 4220579 50.8 16.2823 0.1377 15.1 0 0.0157 3.4

12 QRft-6B.1 ♂ 6B 5619416 53.8 4.8811 0.0634 4.2 0 0.0034 4.5

13 QRft-6B.2 ♂ 6B 3203353 110.8 5.256 0.1004 11.7 0.0294 0.0137 2.5

1A 1B 1R 2A 2B 2R 3A 3B 3R 4A 4B 4R 5A 5B 5R 6A 6B 6R 7A 7B 7R

Fig. 2 Results of compositeinterval mapping (CIM) of malesterility and pollen fertility in theRIL F6: HT352 (N) × Borwo.Horizontal and vertical axes rep-resent triticale chromosomes andthe log of odds (LOD) value ofthe tested trait, respectively. Thered line indicates the level of thecritical value (threshold). Blackbars indicate the QTL region

66 J Appl Genetics (2021) 62:59–71

Table3

Multip

leintervalmapping

(MIM

)of

pollensterility

intriticalewith

CMSTtfor

theRIL

F6:

HT352(N

)×Borwomapping

populatio

n.The

arrangem

ento

ftheMIM

.Rec-L,and

Rec-R

isthe

recombinatio

nfrequencybetweenQTLmaxim

umandadjusted

marker;h2

(%)heritability(narrowsense).T

hecommon

QTLsevaluatedin

CIM

andMIM

have

thesamenames

Maineffects

Epistaticeffects

NrQTLeffect

Value

Chrom

osom

eQTLnameMarkercode

Rec-L

Rec-R

Chrom

osom

eQTLnameMarkercode

Rec-L

Rec-R

Position

onthemap

(cM)LOD

h2(%

)

1A

−0.0789

1BQRft-1B

4259987

0.0001

0.01183

45.649

4.76

8.5

2A

−0.0567

2AQRft-2A

5041728|F|0--50:G>A

0.0001

0.0059

94.865

2.17

1.5

3A

−0.0415

2RQRft-2R

D11912386|F|0--10:T

>C

0.0001

0.00767

3.352

1.34

1.4

4A

−0.0512

3RQRft-3R

4342600

0.0001

0.0058

48.214

2.15

3.3

5A

−0.0696

4BQRft-4B

D10514695|F|0--6:C>T

0.03006

0.00790

7.104

3.23

5.3

6A

−0.0692

5BQRft-5B

4342824

0.02009

0.0040

47.831

1.81

6.9

7A

−0.0517

5RQRft-5R.2

2359223

0.0001

0.0031

37.918

1.68

0.6

8A

0.1148

5RQRft-5R.3

4220579

0.0001

0.01590

50.805

7.44

11.5

9A

0.0722

6BQRft-6B.1

5619416

0.0001

0.00302

53.83

3.62

7.5

10A

0.1161

6BQRft-6B.2

3203353

0.03006

0.01408

110.784

2.43

16

2AA

−0.0489

3RQRft-3R

D3618167|F|0--18:C>T

0.0001

0.00592

4BQRft-4B

D10514695|F|0--6:C>T

0.03006

0.00790

1.639

3.5

2AA

0.0382

3RQRft-3R

D3618167|F|0--18:C>T

0.0001

0.00592

5RQRft-5R.3

4220579

0.0001

0.01590

1.099

2.8

67J Appl Genetics (2021) 62:59–71

of some rye (1R, 2R, and 7R) and wheat (2A, 3A, 6A, 7A, 4B,and 6B) genomes, compared with the maps of respective spe-cies used in the study. The observed reduction could be evi-denced based on Durum wheat consensus map (Maccaferriet al. 2015) covering over 2631 cM or wheat consensus mapcovering 2173.82 cM (DArT-P/L DAT 2016), whereas alllinkage groups representing the wheat genome of the RILF6: HT352 (N) × Borwo genetic map of triticale cover1987.8 cM. In the case of the rye genome, the genetic mapof the RIL F6: HT352 (N) × Borwo population covers561.2 cM, whereas the consensus map of rye (Milczarskiet al. 2011), 1593 cM. The observed phenomenon may reflectgenome changes during the establishment of triticale (Wilson1875). Remarkably, the reduction of the rye genome size intriticale was evidenced by Tyrka et al. (2015). The rye genomeequalled to 1084.5 cM (the 7R chromosome was missing),coinciding with the data demonstrating that some rye chromo-somes in triticale showed an apparent reduction in the size ofC-bands at one or more telomeres, compared with normal rye(Seal and Bennet 2011). Thus, the shorter gaps and the elim-ination of some genomic regions in triticale due to, i.e. theformation of a new species, may also explain why the ryegenome of the RIL F6: HT352 (N) × Borwo population wasshorter than in the case of respective rye (Bolibok-Brągoszewska et al. 2009).

One of the aspects of our study was the comparative anal-ysis of the RIL F6: HT352 (N) × Borwo genetic map with theconsensus (DArT-P/L DAT 2016; Marone et al. 2012) andphysical map of wheat (www.wheatgenome.org 2018), thegenetic map of rye (Bauer et al. 2017; Milczarski et al.2016), and the triticale (Tyrka et al. 2015). The consensusmap of wheat (DArT-P/L DAT 2016) (DArT-P/L DAT2016) and the map of triticale (Tyrka et al. 2015) resulted inthe sufficient number of shared markers allowing linkagegroup orientation assessment based on the known locationof SNP and silicoDArT markers on short (S) and long (L)arms of wheat genome as well as the S-L orientation of ryechromosomes. Comparison of wheat consensus (DArT-P/LDAT 2016), as well as physical (www.wheatgenome.org2018) maps and our triticale (Tyrka et al. 2015) one, showedthat the respective wheat chromosomes were most highly cor-related. The only exception was the 5B linkage group as itexhibited a negative correlation with the respective LG onDH-T triticale map (Tyrka et al. 2015). As we have observed,the positive marker order correlation of our linkage group withthe 5B chromosomes of wheat consensus (DArT-P/L DAT2016) and physical (www.wheatgenome.org 2018) map andthe S-L orientation of the 5B group remained unchanged.Similarly, the RIL F6: HT352 (N) × Borwo–based triticalerye genome map correlated well with rye (Bauer et al. 2017)and DH-T (Tyrka et al. 2015) maps.

At last, the RIL F6: HT352 (N) × Borwo genetic map oftriticale is well saturated with molecular markers of low

heterozygosity, indicating that the lines are quite uniform.Assuming that 10 cM marker density (in our case, it is1.7 cM) allows an accurate estimation of QTL positions inthe case of a population size between 100 and 200 (Li et al.2010), the RIL F6: HT352 (N) × Borwo genetic map based onSNP and silicoDArT markers distributed among 21 linkagegroups assigned to triticale chromosomes fulfils the require-ments for QTL analysis.

By now, most of the materials used for molecular studies ofpollen fertility restoration QTLs in rye (Stojałowski et al.2004) and triticale (Stojałowski et al. 2013) are based on theF2 progeny of biparental mapping populations (Miedaneret al. 2000; Stojałowski et al. 2004). Such an approach limitsthe number of putative crossing-overs leading to the necessityof using large mapping populations if a high resolution ofgenetic maps is required (Stojałowski et al. 2013). To avoidgenotyping expenses and to accumulate recombinationevents, recombinant inbred lines could be exploited(Stojałowski et al. 2011). In our study, the F6 progeny of thebiparental mapping population based on maintainer on non-sterilizing (N) cytoplasm and restorer line on CMS Tt wasutil ized. As the F1 plants of the RIL populationHT352(N) × Borwo were on normal cytoplasm, further gen-erations were easily evaluated via SSD approach without lossof QTLs responsible for male sterility and pollen fertility traitin triticale with CMS Tt. Although RILs are often used instudies on quantitative traits, to our best knowledge, in thecase of CMS Tt, they were employed for the first time.

The evaluation of male sterility and pollen fertility trait viavisual pollen fertility scale proposed by Geiger andMorgenstern (1975) or Góral (2002) seems to be hardly pos-sible in triticale as plant materials based on CMS Tt are easyfor restoration, and the visual scale is often binary. The alter-native option is to score the number of seeds per spike. Suchan approach is also supported by the fact that in contrast to rye,triticale is self-compatible, reducing the possibility of overes-timation of sterile plants. It should be noted that the number ofseeds per spike may depend on the number of spikelets.However, this is not the case in our study as the number ofspikelets varied from 15 to 16, and the normalization forspikelets was omitted. Finally, we observed that male sterilityof maternal line (HT352 cms Tt) was preserved with minorfluctuations in different environments (from time to time, twoto three seeds were observed in the case of some spikes ofplants) which is in agreement with results presented by Góralet al. (2006). All phenotypic data were evaluated from a singleenvironment without repetition in successive years but is no-ticeable that the variation within most hybrid combinations ismoderate, which suggests that results are trustworthy.Nevertheless, the interpretation of the data in terms of QTLanalysis must be treated with caution.

It should be stressed that under bag-isolators, brown rustmay infect spikes, reducing or blocking seed setting. Special

68 J Appl Genetics (2021) 62:59–71

care needs to be taken to avoid or minimize the extent of suchinfections. In the case of the F1 progeny of the HT352(Tt) × (RIL F6: HT352 (N) × Borwo), some spikes were se-verely damaged by brown rust, resulting in lack of or smalldegenerated seeds with a wrinkled surface. Seeds of suchspikes or infected spikes with missing seeds were treated asmissing data. To have phenotypic data corresponding to asmany as possible RIL F6: HT352 (N) × Borwo lines, the datafrom moderately and partly infected materials with normalseeds were scored. Thus, due to brown rust infection, we havelost about 15% of phenotypic data which should not be aproblem assuming that our mapping population consisted of182 RILs.

In order to make our results comparable with other data, wehave performed normalization expressing male sterility rela-tive to the number of seeds of the most successful line.Furthermore, raw data were Bliss-transformed to fulfil re-quirements of normal distribution of the trait (Broman2001). The transformed phenotypic data followed normal dis-tribution based on Kolmogorov-Smirnov test (SupplementaryFile 1).

By now, the QTLs responsible for pollen fertility resto-ration in triticale with CMS Tt were mapped to 6RL and4RL chromosomes (Curtis and Lukaszewski 1993). Theimportance of the 6R chromosome was suggested byStojałowski (Stojałowski et al. 2013). The same authorsindicated that the trait is covered by the remaining chro-mosomes of the sixth homology group (6A and 6B). It isnoteworthy, that we did not identify a QTL for 6R that issupposed to be a good restorer for the Tt cytoplasm. Mostprobably, this QTL was eliminated during breeding pro-grams since we did not identify it using distinct parentallines in other currently running projects. The other putativeQTLs were also suggested on the 1B and 3A (3B) chromo-somes (Stojałowski et al. 2013). Interestingly, the presenceof effective restorer gene on 1BS (as well as those on 2Aand 4B) described in wheat with CMS Tt (Ahmed et al.2001; Kojima et al. 1997; Ma and Sorrells 1995) was notunivocally confirmed in triticale (Stojałowski et al. 2013).In our materials, the 1B QTL confers male sterility ratherthan pollen fertility and was detected only by CIM. Thelack of the 1BS pollen fertility QTL might be the effect ofparental forms that missed the QTL, or it was eliminatedduring selection programs. Not surprisingly, our resultscomprising composite and multiple interval mapping dem-onstrated that the trait is determined by numerous QTLslocated on both wheat and rye chromosomes. Moreover,one cannot exclude that some of the Rf genes may notrestore fertility on their own and that at least some of theQTLs on subgenomes A and B are modifiers as suggestedin common wheat (Würschum et al. 2017). All of the QTLsexplain limited percentages of phenotypic variance withthe highest values of the LOD function for the QRft-5R.2

(13.2) and QRft-5R.3 (16.2) ones. Interestingly, the QRft-5R.3 (identified both by CIM and MIM), in contrast to theQRft-5.2 (CIM and MIM), had a relatively high value ofthe heritability (h2, 11.5 vs 0.6). Moreover, the QRft-5R.3covers only 3.4 cM and has tightly linked markers, whichmay suggest this chromosome region as a putative QTL forMAS purposes. Interestingly, our analysis showed that thetrait is expressed by some QTLs located on rye genome notidentified earlier (Curtis and Lukaszewski 1993), namely,those on the 2R, 3R, and 5R chromosomes. On the otherhand, our results are congruent with previous studies(Ahmed et al. 2001; Curtis and Lukaszewski 1993;Kojima et al. 1997; Ma and Sorrells 1995; Stojałowskiet al. 2013), indicating the multigenic nature of male ste-rility and pollen fertility trait in triticale with CMS Tt. Theyalso support the notion that individual QTLs conferring thetrait explain a small fraction of phenotypic variance(Stojałowski et al. 2013). However, this may not alwaysbe the case as the QRft-5B and QRft-5R.3 explained ca15.1 and 17.8 of variance based on R2 values, respectively.It should be stressed that our study confirmed the role ofthe QTLs located on the sixth homology group as two ofthem (QRft-6B.1 and QRft-6B.2) were identified. TheQRft-6B.2 was the one with the highest heritability value(h2 = 16) but with low level of the LOD score (4.8) asindicated by MIM and CIM, respectively. Both QTLs cov-ered a small genome. Multiple interval mapping methodhas also demonstrated the presence of epistasis betweenthe QRft-3R and QRft-4B and the QRft-3R and QRft-5R.3, but the effects were weak. QTL analyses are congru-ent with phenotypic data indicating multigenic nature of atrait.

In the given study, the RIL F6: HT352 (N) × Borwo–based genetic map dedicated to the identification of pollensterility QTLs in the CMS Tt system of triticale was con-structed. The map is saturated with numerous SNP andsilicoDArT markers, and the evaluated linkage groupshave acceptable linkage disequilibrium values, whichmaps the map useful for the QTL mapping purposes. Aset of male sterility and pollen fertility QTLs was evaluatedby CIM and MIM procedures indicating the prevailing roleof the 5R (QRft-5R.2) and 6B (QRft-6.1) chromosomes inthe trait expression in triticale with CMS Tt. Moreover,some of the QTLs might be considered as the candidatesfor the MAS if SNP markers are converted to the specificPCR conditions and work on a wide range of materials.Our study is the first one to demonstrate not only that thetrait may be expressed by numerous QTLs explaining asmall fraction of the phenotypic variance of the trait butalso that some QTLs might be relatively robust. We cannotexclude, however, that the identification of such QTLs isthe kind of bias reflecting parental effect. Thus, this hy-pothesis needs to be tested on other materials.

69J Appl Genetics (2021) 62:59–71

Supplementary Information The online version contains supplementarymaterial available at https://doi.org/10.1007/s13353-020-00595-z.

Acknowledgements We would like to thank Janina Woś, the breederfrom Plant Breeding Company-Strzelce, Breeding Department Borowofor running part of the field experiments and phenotyping data collecting.

Author contribution PB designed the experiment, performed geneticmapping and QTL analysis, and wrote the manuscript; MW performedfield evaluation of RILs and phenotyping of the F1 plants, run molecularexperiments, participated in data analyses, and wrote the manuscript; HWdelivered parental forms of the RIL F6 mapping population and run partof the field experiments; AN and MP were partially responsible for thefield; and AN also proofread and edited the manuscript. All authors readand approved the final manuscript.

Funding This research was supported by TheMinistry of Agriculture andRural Development of the Republic of Poland (No. HOR hn 801-PB-13).

Compliance with ethical standards

Conflict of interest The authors declare that they have no conflict ofinterest.

Open Access This article is licensed under a Creative CommonsAttribution 4.0 International License, which permits use, sharing,adaptation, distribution and reproduction in any medium or format, aslong as you give appropriate credit to the original author(s) and thesource, provide a link to the Creative Commons licence, and indicate ifchanges weremade. The images or other third party material in this articleare included in the article's Creative Commons licence, unless indicatedotherwise in a credit line to the material. If material is not included in thearticle's Creative Commons licence and your intended use is notpermitted by statutory regulation or exceeds the permitted use, you willneed to obtain permission directly from the copyright holder. To view acopy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

References

Ahmed TA, Tsujimoto H, Sasakuma T (2001) QTL analysis of fertility-restoration against cytoplasmic male sterility in wheat. Genes GenetSyst 76:33–38

Alheit KV, Reif JC, Maurer HP, Hahn V, Miedaner T, Würschum T(2011) Detection of segregation distortion loci in triticale (xTriticosecale Wittmack) based on a high-density DArT marker con-sensus genetic linkage map. BMC Genomics 12:380

Alheit KV, Mauer HP, Reif JC, Tucker MR, Hahn V, Weissmann EA,Würschum T (2012) Genome-wide evaluation of genetic diversityand linkage disequilibrium in winter and spring triticale(×Triticosecale Wittmack). BMC Genomics 13:235

Bauer E, Schmutzer T, Barilar I, Mascher M, Gundlach H, Martis MM,Twardziok SO, Hackauf B, Gordillo A, Wilde P (2017) Towards awhole-genome sequence for rye (Secale cereale L.). Plant J 89:853–869

Bolibok-Brągoszewska H, Heller-Uszyńska K, Wenzl P, Uszyński G,Kilian A, Rakoczy-Trojanowska M (2009) DArT markers for therye genome - genetic diversity and mapping. BMC Genomics 10:578

Broman KW (2001) Review of statistical methods for QTL mapping inexperimental crosses. Lab Anim 30:44–52

Curtis CA, Lukaszewski AJ (1993) Localization of genes in rye thatrestore male fertility to hexaploid wheat with timopheevi cytoplasm.Plant Breed 111:106–112

DArT-P/L DAT (2016) A consensus map of wheat V 4.0 (“The Map).https://www.diversityarrays.com/technology-and-resources/genetic-maps. Accessed 20 June 2016

FAOSTAT (2018) http://www.fao.org/faostat/en/#data/QC. Accessed 26January 2020

Geiger H, Miedaner T (2009) Rye breeding. Cereals 3:157–181Geiger HH, Morgenstern K (1975) Applied genetic studies on cytoplas-

mic pollen sterility in winter rye. Theor Appl Genet 46:269–276Geiger H, Yuan Y, Miedaner T, Wilde P (1995) Environmental sensitiv-

ity of cytoplasmic genic male sterility (CMS) in Secale cereale L.Fortschritte der Pflanzenzuechtung 18:7–18

González JM, Muñiz LM, Jouve N (2005) Mapping of QTLs for andro-genetic response based on a molecular genetic map of ×Triticosecale Wittmack. Genome 48:999–1009

Góral H (2002) Production of triticale (X Triticosecale Wittm.) hybridseeds using the sterilizing cytoplasm of Triticum timopheevi. CerealRes Commun 30:31–38

Góral H (2004) Effect of distance from the pollen source on seed set inmale sterile triticale lines. Biuletyn Instytutu Hodowli iAklimatyzacji Roślin 231:157-164 (Abstract in English)

Góral H (2013) Male fertility of winter triticale depending on the cyto-plasm and male parent. Biuletyn Instytutu Hodowli i AklimatyzacjiRoślin 269:15-20 (Abstract in English)

Góral H, Spiss L (1998) Effect of sterile cytoplasm of T. timopheevi onagronomic traits of winter triticale hybrids. Biuletyn InstytutuHodowli i Aklimatyzacji Roślin 205/206:157–162

Góral H, Warzecha T, Stojałowski S, Pojmaj M, Kurleto D, Trąbka A,Spiss L (2006) Stability of male sterylity and fertility restoration incms T. timopheevi system in triticale. Folia UniversitatisAgriculturae Stetinensis. Agricultura 247(100):55–62

Jannink J-L, Lorenz AJ, Iwata H (2010) Genomic selection in plantbreeding: from theory to practice. Brief Funct Genomics 9:166–177

Kiss A (1971) Origin of the preliminary released Hungarian hexaploidvarieties no. 57 and 64. Wheat Info Serv 32:20–22

Kojima T, Tsujimoto H, Ogihara Y (1997) High-resolution RFLP map-ping of the fertility restoration (Rf3) gene against Triticumtimopheevi cytoplasm located on chromosome 1BS of commonwheat. Genes Genet Syst 72:353–359

Kuleung C, Baenzinger PS, Kachman SD, Dweikat I (2006) Evaluatingthe genetic diversity of triricale with wheat and rye SSR markers.Crop Sci 46:1692–1700

Li H, Hearne S, Bänziger M, Li Z, Wang J (2010) Statistical properties ofQTL linkage mapping in biparental genetic populations. Heredity105:257–267

Losert D,Maurer HP,Marulanda JJ, Würschum T (2017) Phenotypic andgenotypic analyses of diversity and breeding progress in Europeantriticale (× Triticosecale Wittmack). Plant Breed 136:18–27

Luciano Da Costa ES, Wang S, Zeng Z-B (2012) Composite intervalmapping and multiple interval mapping: procedures and guidelinesfor using Windows QTL Cartographer. Methods Mol Biol 871:75–119

Ma Z-Q, Sorrells ME (1995) Genetic analysis of fertility restoration inwheat using restriction fragment length polymorphisms. Crop Sci35:1137–1143

Maccaferri M, Ricci A, Salvi S, Milner SG, Noli E, Martelli PL, CasadioR, Akhunov E, Scalabrin S, Vendramin V, Ammar K, Blanco A,Desiderio F, Distelfeld A, Dubcovsky J, Fahima T, Faris J, Korol A,Massi A,Mastrangelo AM,MorganteM, Pozniak C, N'DiayeA, XuS, Tuberosa R (2015) A high-density, SNP-based consensus map oftetraploid wheat as a bridge to integrate durum and bread wheatgenomics and breeding. Plant Biotechnol J 13:648–663

Marone D, Laidò G, Gadaleta A, Colasuonno P, Ficco DBM, GiancasproA, Giove S, Panio G, Russo MA, De Vita P, Cattivelli L, Papa R,

70 J Appl Genetics (2021) 62:59–71

https://doi.org/10.1007/s13353-020-00595-zhttps://doi.org/https://www.diversityarrays.com/technology-and-resources/genetic-mapshttps://www.diversityarrays.com/technology-and-resources/genetic-mapshttp://www.fao.org/faostat/en/#data/QC

Blanco A, Mastrangelo AM (2012) A high-density consensus mapof a and B wheat genomes. Theor Appl Genet 125:1619–1638

Martis MM, Zhou R, Haseneyer G, Schmutzer T, Vránad J, KubalákováM, König S, Kugler KG, Scholz U, Hackauf B, Korzun V, SchöncCC, Doležel J, Bauer E, Mayer KFX, Stein N (2013) Reticulateevolution of the Rye genome. Plant Cell 25:3685–3698

Meuwissen THE, Hayes BJ, Goddard ME (2001) Prediction of totalgenetic value using genome-wide dense marker maps. Genetics157:1819–1829

Miedaner T, Glass C, Dreyer F, Wilde P, Wortmann H, Geiger HH(2000) Mapping of genes for male-fertility restoration in ‘Pampa’CMS winter rye (Secale cereale L.). Theor Appl Genet 101:1226–1233

Milczarski P, Bolibok-Bragoszewska H, Myśków B, Rakoczy-Trojanowska M (2011) A high density consensus map of Rye(Secale cereale L.) based on DArT markers. PLoS One 6:e28495

Milczarski P, Hanek M, Tyrka M, Stojałowski S (2016) The applicationof GBS markers for extending the dense genetic map of rye (Secalecereale L.) and the localization of the Rfc1 gene restoring malefertility in plants with the C source of sterility-inducing cytoplasm.J Appl Genet 57:439–451

Myer RO, Barnett RD (2000) Triticale grain in swine diets. University ofFlorida Cooperative Extension Service, Institute of Food andAgriculture Sciences, EDIS. https://ufdcimages.uflib.ufl.edu/IR/00/00/42/45/00001/AN03200.pdf

Niedziela A, Bednarek PT, Labudda M,Mankowski DR, Aniol A (2014)Genetic mapping of a 7R Al tolerance QTL in triticale (xTriticosecale Wittmack). J Appl Genet 55:1–14

Niedziela A, Mankowski D, Bednarek PT (2015) Diversity arraystechnology-based PCR markers for marker assisted selection of alu-minum tolerance in triticale (x Triticosecale Wittmack). Mol Breed35:209

Niedziela A, Orlowska R, Machczynska J, Bednarek PT (2016) Thegenetic diversity of triticale genotypes involved in polish breedingprograms. Springerplus 5:355

PenaR (2004) Food uses of triticale. In:MergoumM,Gómez-MacphersonH(eds) Triticale Improvement and Production. Food and AgricultureOrganisation of the United Nations, Rome, pp 37–48

Ronin Y, Minkov D, Mester D, Akhunov E, Korol A (2015) Buildingultra-dense genetic maps in the presence of genotyping errors andmissing data. Advances in wheat genetics: from genome to field.Springer, Tokyo, pp 127–133

Sánchez-Sevilla JF, Horvath A, BotellaMA, GastonA, Folta K, Kilian A,Denoyes B, Amaya I (2015) Diversity arrays technology (DArT)marker platforms for diversity analysis and linkage mapping in acomplex crop, the Octoploid cultivated strawberry (Fragaria ×ananassa). PLoS One 10:e0144960

Seal AG, Bennet MD (2011) The rye genome in winter hexaploid triti-cale. Genome 23:647–653

Silva Lda C, Wang S, Zeng ZB (2012) Composite interval mapping andmultiple interval mapping: procedures and guidelines for using win-dows QTL cartographer. Methods Mol Biol 871:75–119

Sokal RR, Rohlf FJ (1981) Biometry: the principles and practice of sta-tistics in biological research. W. H. Freeman, New York

Stojałowski S, Łapiński M, Masojć P (2004) RAPD markers linked withrestorer genes for the C-source of cytoplasmic male sterility in rye(Secale cereale L). Plant Breed 123:428–433

Stojałowski SA, Milczarski P, Hanek M, Bolibok-Bragoszewska H,Myśków B, Kilian A, Rakoczy-Trojanowska M (2011) DArTmarkers tightly linked with the Rfc1 gene controlling restorationof male fertility in the CMS-C system in cultivated rye (Secalecereale L.). J Appl Genet 52:313–318

Stojałowski S, Bobrowska A, Hanek M, Myśków B (2013) The impor-tance of chromosomes from the sixth homeologic group in the res-toration of male fertility in winter triticale with Triticum timopheeviicytoplasm. J Appl Genet 54:179–184

Tams SH, Melchinger AE, Bauer E (2005) Genetic similarity amongEuropean winter triticale elite germplasms assessed with AFLPand comparison with SSR and pedigree data. Plant Breed 124:154–160

Tyrka M, Bednarek PT, Kilian A, Wedzony M, Hura T, Bauer E (2011)Genetic map of triticale compiling DArT, SSR, and AFLP markers.Genome 54:391–401

TyrkaM, Tyrka D, WędzonyM (2015) Genetic map of Triticale integrat-ing microsatellite, DArT and SNP markers. PLoS One 10(12):e0145714

Veyrieras J-B, Camus-Kulandaivelu L, Gouesnard B, Manicacci D,Charcosset A (2007) Bridging genomics and genetic diversity:Linkage disequilibrium structure and association mapping in maizeand other cereals. Crop Sci 47(Suppl Dec):60–71

Wilson A (1875) On wheat and rye hybrids. Trans Proc Bot Soc 12:286–288

Würschum T, Leiser WL, Weissmann S, Maurer HP (2017) Geneticarchitecture of male fertility restoration of Triticum timopheevii cy-toplasm and fine-mapping of the major restorer locus Rf3 on chro-mosome 1B. Theor Appl Genet 130:1253–1266

www.wheatgenome.org (2018) InternationalWheat Genome SequencingConsortium. Generating a high quality genome sequence of breadwheat. Accessed 24 Sept 2018

XlStat (2019) https://www.xlstat.com/en/solutions/pre-mium. pAccessed 10 December 2018

Publisher’s note Springer Nature remains neutral with regard to jurisdic-tional claims in published maps and institutional affiliations.

71J Appl Genetics (2021) 62:59–71

https://ufdcimages.uflib.ufl.edu/IR/00/00/42/45/00001/AN03200.pdfhttps://ufdcimages.uflib.ufl.edu/IR/00/00/42/45/00001/AN03200.pdfhttps://www.xlstat.com/en/solutions/pre-mium

Genetic mapping of male sterility and pollen fertility QTLs in triticale with sterilizing Triticum timopheevii cytoplasmAbstractIntroductionMaterials and methodsPlant materialsPhenotypingDNA isolationGenotypingLinkage mapChromosomal assignment and arm orientationGenetic map comparisonComposite and multiple interval mapping

ResultsPhenotypingGenetic mapComposite and multiple interval mapping of pollen sterility

DiscussionReferences