Detection of QTLs for traits associated with pre-harvest sprouting resistance in bread wheat (Triticum aestivum L.)

Abstract

Pre-harvest sprouting (PHS) is one of the serious problems for wheat production, especially in rainy regions. Although seed dormancy is the most critical trait for PHS resistance, the control of heading time should also be considered to prevent seed maturation during unfavorable conditions. In addition, awning is known to enhance water absorption by the spike, causing PHS. In this study, we conducted QTL analysis for three PHS resistant related traits, seed dormancy, heading time and awn length, by using recombinant inbred lines from ‘Zenkouji-komugi’ (high PHS resistance) × ‘Chinese Spring’ (weak PHS resistance). QTLs for seed dormancy were detected on chromosomes 1B (QDor-1B) and 4A (QDor-4A), in addition to a QTL on chromosome 3A, which was recently cloned as TaMFT-3A. In addition, the accumulation of the QTLs and their epistatic interactions contributed significantly to a higher level of dormancy. QDor-4A is co-located with the Hooded locus for awn development. Furthermore, an effective QTL, which confers early heading by the Zenkouji-komugi allele, was detected on the short arm of chromosome 7B, where the Vrn-B3 locus is located. Understanding the genetic architecture of traits associated with PHS resistance will facilitate the marker assisted selection to breed new varieties with higher PHS resistance.

Introduction

Pre-harvest sprouting (PHS) in bread wheat (Triticum aestivum L.) is characterized by the germination of seeds in the spikes before harvest. PHS is a serious problem that decreases yields and end use quality of grains, particularly in regions with wet and humid weather conditions at the harvest season. Therefore, resistance to PHS has been one of the major breeding objectives in the many countries, including Japan.

Resistance to PHS is known to be associated with various physiological, morphological and developmental characteristics of the seed and the spike (Groos et al. 2002, King and Richards 1984, Munkvold et al. 2009). In wheat breeding programs, the most attention has been paid to selection for seed dormancy, which is one of the most critical factors for PHS resistance. The control of phenology or heading time also should be considered important to prevent seed maturation during the wet and humid season (Derera 1989, Mares and Mrva 2014). In Japan, the early heading trait is preferable because late heading results in the maturation of seeds in the rainy season. In addition, ear morphology, such as awning and ear nodding angle are associated with water absorption by the spike, affecting the degree of PHS (King 1989). The awnless characteristic causes less water uptake of the spike, resulting in decreased PHS (King and Richards 1984). The understanding of the genetic architecture of various traits associated with PHS resistance should contribute the breeding of new varieties more resistant to PHS.

Regarding the genetic control of PHS resistance, a number of studies for quantitative trait locus (QTL) analysis for seed dormancy have been done, indicating that seed dormancy is controlled by multiple genes or QTLs, which are distributed on all 21 chromosomes of the wheat genome (see review in Mares and Mrva 2014). Major genes or QTLs were reported on chromosome 2B (Chao et al. 2010, Munkvold et al. 2009, Somyong et al. 2014), 3A (Liu et al. 2013, Mori et al. 2005) and 4A (Chen et al. 2008, Mares et al. 2005, Mohan et al. 2009, Ogbonnya et al. 2008, Rasul et al. 2009, Torada et al. 2008,). Recently, one of the major QTLs on chromosome 3A, QPhs.ocs-3A.1, was molecularly characterized, revealing that a causal mutation in the promoter region of MOTHER OF FT and TFL 1 (MFT) gene affects seed dormancy (Nakamura et al. 2011). In addition, multi-allelic variation of the TaMFT-3A locus was found to cause diverse phenotype for PHS resistance (Liu et al. 2013); however, a single major gene or QTL cannot fully explain the genetic diversity of wheat varieties. In addition, seed dormancy is a complex trait, in which the effects of QTLs depend on the genetic background or gene combinations, as well as environmental conditions (Jaiswal et al. 2012, Kulwal et al. 2004, 2012). Therefore, it is important for breeding to explicitly identify the genetic network among QTLs, not only for seed dormancy, but also for various other traits associated with PHS resistance in wheat.

Previously, we reported the QTL analysis for seed dormancy using recombinant inbred lines (RILs) from a cross between ‘Zenkouji-komugi’ (Zen), a Japanese variety with higher PHS resistance, and ‘Chinese Spring’ (CS), a strain susceptible to PHS (Mori et al. 2005, Osa et al. 2003). One major QTL on chromosome 3A (QPhs.ocs-3A.1) in addition to the two minor QTLs were detected; however, the analysis was limited to only three chromosomes, 3A, 4A and 4B (Mori et al. 2005). In this study, genome-wide QTL analysis for seed dormancy, as well as awning and heading time were conducted using RILs from Zen × CS. We focused on three QTLs on chromosomes 1B, 3A and 4A and how their epistatic interactions contribute to higher PHS resistance.

Materials and Methods

Plant materials

Recombinant inbred lines (RILs) were developed form the cross between ‘Zenkouji-komugi’ (Zen) and ‘Chinese spring’ (CS) by the single-seed descent method described in Osa et al. (2003). Zen is a Japanese spring wheat variety showing an extremely high level of seed dormancy (Miura et al. 1997, Osanai and Amano 1993) and CS is a strain with weak seed dormancy (Flintham 2000, Warner et al. 2000). For the experiments 127 RILs in F₉ or later generations were used.

For the analysis of the Vrn-B3 locus, KT742, a chromosomal substitution line of the chromosome 7B of CS to that of ‘Hope’ were used. KT742 was provided by the National BioResource Project-Wheat with support in part by the National BioResource Project of the Ministry of Education, Culture, Sports, Science & Technology (MEXT) Japan.

Cultivation

RILs and their parental strains (CS and Zen) were cultivated on the experimental farm of Obihiro University of Agriculture and Veterinary Medicine, in Obihiro, Japan. Seeds were sown in late April with 10 cm between plants and 30 cm between rows, and were grown under standard field management. Twelve plants each of RIL and 72 plants for each parental strain were grown. After anthesis, the experimental plots were covered with a transparent plastic roof to prevent rain damage. Spikes were harvested at physiological maturity (loss of green color in the seeds and spikes, and the moisture content of the grains was approximately 20–22%). Harvested spikes were allowed to air-dry for about 4 days until the moisture content of the grain was approximately 14%. Dried spikes were gently hand-threshed and seeds were used for the germination test described below. Because of disease damage and poor growth, 117 out of 127 RILs were used for evaluation of seed dormancy.

For the evaluation of heading time and awn length, 127 RILs and their parental strains were grown in a plastic box (60 × 15 cm). Twelve plants per plastic box were raised. Seeds were sown at the late April and seedlings were grown in a greenhouse for 3 weeks. After that, plastic boxes were placed outside until heading. Four plants (replicates) for each RIL and 12 plants (replicates) for parental strains were raised.

Phenotypic evaluation

The degree of seed dormancy was evaluated by germination tests on petri dishes (90 mm diameter) under two germination temperatures: 15°C and 20°C. About 30–40 grains from one or two spikes were placed on a single layer of filter paper wetted with water. The dishes were incubated under darkness for 10 days. Each dish was examined daily, and grains where the seed coat was ruptured by the plumule or radicle were counted as germinated seeds. The degree of seed dormancy was checked during 10 days after incubation and expressed as a cumulative percentage of germination (GR: germination rate). The degree of seed dormancy was evaluated as follow: Seed dormancy rate (%) = 100 − GR (%). The experiment was conducted with two replicates for each RIL and six replicates for parental strains.

For the evaluation of awn length, three spikes for each RIL were harvested at maturity and the longest awn of each spike was measured.

Marker analysis and linkage map construction

The total DNA was extracted from fresh leaves collected from a single plant of each RIL. Small pieces of wheat leaf were crushed in a 2.0 ml tube, and 500 μl of the DNA extraction buffer reported by Monna et al. (2002) was added. Four hundred microliter of chloroform : isoamyl alcohol (24:1) was added, and mixed with the solution by inversion. After mixing, the tube was spun at 13000 rpm for 15 min. Crude DNA in the centrifuged supernatant was precipitated by isopropanol, and the pellet was re-dissolved in TE. For the construction of genome-wide linkage map, 15 restriction fragment length polymorphism (RFLP) markers, 158 simple sequence repeat (SSR) markers, and one derived cleaved amplified polymorphic sequence (dCAPS) marker and three gene specific markers were used. Previously, Mori et al. (2005) created a linkage map of chromosomes 3A, 4A, and 4B in the same population and the genotype data of markers were re-analysed for this study. SSR markers were selected from a public database of polymorphism of markers (http://www.shigen.nig.ac.jp/wheat/komugi/strains/aboutNbrpMarker.jsp and Somers et al. 2004). The dCAPS marker UCW99, which is known to be tightly linked to the Vrn-B3 locus on chromosome 7B (Yan et al. 2006), was also added. Three gene specific markers for TaMFT-3A, Ppd-D1 and R-B1 were used for map construction. For the TaMFT-3A loci, the cleaved amplified polymorphic marker (CAPS) marker reported in Nakamura et al. (2011) was used. For the series of Ppd-1 loci, the allelic difference between CS and Zen was reported at the Ppd-1B and Ppd-1D loci (Nishida et al. 2013, Tanio and Kato 2007). A polymorphism was found for the marker of the Ppd-D1 locus developed by Tanio and Kato (2007) and used for the map construction. Red grain color is controlled by the three dominant R-1 genes located on the long arm of chromosomes 3A (R-A1), 3B (R-B1), and 3D (R-D1) and they were cloned as Tamyb10-A1, Tamyb10-B1, and Tamyb10-D1 genes, respectively (Himi et al. 2011). Both Zen and CS showed red grains and the genotypes of Zen and CS were assumed to be R-A1a R-B1b R-D1b and R-A1a R-B1a R-D1b, respectively (Himi et al. 2011, Miura et al. 2002). The Tamyb10-B1 (R-B1) marker developed by Himi et al. (2011) was added.

The linkage map was constructed from marker genotypes of 127 RILs by using JoinMap^® 4 (van Ooijen 2006). The map distance in centimorgans (cM) was determined using the Kosambi map function (Kosambi 1943).

The insertion of retrotransposon in the promoter region of the Vrn-B3 (TaFT1), was investigated based on Yan et al. (2006). Three primer-pairs developed by Yan et al. (2006), FT-B-INS-F (CATAATGCCAAGCCGGTGAGTAC) and FT-B-INS-R (ATGTCTGCCAATTAGCTAGC), FT-B-NOINS-F (ATGCTTTCGCTTGCCATCC) and FT-B-NOINS-R (CTATCCCTACCGGCCATTAG), FT-B-NOINS-F2 (GCTGTGTGATCTTGCTCTCC) and FT-B-NOINS-R, were used for the analysis.

QTL analysis

QTL analysis was conducted by simple interval mapping and MQM (multiple QTL model) mapping methods by using MapQTL^® 6 (van Ooijen 2002). In the first step, putative QTLs were identified by interval mapping. Then, one marker at each putative QTL was selected as a cofactor, and the selected markers were used as genetic background controls in an approximate multiple QTL model of Map QTL. The markers closest to the significant QTL, which maximized the LOD score, were finally selected as cofactors. The genome-wide LOD thresholds for a significant QTL were estimated by performing a permutation test for each trait (Churchill and Doerge 1994) in MapQTL. The each trait data of RILs were permuted 1000 times over the genotypes, and the empirical LOD thresholds corresponding to the genome-wide significance at 5% were estimated for the all traits examined. The additive effect and percentage of phenotypic variance explained by each QTL were obtained using MapQTL in the final multiple QTL model in which one cofactor marker was fixed per QTL.

Data analysis

For the detection of epistatic interactions among the three QTLs for seed dormancy, the eight classes of genotypes consisting of different combinations of alleles were analyzed by three-way analysis of variance (ANOVA). The differences among genotypes were tested by Tukey-Kramer test for multiple comparisons. All statistical analyses were conducted using PASW^® statistics 18 (SPSS Inc.). For three-way ANOVA and QTL analyses, the data of germination rate were arcsin-transformed to improve normality and heteroscedasticity.

Results

Segregation of three traits associated with PHS resistance

Three traits associated with PHS resistance (seed dormancy, heading time and awn length) were evaluated in RILs (Fig. 1). Seed dormancy rate (Dor) was examined under two temperature conditions (15°C and 20°C). The Dor of Zen was significantly higher than that of CS and RILs had a continuous frequency distribution of Dor under both conditions, whereas Dor at 20°C in the RILs and their parental strains was higher than at 15°C. Significant transgressive segregations in the RILs were observed in both directions. Heading time of Zen was about one week earlier than CS. The RILs showed continuous frequency distributions between parental values, although a small part of the RILs transgressed the parental strains. Regarding awn length, both parental strains either had no awns or very short ones; however, segregants showing a long awn phenotype appeared in the RILs.

Fig. 1

Frequency distributions for three traits in RILs. (A) Seed dormancy rate at 20°C, (B) Seed dormancy rate at 15°C, (C) Days to heading, (D) Awn length. Black and white arrows indicate the mean values of Zen and CS, respectively.

Phenotypic correlations among traits are shown in Table 1. A high positive correlation was observed between the Dor at 15°C and 20°C. The Dor and awn length did not correlate with heading time (days to heading), indicating an independent genetic base. In contrast, significant positive correlations were observed between Dor (both at 15°C and 20°C) and awn length while correlation coefficients were relatively low, indicating a partial genetic relationship between the two traits.

Table 1 Correlation coeffcients among the traits associated with PHS resistance in the RILs

	Dor 15	Awn	Ht
Dor 20	0.69 ***	0.25 **	−0.14
Dor 15		0.29 **	−0.21
Awn			−0.17

Ht: Heading time (Days to heading), Awn: Awn length, Dor 20: Seed dormancy rate at 20°C, Dor 15: Seed dormancy rate at 15°C.

** and ***  indicate the significance at the 1 and 0.1% levels, respectively.

QTL analysis

To understand the genetic control for traits associated with PHS resistance, QTL analysis for three traits (seed dormancy, heading time and awn length) was conducted. A genome-wide linkage map was constructed on the basis of genotypes of 127 RILs. A total of 177 polymorphic marker loci consisting of 158 SSR, 15 RFLP, one dCAPS and three gene specific markers, were distributed on 21 chromosomes, which covered the total chromosome of the wheat genome; however, markers were scarce in some regions (Fig. 2).

Fig. 2

Linkage map showing the positions of putative QTLs detected in the RILs. SSR and gene specific markers are shown on the left of chromosomes. QTL positions are represented as boxes with two-LOD support intervals with the LOD peaks indicated by arrowheads. Filled and white boxes indicate the effects of the QTLs to increase the trait values of the Zen and CS alleles, respectively.

Three significant QTLs (QDor-1B, QDor-3A and QDor-4A) were detected for seed dormancy (Fig. 2, Table 2). QDor-3A and QDor-4A were detected under both temperature conditions, whereas QDor-1B had the effect only at 20°C. QDor-3A and QDor-4A had relatively large effects in conferring high seed dormancy (decreasing the GR) by the Zen allele. In contrast, QDor-1B had the opposite effect (conferring high seed dormancy by the CS allele). A QTL with the large effect on heading time was detected on chromosome 7B (QHt-7B), which conferred early heading by the Zen allele (Fig. 2, Table 2). Two QTLs for awn length (QAwn-4A and QAwn-6B) were found on chromosome 4A and 6B (Fig. 2, Table 2). QAwn-4A and QAwn-6B suppressed awn development by the CS allele.

Table 2 Results of MQM mapping for three traits assoicated with PHS resistance in the RILs

Trait	QTL	Nearest marker	LOD	PVE (%)	Add
Seed Dormancy (20°C)
	QDor-1B	barc181	3.0	8.4	4.5
	QDor-3A	TaMFT	6.3	18.9	−6.2
	QDor-4A	cfa2256	5.0	14.3	−5.7
Seed Dormancy (15°C)
	QDor-3A	TaMFT	4.6	15.3	−7.3
	QDor-4A	cfa2256	3.8	12.4	−7.1
Heading time
	QHt-7B	UCW99	12.5	36.5	5.7
Awn length
	QAwn-4A	cfa2256	7.8	19.8	−1.1
	QAwn-6B	wmc397	5.8	14.6	−1.0

Nearest markers, LOD values, PVE (phenotypic variance explained by each QTL), and Add (additive effect) are listed. Positive and negative values of Add indicate the effects increasing trait values by the CS and Zen alleles, respectively.

Epistatic interaction among QTLs for seed dormancy

To elucidate the genetic interaction among QTLs for seed dormancy, RILs were classified into eight genotypes having different combinations of alleles at three QTLs (Fig. 3). The alleles of QTLs were estimated by two flanking markers for QDor-1B (barc137 and barc181) and QDor-4A (cfa2256 and gpw2271), and a CAPS marker for the TaMFT-3A gene. The seed dormancy rate (Dor) was compared among genotypes. Three-way ANOVA indicated that all the main effects of the three QTLs were significant in both condition (15°C and 20°C) (Table 3). The effect of QDor-1B was not significant at 15°C in QTL analysis, which might be due to the higher resolution of ANOVA. The interactions among the two (QDor-1B × QDor-4A) and three QTLs (QDor-1B × QDor-3A × QDor-4A) at the 20°C condition were significant (P = 0.006 and P = 0.001, respectively) (Table 3), indicating epistatic interactions among QTLs for seed dormancy. The Dor of the Type II–IV having one of the dormant-alleles at three QTLs (CS allele for QDor-1B, and Zen allele for QDor-3A and QDor-4A) showed no significant differences from the Type I having no dormant-alleles at all; however, RILs having the combinations of two (Type V–VII at 20°C and Type VII at 15°C) or three (Type VIII) dormant-alleles showed significantly higher Dor than Type I (Fig. 3).

Fig. 3

Interactions among three QTLs for seed dormancy on chromosomes 1B (QDor-1B), 3A (QDor-3A) and 4A (QDor-4A). (A) Seed dormancy rate at 20°C, (B) Seed dormancy rate at 15°C. The alleles of the QTLs were estimated by using their flanking markers (see text). C and Z indicate the CS and Zen alleles, respectively, and the alleles with underline indicate the dormant-alleles. The bar plots show seed dormancy rate (%) ± SD. Different letters above the bars indicate a significant difference by Tukey-Kramer test.

Table 3 Three-way ANOVA for the effects of three QTLs for seed dormancy in the RILs

Factor	d.f	Dor 20			Dor 15
		MS	F-value	P	MS	F-value	P
QDor-1B	1	2425.70	10.92	0.001 **	2653.72	5.50	0.022 *
QDor-3A	1	7719.48	34.74	0.000 ***	6823.33	14.14	0.000 ***
QDor-4A	1	5905.99	26.58	0.000 ***	6940.32	14.39	0.000 ***
QDor-1B × QDor-3A	1	15.13	0.07	0.795 ns	3.50	0.01	0.932 ns
QDor-1B × QDor-4A	1	1765.83	7.95	0.006 **	485.72	1.01	0.319 ns
QDor-3A × QDor-4A	1	77.91	0.35	0.556 ns	69.33	0.14	0.706 ns
QDor-1B × QDor-3A × QDor-4A	1	2803.24	12.62	0.001 ***	1266.70	2.63	0.109 ns
Error	75	222.19			482.41

Dor 20: Seed dormancy rate at 20°C, Dor 15: Seed dormancy rate at 15°C.

*, ** and ***  indicate the significance at the 5, 1 and 0.1% levels, respectively.

Comparison of the chromosomal locations among QTLs and genes

The chromosomal locations of the QTLs detected in this study were compared with previously reported QTLs or genes. The location of QDor-3A was in the region including TaMFT-3A (Nakamura et al. 2011), which is the causal gene of QPhs.ocs-3A.1 found in the same mapping population by Mori et al. (2005). On chromosome 1B, a few QTLs have been reported, so far (Fig. 4A). On the chromosome 4A, a number of QTLs for seed dormancy were detected in the region including the centromere and on the long arm of the chromosome (Fig. 4B). A major gene (Phs-1; Torada et al. 2008) and QTLs with large effects (Chen et al. 2008, Mares et al. 2005, Mohan et al. 2009, Ogbonnaya et al. 2008, Tan et al. 2006) were frequently found in the region between Xbarc170 and Xgwm397, indicating that QDor-4A should be a different locus.

Fig. 4

Comparative maps of locations of QTLs or the gene on chromosomes 1B (A) and 4A (B). The linkage map constructed in this study is shown on the left and the framework map based on Somers et al. (2004) is shown on the right. QTLs detected in this study are shown on the left linkage map as boxes with two-LOD support intervals with the LOD peaks indicated by arrowheads. The locations of reported QTLs or genes are shown on the right linkage map. Bars represent the regions including the LOD peaks of QTLs. a) Flintham et al. (2002), b) Singh et al. (2014), c) Zanetti et al. (2000), d) Torada et al. (2008), e) Sourdille et al. (2002), f) Chen et al. (2008), g) Imtiaz et al. (2008), h) Kato et al. (2001), i) Mares and Mrva (2001), j) Mares et al. (2005), k) Mohan et al. (2009), l) Ogbonnaya et al. (2008),m) Rasul et al. (2009), n) Tan et al. (2006). A shaded zone on each chromosome indicates the approximate position of the centromere based on Somers et al. (2004).

The LOD peak of QHt-7B was detected around marker UCW99 and Yan et al. (2006) found it to be closely linked to the Vernalization-B3 (Vrn-B3) locus (Fig. 5A). The Vrn-B3 locus was cloned as the FLOWERING LOCUS T1 (FT1) gene (TaFT1) and the insertion of retrotransposon in the promoter region was found to cause early heading in ‘Hope’ whereas there is no insertion in CS (Yan et al. 2006). The insertion of retrotransposon was examined based on the analysis of the PCR-product size by using primers in Yan et al. (2006) (Fig. 5B). For the primer pairs FT-B-INS-F and FT-B-INS-R, a 1.2-kb fragment is amplified when the insertion is present, and no fragment is found when the insertion is absent (Yan et al. 2006). In contrast, for the primers pairs FT-B-NOINS-F and FT-B-NOINS-R or FT-B-NOINS-F2 and FT-B-NOINS-R, the fragments of 1,140 bp or 691 bp are amplified, respectively, when the insertion is absent, and no fragment is found when the insertion is present (Yan et al. 2006). The results indicate that Zen has no-insertion of retrotransposon as shown in CS.

Fig. 5

(A) Comparative maps of locations of QTLs or the gene on the chromosome 7B. The linkage map constructed in this study is shown on the left and the framework map based on Somers et al. (2004) is shown in the center. The linkage map including the Vrn-B3 (TaFT1) locus (Yan et al. 2006) is shown on the right. The QTL detected in this study is shown on the left linkage map as a box with two-LOD support interval with the LOD peak indicated by an arrowhead. The locations of reported QTLs are shown on the central linkage map. Bars represent the regions including the LOD peaks of QTLs. a) Bennett et al. (2012), b) Griffiths et al. (2009), c) Hanocq et al. (2007), d) Khlestkina et al. (2009), e) Kulwal et al. (2003), f) Lin et al. (2008), g) Maccaferri et al. (2008). h) Manickavelu et al. (2011). i) Sourdille et al. (2003). A shaded zone on the chromosome indicates the approximate position of the centromere based on Somers et al. (2004). (B) PCR analysis for detection of the insertion of retrotransposon in the promotor region of the Vrn-B3 (TaFT1) gene based on Yan et al. (2006). PCR products in K742, CS and Zen by using primer pairs (a) FT-B-INS-F and FT-B-INS-R, (b) FT-B-NOINS-F and FT-B-NOINS-R, and (c) FT-B-NOINS-F2 and FT-B-NOINS-R (see text).

Regarding awn development, three dominant inhibitor gene, Hd (Hooded) on chromosome 4A, B1 on chromosome 5A, and B2 on chromosome 6B, were reported (McIntosh et al. 2013). CS and Zen show awnless and very-short awn phenotypes, respectively, which suggests that both strains have inhibitor gene(s). CS is known to have Hd and B2 (Sourdille et al. 2002), and based on the location and the direction of effects, two QTLs detected in this study assumed to correspond to Hd and B2 carried by CS. Although no significant QTL was found on chromosome 5A, a weak LOD score (LOD = 2.3) detected in the terminal region of chromosome 5AL suggests the Zen allele decreases awn length, and might correspond to the B1 gene carried by Zen. The QDor-4A was co-located with QAwn-4A (Fig. 4B).

Discussion

Pre-harvest sprouting (PHS) resistance is a complex trait involving various physiological, morphological and developmental characteristics in cereal crops (Mares and Mrva 2014). Seed dormancy is the most critical factor for PHS resistance in wheat. Previously, Mori et al. (2005) detected a major QTL for seed dormancy on chromosome 3A (QPhs.ocs-3A.1) with two minor QTLs on chromosome 4A and 4B in RILs from the cross between Zen × CS, although QTL analysis was limited on the three chromosomes (chromosome 3A, 4A and 4B). In this study, we constructed a linkage map of 21 chromosomes and did a genome-wide QTL analysis using the same mapping population. In addition, we also conducted a QTL analysis for awn length and heading time, which could be the traits associated with PHS resistance. The understanding of the genetic architecture not only of seed dormancy but also of other traits associated with PHS resistance will contribute to breeding new varieties with higher PHS resistance.

The ability to synchronize the heading time with favorable environmental conditions is the one of the key factors for PHS resistance as well as creating regional adaptability (Kamran et al. 2014). In Japan, harvesting before the rainy season is necessary to prevent PHS. Allelic variation at the Photoperiod-1 (Ppd-1) locus on the group 2 chromosomes and the Vernalization-1 (Vrn-1) locus on the group 5 chromosomes are the major sources for the variation of heading time in diverse germplasm in bread wheat (Kamran et al. 2014). The genotype at the Ppd-1 series loci of Zen and CS are Ppd-A1b Ppd-B1a Ppd-D1b and Ppd-A1b Ppd-B1b Ppd-D1a, respectively (Nishida et al. 2013, Tanio and Kato 2007), whereas Zen and CS have the same alleles (vrn-A1 vrn-B1 Vrn-D1) at the Vrn-1 series loci (Bentley et al. 2013, Tanio et al. 2005). However, no QTL was detected on chromosomes 2B and 2D under the present environmental conditions, despite the allelic differences at the Ppd-B1 and Ppd-D1 loci between Zen and CS. Instead, a major QTL for heading time was detected on chromosome 7B (QHt-7B), which causes early heading by the Zen allele. QHt-7B was found in the region where Vrn-B3 is known to be located. Yan et al. (2006) reported that the causal gene of the Vrn-B3 locus is TaFT1, which encodes the protein known as florigen in plants. The insertion of retrotransposon in the promoter region of TaFT1 causes early heading in ‘Hope’ whereas there is no insertion in CS (Yan et al. 2006). In hexaploid wheat strains surveyed, the insertion of retrotransposon has been found in only a few strains, so far (Chen et al. 2013, Yan et al. 2006). Our results indicate that Zen also has no-insertion, in spite of the Zen allele at QHt-7B conferring earlier heading compared to the CS allele. A number of QTLs for traits associated with heading time, such as vernalization and photoperiod responses, were reported on the short arm of chromosome 7B (see review in Kamran et al. 2014, Fig. 5), whereas it is still controversial whether these QTLs were caused by the same gene or multiple genes. Regarding QHt-7B, there are two possibilities that QHt-7B is a novel allele at the Vrn-B3 locus or different locus linked to Vrn-B3. Further fine-mapping is needed to delimitate the location of QHt-7B.

The retention of water in the spike is one of the factors causing PHS and awns trap more water and thereby accelerate PHS as well as other disease problems (King and Richards 1984). The awnless characteristic may cause less water absorption of the spike, resulting in a decrease in PHS (King 1989, King and Richards 1984, Mares and Mrva 2014). In addition, awns sometimes impede harvesting and processing. However, awns may enhance spike photosynthesis and provide a source of assimilates to developing seeds (Maydup et al. 2014). Three dominant inhibitors for awn development, Hooded (Hd), B1 and B2 on chromosomes 4A, 5A and 6B, respectively, have been well analyzed in wheat (McIntosh et al. 2013). From the results of this study, the genotypes of CS and Zen were tentatively assumed to be Hd b1 B2 and hd B1 b2, respectively, whereas allelic states of B1 locus are uncertain because of multiple allelic variation at the B1 locus (Watkins and Ellerton 1940). In Japan, awn length has not been the target of intensive selection during the recent breeding program of wheat; however, caution is needed in the breeding of awnless varieties with high PHS resistance because QDor-4A was colocated with QAwn-4A (Hd locus) and the dormant-allele at QDor-4A might be closely linked to the allele enhancing awn development (hd gene) in Zen. The dissection of the QDor-4A-QAwn-4A regions is needed to utilize the useful combinations of alleles for breeding, whereas there remains a possibility that the awning gene pleiotropically affects seed dormancy.

Recent progress in our understanding of genes associated with seed dormancy in various plant species reveals that seed dormancy is a complex phenomenon involving various physiological and molecular mechanisms (Gao and Ayele 2014, Graeber et al. 2012). In natural variation for the levels of seed dormancy in wheat, a number of QTLs were reported in all 21 chromosomes of the genome. Among them, QPhs.ocs-3A.1, a QTL detected on chromosome 3A (the same locus as QDor-3A in this study), was molecularly characterized as TaMFT-3A (Nakamura et al. 2011). The MFT gene is one of the key factors in the genetic control for seed dormancy in plants (Gao et al. 2014, Graeber et al. 2012) and multi-allelic variation at the TaMFT-3A locus was revealed to cause diverse phenotype of seed dormancy (Liu et al. 2013, Nakamura et al. 2011). Although Zen has the strongest allele for seed dormancy reported at the TaMFT-3A locus, so far (Liu et al. 2013, Nakamura et al. 2011), the TaMFT-3A locus could explain only the small part of the phenotypic variation observed in RILs. The present study detected two other QTLs on chromosomes 4A and 1B, and high level of seed dormancy was achieved by the accumulation of dormant-alleles at these three QTLs.

Epistasis, which describes the non-additivity of effects between loci, has been known to play important roles in phenotypic variation in plants including crop species (Cooper et al. 2009, Lynch and Walsh 1998). Regarding PHS resistance, some evidences of epistatic interactions among QTLs have been reported in wheat (Imtiaz et al. 2008, Jaiswal et al. 2012, Kulwal et al. 2004, 2012); however, it remains unclear to what extent the epitasis contributes the diversification of PHS resistance. Present studies have detected the epistatic interaction among three QTLs, QDor-3A (TaMFT-3A), QDor-4A and QDor-1B, which suggests that it is important to introduce useful QTLs, combined with proper genes and genetic background, to breed varieties with higher PHS resistance. Although the epistatic interactions among the QTLs detected in this study do not mean interactions at the molecular and physiological levels, these results imply a complex gene network for seed dormancy in wheat. Further studies to the delimitate the chromosomal location and map-base cloning of QDor-4A and QDor-1B will provide new insights for the genetic control of seed dormancy in wheat, which will facilitate the marker assisted selection to breed new varieties with higher PHS resistance.

Acknowledgement

We thank the Japanese National BioResource Project-wheat (NBRP-wheat) for providing the information of markers used in this study. This work was supported by a grant from the Ministry of Agriculture, Forestry, and Fisheries of Japan (Genomics-based Technology for Agricultural Improvement, TRS1001).

Literature Cited

•  Bennett,  D.,  A.  Izanloo,  J.  Edwards,  H.  Kuchel,  K.  Chalmers,  M.  Tester,  M.  Reynolds,  T.  Schnurbusch and  P.  Langridge (2012) Identification of novel quantitative trait loci for days to ear emergence and flag leaf glaucousness in a bread wheat (Triticum aestivum L.) population adapted to southern Australian conditions. Theor. Appl. Genet. 124: 697–711.
•  Bentley,  A.R.,  R.  Horsnell,  C.P.  Werner,  A.S.  Turner,  G.A.  Rose,  C.  Bedard,  P.  Howell,  E.P.  Wilhelm,  I.J.  Mackay,  R.M.  Howells et al. (2013) Short, natural, and extended photoperiod response in BC₂F₄ lines of bread wheat with different Photoperiod-1 (Ppd-1) alleles. J. Exp. Bot. 64: 1783–1793.
•  Chao,  S.,  S.S.  Xu,  E.M.  Elias,  J.D.  Faris and  M.E.  Sorrells (2010) Identification of chromosome locations of genes affecting preharvest sprouting and seed dormancy using chromosome substitution lines in tetraploid wheat (Triticum turgidum L.). Crop Sci. 50: 1180–1187.
•  Chen,  C.,  S.  Cai and  G.  Bai (2008) A major QTL controlling seed dormancy and pre-harvest sprouting resistance on chromosome 4A in a Chinese wheat landrace. Mol. Breed. 21: 351–358.
•  Chen,  F.,  M.  Gao,  J.  Zhang,  A.  Zuo,  X.  Shang and  D.  Cui (2013) Molecular characterization of vernalization and response genes in bread wheat from the Yellow and Huai valley of China. BMC Plant Biol. 13: 199.
•  Churchill,  G.A. and  R.W.  Doerge (1994) Empirical threshold values for quantitative trait mapping. Genetics 138: 963–971.
•  Cooper,  M.,  F.A.  van Eeuwijk,  G.L.  Hammer,  D.W.  Podlich and  C.  Messina (2009) Modeling QTL for complex traits: detection and context for plant breeding. Curr. Opin. Plant Biol. 12: 231–240.
•  Derera,  N.F. (1989) Breeding for preharvest sprouting tolerance, In:  Derera,  N.F. (ed.) Preharvest field sprouting in cereals, CRC Press, Inc. Boca Raton, FL, USA, pp. 111–128.
•  Flintham,  J.E. (2000) Different genetic components control coat-imposed and embryo-imposed dormancy in wheat. Seed Sci. Res. 10: 43–50.
•  Flintham,  J.,  R.  Adlam,  M.  Bassoi,  M.  Holdsworth and  M.  Gale (2002) Mapping genes for resistance to sprouting damage in wheat. Euphytica 126: 39–45.
•  Gao,  F. and  B.T.  Ayele (2014) Functional genomics of seed dormancy in wheat: advances and prospects. Fron. Plant Sci. 5: 458–469.
•  Graeber,  K.,  K.  Nakabayashi,  E.  Miatton,  G.L.  Metzger and  W.J.  Soppe (2012) Molecular mechanisms of seed dormancy. Plant Cell Environ. 35: 1769–1786.
•  Griffiths,  S.,  J.  Simmonds,  M.  Leverington,  Y.  Wang,  L.  Fish,  L.  Sayers,  L.  Alibert,  S.  Orford,  L.  Wingen,  L.  Herry et al. (2009) Meta-QTL analysis of the genetic control of ear emergence in elite European winter wheat germplasm. Theor. Appl. Genet. 119: 383–395.
•  Groos,  C.,  G.  Gay,  M.R.  Perretant,  L.  Gervais,  M.  Bernard,  F.  Dedryver and  G.  Charmet (2002) Study of the relationship between pre-harvest sprouting and grain color by quantitative trait loci analysis in a white × red grain bread-wheat cross. Theor. Appl. Genet. 104: 39–47.
•  Hanocq,  E.,  A.  Laperche,  O.  Jaminon,  A.-L.  Lainé and  J.  Le Gouis (2007) Most significant genome regions involved in the control of earliness traits in bread wheat, as revealed by QTL meta-analysis. Theor. Appl. Genet. 114: 569–584.
•  Himi,  E.,  M.  Maekawa,  H.  Miura and  K.  Noda (2011) Development of PCR marker for Tamyb10 related to R-1, red grain color gene in wheat. Theor. Appl. Genet. 122: 1561–1576.
•  Imtiaz,  M.,  F.C.  Ogbonnaya,  J.  Oman and  M.  van Ginkel (2008) Characterization of quantitative trait loci controlling genetic variation for preharvest sprouting in synthetic backcross-derived wheat lines. Genetics 178: 1725–1736.
•  Jaiswal,  V.,  R.R.  Mir,  A.  Mohan,  H.S.  Balyan and  P.K.  Gupta (2012) Association mapping for pre-harvest sprouting tolerance in common wheat (Triticum aestivum L.). Euphytica 188: 89–102.
•  Kamran,  A.,  M.  Iqbal and  D.  Spaner (2014) Flowering time in wheat (Triticum aestivum L.): a key factor for global adaptability. Euphytica 197: 1–26.
•  Kato,  K.,  W.  Nakamura,  T.  Tabiki,  H.  Miura and  S.  Sawada (2001) Detection of loci controlling seed dormancy on group 4 chromosomes of wheat and comparative mapping with rice and barley genomes. Theor. Appl. Genet. 102: 980–985.
•  Khlestkina,  E.K.,  A.  Giura,  M.S.  Röder and  A.  Börner (2009) A new gene controlling the flowering response to photoperiod in wheat. Euphytica 165: 579–585.
•  King,  R.W. and  R.A.  Richards (1984) Water uptake in relation to pre-harvest sprouting damage in wheat: ear characteristics. Aust. J. Agric. Res. 35: 327–336.
•  King,  R.W. (1989) Physiology of sprouting resistance, In:  Derera,  N.F. (ed.) Preharvest field sprouting in cereals, CRC Press, Inc. Boca Raton, FL, USA, pp. 27–60.
•  Kosambi,  D.D. (1943) The estimation of map distances from the recombination values. Ann. Eugen. 12: 172–175.
•  Kulwal,  P.L.,  J.K.  Roy,  H.S.  Balyan and  P.K.  Gupta (2003) QTL mapping for growth and leaf characters in bread wheat. Plant Sci. 164: 267–277.
•  Kulwal,  P.L.,  R.  Singh,  H.S.  Balyan and  P.K.  Gupta (2004) Genetic basis of pre-harvest spouting tolerance using single-locus and two-locus QTL analyses in bread wheat. Funct. Integr. Genomics 4: 94–101.
•  Kulwal,  P.,  G.  Ishikawa,  D.  Benscher,  Z.  Feng,  L.X.  Yu,  A.  Jadhav,  S.  Mehetre and  M.E.  Sorrells (2012) Association mapping for pre-harvest sprouting resistance in white winter wheat. Theor. Appl. Genet. 125: 793–805.
•  Lin,  F.,  S.L.  Xue,  D.G.  Tian,  C.J.  Li,  Y.  Cao,  Z.Z.  Zhang,  C.Q.  Zhang and  Z.Q.  Ma (2008) Mapping chromosomal regions affecting flowering time in a spring wheat RIL population. Euphytica 164: 769–777.
•  Liu,  S.,  S.K.  Sehgal,  J.  Li,  M.  Lin,  H.N.  Trick,  J.  Yu,  B.S.  Gill and  G.  Bai (2013) Cloning and characterization of a critical regulator for preharvest sprouting in wheat. Genetics 195: 263–273.
•  Lynch,  M. and  B.  Walsh (1998) Genetics and analysis of quantitative traits. Sinauer Associates, Inc., Sunderland, MA, USA, p. 980.
•  Maccaferri,  M.,  M.C.  Sanguineti,  S.  Conrneti,  J.L.A.  Ortega,  M.B.  Salem,  J.  Bort,  E.  DeAmbrogio,  L.F.  del Moral,  A.  Demontis,  A.  El-Ahmed et al. (2008) Quantitative trait loci grain yield and adaptation of durum wheat (Triticum durum Desf.) across a wide range of water availability. Genetics 178: 489–511.
•  Manickavelu,  A.,  K.  Kawaura,  H.  Imamura,  M.  Mori and  Y.  Ogihara (2011) Molecular mapping of quantitative trait loci for domestication traits and β-glucan content in a wheat recombinant inbred line population. Euphytica 177: 179–190.
•  Mares,  D.J. and  K.  Mrva (2001) Mapping quantitative trait loci associated with variation in grain dormancy in Australian wheat. Crop Pas. Sci. 52: 1257–1265.
•  Mares,  D.,  K.  Mrva,  J.  Cheong,  K.  Williams,  B.  Watson,  E.  Storlie,  M.  Sutherland and  Y.  Zou (2005) A QTL located on chromosome 4A associated with dormancy in white- and red-grained wheats of diverse origin. Theor. Appl. Genet. 111: 1357–1364.
•  Mares,  D.J. and  K.  Mrva (2014) Wheat grain preharvest sprouting and late maturity alpha-amylase. Planta 204: 1167–1178.
•  Maydup,  M.L.,  M.  Antonietta,  C.  Graciano,  J.J.  Guiamet and  E.A.  Tambussi (2014) The contribution of the awns of bread wheat (Triticum aestivum L.) to grain filling: Responses to water deficit and the effects of awns on ear temperature and hydraulic conductance. Field Crops Res. 167: 102–111.
•  McIntosh,  R.A.,  J.  Dubcovsky,  W.J.  Rogers,  C.  Morris,  R.  Appels and  X.C.  Xia (2013) Catalogue of gene symbols for wheat. Proc. 12th Int. Wheat Genet. Symp.
•  Miura,  H.,  Y.  Fukuda and  S.  Sawada (1997) Expression of seed dormancy in diallel F₁ and F₂ seed of wheat ripened under controlled environment. J. Genet. Breed. 51: 195–200.
•  Miura,  H.,  N.  Sato,  K.  Kato,  Y.  Amano and  R.A.  Mcintosh (2002) Detection of chromosomes carrying genes for seed dormancy of wheat using the backcross reciprocal monosomic method. Plant Breed. 121: 394–399.
•  Mohan,  A.,  P.  Kulwal,  R.  Singh,  V.  Kumar,  R.R.  Mir,  J.  Kumar,  M.  Prasad,  H.S.  Balyan and  P.K.  Gupta (2009) Genome-wide QTL analysis for preharvest sprouting tolerance in bread wheat. Euphytica 168: 319–329.
•  Monna,  L.,  H.X.  Lin,  S.  Kojima,  T.  Sasaki and  M.  Yano (2002) Genetic dissection of a genomic region for a quantitative trait locus, Hd3, into two loci, Hd3a and Hd3b, controlling heading date in rice. Theor. Appl. Genet. 104: 772–778.
•  Mori,  M.,  N.  Uchino,  M.  Chono,  K.  Kato and  H.  Miura (2005) Mapping QTLs for grain dormancy on wheat chromosome 3A and the group 4 chromosomes, and their combined effect. Theor. Appl. Genet. 110: 1315–1323.
•  Munkvold,  J.D.,  J.  Tanaka,  D.  Benscher and  M.E.  Sorrells (2009) Mapping quantitative trait loci for preharvest sprouting resistance in white wheat. Theor. Appl. Genet. 119: 1223–1235.
•  Nakamura,  S.,  F.  Abe,  H.  Kawahigashi,  K.  Nakazono,  A.  Tagiri,  T.  Matsumoto,  S.  Utsugi,  T.  Ogawa,  H.  Handa,  H.  Ishida et al. (2011) A wheat homolog of MOTHER OF FT AND TFL1 acts in the regulation of germination. Plant Cell 23: 3215–3229.
•  Nishida,  H.,  T.  Yoshida,  K.  Kawakami,  M.  Fujita,  B.  Long,  Y.  Akashi,  D.A.  Lauri and  K.  Kato (2013) Structural variation in the 5′ upstream region of photoperiod-insensitive alleles Ppd-A1a and Ppd-B1a identified in hexaploid wheat (Triticum aestivum L.), and their effect on heading time. Mol. Breed. 31: 27–37.
•  Ogbonnaya,  F.C.,  M.  Imtiaz,  G.  Ye,  P.R.  Hearnden,  E.  Hernandez,  R.F.  Eastwood,  M.  van Ginkel,  S.C.  Shorter and  J.M.  Winchester (2008) Genetic and QTL analyses of seed dormancy and preharvest sprouting resistance in the wheat germplasm CN10955. Theor. Appl. Genet. 116: 891–902.
•  Osa,  M.,  K.  Kato,  M.  Mori,  C.  Shindo,  A.  Torada and  H.  Miura (2003) Mapping QTLs for seed dormancy and the Vp1 homologue on chromosome 3A in wheat. Theor. Appl. Genet. 106: 1491–1496.
•  Osanai,  S. and  Y.  Amano (1993) Selection for tolerance to low temperature germinability. In:  Walker-simmons,  M.K. and  J.L.  Ried (eds.) Per-harvest sprouting in cereals 1992, American Association of Cereal Chemistry, Minnesota, USA, pp. 76–83.
•  Rasul,  G.,  D.G.  Humphreys,  A.  Brûlé-Babel,  C.A.  McCartney,  R.E.  Knox,  R.M.  DePauw and  D.J.  Somers (2009) Mapping QTLs for pre-harvest sprouting traits in the spring wheat cross ‘RL4452/AC Domain’. Euphytica 168: 363–378.
•  Singh,  A.K.,  R.E.  Knox,  J.M.  Clarke,  F.R.  Clarke,  A.  Singh,  R.M.  DePauw and  R.D.  Cuthbert (2014) Genetics of pre-harvest sprouting resistance in a cross of Canadian adapted durum wheat genotypes. Mol. Breed. 33: 919–929.
•  Somers,  D.J.,  P.  Isaac and  K.  Edwards (2004) A high-density microsatellite consensus map for bread wheat (Triticum aestivum L.). Theor. Appl. Genet. 109: 1105–1114.
•  Somyong,  S.,  G.  Ishikawa,  J.D.  Munkvold,  J.  Tanaka,  D.  Benscher,  Y.G.  Cho and  M.E.  Sorrells (2014) Fine mapping of a preharvest sprouting QTL interval on chromosome 2B in white wheat. Theor. Appl. Genet. 127: 1843–1855.
•  Sourdille,  P.,  T.  Cadalen,  G.  Gay,  B.  Gill and  M.  Bernard (2002) Molecular and physical mapping of genes affecting awning in wheat. Plant Breed. 121: 320–324.
•  Sourdille,  P.,  T.  Cadalen,  H.  Guyomarc’h,  J.W.  Snape,  M.R.  Perretant,  G.  Charmet,  C.  Boeuf,  S.  Bernard and  M.  Bernard (2003) An update of the Courtot × Chinese Spring intervarietal molecular marker linkage map for the QTL detection of agronomic traits in wheat. Theor. Appl. Genet. 106: 530–538.
•  Tan,  M.K.,  P.J.  Sharp,  M.Q.  Lu and  N.  Howes (2006) Genetics of grain dormancy in a white wheat. Aust. J. Agric. Res. 57: 1157–1165.
•  Tanio,  M.,  K.  Kato,  N.  Ishikawa,  Y.  Tamura,  M.  Sato,  H.  Takagi and  M.  Matsuoka (2005) Genetic analysis of photoperiod response in wheat and its relation with the earliness of heading in the southwestern part of Japan. Breed. Sci. 55: 327–334.
•  Tanio,  M. and  K.  Kato (2007) Development of near-isogenic lines for photoperiod-insensitive genes, Ppd-B1 and Ppd-D1, carried by the Japanese wheat cultivars and their effect on apical development. Breed. Sci. 57: 65–72.
•  Torada,  A.,  M.  Koike,  S.  Ikeguchi and  I.  Tsutsui (2008) Mapping of a major locus controlling seed dormancy using backcrossed progenies in wheat (Triticum aestivum L.). Genome 51: 426–432.
•  van Ooijen,  J.W. (2002) MapQTL^® 6, Software for the mapping of quantitative trait loci in experimental populations of diploid species. Kyazma B.V., Wageningen, The Netherlands.
•  van Ooijen,  J.W. (2006) JoinMap^® 4, Software for the calculation of genetic linkage maps in experimental population. Kyazma B.V., Wageningen, The Netherlands.
•  Warner,  R.L.,  D.A.  Kudrna,  S.C.  Spaeth and  S.S.  Jones (2000) Dormancy in white-grain mutants of Chinese Spring wheat (Triticum aestivum L.). Seed Sci. Res. 10: 51–60.
•  Watkins,  A.E. and  S.  Ellerton (1940) Variation and genetics of the awn in Triticum. J. Genetics 40: 243–270.
•  Yan,  L.,  D.  Fu,  C.  Li,  A.  Blechl,  G.  Tranquilli,  M.  Bonafede,  A.  Sanchez,  M.  Valarik,  S.  Yasuda and  J.  Dubcovsky (2006) The wheat and barley vernalization gene VRN3 is an orthologue of FT. Proc. Natl. Acad. Sci. USA 103: 19581–19586.
•  Zanetti,  S.,  M.  Winzeler,  M.  Keller,  B.  Keller and  M.  Messmer (2000) Genetic analysis of pre-harvest sprouting resistance in a wheat × spelt cross. Crop Sci. 40: 1406–1417.