Quantitative Trait Loci Associated with ...

RESEARCH

Quantitative Trait Loci Associated with Deoxynivalenol Content and Kernel Quality in the Soft Red Winter Wheat ‘Ernie’ Z. A. Abate, S. Liu, and A. L. McKendry*

ABSTRACT Fusarium head blight (FHB) caused by Fusarium graminearum Schwabe [telomorph: Gibberella zeae (Schwein.) Petch] reduces quality in wheat (Triticum aestivum L.) when deoxynivalenol (DON) accumulates in the grain. ‘Ernie’ soft red winter wheat has been used as a source of FHB resistance in many U.S. breeding programs because of its type II resistance, low DON, and good kernel quality under heavy disease pressure. The objectives of this study were to identify quantitative trait loci (QTL) associated with both low DON and kernel quality and to determine their association with QTL previously identified for type II resistance in Ernie. Data for DON and Fusarium damaged kernels (FDK) were collected from 243 recombinant inbred lines of the cross Ernie/MO 94-317 grown in inoculated greenhouse experiments conducted in 2002 and 2003. A total of 420 simple sequence repeat and 64 EcoRI/MseI amplified fragment length polymorphism primers were used for parental screening. Three QTL were consistently identified on chromosomes 3BSc, 4BL, and 5AS for both DON and FDK that explained 31 and 42% of the total phenotypic variation in these two traits, respectively. A fourth minor QTL on 2B (R2 = 4%) was also detected for FDK but was not significant for DON. These QTL were coincidental with those previously identified for type II resistance in Ernie, which suggested that the inheritance of these three resistance factors is interdependent in this variety. These findings should enhance the efficiency of using this source of resistance for FHB-resistance breeding.

Z.A. Abate and A.L. McKendry, Division of Plant Sciences, Univ. of Missouri-Columbia, Columbia, MO 65211; S. Liu, Crop and Soil Environmental Sciences Dep., Virginia Polytechnic Inst. and State Univ., Blacksburg, VA 24061-0404. Received 18 Jan. 2008. *Corresponding author ([email protected]). Abbreviations: AFLP, amplified fragment length polymorphism; CIM, composite interval mapping; FDK, Fusarium damaged kernels; FHB, Fusarium head blight; DON, deoxynivalenol; MIM, multiple interval mapping; QTL, quantitative trait loci; RILs, recombinant inbred lines; SSR, simple sequence repeat.

F

usarium head blight (FHB) primarily caused by Fusarium graminearum Schwabe [telomorph: Gibberella zeae (Schwein.) Petch] continues to be a significant problem affecting wheat (Triticum aestivum L.) in the north-central United States. Losses are not only associated with reduced grain yield but also with reductions in grain quality due to low test weight and contamination of grain with trichothecene mycotoxins, especially deoxynivalenol (DON), rendering the grain unsuitable for end-use products. Deoxynivalenol is linked to feed refusal in livestock (Meronuck and Xie, 2000). Although data are limited, DON may also cause gastrointestinal symptoms including vomiting and depression of the immune system in humans (Maresca et al., 2002; Pestka et al., 2004; Bryden, 2007). Mesterházy (1995) extended the components of resistance originally proposed by Schroeder and Christensen (1963) to include components related to resistance to both DON accumulation and kernel infection. Today, sources of low DON and kernel quality retention are available to wheat breeders developing FHB-resistant varieties in several wheat backgrounds including those of Chinese descent such as ‘Sumai 3’ (Bai and Shaner, 2004), ‘Wangshuibai’

Published in Crop Sci. 48:1408–1418 (2008). doi: 10.2135/cropsci2007.07.0411 © Crop Science Society of America 677 S. Segoe Rd., Madison, WI 53711 USA All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher.

1408

WWW.CROPS.ORG

CROP SCIENCE, VOL. 48, JULY– AUGUST 2008

(Ma et al., 2006), and ‘W14’ (Bai et al., 2001; Chen et al., 2006) as well as those in European (Mesterházy et al., 1999) and North American (Bai et al., 2001; McKendry et al., 1995, 2005, 2007) germplasm. However, selection strategies for these traits have not been well investigated. The possibility of indirectly selecting wheat lines with low DON accumulation through selection for types I and/or II FHB resistances under field conditions has been demonstrated but the interrelationships among these resistance factors and DON have been inconsistent. Arseniuk et al. (1999) found significant deviations from linearity when FHB symptoms were regressed on DON concentration and concluded that the mechanisms for FHB resistance and DON accumulation were independent in wheat. Mesterházy et al. (1999) determined that among highly resistant cultivars, DON levels closely paralleled visual resistance symptoms; however, variation did exist in moderately resistant cultivars that had low visual symptoms but higher DON levels and in moderately susceptible lines, where DON accumulation was lower than expected. Bai et al. (2001) reported similar fi ndings among 116 breeding lines and cultivars. Measures of percentage of scabby spikelets and seed quality were highly correlated and each was also correlated with DON in both greenhouse and field studies, but again exceptions were observed among cultivars with moderate levels of resistance or susceptibility. Conclusions of both studies were that breeders could indirectly select for low DON by selecting for high levels of FHB resistance using visual selection parameters (Mesterházy et al., 1999; Bai et al., 2001). Using both spring and winter populations, Wilde and Miedaner (2006) confi rmed that phenotypic selection for high levels of FHB resistance resulted in a significant reduction in DON. Despite their potential importance for FHB-resistance breeding, there are few studies in the literature to date reporting quantitative trait loci (QTL) associated with low DON, particularly in germplasm adapted to the United States. Additionally, the interrelationships among QTL associated with types I and II resistance and low DON have been inconsistent and their relationship with kernel quality retention remains largely unknown. Somers et al. (2003) were among the fi rst to directly report QTL associated with DON accumulation using double haploid lines derived from the cross of ‘Wuhan’/‘Maringa’. Although the range of DON was relatively small in this population, they reported significant QTL on chromosomes 2DS, 3BS, and 5AS. The QTL on 2DS and 5AS were associated with low DON and were independent of FHB severity, while that on 3BS appeared to condition both DON and type II FHB resistance. Lines with favorable alleles on 3BS and 5AS reduced DON accumulation by 17%. Their data suggested that low DON was partially independent of type II resistance. CROP SCIENCE, VOL. 48, JULY– AUGUST 2008

Partial independence of QTL effects for DON and types I and II resistance was also reported for Wangshuibai. Using a set of recombinant inbred lines (RILs) developed from the cross Wangshuibai/‘Annong 8455’, Ma et al. (2006) reported three QTL on chromosomes 5A, 2A, and 3B from Wangshuibai that were associated with low DON. Although QTL effects for both DON and type II resistance were interrelated, the magnitude of the effects differed for each trait. The 3B QTL, which is probably allelic to that in Sumai 3, showed the largest effect on type II resistance (R 2 = 17%), but had a relatively small effect on DON (R 2 = 6.2%), while that on 5A had a relatively large effect on DON (R 2 = 12.4%) but was not consistently significant for type II resistance. The 2A QTL was intermediate and impacted both severity (R2 = 11.5%) and DON (R 2 = 8.5%). Semagn et al. (2007) also reported partial independence of QTL for FHB and DON in a mapping population derived from a cross between the European resistant line ‘Arina’ and the Norwegian breeding line ‘NK94603’. Two QTL, on 1AL, and 2AS, explained 27.9 and 26.7% of the phenotypic variation for DON, respectively. While the 1AL QTL impacted both FHB resistance and DON, that on 2AS was only associated with low DON. In the Sumai 3 genetic background, however, DON and FHB do not appear to be independent. Lemmens et al. (2005) identified a single QTL on 3BS associated with reduced DON in a set of doubled haploid lines developed from the cross of ‘CM82036’ and the susceptible parent ‘Remus’. This locus, which also conditions type II resistance in Sumai 3 (Waldron et al., 1999), accounted for 92.6% of the observed variation in DON. Kernel infection is an important component of resistance (Mesterházy, 1995) impacting the marketability of grain infected with FHB; however, the only known report of QTL associated with kernel infection is that of Chen et al. (2006). In a greenhouse validation study of the 3BS and 5AS QTL identified in the Chinese landrace W14 they found that these QTL had a significant impact on type II resistance, DON content, and kernel infection, together explaining 33, 35, and 31% of the phenotypic variation in these traits, respectively. This study agrees with earlier findings of a strong interdependence of resistance factors in Chinese germplasm carrying the 3BS QTL. Ernie, a soft red winter wheat developed and released by the University of Missouri (McKendry et al., 1995) has moderately high type II resistance, coupled with low DON and good kernel quality retention under FHB pressure. It serves as the early resistant check in both the U.S. Northern and Southern Winter Wheat Scab Nurseries coordinated through the U.S. Wheat and Barley Scab Initiative. As a breeding line, Ernie has the added value of being adapted to the U.S. soft red winter wheat region, thus breeding populations may be developed using Ernie that do not suffer from the lack of adaptation associated with the use of

WWW.CROPS.ORG

1409

Sumai 3 and its derivatives. Recently, QTL were identified that were associated with the type II resistance in Ernie (Liu et al., 2007) and it was determined that the FHB resistance in Ernie differs from that in Sumai 3. However, low DON and kernel quality retention in Ernie have not been mapped. The objectives of this study were to identify QTL associated with low DON and kernel quality retention and to determine whether interrelationships exist among these QTL and those identified for type II resistance in this line.

MATERIALS AND METHODS Mapping Population A set of 243 F8 RILs developed from the cross Ernie/MO 94-317 was used to map QTL for both DON and kernel quality. Ernie originated from the cross ‘Pike’/MO9965. MO9965, an unreleased Missouri breeding line, was derived from the cross ‘Stoddard’/‘Blueboy’//Stoddard/D1707 (McKendry et al., 1995). The highly susceptible parent, MO 94-317 is a Missouri breeding line derived from the cross ‘Agripro Traveler’/‘Pioneer Variety 2555’. Fusarium head blight severity in this line ranges from 70 to 90%. Ernie has been shown to have moderately low to low scores for both DON and Fusarium damaged kernels (FDK) under inoculation in both the field and greenhouse (Bai et al., 2001; Sneller et al., 2006). Development of the population has been previously described (Liu et al., 2005). Briefly, 1000 F3 plants were advanced to the F6 by single seed descent at which time, F7 RILs were derived. Because lines were initiated from the F3 selections rather than from F2 selections some F2 plants in theory may have contributed more than once to the fi nal mapping population. If this was the case, our power to detect QTL and QTL effects may have been diminished. We believe that the probability of redundancy was small because in excess of 5000 F2 plants were advanced to the F3 from which 1000 plants were randomly selected. To provide additional statistical confidence, the genetic similarity between RILs was tested with JoinMap using all amplified fragment length polymorphism (AFLP) and simple sequence repeat (SSR) data collected on the set of 243 RILs. This analysis indicated with a 95% confidence limit, that no two lines were genetically similar and therefore, we concluded that the probability of bias in this data set was low.

Phenotypic Evaluation Experiments were conducted in the greenhouse environment to ensure precision in inoculation technique, a consistently high level of disease pressure and the absence of confounding effects that can often impact field data. Although we acknowledge that this work needs to be verified in the field environment, we elected to conduct this initial work in the greenhouse under more controlled environmental conditions. Eight plants per RIL per replication were arranged in a randomized complete block design with three replications in 2002 and four replications in 2003. Within replication, RILs were randomized by line. Procedures for inoculation are described in Liu et al. (2005). Infected heads from each plant were harvested at maturity and hand-threshed to ensure all diseased kernels were collected. Kernels from each of the eight plants per replication were separated visually into two groups consisting of 1410

sound kernels and a second group of FDK that included from slightly to highly shriveled kernels and tombstones. The number of kernels in each group was counted to precisely determine the proportion of FDK in the infected head. Percentage FDK was determined as the total number of kernels showing FHB symptoms divided by the total number of kernels in the inoculated head, expressed as a percentage. Kernels from the eight heads per RIL within each replication were then bulked to ensure an adequate sample for DON determination. Bulked seed was ground using a Braun coffee grinder and analyzed for DON content at the U.S. Wheat and Barley Scab Initiative– funded laboratory at Michigan State University in East Lansing, MI. Deoxynivalenol content was quantified in micrograms per gram (μg g−1) using the mycotoxin extraction kit Veratoxin for DON 5/5 (Veratox, Lansing, MI).

Molecular Marker Analysis The map for this population was developed by Liu et al. (2007). A total of 420 SSR markers including Xgwm and Xbarc markers (Röder et al., 1998; Plaschke et al., 1995; Song et al., 2005) and 64 EcoRI/MseI AFLP primer pairs were used to identify polymorphic loci between the two parents, Ernie and MO 94-317. For both DON and FDK, 162 AFLP and 100 SSR polymorphic loci were used to construct the genetic linkage groups.

Statistical Analysis Before conducting the analyses of variance (ANOVA) data for each trait were tested for normality using PROC UNIVARIATE NORMAL PLOT (SAS Institute, 2006). Deoxynivalenol data were transformed using a log (x + 1) transformation where x represented DON (μg g–1). Error variances were tested for homogeneity using Bartlett’s test to determine whether or not data could be combined across experiments. Analyses of variance were conducted for both traits on data from individual years and on data combined over years using PROC MIXED (SAS Institute, 2006). The statistical model used considered RILs fi xed while years, replications, and the year × RIL interaction were considered random effects. Variation due to years and RILs was tested for significance against the RILs × year interaction while the RILs × year interaction was tested against the experimental error. Significance levels for ANOVA were declared at the 5% level of probability. Correlation coefficients among FHB-resistance traits were estimated using PROC CORR (SAS Institute, 2006). Broad-sense heritability (H 2BS) for log DON and FDK were estimated from the ANOVA for each individual year and for data combined over years. Heritability was determined as: H 2BS = σ2g/(σ2g + σ2g×y/y + σ2e/ry), where σ2g is the genetic variance among RILs, σ2g×y is variance due to genotype × year interaction, σ2e is variance due to error, and r and y are replications and years, respectively. Exact 95% confidence limits of the estimated heritability were calculated following the procedure of Knapp et al. (1985). The minimum number of genes was estimated using Cockerham’s (1983) modification of Wright’s (1968) formula.

Linkage Map Construction and QTL Analyses Linkage maps were constructed using MapMaker 3.0 (Lander et al., 1987). Initial chromosome locations of SSRs were

WWW.CROPS.ORG


determined based on the reference map of the International Triticeae Mapping Initiative (Röder et al., 1998; Song et al., 2005). The Kosambi mapping function was used to estimate the distance between markers (Kosambi, 1944). Markers were grouped using a logarithms of odds (LOD) value of 3.0 and distance 5% were included in the model. Phenotypic effects of combinations of the QTL alleles of the two parents were estimated for both DON and FDK using the closest marker for QTL that were significant across the two experimental years. Percentage reductions in DON and FDK due to QTL allele combinations were determined.

the assumption of normalcy necessary for both ANOVA and QTL analyses (Miedaner et al., 2006; Semagn et al., 2007). Over years, error variances for both DON and FDK were homogeneous; therefore data for individual years were combined for analyses. Significant genotypic effects (P < 0.001) among RILs were detected for DON and FDK (Table 1). Although the effect of years was not significant, the RIL × year interaction was significant for both traits. Deoxynivalenol levels for Ernie and MO 94-317 were 5.0 and 76.6 μg g–1, respectively, and over years averaged 36.8 μg g−1 for the population (Table 2). The untransformed DON values presented in Table 2 illustrate the genetic difference between parents for this trait. Disease levels in 2003 were higher than those in 2002 due to the known effects of photoperiod and temperature on FHB severity, which led to higher DON levels in the second year of this study. Over years, 14 RILs (5.7% of the population) had DON levels lower than the resistant parent, while 26 RILs (10.7% of the population) had DON levels higher than the susceptible parent.

RESULTS AND DISCUSSION Analyses of Phenotypic Data Distribution of RILs for experiments conducted in 2002 and 2003 showed continuous variation for both DON and FDK indicating quantitative inheritance of the two traits (Fig. 1). Data for DON were not normally distributed thus analyses were conducted on data that had been normalized using a log (x + 1) transformation where x represented DON (μg g–1). Because DON is highly sensitive to environmental factors that may over- Figure 1. Frequency distributions for log deoxynivalenol (DON; μg g−1) and Fusarium estimate the error variance, it is often damaged kernels (FDK; %) for 243 recombinant inbred lines (RILs) of the soft red winter necessary to transform the data to meet wheat cross Ernie/MO 94-317 point-inoculated in the greenhouse in 2002 and 2003 with Fusarium graminearum.


WWW.CROPS.ORG

1411

in the population, we concluded that these heritability estimates reflected largely additive genetic variance. Three to four genetic factors conditioned both traits (Table 2). These estimated gene numbers are consistent with those reported earlier for type II resistance in Ernie (Liu et al., 2005). Correlation coefficients between DON and FDK both within and across years are given in Table 3. Despite higher disease levels in 2003, DON was significantly correlated (r = 0.41) across years as was FDK (r = 0.63). Deoxynivalenol was highly correlated with FDK in both 2002 (r = 0.75) and 2003 (r = 0.70) and each was highly correlated Table 1. Mean squares for log deoxynivalenol (DON) and Fusarwith the FHB severity data from this population reported ium damaged kernels (FDK) of 243 F3 –derived F8 and F9 recomearlier (Liu et al., 2007). These results indicate the interbinant inbred lines (RILs) from the soft red winter wheat cross Ernie/MO 94-317 following greenhouse point inoculation with dependence of both traits and their relationship with FHB Fusarium graminearum. The experiment was conducted in 2002 severity in an Ernie genetic background. Although deviaand repeated in 2003 at the University of Missouri, Columbia. tions from linearity occurred, our results are consistent Mean square with those of Bai et al. (2001) who reported significant Source of Year variation df† Log DON df† FDK correlations under greenhouse conditions between FHB 2002 RILs 242 0.586*** 242 1553*** ratings, DON accumulation, and visual seed quality traits Replication 2 0.104 2 1456*** in a diverse set of germplasm representing major sources Error 461 0.123 484 349 of FHB resistance. Similar correlations have also been 2003 RILs 242 0.605*** 242 2188*** reported under field, spray-inoculated conditions (MesterReplication 3 2.447*** 3 1417* házy et al., 1999; Arseniuk et al., 1999; Miedaner et al., Error 669 0.206 724 457 2003; 2004). The most thorough report on these interrelaCombined Years 1 8.727 1 6235 tionships was a meta-analysis of 163 studies conducted by Rep(years) 5 1.506*** 5 1433*** Paul et al. (2005) in which they found highly significant RILs 242 0.886** 242 2983*** correlations between DON and FDK and between each of RILs × years 242 0.232*** 242 670** these traits and severity of FHB in the diseased head. Our Error 1430 0.172 1208 414 data suggest that in Ernie there is a somewhat stronger relationship between FDK and FHB severity (r = 0.86 in both *Significant at P = 0.05. **Significant at P = 0.01. 2002 and 2003) than between DON and FDK (r = 0.68 and ***Significant at P = 0.001. 0.64 in 2002 and 2003, respectively). The high heritability † For combined ANOVA, degrees of freedom for replications within years and pooled of both DON, and FDK as well as that for FHB severity error were obtained using Satterthwaite approximation due to unequal replication for the 2002 (three replications) and 2003 (four replications) experiments. (Liu et al., 2007) along with the relatively equal number of genetic factors controlling all three Table 2. Mean deoxynivalenol (DON) and Fusarium damaged kernel (FDK) of resistraits suggest similar genetic control of tant (Ernie) and susceptible (MO 94-317) parents along with means and range of the two traits for 243 recombinant inbred lines (RILs) derived from the cross of Ernie/MO these resistance factors in Ernie. How94-317. The experiment was carried out in 2002 and 2003 using point-inoculation at ever, further study is needed before this University of Missouri, Columbia. statement can be made conclusively.

Combined means for FDK in Ernie and MO 94-317 were 25 and 83%, respectively. As with DON, FDK data were higher in 2003 than in 2002. Over years RIL means averaged 48.2% FDK with 46 RILs (19%) having a lower FDK than the resistant parent, Ernie (Table 2; Fig. 1). Estimated heritabilities varied from 59 to 82% for DON and from 73 to 82% for FDK with mean heritabilities across years of 74% and 78% for DON and FDK, respectively (Table 2). Because these data were determined from inbred lines, and significant transgressive segregation was observed

Traits

Year

Ernie

MO 94-317

RIL mean ± SE

H2†

95% CI for H2‡

No. of factors§

DON (μg g –1)¶ 2002

3.6

69.2

39.5 ± 1.52

0.79

0.75–0.82

4

2003

6.4

84.2

34.7 ± 1.60

0.66

0.59–0.71

4

Combined

5.0

76.6

36.8 ± 1.12

0.74

0.69– 0.78

4

FDK (%)

†

#

2002

13

89

45.1 ± 1.0

0.78

0.73–0.81

3

2003

38

77

49.7 ± 1.0

0.79

0.75–0.82

3

Combined 25

83

48.2 ± 0.7

0.78

0.73– 0.81

4

Broad-sense heritability (H2) was estimated based on genetic, phenotypic, and error variances calculated from expected mean squares of ANOVA.

‡

95% confidence interval for heritability was based on Knapp et al. (1985) formula.

§

Minimum number of genes controlling DON and FDK calculated following (Wright, 1968) as modified by Cockerham (1983).

¶

#

DON content of RILs was quantified from bulked seed of inoculated spikes using mycotoxin extraction kit (Veratoxin, DON 5/5, Veratox) at Michigan State University.

FDK was assessed by grouping kernels from the inoculated spikes into two groups, those that were unaffected by FHB and those that were shriveled or tombstones due to FHB infection. Fusarium damaged kernels were expressed as a percentage of those on the inoculated head.

1412

WWW.CROPS.ORG

QTL Analysis Molecular map construction for this population has been thoroughly described (Liu et al., 2007); however, it is worth repeating here that marker density in this population was lower than expected because SSR polymorphism (22%) was lower than that reported for other populations. This was in part because Ernie is an adapted soft red winter wheat cultivar and therefore has a greater coefficient of parentage with MO 94-317 than would be the case in populations developed from Chinese CROP SCIENCE, VOL. 48, JULY– AUGUST 2008

Table 3. Correlation coefficients among deoxynivalenol (DON), resistant germplasm and adapted susceptible cultivars. We Fusarium damaged kernels (FDK), and Fusarium head blight included AFLPs in this study to offset this problem but severity (FHBS; type II resistance) of 243 recombinant inbred still had low marker coverage on some chromosome arms. lines derived from the soft red winter wheat cross Ernie/MO As such, low marker densities on some chromosomes may 94-317. Experiments were conducted in 2002 and 2003. have led to an under- or overestimation of the magnitude DON FDK Resistance traits of QTL effects in these chromosome regions. 2002 2003 2002 2003 Based on data combined over years, three regions DON-2003 0.41*** were identified on chromosomes 3B, 4B, and 5A that were FDK-2002 0.75*** 0.37*** associated with reduced DON (Table 4, Fig. 2). Across FDK-2003 0.58*** 0.70*** 0.63*** years, these three QTL explained 31% of the total phe0.68*** 0.45*** 0.86*** 0.68*** FHBS-2002† notypic variation in DON content. Quantitative trait loci † 0.57*** 0.64*** 0.61*** 0.86*** FHBS-2003 effects were predominantly additive and all alleles were *** Significant at P < 0.0001. from the resistant parent Ernie. A forth QTL on 2B was † FHBS is the Fusarium head blight severity, determined as the number of diseased spikelets/total spikelets in the inoculated head, expressed as a percentage. Data significant in 2002 but failed to reach significance in 2003 were previously reported (Liu et al., 2007). or when data were combined across years. The major QTL on 3B, which, across years accounted 3. The location of the QTL from Ernie and its associated for 14% of the phenotypic variation in DON accumulation markers (Fig. 2) led us to conclude, however, that the 3BSc (Table 4), was centromeric, flanked by Xgwm285 and the QTL in our population differs from the 3BS QTL associAFLP marker Xe41m50_6. The QTL peak (107.3 cM) was ated with Xgwm533 reported for DON accumulation in the located 8 cM from Xgwm285. Consistent with Somers et al. germplasm described above. (2003), we have referred to this location as 3BSc to differThe 4BL QTL, which is flanked by Xgwm495 and entiate this centromeric QTL from the well-known QTL Xgwm149 (Fig. 2) accounted for 7% of the variation in from Sumai 3 that is distally located on 3BS. The 3BSc QTL DON. These flanking markers are 5.5 cM apart and the for DON accumulation is coincident with the 3BSc QTL, QTL peak (0.1 cM) is 0.1 cM from Xgwm495 suggesting Q fhs.umc-3B, previously reported for type II resistance in that this marker once validated, could be immediately useErnie (Liu et al., 2007). Although these data do not enable ful for marker-assisted selection for low DON. As was the us to resolve whether or not there is a pleiotropic effect of case with the 3BSc QTL, the 4BL QTL in Ernie is coincithis allele, the data suggest that in Ernie, this allele impacts dent with the 4BL QTL, Q fhs.umc-4BL, previously reported both visual type II symptoms and DON accumulation. A for type II resistance in this cultivar (Liu et al., 2007). To 3BSc QTL was detected for FHB field severity in a cross of our knowledge, a 4BL QTL has not been reported for DON Wuhan 1/Maringa, but was not detected with greenhouse content. Somers et al. (2003) reported a 4BL QTL linked inoculations (Somers et al., 2003). Somers et al. (2003) did, however, report a distal 3BS QTL associated with reduced DON accumula- Table 4. Quantitative trait loci (QTL) associated with deoxynivalenol (DON) accumution in this population that appeared to lation and Fusarium damaged kernels (FDK) of recombinant inbred lines of the soft be consistent with the widely studied red winter wheat cross, Ernie/MO 94-317 following greenhouse inoculation with F. Sumai 3 3BS QTL allele (Waldron et graminearum. Analyses were conducted for individual experimental years (2002 and al., 1999; Anderson et al., 2001; Zhou 2003) and for data combined over years. 2002 2003 Combined et al., 2002). The interdependence of Chromosome Marker † ‡ 2§ 2 location LOD A R LOD A R LOD A R2 FHB severity and DON associated % % % with this QTL has been reported in other germplasm carrying this allele. DON Xgwm276b 3.4 –8.0 5 – – – – – – Lemmens et al. (2005) found an inter- 2B Xgwm285 3.9 –7.6 8 4.8 –9.2 13 6.4 –9.8 14 dependence between FHB spread and 3BSc Xgwm495 4.4 –7.2 6 2.0 – – 4.4 –5.3 7 reduced DON accumulation associated 4BL Xbarc56 6.5 –11.1 19 3.2 –7.3 6 4.0 –5.1 10 with the 3BS QTL in a cross involving 5AS CM82036, a Sumai 3 derivative, while FDK Xgwm271 3.4 –5.9 5 – – – 4.1 –5.8 4 both Ma et al. (2006) and Chen et al. 2B Xgwm285 5.3 –7.5 11 7.3 –9.3 16 8.8 –9.2 18 (2006) found the same interrelation- 3BSc Xgwm495 3.9 –5.7 6 3.5 –5.6 4 5.2 –5.8 6 ship of effects for the 3BS QTL from 4BL 5AS Xbarc165 7.0 –10.1 21 3.7 –7.3 10 6.8 –8.8 18 Wangshuibai and W14, respectively. In both of these genotypes, the 3BS †Marker linked to the QTL peak. QTL allele is thought to be either the ‡Additive effects−1 from the marker at the peak LOD. For DON, values of the additive effects reflect actual DON values in μg g . same as or allelic with that from Sumai §Proportion of the phenotypic variation explained by each QTL based on multiple interval mapping (MIM) estimations. CROP SCIENCE, VOL. 48, JULY– AUGUST 2008

WWW.CROPS.ORG

1413

1414

WWW.CROPS.ORG


Figure 2. Quantitative trait loci (QTL) associated with deoxynivalenol (DON) content and Fusarium damaged kernels (FDK) in the soft red winter wheat cross Ernie/MO 94-317 detected by composite interval analysis. The LOD scores were plotted against centimorgans on the chromosome. Data are combined over 2 yr of evaluation in 2002 and 2003. The black square inside the chromosome indicates the position of the centromere. For chromosomes 3B and 5A, three and two unlinked segments, respectively, were obtained. Anchored simple sequence repeat (SSR) marker locations on chromosomes are based on the reference International Triticeae Mapping Initiative mapping population (Röder et al., 1998; Song et al., 2005).

with Xwmc238 that was associated field FHB resistance in the cross Wuhan 1/Maringa, but failed to detect this QTL associated with DON. Finally, the 5A QTL, associated with Xbarc56 on 5AS accounted for 10% of the variation in DON. The flanking markers, Xbarc56 and Xbarc165 are approximately 21 cM apart and Xbarc56 is 14 cM from the QTL peak (45 cM). Again, this QTL was coincident with a 5AS QTL for type II resistance in Ernie (Liu et al., 2007). Although not conclusive because of the low marker density in this region the data again suggest the interdependence of visual FHB symptoms and DON in this cultivar. Additional markers are being developed to saturate this region and refine this QTL position. The 5AS chromosome region has been found to be involved in FHB resistance in widely diverse germplasm. Quantitative trait loci for type I or field resistance in CM82036 (Buerstmayr et al., 2003), Renan (Gervais et al., 2003), ‘Frontana’ (Steiner et al., 2004), W14 (Chen et al., 2006), and Wangshuibai (Lin et al., 2006) have been detected in this region as well as for type II resistance in ‘Fundulea 201 R’ (Shen et al., 2003), and Ernie (Liu et al., 2007). Our data for Ernie agree with those of Chen et al. (2006), who found interdependence among types of resistance conferred by alleles in the 5AS region. In their study of W14, a 5AS QTL was associated with reduced field incidence, severity, and DON. Flanking markers for this QTL (Xbarc117 and Xbarc186) suggest that this QTL may be the same as that for DON and type II resistance in Ernie, although the QTL effect is somewhat lower. These data, however, conflict with those of Somers et al. (2003) who reported interdependence of DON and type II resistance at the 3BS QTL but not at the 5AS QTL. The latter allele from Maringa had no effect on either type II or field resistance but reduced DON content by 36%. Although the 5AS QTL in Ernie was in the same region as that in Maringa its linkage with Xbarc56 positioned it slightly more distally than that in Maringa, which was linked to Xgwm96 (Somers et al., 2004). Thus, it is unclear whether or not they reflect the same or different genes. Ma et al. (2006) also reported a 5A QTL that was associated with reduced DON in Wangshuibai. However, based on the consensus map (Somers et al., 2004) this QTL, which is linked to Xgwm156, is on the long arm of chromosome 5A, and therefore appears to differ from the 5AS QTL reported here. A significant epistatic interaction between QTL alleles on 3B and 5A was detected using ANOVA (P < 0.05) that accounted for 8% of the variation in DON. Using the epistasis function in CIM, an interaction between these two alleles was also detected, however, it failed to reach the LOD threshold for significance (LOD = 2.8) and therefore was not reported in Table 4. For FDK, CIM revealed four chromosome regions on 2B, 3BSc, 4BL, and 5AS, which together over years accounted for 46% of the observed variability in kernel CROP SCIENCE, VOL. 48, JULY– AUGUST 2008

quality. Except for the 2B QTL, all QTL were consistently detected over years. The 2B QTL was significant in both 2002 and in data combined over years but failed to meet the significance threshold in 2003. As with DON, these QTL exhibited predominantly additive effects and QTL × QTL interactions were not significant. All QTL alleles were from the resistant parent Ernie. The three major QTL on 3BSc, 4BL, and 5AS accounted for 18, 6, and 18% of the observed variability in FDK, respectively (Fig. 2; Table 4). These QTL, linked to Xgwm285, Xgwm495, and Xbarc56 were coincidental with those for DON. They also co-localized with QTL in the same regions for type II resistance (Liu et al., 2007), which confirmed the interdependence of these sources of resistance in Ernie. Reports of QTL for FDK are limited in the literature and to our knowledge this is the first report of QTL for kernel quality on 3BSc and 4BL. Our data for the 5AS QTL are consistent with those of Chen et al. (2006) who reported a QTL in W14 in the same region of 5AS that confers reduced kernel infection. Chen et al. (2006) also found that this QTL conferred type II resistance and reduced DON, although its effect on greenhouse data was smaller than that observed for Ernie (Liu et al., 2007). The centromeric 2B QTL had a smaller effect on FDK (R 2 = 4%) and although it was in the same chromosome region as that for DON, was linked to Xgwm276b with the QTL peak estimated to be 8.2 cM proximal to that for DON. We were unable to resolve whether or not this is the same allele as that for DON, however, the data suggest that these two QTL are coincidental. A 2B QTL with a small effect, Q fhs.umc-2B, linked to Xgwm276b has also been reported for type II resistance in Ernie (Liu et al., 2007). We believe this is the first QTL on 2B reported for FDK. Both ANOVA and CIM were again used to detect epistatic interactions among QTL alleles. A significant three-way interaction (P < 0.05) among QTL alleles on 2B, 3B, and 5A was detected using ANOVA, which accounted for 6% of the genetic variance, however, this interaction was not detected with CIM. Because of the inconsistency across analytical techniques, this interaction is not presented in Table 4.

Phenotypic Effects of QTL Allele Combinations To examine allelic effects of QTL for each trait, phenotypic effects of allele combinations were compared using combined data for RILs carrying resistant or susceptible alleles from Ernie (E) or MO 94-317 (M), respectively (Table 5). For DON, effects of significant QTL on 3BSc, 4BL, and 5AS were considered individually and then in combination with other desirable alleles from Ernie based on associated markers with the largest QTL effect. Variance in the DON data was high as expected, yet significant effects of desirable alleles were observed, which indicate

WWW.CROPS.ORG

1415

the additive effect of these three QTL alleles and their combinations. Individually, the 5AS, 4BL, and 3BSc QTL alleles reduced DON by 11.8, 16.0, and 20.8%, respectively compared with RILs carrying alleles from MO 94-317 (Table 5). Where lines carried two of the favorable alleles, the combination of QTL on 3BSc and 5AS had the largest effect, reducing DON by 47% while three favorable alleles from Ernie reduced DON by 78% compared to RILs carrying none of the favorable alleles. The magnitude of the effect of 3BSc and 5AS combination may in part reflect an interaction between these two QTL alleles that was detected by CIM but did not reach the LOD threshold required for significance. Although QTL for DON on 3BSc and 4BL have not been previously reported, a similar analysis by Somers et al. (2003) revealed a 6% reduction in DON associated with a 5AS QTL from Maringa that may reflect the same gene as that in Ernie. Whether or not this difference is meaningful is questionable because of the lower range of DON in the Somers study, and the large variance associated with the DON levels themselves.

It should be noted, that the DON level (50.0 μg g–1) of RILs carrying alleles from the susceptible parent MO 94-317 (M 2M 2; M3M3; M4M4) was lower than expected and significantly lower than that for the susceptible parent itself (Tables 2 and 5). When the 2B QTL, which was significant in 2002, was included in the analysis it further reduced DON (data not shown). The mean DON level of the eight RILs that contained the four alleles from MO 94-317 was 64.7 ± 7.5 μg g–1, which suggested that there may indeed be a significant effect of the 2B QTL on DON content. Similarly, when it was included with the 3BSc, 4BL, and 5AS alleles from Ernie, the mean DON value of the 10 RILs carrying the four favorable QTL alleles was further reduced to 7.8 ± 6.7 μg g−1. Although in 2003 the 2B QTL (LOD = 1.1) failed to reach the threshold for significance (LOD = 3.0) a QTL peak was observed. Its lack of significance in the combined data may be related to the inherent variability in DON data. For FDK, four significant QTL were identified in data combined across years. In an analysis similar to that

Table 5. Phenotypic effects of quantitative trait loci (QTL) allele combinations on deoxynivalenol (DON) content and the percentage of Fusarium damaged kernels (FDK) in recombinant inbred lines (RILs) of the cross Ernie/MO 94-317 following greenhouse inoculation with Fusarium graminearum. Data are from experiments combined across two experimental years (2002 and 2003). QTL combination†

Markers linked to QTL ‡

No RILs with QTL

Trait

M4 M4

46

50.0 ± 3.2 44.1 ± 3.8

Xgwm276b

Xgwm285

Xgwm495

Xbarc56

–

M 2M 2

M3M3

DON (μg g−1) –,–,– –,–, 5A

–

M 2M 2

M3M3

E4E4

36

–, 4B,–

–

M 2M 2

E3 E3

M4 M4

26

37.0 ± 4.4

3B,–,–

–

E 2E 2

M3M3

M4 M4

30

39.6 ± 4.0

–,4B,5A

–

M 2M 2

E3 E3

E4E4

25

34.0 ± 4.5

3B,–,5A

–

E 2E 2

M3M3

E4E4

21

26.5 ± 4.9

3B,4B,–

–

E 2E 2

E3 E3

M4 M4

17

26.7 ± 5.4

3B,4B,5A

–

E 2E 2

E3 E3

E4E4

18

11.0 ± 5.4 71.8 ± 3.2

FDK (%) –,–,–,–

M1M1

M 2M 2

M3M3

M4 M4

9

–,–,–, 5A

M1M1

M 2M 2

M3M3

E4E4

6

57.0 ± 5.9

–,–, 4B,–

M1M1

M 2M 2

E3 E3

M4 M4

8

64.1 ± 5.1

2B,–,–,–

E1E1

M 2M 2

M3M3

M4 M4

36

55.9 ± 2.4

–, 3B,–,–

M1M1

E 2E 2

M3M3

M4 M4

7

51.7 ± 5.4

2B, 3B,–,–

E1E1

E 2E 2

M3M3

M4 M4

22

49.3 ± 3.0

–,–, 4B, 5A

M1M1

M 2M 2

E3 E3

E4E4

10

47.2 ± 4.6

2B,–,–, 5A

E1E1

M 2M 2

M3M3

E4E4

29

49.4 ± 2.7

2B,–, 4B,–

E1E1

M 2M 2

E3 E3

M4 M4

17

47.6 ± 3.5

–3B,–, 5A

M1M1

E 2E 2

M3M3

E4E4

5

55.6 ± 6.4

–, 3B, 4B,–

M1M1

E 2E 2

E3 E3

M4 M4

6

38.5 ± 5.9

2B, 3B, 4B,–

E1E1

E 2E 2

E3 E3

M4 M4

10

36.2 ± 4.5

2B, 3B,–, 5A

E1E1

E 2E 2

M3M3

E4E4

16

33.8 ± 3.8

2B,–, 4B, 5A

E1E1

M 2M 2

E3 E3

E4E4

15

43.5 ± 3.9

–, 3B, 4B, 5A

M1M1

E 2E 2

E3 E3

E4E4

5

31.5 ± 6.4

2B, 3B, 4B, 5A

E1E1

E 2E 2

E3 E3

E4E4

13

20.8 ± 4.1

†

Chromosome locations of QTL associated with resistance alleles from Ernie.

‡

‘E’ allele from Ernie and ‘M’ allele from MO 94-317 for all markers. Subscripts represent QTL alleles from 2B, 3B, 4BL, and 5A linked to Xgwm276b, Xgwm285, Xgwm495, and Xbarc56, respectively.

1416

WWW.CROPS.ORG


conducted for DON, individual and combined allelic effects were determined (Table 5). Mean FDK of RILs carrying only those alleles from the susceptible parent (M1M1; M2M2; M3M3; M4M4) was 71.8% whereas that for RILs carrying Ernie alleles for all four QTL (E1E1; E2E2; E3E3; E4E4) was 20.8%. With the inclusion of the 2B QTL, which was significant in the combined data, this range more closely paralleled that of the parents than was the case with DON. Individually, the 3BSc QTL allele had the largest effect on kernel quality, reducing FDK by 28%, while the 4BL QTL allele had the smallest effect, reducing FDK by 10.7%. Interestingly, when these two alleles were combined, either with other susceptible alleles or in combination with either or both of the 2B and 3BSc alleles from Ernie their effect appeared to be synergistic. No significant interactions between QTL alleles on 2B and 3BSc, however, were detected by either ANOVA or CIM. When combined with the 5AS allele, FDK was reduced by 56% and where RILs contained all four favorable alleles from Ernie, FDK was reduced by 71% compared to RILs carrying only the susceptible alleles. This data set further suggests the additive effects of these QTL combinations on FDK. In summary, these results agree with those of Liu et al. (2007) that suggest resistance in Ernie differs from Sumai 3 and other sources of resistance commonly in use in U.S. breeding programs. On-going research will seek to verify these QTL in the field environment and validate them in Ernie-derived populations near-isogenic for these QTL alleles. Additionally, we are working to add markers to the map to saturate regions to refine QTL positions and identify markers linked more closely to the QTL peak. Once this research is completed these markers should be a valuable resource for marker-assisted selection. These data, along with those of Liu et al. (2007), indicate that Ernie will complement other known sources of FHB resistance in breeding programs aimed at pyramiding genetically different FHB alleles into winter wheat varieties. Colocalization of QTL for DON and FDK with those for type II resistance suggests that in this cultivar, resistances to FHB are interdependent. From a breeding perspective, therefore, selection for QTL for type II resistance in Ernie-derived populations should concurrently result in low DON and improved kernel quality retention. Acknowledgments The authors wish to thank Dr. Pat Hart, Michigan State University for his help with DON analyses. This material is based on work supported in part by the U.S. Department of Agriculture under Award No. 59-0790-9-052. This is a cooperative project with the U.S. Wheat and Barley Scab Initiative. Any opinions, fi ndings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the view of the U.S. Department of Agriculture.


References Anderson, J.A., R.W. Stack, S. Liu, B.L. Waldron, A.D. Field, C. Coyne, B. Moreno-Sevilla, J.M. Fetch, Q.J. Song, P.B. Cregan, and R.C. Frohberg. 2001. DNA markers for Fusarium head blight resistance QTLs in two wheat populations. Theor. Appl. Genet. 102:1164–1168. Arseniuk, E., E. Foremska, T. Goral, and J. Chelkowski. 1999. Fusarium head blight reactions and accumulation of deoxynivalenol (DON) and some of its derivatives in kernels of wheat, triticale, and rye. J. Phytopathol. 147:577–590. Bai, G.H., R. Plattner, A. Desjardins, and F. Kolb. 2001. Resistance of Fusarium head blight and deoxynivalenol accumulation in wheat. Plant Breed. 120:1–6. Bai, G.H., and G.E. Shaner. 2004. Management and resistance in wheat and barely to Fusarium head blight. Annu. Rev. Phytopathol. 42:135–161. Bryden, W.L. 2007. Mycotoxins in the food chain: Human health implications. Asia Pac. J. Clin. Nutr. 16:95–101. Buerstmayr, H., B. Steiner, L. Hartl, M. Griesser, N. Angerer, D. Lengauer, T. Miedaner, B. Schneider, and M. Lemmens. 2003. Molecular mapping of QTLs for Fusarium head blight resistance in spring wheat: II. Resistance to fungal penetration and spread. Theor. Appl. Genet. 107:503–508. Chen, J., C.A. Griffey, M.A. Saghai Maroof, E.L. Stromberg, R.M. Biyashev, W. Zhao, M.R. Chappell, T.H. Pridgen, Y. Dong, and Z. Zeng. 2006. Validation of two major quantitative trait loci for Fusarium head blight resistance in Chinese wheat line W14. Plant Breed. 125:99–101. Cockerham, C.C. 1983. Covariance of relatives from self-fertilization. Crop Sci. 23:1177–1180. Doerge, R.W., and G.A. Churchill. 1994. Permutation tests for multiple loci affecting a quantitative character. Genetics 142:285–294. Gervais, L., F. Dedryver, J.-Y. Morlais, V. Bodusseau, S. Negre, M. Bilous, C. Groos, and M. Trottet. 2003. Mapping of quantitative trait loci for field resistance to Fusarium head blight in a European winter wheat. Theor. Appl. Genet. 106:961–970. Knapp, S.J., W.W. Stroup, and W.W. Ross. 1985. Exact confidence intervals for heritability on a progeny mean basis. Crop Sci. 25:192–194. Kosambi, D.D. 1944. The estimation of map distances from recombination values. Ann. Eugen. 12:172–175. Lander, E.S., P. Green, J. Abrahamson, A. Barlow, M.J. Daly, S.E. Lincoln, and L. Newburg. 1987. MAPMAKER: An interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics 1:174–181. Lemmens, M., U. Scholz, F. Berthiller, C. Dall’Asta, A. Koutnik, R. Schuhmacher, G. Adam, H. Buerstmayr, A. Mesterházy, R. Krska, and P. Ruckenbauer. 2005. The ability to detoxify the mycotoxin deoxynivalenol co-localize with a major quantitative trait locus for Fusarium head blight resistance in wheat. Mol. Plant Microbe Interact. 18(12):1318–1324. Lin, F., S.L. Xue, Z.Z. Zhang, C.Q. Zhang, Z.X. Kong, G.Q. Yao, D.G. Tian, H.L. Zhu, C.J. Li, Y. Cao, J.B. Wei, Q.Y. Luo, and Z.Q. Ma. 2006. Mapping QTL associated with resistance to Fusarium head blight in the Nanda2419 × Wangshuibai population: II. Type I resistance. Theor. Appl. Genet. 112:528–553. Liu, S., Z.A. Abate, H. Lu, T. Musket, G.L. Davis, and A.L. McKendry. 2007. QTL associated with Fusarium head blight resistance in the soft red winter wheat Ernie. Theor. Appl. Genet. 115:417–427.

WWW.CROPS.ORG

1417

Liu, S., Z.A. Abate, and A.L. McKendry. 2005. Inheritance of Fusarium head blight resistance in the soft red winter wheat Ernie. Theor. Appl. Genet. 110:454–461. Ma, H.X., K.M. Zhang, L. Gao, G.H. Bai, H.G. Chen, Z.X. Cai, and W.Z. Liu. 2006. Quantitative trait loci for resistance to Fusarium head blight and deoxynivalenol accumulation in Wangshuibai wheat under field condition. Plant Pathol. 55:739–745. Maresca, M., R. Mahfoud, M. Garmy, and J. Fantini. 2002. The mycotoxin deoxynivalenol affects nutrient absorption in human intestinal epithelial cells. J. Nutr. 132:2723–2731. McKendry, A.L., J.E. Berg, D.N. Tague, and K.D. Kephart. 1995. Registration of Ernie wheat. Crop Sci. 35:1513–1514. McKendry, A.L., D.N. Tague, R.L. Wright, and J.A. Tremain. 2007. Registration of Bess wheat. J. Plant Reg. 1:21–23. McKendry, A.L., D.N. Tague, R.L. Wright, J.A. Tremain, and S.P. Conley. 2005. Registration of Truman wheat. Crop Sci. 45:421–422. Meronuck, R., and W. Xie. 2000. Mycotoxins in feed. Feedstuff 72:95–100. Mesterházy, A. 1995. Types and composition of resistance to Fusarium head blight of wheat. Plant Breed. 114:377–386. Mesterházy, A., T. Bartok, C.J. Mirocha, and R. Komoróczy. 1999. Nature of wheat resistance to Fusarium head blight and the role of deoxynivalenol for breeding. Plant Breed. 118:97–110. Miedaner, T., N. Heinrich, B. Schneider, G. Oettler, S. Rohde, and F. Rabenstein. 2004. Estimation of deoxynivalenol (DON) content by symptom rating and exoantigen content for resistance selection in wheat and triticale. Euphytica 139:123–132. Miedaner, T., B. Schneider, and H.H. Geiger. 2003. Deoxynivalenol (DON) content and Fusarium head blight resistance in segregating population of winter rye and winter wheat. Crop Sci. 43:519–526. Miedaner, T., F. Wilde, B. Steiner, H. Buerstmayr, V. Korzun, and E. Ebmeyer. 2006. Stacking quantitative trait loci (QTL) for Fusarium head blight resistance from non-adapted sources in an European elite spring wheat background and assessing their effect on deoxynivalenol (DON) content and disease severity. Theor. Appl. Genet. 112:562–569. Paul, P.A., P.E. Lipps, and L.V. Madden. 2005. Relationship between visual estimates of Fusarium head blight intensity and deoxynivalenol accumulation in harvested wheat grain: A meta analysis. Phytopathology 95:1225–1236. Plaschke, J., M.W. Ganal, and M.S. Röder. 1995. Detection of genetic diversity in closely related bread wheat using microsatellite markers. Theor. Appl. Genet. 91:1001–1007. Pestka, J.K., R.R. Zhou, Y. Moon, and Y.J. Chung. 2004. Cellular and molecular mechanisms for immune modulation by deoxynivalenol and other trichothecenes; unraveling a paradox. Toxicol. Lett. 153:61–73. Röder, M.S., V. Korzyn, K. Wendehake, J. Plaschke, H. Tixier, L. Leroy, and M.W. Ganal. 1998. A microsatellite map of wheat. Genetics 149:2007–2023.

1418

SAS Institute. 2006. SAS/STAT user’s guide. Version 9.1. SAS Inst., Cary, NC. Schroeder, H.W., and J.J. Christensen. 1963. Factors affecting resistance of wheat to scab caused by Gibberella zeae. Phytopathology 53:831–838. Semagn, K., H. Skinner, Å. Bjørnstad, Å.G. Marøy, and Y. Tarkegne. 2007. Quantitative trait loci controlling Fusarium head blight resistance and low deoxynivalenol content in hexaploid wheat population from ‘Arina’ and NK93604. Crop Sci. 47:294–303. Shen, X., M. Ittu, and H.W. Ohm. 2003. Quantitative trait loci conditioning resistance to Fusarium head blight in wheat line F201R. Crop Sci. 43:850–857. Sneller, C., P. Paul, L. Herald, B. Sugerman, and A. Johnston. 2006. Report on the 2005–06 Northern Uniform Winter Wheat Scab Nurseries. p. 121–125. In S.M. Canty et al. (ed.) Proc. of the 2006 National Fusarium Head Blight Forum, Research Triangle Park, NC. 10–12 Dec. 2006. Michigan State University, East Lansing. Somers, D.J., G. Fedak, and M. Savard. 2003. Molecular mapping of novel genes controlling Fusarium head blight resistance and deoxynivalenol accumulation in spring wheat. Genome 46:555–564. 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. Song, Q.J., J.R. Shi, S. Singh, E.W. Fickus, J.M. Costa, J. Lewis, B.S. Gill, R. Ward, and P.B. Cregan. 2005. Development and mapping of microsatellite (SSR) markers in wheat. Theor. Appl. Genet. 110:550–560. Steiner, B., M. Lemmens, M. Griesser, U. Scholz, J. Schondelmaier, and H. Buerstmayr. 2004. Molecular mapping of resistance to Fusarium head blight in the spring wheat cultivar Frontana. Theor. Appl. Genet. 109:215–224. Waldron, B.L., B. Moreno-Sevilla, J.A. Anderson, R.W. Stack, and R.C. Frohberg. 1999. RFLP mapping of QTL for Fusarium head blight resistance in wheat. Crop Sci. 39:805–811. Wang, S., C.J. Basten, and Z.B. Zeng. 2006. Windows QTL Cartographer 2.5. Available at statgen.ncsu.edu/qtlcart/WQTLCart.htm (verified 28 May 2008). Dep. of Statistics, North Carolina State Univ., Raleigh. Wilde, F., and T. Miedaner. 2006. Selection for Fusarium head blight resistance in early generation reduces the deoxynivalenol (DON) content in grain of winter and spring wheat. Plant Breed. 125:96–98. Wright, S. 1968. Evolution and genetics of populations. Vol. 1. Genetics and biometric foundation. Univ. of Chicago Press, Chicago, IL. Zhou, W., F.L. Kolb, G.H. Bai, G.E. Shaner, and L.L. Domier. 2002. Genetic analysis of scab resistance QTL in wheat with microsatellite and AFLP markers. Genome 45:719–727.

WWW.CROPS.ORG
