Professional Documents
Culture Documents
RESEARCH
E.W. Tessmann and D.A. Van Sanford, Dep. of Plant and Soil Science,
ABSTRACT Univ. of Kentucky, Lexington, KY 40546; Y. Dong, Dep. of Plant
Fusarium head blight (FHB) is an important Pathology, Univ. of Minnesota, St. Paul, MN, 55108. Received 14 Aug.
disease of wheat (Triticum aestivum L.) that has 2018. Accepted 12 May 2019. *Corresponding author (dvs@uky.edu).
caused billions of dollars in losses in recent Assigned to Associate Editor Toi Tsilo.
decades. Although a massive breeding effort
Abbreviations: DON, deoxynivalenol; FDK, Fusarium-damaged
has been undertaken on multiple continents,
kernels; FHB, Fusarium head blight; GS, genomic selection; GWAS,
there are no wheat cultivars with immunity
genome-wide association study; HD, heading date; INC, incidence;
to the disease. Resistance is conditioned by
KASP, Kompetitive allele specific polymerase chain reaction; PH, plant
multiple loci and is further complicated by the
height; QQ, quantile-quantile; QTL, quantitative trait locus/loci;
role of the environment in expression of the
SEV, severity; SNP, single nucleotide polymorphism; TCAP, Triticeae
disease phenotype. The objectives of our study
Coordinated Agricultural Project.
were (i) to evaluate the phenotypic response to
FHB in a large, diverse soft red winter wheat
mapping panel; and (ii) to identify promising
quantitative trait loci (QTL) associated with FHB
C limate change brings to the agricultural sector the uncer-
tainty of yield production, unpredictable rain patterns, and
changes in disease and insect pressure, to name a few challenges.
resistance based on a genome-wide associa-
Grain production is estimated to decrease by 6% for wheat (Triticum
tion study (GWAS). We evaluated the mapping
aestivum L.) in the next decades for each increase of 1°C (Zhao et
panel in 2014–2015 and 2015–2016 in an irri-
gated, inoculated scab nursery near Lexington,
al., 2017). Climate change is also expected to affect occurrence,
KY. Traits evaluated were heading date, plant distribution and intensity of plant diseases such as Fusarium head
height, FHB rating, severity, incidence, index, blight (FHB) (Audenaert et al., 2013). Using a modeling approach,
Fusarium-damaged kernels (FDK), and deoxyni- Backhouse (2014) found a positive correlation between climate
valenol (DON). There were significant (p < 0.05) and distribution of pathogenic species of Fusarium, which includes
differences among genotypes for all traits Fusarium graminearum Schwabe. Their research predicts wide distri-
measured. The GWAS (based on 2-yr entry bution in countries where this disease already occurs, and new
means) identified 16 significant (p < 0.001) single regions experiencing epidemics of FHB such as Mexico, North
nucleotide polymorphisms (SNPs) associated Africa, Ethiopia, and western Siberia (Backhouse, 2014). Devel-
with disease traits on multiple chromosomes. oping genetically improved resistant cultivars in FHB-prone
Single nucleotide polymorphism association
environments will be fundamental to meeting future food demand.
ranged from −2.14 to 4.01% of the mean of a
Plant diseases are estimated to cause 10 to 16% of yield losses
given trait. We detected SNPs associated with
FDK and DON on chromosomes 4A, 5B and
globally (Chakraborty and Newton, 2011). Fusarium head blight
6B, and these SNPs decreased DON levels by is one of the most important diseases of wheat and other small
1.5, 2.1 and 3.2 mg kg−1, respectively. Our study grains. In the 1990s, this disease caused US$4.8 billion in losses
demonstrated that even small-effect QTL can in the United States ( Johnson et al., 2003). During the epidemic
potentially decrease disease levels and thus be years of 1998 through 2000, an estimated $2.7 billion in direct
useful in breeding programs.
Published in Crop Sci. 59:1823–1837 (2019).
doi: 10.2135/cropsci2018.08.0492
Table 1. Means of scab traits for the 2015–2016 study of 256 soft red winter wheat lines grown in Lexington, KY. Below the
means, mean squares and levels of significance for genotype, year, and genotype ´ year (G ´ Y), broad-sense heritability (h2),
and the 90% confidence interval (lower limit [LL] and upper limit [UP]) are shown for each trait.
Trait†
Parameter HD PH Rating SEV INC Index FDK DON
Julian d cm 0–9 ———————————————— % ———————————————— mg kg−1
Means
2015 129.8a‡ 87.7a 6.8a 37.3a 80.3a 31.5a 6.9a 8.9b
2016 123.4b 83.7b 3.8b 22.5b 77.0b 17.3b 5.3b 13.2a
ANOVA
Genotype 28.9* 156.5* 4.4* 243.0* 364.5* 276.5* 24.3* 60.4*
Year 10,554.1* 4207.6* 2355.9* 56,177.3* 2899.2* 50,892.7* 643.7* 4522.4*
G ´ Y 10.5* 33.1* 1.7* 166.8* 288.5* 196.8* 7.6* 14.4*
CV 1.6 5.2 20.0 33.1 16.7 41.3 27.7 28.0
Broad-sense heritability
h2 0.64 0.79 0.60 0.31 0.20 0.28 0.69 0.77
LL 0.59 0.75 0.55 0.21 0.08 0.17 0.64 0.73
UL 0.71 0.83 0.69 0.45 0.35 0.42 0.75 0.81
* Significant at the 0.05 probability level.
† HD, heading date; PH, plant height; rating, Fusarium head blight rating; SEV, severity; INC, incidence; FDK, Fusarium-damaged kernels; DON, deoxynivalenol.
‡ Within columns, means followed by a common letter are not significantly different according to a t test (0.05).
Table 2. Pearson correlations among traits evaluated in a 256 entry wheat mapping panel in 2015 and 2016 in Lexington, KY.
Correlations from 2015 are above the diagonal; 2016 correlations are below the diagonal.
2015
Year Traits† HD PH Rating SEV INC Index FDK DON
2016 HD – 0.26** 0.08ns‡ 0.34** 0.20** 0.33** 0.06ns 0.13**
PH 0.51** – −0.38** −0.20** −0.30** −0.25** −0.32** −0.29**
Rating −0.22** −0.35** – 0.59** 0.73** 0.67** 0.50** 0.43**
SEV −0.30** −0.34** 0.50** – 0.56** 0.97** 0.52** 0.46**
INC 0.14** 0.04ns 0.11* 0.03ns – 0.72** 0.46** 0.42**
Index −0.17** −0.26** 0.47** 0.86** 0.51** – 0.56** 0.50**
FDK −0.29** −0.43** 0.56** 0.50** −0.01ns 0.42** – 0.71**
DON 0.11* −0.13** 0.35** 0.36** −0.02ns 0.29** 0.59** –
* Significant at the 0.05 probability level.
† HD, heading date; PH, plant height; rating, Fusarium head blight rating; SEV, severity; INC, incidence; FDK, Fusarium-damaged kernels; DON, deoxynivalenol.
‡ ns, nonsignificant.
‡ Chr., chromosome.
−5.04% of the mean for the trait. Two SNPs were associ- are either important scab resistance genes (Fhb1) or funda-
ated with INC on chromosomes 2B and 7B, with effects mental growth and development genes that have been
of −3.91 and 5.02%, respectively, in 2015. The SNPs asso- implicated in FHB studies (e.g., Ppd-D1; Islam et al.,
ciated with FDK showed effects of −1.05 and −1.09% for 2016). Using the average of each trait, the population was
chromosomes 2A and 5B in 2015, respectively. classified for each allelic form, and levels of PH, HD, FHB
In 2016, the GWAS highlighted a total of 16 SNPs rating, FDK, and DON were calculated for each QTL.
associated with the traits in this study. Four SNPs on The remaining traits and a complete QTL description for
chromosomes 1B and 5A were associated with rating in each line are in the supplemental material (Supplemental
2016, with effects ranging from −0.37 to 0.38%. These Tables S4, S5, and S6). It would have been desirable to test
results differ from the 2-yr GWAS, where SNPs on chro- for the presence of resistance alleles at the newly identi-
mosomes 7A and 7B were associated with rating. Similar fied QTL from ‘Bess’ and ‘Neuse’ reported in Petersen
to 2015, four SNPs associated with SEV in 2016 had et al. (2016). However, when the TCAP population was
effects ranging from −2.51 to 1.37%. Two small-effect genotyped in 2013, these markers were not available (Gina
SNPs were associated with FDK in 2016, explaining 0.63 Brown-Guedira, personal communication, 2019).
and 0.78% of the variation observed. For DON, SNPs on Two hundred and twenty-nine genotypes did not
chromosomes 7D and 5B with effects of 2.53 and −1.47%, have resistance alleles at Fhb1, whereas only 19 panel
respectively, were detected in 2016. entries did. There was no significant difference between
The impact of known QTL on the genotypic response the group of genotypes with Fhb1-S alleles and the group
to FHB rating, FDK, and DON are presented in Table 4. with Fhb1-R alleles for the traits evaluated.
The QTL analyzed in the population were Fhb1, height In the mapping panel, lines with the dwarfing allele
(Rht-B1 and Rht-D1), vernalization (Vrn-A1, Vrn-B1, and Rht-B1b, on average, showed 7.3 and 16.5% lower FHB
Vrn-D3), and photoperiod (Ppd-A1, Ppd-B1, and Ppd-D1) rating and DON, respectively, than lines with the wild-
genes. These QTL were chosen for analysis because they type alleles. Surprisingly, no differences were found
Table 5. Means for scab traits for contrasting alleles at single nucleotide polymorphisms (SNPs) M9432, M6959, and M11423,
associated with Fusarium-damaged kernels (FDK) and deoxynivalenol (DON) in 250 soft red winter wheat lines grown in 2015
and 2016 in Lexington, KY. The number of lines for TT and AA genotypes at each SNP is shown in parentheses.
M9432 (Chr. 5B) M6959 (Chr. 4A) M11423 (Chr. 6B)
Trait† TT (174) AA (76) TT (183) AA (65) TT (213) AA (35)
FDK (%) 5.7b‡ 6.9a 6.0a 6.3a 5.9b 7.2a
DON (mg kg−1) 10.4b 12.5a 10.7b 12.2a 10.6b 13.8a
Rating (0–9) 5.2a 5.5a 5.3a 5.3a 5.2a 5.5a
SEV (%) 29.1a 31.2a 29.6a 30.3a 29.3b 32.8a
INC (%) 78.3a 79.3a 78.4a 79.1a 77.7b 83.2a
Index (%) 23.7a 25.8a 24.1a 25.0a 23.6b 28.0a
† Rating, Fusarium head blight rating; SEV, severity; INC, incidence.
‡ Within rows at each SNP, means followed by a common letter are not significantly different according to a t test (p < 0.05).
Table 6. Performance of the 20 soft red winter wheat breeding lines and cultivars with the lowest deoxynivalenol (DON)
concentration and their respective plant height (PH), heading date (HD), and genotypes at these quantitative trait loci (QTL):
Rht-B1, Rht-D1, Vrn-A1, Vrn-B1, Vrn-D3, Ppd-A1, Ppd-B1, Ppd-D1, and Fhb1. Plant height and heading date are average values
for 2015 and 2016 in Lexington, KY.
QTL†
Genotype DON Rank PH HD Rht-B1 Rht-D1 Vrn-A1 Vrn-B1 Vrn-D3 Ppd-A1 Ppd-B1 Ppd-D1 Fhb1
mg kg−1 cm Julian d
IL06-13708 3.7 1 90.2 123.0 b† a W W vrn-D3a-early b b b S
04620A1-1-7-4 4.0 2 85.7 124.5 b a W W vrn-D3a-early b a b S
KY02C-2215–02 4.5 3 83.2 126.8 a a W-short W vrn-D3b a b a S
MO080584 4.6 4 110.5 132.5 b a W W vrn-D3a-early a b b S
07287RA1-14 4.7 5 84.5 123.3 b a W W vrn-D3a-early a b a S
KY02C-1121-75 4.7 6 89.5 127.8 a b W W vrn-D3b a b b S
IL08-34020 4.8 7 82.6 125.3 b a W W vrn-D3a-early a b a S
KY02C-1058-03 4.8 8 87.0 128.3 a b W W vrn-D3b b b a S
MO081699 5.0 9 93.4 124.3 a a W W vrn-D3b a b a S
05287A1-1-13 5.0 10 83.2 123.0 b a W W vrn-D3b a b b S
OH08-207-33 5.1 11 83.8 123.3 a a W W vrn-D3a-early b a b S
MO100745 5.1 12 100.3 133.3 b a W W vrn-D3a-early a b b S
IL06-7550 5.2 13 90.2 126.8 b a W W vrn-D3a-early b b b S
04719A1-16-1-1-7 5.2 14 79.4 124.5 b a W W vrn-D3a-early a b a S
KY05C-1381-77-7-5 5.3 15 87.6 124.8 b a W W vrn-D3b b b a S
MO090574 5.4 16 95.3 124.8 a a W W vrn-D3b a b a S
OH08-234-4 5.5 17 90.2 125.8 a a W W vrn-D3a-early a – b S
KY06C-1003-139-8-3 5.6 18 80.0 124.8 b a W W vrn-D3b b b a S
IL01-11934 5.7 19 76.2 124.0 b a W W vrn-D3a-early a b b S
KY03C-2047-06 5.7 20 80.7 124.0 a b W W vrn-D3b a b a S
Avg. 5.0 87.7 125.7
† a, wild-type allele for Rht-B1 and Rht-D1; b, semidwarf allele for Rht-B1 and Rht-D1; W, winter allele for Vrn-A1, Vrn-B1; W-short, short vernalization requirement; Vrn-D3a-
early, early flowering; Vrn-D3b, wild type allele; a, photoperiod-insensitive allele for Ppd-A1, Ppd-B1, and Ppd-D1; b, photoperiod-sensitive allele for Ppd-A1, Ppd-B1, and
Ppd-D1; Fhb1-S, susceptible allele for Fhb1; –, no genotypic data.