Content uploaded by Xiaoli Fan
Author content
All content in this area was uploaded by Xiaoli Fan on Apr 10, 2015
Content may be subject to copyright.
QTLs for flag leaf size and their influence on yield-related
traits in wheat (Triticum aestivum L.)
Xiaoli Fan •Fa Cui •Chunhua Zhao •Wei Zhang •
Lijuan Yang •Xueqiang Zhao •Jie Han •Qiannan Su •
Jun Ji •Zongwu Zhao •Yiping Tong •Junming Li
Received: 31 July 2014 / Accepted: 29 December 2014
ÓSpringer Science+Business Media Dordrecht 2015
Abstract Flag leaf-related traits (FLRTs) are deter-
minant traits affecting plant architecture and yield
potential in wheat (Triticum aestivum L.). In this
study, a recombinant inbred line population with 188
lines, derived from the cross between Kenong9204
and Jing411, was developed to identify quantitative
trait loci (QTL) for flag leaf width (FLW), length
(FLL), and area (FLA) under both low nitrogen and
high nitrogen treatments. A total of 38 QTLs were
detected in eight environments (year 9loca-
tion 9treatment). Of these, two QTLs for FLW on
chromosomes 4B and 6B (qFlw-4B.3 and qFlw-6B.2)
and one for FLA on chromosome 5B (qFla-5B) were
major stable QTLs. Both phenotypic and QTL map-
ping analyses indicated that FLW was the major
contributor to flag leaf size. To investigate the genetic
relationship between FLRTs and yield-related traits
(YRTs) at the QTL level, both unconditional and
multivariable conditional QTL mapping for YRTs
with respect to FLRTs were conducted. Twelve QTL
clusters simultaneously controlling FLRTs and YRTs
were identified. In comparison with unconditional
QTL mapping, conditional QTL mapping analysis
revealed that most but not all the QTL for YRTs were
improved by FLRTs. At the QTL level, FLA had the
greatest contribution to YRTs, followed by FLW and
FLL. This study provided a genetic foundation from
which to obtain desirable plant architecture and
improve yield potential in wheat breeding programs.
Xiaoli Fan and Fa Cui have contributed equally to this work.
Electronic supplementary material The online version of
this article (doi:10.1007/s11032-015-0205-9) contains supple-
mentary material, which is available to authorized users.
X. Fan F. Cui (&)C. Zhao W. Zhang
J. Ji J. Li (&)
Center for Agricultural Resources Research, Institute of
Genetics and Developmental Biology, Chinese Academy
of Sciences, Shijiazhuang 050022, China
e-mail: facui@sjziam.ac.cn
J. Li
e-mail: ljm@sjziam.ac.cn
X. Fan
e-mail: fanjie198656@aliyun.com
X. Fan J. Han Q. Su
University of Chinese Academy of Sciences,
Beijing 100049, China
F. Cui C. Zhao W. Zhang X. Zhao
J. Ji Y. Tong J. Li
State Key Laboratory of Plant Cell and Chromosome
Engineering, Chinese Academy of Sciences,
Beijing 100101, China
L. Yang Z. Zhao
Xinxiang Academy of Agricultural Sciences,
Xinxiang 453000, China
123
Mol Breeding (2015) 35:24
DOI 10.1007/s11032-015-0205-9
Keywords Flag leaf-related traits Yield potential
Quantitative trait locus Conditional QTL mapping
Wheat
Abbreviations
FLRTs Flag leaf-related traits
FLW Flag leaf width
FLL Flag leaf length
FLA Flag leaf area
YRTs Yield-related traits
SL Spike length
SN Spikelet number
KN Kernel number
KW Kernel weight per spike
KJ-RILs Recombinant inbred line population
derived from the cross between
Kenong9204 and Jing411
Introduction
Wheat (Triticum aestivum L.) is a staple food world-
wide, and the improvement of its yield potential is an
ultimate goal of wheat breeding programs. As they
exhibit important morphological traits for maintaining
the ideotype, leaves play a crucial role in determining
yield and facilitating photosynthesis (Tsukaya 2006).
The flag leaf is the last leaf to emerge before heading
and is regarded as the major source of carbohydrates
stored in grains (Gladun and Karpov 1993). Therefore,
understanding the genetic mechanism underlying flag
leaf characteristics is necessary to increase the grain
output in wheat breeding programs.
Flag leaf traits are controlled by multiple genes and
strongly influenced by environmental factors (Kobay-
ashi et al. 2003). In recent decades, numerous
quantitative trait loci (QTLs) for flag leaf-related
traits (FLRTs) have been reported in cereal crops
(Kobayashi et al. 2003; Verma et al. 2004; Abdelkha-
lik et al. 2005; Xue et al. 2008,2013; Zeng et al. 2009;
Wang et al. 2012). A large proportion of these reports
focused on the genetic basis of physiological traits,
such as photosynthesis capacity (Zhao et al. 2008),
chlorophyll content (Guo et al. 2008; Xue et al. 2008),
and leaf senescence (Verma et al. 2004). Considering
its determinant effects on photosynthesis and the
accumulation of assimilates (Horton 2000), assessing
the optimal source size and sink capacity has attracted
increasing attention; thus, several QTL mapping
analyses for flag leaf morphological traits, such as
flag leaf width (FLW), length (FLL), and area (FLA),
were also conducted in rice (Zeng et al. 2009; Wang
et al. 2012), maize (Tian et al. 2011), barley (Xue et al.
2008), etc. In wheat, Xue et al. (2013) reported the fine
mapping of a QTL for FLW on chromosome 5A,
which was tightly linked to a major QTL for type I
resistance to Fusarium head blight.
Numerous studies have shown a high correlation
between FLRTs and yield components. Mei et al.
(2005) detected two pleiotropic genomic regions on
chromosomes 3 and 4 in rice, which were associated
with FLL, FLW, plant height, and spikelet number.
Ding et al. (2011) reported the fine mapping of a QTL
that simultaneously affects FLW, spikelet number,
and root volume in rice. Wang et al. (2011,2012)
detected four QTL clusters that contribute to FLL,
FLW, panicle weight, and secondary branch number
in rice, and this group subsequently fine mapped one
major pleiotropic QTL for FLL and yield-related
traits. These results implied that pleiotropic and linked
loci may simultaneously affect FLRTs and yield
potential; however, few results support this hypothesis
in wheat.
Multivariable conditional analysis was recently
used to investigate the genetic relationships between
related traits at the QTL level (Wen and Zhu 2005),
such as yield and its related traits (Guo et al. 2005),
oil content and its related traits (Zhao et al. 2006),
plant height and its related traits (Cui et al. 2011),
etc. For providing the cogent evidence to understand
the interdependence between FLRTs and yield, it is
an available method to conduct conditional QTL
analysis for these traits. This evidence could support
a theoretical basis from which breeding programs
could select desirable cultivars and obtain desired
yields.
In this study, unconditional and conditional QTL
mapping for FLW, FLL, FLA, and four yield-related
traits (YRTs) in wheat in multiple environments was
conducted. The objectives of this study were to (1)
identify major and stable QTLs for FLRTs in different
environments, (2) specify the overlapping genomic
regions simultaneously affecting FLRTs and YRTs,
and (3) dissect the genetic effect of FLRTs on the
expression of QTL for YRTs at the individual QTL
level.
24 Page 2 of 16 Mol Breeding (2015) 35:24
123
Materials and methods
Experimental materials and evaluation
A recombinant inbred line (RIL) population derived
from a cross between Kenong9204 (KN9204) and
Jing411 (J411) was used in this study (represented by
KJ-RILs). KN9204 is a representative cultivar in the
North China Plain and was released in 2002 by the
Center for Agricultural Resources Research, Institute
of Genetics and Developmental Biology, Chinese
Academy of Sciences, Hebei, China. This cultivar has
a larger FLA than J411 (Figure S1). In the present
study, 188 randomly sampled lines of the 427 KJ-RILs
were used for genotyping and phenotyping.
The RILs and their parents were evaluated in
Shijiazhuang in 2011–2012 (T1: 37°530N, 114°410E,
altitude 54 m) and in 2012–2013 (T2), in Beijing in
2012–2013 (T3: 40°060N, 116°240E, altitude 41 m),
and in Xinxiang in 2012–2013 (T4: 35°270N,
113°480E, altitude 95 m). Two nitrogen treatments
were applied in each trial [low nitrogen (LN) and high
nitrogen (HN)] for a total of eight environments
(year 9location 9treatment) designated E1, E2, E3,
E4, E5, E6, E7, and E8. In each HN plot, 300 kg ha
-1
of diamine phosphate and 225 kg ha
-1
of urea were
applied before sowing, and 150 kg ha
-1
of urea was
applied at the elongation stage every year. In the LN
plots, no N fertilizer (N-deficient) was applied during
the growing period. A randomized block design with
two replications was used in each of the eight
environments, and 40 seeds were hand-planted in
each row of a two-row plot with 2-m-long rows spaced
0.25 m apart. All of the recommended agronomic
practices were followed in each of the trials except for
the nitrogen fertilization treatment described above.
Three FLRTs and four YRTs were evaluated in this
study. For each plot, the main shoots of five plants in
the center of each row were randomly chosen to
measure the FLL and FLW at 15 days after flowering
and to investigate the spike length (SL), spikelet
number (SN), kernel number (KN), and kernel weight
per spike (KW) at physiological maturity. The FLL
and FLW measurements were taken at the longest and
widest parts of the flag leaf, respectively. The flag leaf
area (FLA) (= FLL 9FLW 90.83) was also evalu-
ated. SL was measured from the base of the spike to
the tip (excluding the awns). SN was evaluated by
calculating the total number of spikelets on the main
shoot for each plant. KN was measured by calculating
the total number of kernels per spike of the main shoot
for each plant. KW was evaluated by calculating the
total kernel weight of the main shoot for each plant.
The average value of each plot represented the
phenotype value.
Data analysis and QTL mapping
The analysis of variance (ANOVA) and the calcula-
tion of phenotypic correlation coefficients between
traits were performed using SPSS 19.0 (SPSS,
Chicago, MI, USA; http://en.wikipedia.org/wiki/
SPSS). With an aim to estimate genetic variance and
phenotypic variance of the three FLRTs, the data from
each environment were assembled individually
according to the QTLData format of QGAStation 2.0
(http://ibi.zju.edu.cn/software/qga/). The first column
represented the block (two replications), the second
column represented the genotype (188 KJ-RILs), and
the following columns represented the three FLRTs
investigated in this study. The ‘block effect’ was
attributed a value of ‘YES’ to the ‘Ge Var’ analysis.
The output file provided information on genetic vari-
ance, phenotypic variance, and error variance of each
trait in the corresponding environment. The condi-
tional phenotypic values of YRTs with respect to
FLRTs were also evaluated using QGAStation 2.0.
The raw data from each environment were assembled
as follows: The first column represented the block
(two replications), the second column represented the
genotype (188 KJ-RILs), and the following columns
were trait data, specifically the three FLRTs and each
of the four YRTs of SL, SN, KN or KW. ‘Conditional
Final’ was conducted and the output file provided
information on conditional phenotypic values of
y
(YRTs|FLRTs)
, which reflect the genetic variation of
YRTs values without the influence of the FLRTs in the
corresponding environment.
The genetic map, specifically the KJ-derived maps,
which included 591 loci spanning 3,930.7 cM with an
average density of one marker per 6.7 cM, was used to
screen QTLs in the present study. More information on
this map has been provided by Cui et al. (2014). The
directly investigated and the conditional phenotypic
values were used for conventional and conditional
QTL mapping analyses, respectively. The inclusive
composite interval mapping performed with IciMap-
ping 3.3 (http://www.isbreeding.net/) was used to
Mol Breeding (2015) 35:24 Page 3 of 16 24
123
detect putative additive QTLs. The phenotypic values
of the 188 RILs in E1, E2, E3, E4, E5, E6, E7, and E8
were used for individual environment QTL mapping.
The walking speed chosen for all QTL was 1.0 cM,
and the Pvalue inclusion threshold was 0.001. The
threshold LOD scores were calculated using 1,000
permutations with a type 1 error of 0.05.
QTL nomenclature
All of the QTLs were designated as follows: An
italicized lowercase letter ‘q’ denotes ‘QTL’; the
letters following the qand preceding the dash repre-
sent the abbreviation of the corresponding trait; the
letters and numbers following the dash represent the
wheat chromosome on which the corresponding QTLs
are distributed; and if several QTLs associated with a
certain trait were dispersed along a certain chromo-
some, a serial number such as 1, 2, 3, etc., is used after
the chromosome name to describe their order (from
the short arm to the long arm). When two or more
QTLs associated with the same trait with overlapping
confidence intervals were detected in different envi-
ronments, they were considered to be congruent QTLs.
We defined a major QTL as a QTL with an LOD
value of [3.0 and a phenotypic variance contribution
of[10 % (on average); we defined a stable QTL as a
QTL that showed significance in at least five of the
eight environments (E1, E2, E3, E4, E5, E6, E7, and
E8).
Results
Phenotypic variation and correlations
between traits
ANOVA showed that FLW, FLL, and FLA were all
significantly affected by genotype, environment, and
G9E interactions (Table S1). The phenotypic per-
formance and the h2
Bfor the three FLRTs in the KJ-
RILs population and their parents in the eight envi-
ronments are shown in Electronic Supplementary
Material Table S2. KN9204 presented larger, wider,
and longer flag leaves compared with those of J411 in
all eight environments. The coefficients of variation
(CVs) ranged from 6.43 % of FLL in E8 to 16.08 % of
FLA in E3. The absolute values of skewness and
kurtosis for the three FLRTs were less than 1.0 in most
cases. This result indicated that all of the traits were
controlled by multiple genes. Strong transgressive
segregation exceeding the limits of both parents was
observed, indicating that alleles with positive effects
were distributed between the two parents. The esti-
mated broad-sense heritability (h2
B) of the four traits
ranged from 30.65 to 76.94 %.
Based on the mean value of all environments,
highly significant and positive correlations were
observed between FLW and FLA, FLL and FLA,
and FLRTs and most YRTs (SL, SN, KN, and KW)
(Table S3). The correlation coefficients of FLW–FLA
(0.84) were higher than those of FLL–FLA (0.57),
which suggested that FLW should be the main
contributor to affect flag leaf size.
Putative additive QTLs for FLRTs
A total of 38 putative additive QTLs associated with
FLW, FLL, and FLA were detected in the eight
individual environments (Table 1; Fig. 1). These
QTLs were distributed across 12 chromosomes. Of
these, 27 QTLs were mapped to the B genome, six
were mapped to the D genome, and five were mapped
to the A genome. These QTLs individually explained
3.96–27.68 % of the phenotypic variance with LOD
values ranging from 2.51 to 7.68. Fifteen QTL
(39.47 %) were reproducibly detected in at least two
environments. Fifteen QTLs (39.47 %) individually
accounted for more than 10 % of the phenotypic
variance with average LOD values[3.0 (major QTL);
three of these QTLs (qFlw-4B.3,qFlw-6B.2, and qFla-
5B) also showed stability across more than five
environments and were thus characterized as major
stable QTLs.
Fourteen QTLs were identified for FLW, nine of
which carried the favorable alleles from KN9204 that
increase the FLW (Table 1). Of these QTLs, five on
chromosomes 4B (three QTLs) and 6B (two QTLs)
were major QTLs that individually accounted for
[10 % of the phenotypic variance; three QTLs on
chromosomes 2D, 4B, and 6B exhibited high stability
across environments. qFlw-4B.3 and qFlw-6B.2, two
major stable QTLs, both showed significant effects in
seven different environments, individually exhibiting
5.27–22.34 and 10.52–18.66 % of the phenotypic
variance, respectively.
24 Page 4 of 16 Mol Breeding (2015) 35:24
123
Table 1 Unconditional QTLs with significant additive effects for FLL, FLW, and FLA in the KJ population
Traits QTLs
a
Left markers Right markers Environments
b
LOD value PVE%
c
Additive effects
d
Range Mean Range Mean Range Mean
FLW (mm) qFlw-2B XwPt-2430 Xme12em26.1 E6 3.06 3.06 3.96 3.96 0.30 0.30
qFlw-2D.1 XwmpE08 Xwmc601 E1/E3/E4/E5/E6 3.43/3.73/8.73/3.10/4.37 4.67 6.96/5.95/10.37/5.58/5.49 6.87 –0.37/–0.29/–0.50/
–0.38/–0.36
–0.38
qFlw-2D.2 Xmag3596 IN10 E4 5.39 5.39 5.89 5.89 0.38 0.38
qFlw-2D.3 Xksum244 Xksum174.1 E6 3.93 3.93 5.26 5.26 0.35 0.35
qFlw-3B Xbarc229 XwPt-5769 E3 3.72 3.72 6.24 6.24 0.30 0.30
qFlw-4B.1 Leaftype Xme16em26 E4/E8 4.04/4.68 4.36 9.20/18.24 14.03 0.48/0.62 0.55
qFlw-4B.2 XwPt-6149 Xmag2055 E3 7.29 7.29 14.98 14.98 0.46 0.46
qFlw-4B.3 Xcnl10 XwPt-1046 E1/E2/E4/E5/
E6/E7/E8
5.76/5.29/3.91/4.20/
10.98/7.87/2.61
5.80 15.63/10.91/6.96/9.86/
22.34/16.00/5.27
12.42 0.56/0.56/0.41/0.50/
0.72/0.56/0.33
0.52
qFlw-5B XwPt-9103 Xbarc142 E3/E4/E5/E6 5.12/4.27/4.04/8.00 5.36 8.30/5.31/7.83/11.65 8.27 0.34/0.36/0.44/0.52 0.42
qFlw-6A Xme16em12.2 XwPt-664792 E4/E6/E7/E8 2.81/4.81/3.89/3.70 3.80 3.04/6.91/8.56/6.17 6.17 –0.28/–0.41/–0.42/–0.37 –0.37
qFlw-6B.1 Xwmc487 Xme7em19.2 E4 4.79 4.79 10.84 10.84 –0.52 –0.52
qFlw-6B.2 XwPt-5037 Xme9em2.2 E1/E2/E3/E4/
E5/E6/E7/E8
4.80/4.77/10.62/10.31/
6.11/9.86/5.46/6.02
7.24 10.52/10.91/18.66/
14.00/11.97/13.71/
11.67/13.06
13.06 –0.46/–0.56/–0.51/–0.59/
–0.55/–0.56/–0.48/–0.56
–0.53
qFlw-7B XwPt-7108 XwPt-6495 E8 2.51 2.51 4.05 4.05 –0.29 –0.29
qFlw-7D Xbarc126 Xcwem52 E4/E6 4.26/2.72 3.49 4.70/12.08 8.39 0.34/0.53 0.44
FLL (mm) qFll-1B.1 XwPt-1684 XwPt-2988 E3 5.13 5.13 11.39 11.39 4.97 4.97
qFll-1B.2 XwPt-1328 XwPt-4107 E1 7.68 7.68 14.03 14.03 5.05 5.05
qFll-1B.3 Xgwm374 Xwmc406 E2 3.26 3.26 6.72 6.72 3.87 3.87
qFll-2B.1 Xwmc332 XwPt-0473 E2/E4 3.19/5.80 4.50 6.48/13.39 9.94 –3.53/–6.44 –4.99
qFll-2B.2 XwPt-2430 Xme12em26.1 E1 4.89 4.89 10.83 10.83 –4.32 –4.32
qFll-4A.1 Xgpw2331 XwPt-4230 E1 7.28 7.28 13.14 13.14 –4.70 –4.70
qFll-4A.2 Xmag3886 XwPt-9418 E5 2.80 2.80 5.65 5.65 2.82 2.82
qFll-4B.1 Xme16em26 Xwmc657 E5/E6 4.58/5.50 5.04 10.41/13.97 12.19 –3.79/–4.74 –4.27
qFll-4B.2 XwPt-1046 XwPt-5559 E7 2.99 2.99 7.48 7.48 –3.64 –3.64
qFll-5B XwPt-9103 Xbarc142 E2/E6 4.17/2.65 3.41 8.91/5.91 7.41 4.13/3.08 3.61
qFll-5D Xcfd189 Xgwm174 E5 3.22 3.22 6.75 6.75 –3.04 –3.04
FLA (mm
2
)qFla-1B.1 XwPt-1684 XwPt-2988 E2/E3 3.73/5.20 4.47 6.26/6.85 6.56 65.72/80.13 72.93
qFla-1B.2 XwPt-1328 XwPt-4107 E1 6.49 6.49 12.58 12.58 71.95 71.95
qFla-1B.3 Xissr811.3 Xme7em10.2 E4 4.76 4.76 8.16 8.16 115.25 115.25
qFla-2B PPO33 XwPt-2430 E4 3.08 3.08 5.00 5.00 –90.73 –90.73
qFla-2D Xksum244 Xksum174.1 E6 2.58 2.58 4.59 4.59 64.90 64.90
qFla-4A Xgpw2331 XwPt-4230 E1 4.30 4.30 7.71 7.71 –54.16 –54.16
qFla-4B.1 XwPt-6149 Xmag2055 E3/E4/E7 3.31/5.04/2.98 3.78 7.04/12.07/8.68 9.26 70.63/137.93/85.64 98.06
Mol Breeding (2015) 35:24 Page 5 of 16 24
123
Eleven QTLs for FLL were detected (Table 1).
qFll-2B.1,qFll-4B.1, and qFll-5B could be repeatedly
identified in two different environments. Favorable
alleles of the first two QTLs were contributed by J411,
but the alleles of qFll-5B which increased FLL were
from KN9204. qFll-4B.1 was a major QTL that
explained 10.41–13.97 % of the phenotypic variation
with LOD values of 4.58–5.50.
Thirteen QTLs were detected for FLA, five and
seven of which were co-localized with those for FLL
and FLW, respectively (Table 1; Fig. 1). qFla-5B, one
major stable QTL, was effective in E1, E2, E3, E4, E5,
E6, and E8, and co-localized with qFll-5B and qFlw-
5B, with the favorable alleles from KN9204 simulta-
neously increasing FLA, FLL, and FLW. qFla-6B.2,
the other major QTL for FLA, was co-localized with
the major stable QTL for FLW (qFlw-6B.2) and with
J411-derived alleles that increase FLA and FLW.
QTL clusters for FLRTs and YRTs
The conventional QTL mapping for four YRTs (SL,
SN, KN, and KW) was conducted in eight individual
environments. Twelve QTL clusters simultaneously
affected YRTs and FLRTs were identified on chro-
mosomes 1B (three clusters), 2D (two clusters), 4A
(one cluster), 4B (three clusters), 5B (one cluster), 6B
(one cluster), and 7D (one cluster). KN9204- or J411-
derived alleles in C1, C2, C4, C6, C10, C11, and C12
that increased FLRTs were also found to improve
YRTs (Table 2; Fig. 1). Alleles of the remaining QTL
clusters from KN9204 or J411 showed additive effects
that increase/decrease FLRTs but decrease/increase
YRTs.
Of the QTLs for SL in the five QTL clusters, the
major QTL qSl-2D in C5 was detected in three
environments (E1, E2, and E5), with KN9204-derived
alleles decreasing SL but increasing FLW and KW;
qSl-4B, another major QTL in C7, was co-localized
with the other two major QTLs (qFlw-4B.1 and qFll-
4B.1), with the favorable alleles from KN9204 that
increase the SL and FLW and those from J411 that
increase FLL (Tables 2,3; Fig. 1).
Of the QTLs for SN in the seven QTL clusters, qSn-
7D in C12 showed significance in six different
environments and accounted for 7.43–15.81 % of the
phenotypic variance, with alleles from KN9204
increasing FLW, SL, SN, and KW (Tables 2,3;
Fig. 1).
Table 1 continued
Traits QTLs
a
Left markers Right markers Environments
b
LOD value PVE%
c
Additive effects
d
Range Mean Range Mean Range Mean
qFla-4B.2 Xcnl10 XwPt-1046 E2 3.85 3.85 6.21 6.21 64.63 64.63
qFla-5B XwPt-9103 Xbarc142 E1/E2/E3/E4/E5/E6/E8 3.38/7.37/3.87/
5.14/3.61/7.84/3.67
4.48 6.81/13.95/6.63/
9.25/8.11/16.65/9.11
10.07 51.02/96.35/68.43/
120.60/68.13/
123.55/83.11
87.60
qFla-6A Xme16em12.2 XwPt-664792 E6 4.12 4.12 8.00 8.00 –87.78 –87.78
qFla-6B.1 Xwmc756 XwPt-9642 E4/E5/E6 8.23/5.40/6.80 6.81 15.28/11.87/12.13 13.09 –154.46/–82.22/–105.66 –85.94
qFla-6B.2 Xme9em2.2 Xswes199 E1/E2/E3/E7 4.08/8.39/5.45/4.51 5.61 7.27/15.30/11.55/11.55 11.42 –52.64/–100.65/
–89.82/–98.32
–85.36
qFla-6B.3 Xme7em11.2 Xcnl64 E8 6.01 6.01 27.68 27.68 –145.75 –145.75
a
A putative major QTL is marked by bold typeface and is characterized by a mean LOD [3.0 and a mean PVE [10 %; a putative stable QTL is underlined when this locus can
be detected in at least five of the eight environments
b
E1, E2, E3, E4, E5, E6, E7, and E8 indicate trial 1 (2011–2013, Shijiazhuang) LN, trial 1 HN, trial 2 (2012–2013, Shijiazhuang) LN, trial 2 HN, trial 3 (2012–2013, Beijing)
LN, trial 3 HN, trial 4 (2012–2013, Xinxiang) LN, and trial 4 HN, respectively
c
PVE indicates the percentage of explained phenotypic variation
d
A positive sign means that the positive alleles come from the KN9204 parent; a negative sign means that the positive alleles come from the J411 parent
24 Page 6 of 16 Mol Breeding (2015) 35:24
123
A stable QTL for KN (qKn-4A) in C6 showed
significance in all eight environments and clustered
with two other QTLs for YRTs (qSl-4A and qSn-4A)
and two QTLs for FLRTs (qFll-4A.1 and qFla-4A),
with the J411-derived alleles simultaneously increas-
ing these traits. qKn-4B in C9, a major stable QTL for
KN, co-localized with QTLs for the three FLRTs
(qFlw-4B.3,qFll-4B.2, and qFla-4B.2) and one QTL
for KW (qKw-4B.3) (Tables 2,3; Fig. 1).
For KW, only qKw-2D.2 in C5 was reproducibly
identified in different environments, which accounted
for 4.69–9.31 % of the phenotypic variance, with
alleles from KN9204 increasing FLW and KW but
decreasing SL (Tables 2,3; Fig. 1). The remaining
QTLs for KW in the four QTL clusters were signif-
icant in only one environment.
Conditional QTLs for YRTs with respect to FLRTs
To dissect the genetic effects of FLRTs on the
expression of QTLs for the YRTs that were detected
in the above-mentioned 12 clusters, we conducted
conditional QTL mapping analysis for the YRTs with
respect to FLRTs. By comparing the additive effect of
the 21 conditional QTLs for YRTs with that of the
unconditional QTLs in these clusters as shown in
Table 2, three cases could be presented: (1) the QTL
for YRTs showed reduced effects or were undetectable
XwPt-5896
XwPt-1482
XwPt-9598XwPt-9103
Xbarc142
Xwmc28
qFll-5B(E2)
qFll-5B(E6)
qFlw-5B(E3)
qFlw-5B(E4)
qFlw-5B(E5)
qFlw-5B(E6)
qFla-5B(E1)
qFla-5B(E2)
qFla-5B(E3)
qFla-5B(E4)
qFla-5B(E5)
qFla-5B(E6)
qFla-5B(E8)
qKn-5B(E1)
5B
Xcfd18
Xcfd183
Xcfd189
Xgwm174
qFll-5D(E5)
5D
Xme16em12.2
XwPt-730772XwPt-664792
Xwmc754
qFlw-6A(E4)
qFlw-6A(E6)
qFlw-6A(E7)
qFlw-6A(E8)
qFla-6A(E6)
6A
Xwmc487
Xme7em19.2
Xwmc737
Xcnl113
Xwmc756
XwPt-9642
Xbarc198
XwPt-5037
Xme9em2.2
Xswes199
Xme12em13.2
Xme7em11.2
Xcnl64
qFlw-6B.1(E4)
qFla-6B.1(E4)
qFla-6B.1(E5)
qFla-6B.1(E6)
qSl-6B(E4)
qFlw-6B.2(E1)
qFlw-6B.2(E2)
qFlw-6B.2(E3)
qFlw-6B.2(E4)
qFlw-6B.2(E5)
qFlw-6B.2(E6)
qFlw-6B.2(E7)
qFlw-6B.2(E8)
qFla-6B.2(E1)
qFla-6B.2(E2)
qFla-6B.2(E3)
qFla-6B.2(E7)
qFla-6B.3(E8)
6B
Xme11em12.1
XwPt-7108
XwPt-5533wPt-6495
qFlw-7B(E8)
7B
Xbarc126
Xcwem52
XwPt-671530
Xgdm67
qFlw-7D(E4)
qFlw-7D(E6)
qSl-7D(E3)
qSn-7D(E3)
qSn-7D(E4)
qSn-7D(E5)
qSn-7D(E6)
qSn-7D(E7)
qSn-7D(E8)
qKw-7D(E6)
7D
XwPt-669239
XwPt-5281
XwPt-3282
XwPt-1684
XwPt-2988
XwPt-3465
XwPt-1176X wPt-9903
XwPt-5765
XwPt-5745X wPt-7529
XwPt-1328
XwPt-8930X wPt-4107
XwPt-5312
XwPt-2052X gwm374
Xgwm11X wmc406
Xwmc128 Xbarc187
Xcfd59 XwPt-2019
XwPt-2614
XwPt-6985
XwPt-4434X wPt-3177
XwPt-0974
Xcinau172
XwPt-3451
Xissr811.3
Xme10em7Xme9em2.1
Xme7em19.1 Xme7em10.2
Xme11em12.2
qFll-1B.1(E3)
qFla-1B.1(E2)
qFla-1B.1(E3)
qSn-1B.1(E2)
qSn-1B.1(E7) qFll-1B.2(E1)
qFla-1B.2(E1)
qSn-1B.2(E1)
qFll-1B.3(E2)
qSn-1B.3(E5)
qKn-1B(E8) qFla-1B.3(E4)
1B
Xme13em3
Xwmc332
XwPt-0473
XwPt-9736
XwPt-0694
PPO33
XwPt-2430
Xme12em26.1
qFll-2B.1(E2) qFll-2B.2(E1)
qFll-2B.3(E4) qFlw-2B(E6)
qFla-2B(E4)
2B
Xgpw361
XwmpE08
Xwmc601
XwPt-666518
Xcfd233
Xswes61
XwPt-5865XwPt-4223
XwPt-730744
XwPt-6847
Xmag3947
XwPt-665317
Xmag3596
XwPt-667476Xbarc228
IN10
Xmag4059 Xmag4089
STS01
XwPt-1554XwPt-731134
XwPt-664745XwPt-671737
Xgpw5215
Xmag2956.1
Xmag2956.2
Xgpw5215
Xksum244
Xksum174.1
qFlw-2D.1(E1)
qFlw-2D.1(E3)
qFlw-2D.1(E4)
qFlw-2D.1(E5)
qFlw-2D.1(E6)
qSn-2D(E7)
qFlw-2D.2(E4)
qSl-2D(E1)
qSl-2D(E2)
qSl-2D(E5)
qKw-2D(E4)
qKw-2D(E6)
qFlw-2D.3(E6)
qFla-2D(E6)
2D
XwPt-8096
Xbarc229
XwPt-5769
qFlw-3B(E3)
3B
Xwmc161
Xgpw2331
XwPt-4230
Xgpw7543
Xwmc760
XwPt-7064
XwPt-4620
XwPt-8657
XwPt-664749X wmc313
XwPt-731166
XwPt-4064
XwPt-4424
Xgpw3079
Xmag3886
XwPt-1155XwPt-9418
qFll-4A.1(E1)
qFla-4A(E1)
qSl-4A(E1)
qSl-4A(E3)
qSl-4A(E4)
qSn-4A(E1)
qKn-4A(E1)
qKn-4A(E2)
qKn-4A(E3)
qKn-4A(E4)
qKn-4A(E5)
qKn-4A(E6)
qKn-4A(E7)
qKn-4A(E8) qFll-4A.2(E5)
4A
Xcau9
Leaftype
Xme16em26
Xwmc657
Xbarc199
Xcfe89
XwPt-6149
Xmag2055
Xmag4087
Xcnl10
XwPt-1046
XwPt-5559
XwPt-1272
qFll-4B.1(E5)
qFll-4B.1(E6)
qFlw-4B.1(E4)
qFlw-4B.1(E8)
qSl-4B(E2)
qFlw-4B.2(E3)
qFla-4B.1(E3)
qFla-4B.1(E4)
qFla-4B.1(E7)
qSn-4B(E5)
qSn-4B(E6)
qSn-4B(E7)
qSn-4B(E8)
qKw-4B.1(E6)
qKw-4B.2(E8)
qFll-4B.2(E7)
qFlw-4B.3(E1)
qFlw-4B.3(E2)
qFlw-4B.3(E4)
qFlw-4B.3(E5)
qFlw-4B.3(E6)
qFlw-4B.3(E7)
qFlw-4B.3(E8)
qFla-4B.2(E2)
qKn-4B(E4)
qKn-4B(E5)
qKn-4B(E6)
qKn-4B(E7)
qKn-4B(E8)
qKw-4B.3(E7)
4B
Fig. 1 The unconditional QTLs conferring FLSTs and YRTs
detected in eight environments. The QTL intervals are
composed of an inner (LOD value C2.5) and an outer (LOD
value equals 2.0) interval. A black rectangle indicates a QTL
associated with FLRTs; a green rectangle indicates a QTL
associated with YRTs. (Color figure online)
Mol Breeding (2015) 35:24 Page 7 of 16 24
123
in the conditional analysis, indicating that the expres-
sion of these QTLs was partially or completely
improved by FLRTs; (2) the QTLs for YRTs showed
increased effects or were detected as novel QTLs that
could not be identified in traditional QTL mapping
analysis, indicating that the expression of the QTL was
partially or completely suppressed by FLRTs; (3) the
genetic effect of the QTL for YRTs was similar to that
of the corresponding unconditional QTL, suggesting
that flag leaf size had little impact on the expression of
the corresponding QTL for YRTs (Table 3).
Of the five traditional QTLs for SL in the above-
mentioned five clusters, FLRTs made no or little
contribution to the expression of qSl-2D. All three
FLRTs improved the expression of qSl-4A in E3 and
E4; in E1, FLW made no contribution to the expression
of qSl-4A, but both FLL and FLA contributed to the
expression of this QTL largely; in E6, FLL suppressed
the expression of qSl-4A. FLL made no contribution to
the expression of qSl-4B, but FLW and FLA sup-
pressed the expression of this QTL in E2; in E4, E5, E6,
E7, and E8, FLL suppressed the expression of qSl-4B
largely. The expression of qSl-6B was affected by all
three FLRTS. FLW, FLL, and FLA partially contrib-
uted to the expression of qSl-7D in E3; however, FLL
suppressed its expression in E5 (Table 3).
Concerning the seven traditional QTLs for SN in
the above-mentioned seven clusters, all three FLRTs
contributed to the expression of qSn-1B.2,qSn-1B.3,
and qSn-4A. However, for qSn-1B.1 and qSn-2D in E7,
FLL made no contribution to the expression of these
two QTLs, and both FLW and FLA suppressed the
expression of qSn-1B.1. The expression of qSn-4B,a
moderate stable QTL for SN, was partially or
completely contributed by FLW and FLL, except in
E4 (in which its expression was suppressed by FLL)
and E8 (in which FLL had no effect on the expression
of qSn-4B); FLA completely contributed to the
expression of qSn-4B in E6 and E8 and suppressed
its expression in E5, but had no effect on its expression
in E7. The expression of qSn-7D, a stable QTL for SN,
had no relationship with FLRTs in most cases;
however, FLW and FLA in E7 and E8 contributed to
the expression of this QTL, and FLL and FLA partially
contributed to the expression of this QTL in E4 and
E5, respectively (Table 3).
Of the four traditional QTLs for KN in the above-
mentioned four clusters, qKn-1B was independent of
FLL and FLW; however, FLA suppressed the expres-
sion of qKn-1B in E8. qKn-4A was a stable QTL for
KN; FLW and FLL partially or completely contributed
to the expression of this QTL in five of the eight
Table 2 The QTL clusters simultaneously affecting flag leaf size and yield-related traits in this study
Clusters Chromosomes Intervals No. of
QTLs
Traits (additive effect, number of
environments)
a
C1 1B XwPt-1684–XwPt-2988 3FLL(1, 1), FLA(?, 2), SN(?,2)
C2 1B XwPt-1328–XwPt-4107 3FLL(1, 1), FLA(1, 1), SN(?,1)
C3 1B Xgwm374–Xwmc406–Xbarc187–XwPt-2614 3 FLL(?,1), SN(?,1), KN(–,1)
C4 2D XwmpE08–Xwmc601–XwPt-666518 2FLW(–, 5), SN(–, 1)
C5 2D Xmag3596–IN10–Xmag4089 3 FLW(?, 1), SL(–,3), KW(?,2)
C6 4A Xwmc161–Xgpw2331–XwPt-4230–
Xgpw7543
5FLL(–, 1), FLA(–, 1), SL(–, 3), SN(–, 1),
KN(–, 8)
C7 4B Leaftype–Xme16em26–Xwmc657 3FLL(–, 2), FLW(1, 2), SL(1,1)
C8 4B Xcef89–XwPt-6149–Xmag2055 5FLW(1, 1), FLA(?, 3), SN(?, 4),
KW(–, 1), KW(?,1)
C9 4B Xmag4087–Xcnl10–XwPt-1046–XwPt-5559 5 FLL(–, 1), FLW(1, 8), FLA(?, 1),
KN(1,5), KW(?,1)
C10 5B XwPt-9103–Xbarc142–Xwmc28 4 FLL(?, 2), FLW(?, 4), FLA(1,7), KN(?,1)
C11 6B Xcnl113–Xwmc756–XwPt-9642 2FLA(–, 3), SL(–, 1)
C12 7D Xbarc126–Xcwem52–XwPt-671530–Xgdm67 4 FLW(?, 2), SL(?, 1), SN(?, 6), KW(?,1)
a
The traits name in bold type indicates that major QTLs were detected for the corresponding traits; the traits name in underline type
indicates that stable QTLs were detected for the corresponding traits
24 Page 8 of 16 Mol Breeding (2015) 35:24
123
environments; FLA improved the expression of this
QTL in all eight environments. qKn-4B was a major
stable QTL for KN; FLW partially contributed to the
expression of this QTL in E5, E7, and E8; FLL made
no contribution to the expression of this QTL in the
five environments in which the effect of qKn-4B was
significant, with the exception of E5, in which FLL
suppressed the expression of qKn-4B; FLA partially or
completely contributed to the expression of qKn-4B in
most cases. qKn-5B was induced by FLW and FLA;
FLL made no contribution to the expression of this
QTL (Table 3).
Concerning the five traditional QTLs for KW in the
above-mentioned seven clusters, FLW and FLA
largely improved the expression of qKw-2D in E4; in
E5, all three FLRTs suppressed the expression of qKw-
2D; however, qKw-2D was independent of FLW and
FLA, and its expression was suppressed by FLL in E6.
qKw-4B.1 was independent of FLW and FLA; FLL
largely improved the expression of this QTL. The
expression of qKw-4B.2 was suppressed by FLL in E8;
FLW and FLA largely improved the expression of this
QTL in E8; in addition, FLA suppressed the expres-
sion of qKw-4B.2 in E4. All three FLRTs improved the
expression of qKw-4B.3 in E7; however, FLW (in E2
and E4) and FLA (in E2) suppressed the expression of
this QTL. All three FLRTs made large contributions to
the expression of qKw-7D (Table 3).
Discussion
FLRTs: their genetic relationships and important
QTL clusters
Flag leaves, as the most important source leaves,
contribute the 41–43 % of the carbohydrates for grain
filling (Sharma et al. 2003). The morphological
characteristics of flag leaves such as leaf size and leaf
shape, are therefore critical factors in determining a
desirable plant type (Tsukaya 2006). Generally, leaf
size regulation is governed by two main dimensions:
Length and width (Tsukaya 2006) which are sensitive
to environmental factors (Tsukaya 2005). In this study,
half of the 14 QTLs for FLA were co-localized with
QTLs for FLW, which was consistent with the
phenotypic correlation analysis results listed in Elec-
tronic Supplementary Material Table S3 and further
confirmed that FLW was more crucial than FLL in
determining the FLA available for absorbing sufficient
light energy (Tsukaya 2006).
Genes and QTLs for multiple FLRTs that were
linked or showed pleiotropic effects were reported
previously. Yue et al. (2008) detected two intervals
that harbor QTL clusters for FLL and FLW on
chromosomes 3 and 4 in rice; Wang et al. (2012)
reported a hotspot QTL region in rice associated with
FLL and FLW on chromosome 1 and confirmed that it
contained two tightly linked QTLs each for FLL and
FLW, respectively. In the present study, nine QTLs for
FLA were found to co-localize with QTLs for FLW or
FLL, three of which (qFla-2B,qFla-4B.2, and qFla-
5B) were co-located with those for FLW and FLL
simultaneously. Interestingly, the interval XwPt-
9103–Xbarc142 on chromosome 5B harbored qFla-
5B,qFlw-5B,qFll-5B, and qKn-5B, with favorable
alleles from KN9204 that consistently increase all the
corresponding traits (Tables 1,2,3). Based on the KJ-
RILs genetic map reported by Cui et al. (2014), the
interval XwPt-9103–Xbarc142 resides in the C-5BL6-
0.29 bin in which a QTL for resistance to spot blotch
caused by Bipolaris sorokiniana (Kumar et al. 2010)
was found. qFla-5B was the only major stable QTL for
FLA detected in this study, which might account for
the genetic basis of the larger flag leaf area of KN9204
than that of J411. The allelic effect of qFla-5B was
evaluated by classifying the lines into two groups
based on whether they carried KN9204 alleles or not in
the interval XwPt-9103–Xbarc142. The alleles from
KN9204 indeed significantly improved FLA at the
level of 0.001 (Fig. 2). The conditional QTL mapping
indicated that the expression of qKn-5B was induced
by FLW and FLA. Therefore, further fine mapping and
cloning of the putative pleiotropic gene increasing KN
by controlling flag leaf size in this interval were of
great value for breeding.
Two major stable QTLs for FLW (qFlw-4B.3 and
qFlw-6B.2) were identified in the interval Xcnl10–
XwPt-1046 of chromosome 4B and the interval XwPt-
5037–Xme9em2.2 of chromosome 6B, respectively.
The confidence interval of qFlw-4B.3 overlapped that
of another major stable QTL for KN (qKn-4B) (Figure
S2). The LOD peaks of both qFlw-4B.3 and qKn-4B
were at Xcnl10 which was reported in durum wheat to
be strongly associated with Lpx-B1, which encodes
lipoxygenase (Hessler et al. 2002). Based on the
conditional QTL mapping analysis, the expression of
qKn-4B was induced by FLW and FLA (Table 3).
Mol Breeding (2015) 35:24 Page 9 of 16 24
123
Table 3 The conditional QTLs for yield-related traits with respect to flag leaf size
Traits QTLs
a
Clusters
b
Intervals marker
c
Additive[E/PVE (%)]
d
YRTs YRTs|FLW YRTs|FLL YRTs|FLA
SL (cm) qSl-2D C5 IN10–Xmag4089 –0.28 (E1/16.95) –0.27 (E1/15.99) = –0.25 (E1/14.29) – –0.26 (E1/15.38) =
C5 –0.19 (E2/7.32) –0.19 (E2/7.67) = –0.18 (E2/6.26) = 0.19 (E2/7.57) =
C5 –0.23 (E5/9.78) –0.23 (E5/9.12) = –0.22 (E5/9.66) = –0.23 (E5/10.16) =
qSl-4A C6 Xwmc161–Xgpw2331 –0.18 (E1/6.91) –0.17 (E1/6.45) =
C6 –2.31 (E3/9.31) –0.19 (E3/6.52) – –0.21 (E3/8.94) – –0.19 (E3/6.52) –
C6 –2.54 (E4/10.63) –0.21 (E4/7.58) – –0.20 (E4/7.54) – –0.12 (E4/7.58) –
C6 –0.17 (E6/4.54)
qSl-4B C7 Xme16em26–Xwmc657 0.15 (E1/4.17)
C7 0.23 (E2/10.95) 0.20 (E2/8.16) – 0.26 (E2/13.62) = 0.20 (E2/8.16) –
C7 0.19 (E4/6.32)
C7 0.18 (E5/6.42)
C7 0.22 (E6/7.95)
C7 0.18 (E7/5.99)
C7 0.17 (E8/4.91)
qSl-6B C11 Xcnl113–Xwmc756 –1.85 (E4/5.65)
qSl-7D C12 XwPt-671530–Xgdm67 2.36 (E3/9.73) 0.24 (E3/10.17) – 0.22 (E3/9.85) – 0.23 (E3/10.12) –
C12 0.14 (E5/4.11)
SN qSn-1B.1 C1 XwPt-1684–XwPt-2988 0.23 (E2/5.44)
C1 0.28 (E7/7.75) 0.38 (E7/8.40) ?0.30 (E7/5.03) = 0.34 (E7/6.64) ?
qSn-1B.2 C2 XwPt-1328–XwPt-4107 0.30 (E1/6.25)
qSn-1B.3 C3 Xgwm374–Xwmc406 0.21 (E5/2.37)
qSn-2D C4 Xwmc601–XwPt-666518 –0.27 (E7/4.07) –0.28 (E7/4.49) =
qSn-4A C6 Xgpw2331–XwPt-4230 –0.35 (E1/9.14) –0.31 (E1/7.53) –
qSn-4B C8 XwPt-6149–Xmag2055 0.22 (E4/3.23)
C8 0.35 (E5/7.70) 0.39 (E5/10.24) – 0.29 (E5/5.75) ?
C8 0.22 (E6/3.21) 0.29 (E6/5.98) –
C8 0.42 (E7/9.78) 0.33 (E7/6.50) – 0.50 (E7/14.16) – 0.40 (E7/9.75) =
C8 0.31 (E8/4.71) 0.33 (E8/5.18) =
qSn-7D C12 XwPt-671530–Xgdm67 0.49 (E3/15.81) 0.45 (E3/13.14) = 0.45 (E3/13.14) = 0.41 (E3/11.37) =
C12 0.35 (E4/7.82) 0.30 (E4/5.97) = 0.30 (E4/5.97) – 0.28 (E4/5.55) =
C12 0.35 (E5/7.95) 0.39 (E5/10.24) = 0.39 (E5/10.24) = 0.34 (E5/7.90) –
C12 0.33 (E6/7.43) 0.35 (E6/8.55) = 0.35 (E6/8.55) = 0.36 (E6/9.74) =
C12 0.40 (E7/8.79) 0.43 (E7/10.39) =
C12 0.40 (E8/7.99) 0.40 (E8/7.95) =
24 Page 10 of 16 Mol Breeding (2015) 35:24
123
Table 3 continued
Traits QTLs
a
Clusters
b
Intervals marker
c
Additive[E/PVE (%)]
d
YRTs YRTs|FLW YRTs|FLL YRTs|FLA
KN qKn-1B C3 Xbarc187–XwPt-2614 –1.56 (E8/5.20) –1.54 (E8/5.43) = –1.70 (E8/6.00) = –2.00 (E8/9.14) ?
qKn-4A C6 Xgpw2331–XwPt-4230 –1.74 (E1/12.33) –1.60 (E1/10.83) – –1.42 (E1/8.95) – –1.36 (E1/8.44) –
C6 –1.81 (E2/10.04)
C6 –1.85 (E3/9.39) –1.84 (E3/9.88) = –1.76 (E3/9.58) = –1.62 (E3/8.27) –
C6 –2.15 (E4/11.73) –1.98 (E4/11.47) = –1.78 (E4/8.73) –
C6 –1.51 (E5/8.21) –1.07 (E5/4.29) – –1.44 (E5/7.32) = –1.23 (E5/5.79) –
C6 –1.98 (E6/9.45) –1.61 (E6/6.99) – –1.88 (E6/8.30) =
C6 –1.56 (E7/6.57)
C6 –1.91 (E8/8.37) –2.06 (E8/10.78) = –1.62 (E8/6.05) –
qKn-4B C9 Xcnl10–XwPt-1046 1.78 (E4/7.95) 1.91 (E4/10.58) = 1.87 (E4/9.66) =
C9 1.84 (E5/12.13) 1.61 (E5/9.70) – 2.46 (E5/21.30) ?1.83 (E5/12.65) =
C9 2.97 (E6/21.17) 2.86 (E6/19.07) = 3.06 (E6/21.85) = 2.14 (E7/11.11) –
C9 2.54 (E7/17.27) 2.20 (E7/13.40) – 2.78 (E7/20.45) = 2.10 (E712.87) –
C9 2.16 (E8/10.66) 1.62 (E8/6.62) – 2.27 (E8/11.88) =
qKn-5B C10 Xbarc142–Xwmc28 1.19 (E1/5.81) 1.13 (E1/5.59) =
KW (mg) qKw-2D C5 Xmag3596–IN10 0.02 (E4/9.31) 78.00 (E4/7.41) =
C5 51.30 (E5/5.76) 55.90 (E5/6.73) 54.50 (E5/5.65)
C5 0.05 (E6/4.69) 57.60 (E6/5.79) = 64.40 (E6/7.19) 54.50 (E6/5.30) =
qKw-4B.1 C8 Xcef89–XwPt-6149 –0.07 (E6/8,79) –71.10 (E6/8.71) = –67.3 (E6/8.00) =
qKw-4B.2 C8 XwPt-6149–Xmag2055 –126.00 (E4/20.20)
C8 0.01 (E8/5.70) 85.00 (E8/7.99)
qKw-4B.3 C9 Xcnl10–XwPt-1046 –63.90 (E2/6.92) –63.50 (E2/6.98)
C9 –22.60 (E4/18.29)
C9 0.02 (E7/7.19)
qKw-7D C12 XwPt-671530–Xgdm67 0.05 (E6/4.17)
a
QTLs for YRTs located in the 12 clusters in Table 2. A putative major QTL is marked by bold typeface and is characterized by a mean LOD value [3.0 and a mean PVE
[10 %; a putative stable QTL is underlined when this locus can be detected in at least five of the eight environments
b
The clusters containing QTLs affecting FLSTs and YRTs are shown in Table 2
c
Flanking markers of the QTLs
d
Numerals before parentheses are estimates of the additive effects of the QTL. Positive values indicate that KN9204 alleles increase the YRTs. Negative values indicate that
KN9204 alleles reduce YRTs. E and numerals in parentheses indicate the environment in which the QTL was detected and the percentage of phenotypic variance explained by the
additive effects of the mapped QTLs, respectively. A minus, ‘‘-’’, or a plus sign, ‘‘?’’, following the parentheses denotes the additive effect of a conditional QTL, in absolute
values, that reduces or increases more than 10 % compared to the corresponding unconditional QTL, respectively. An equal sign, ‘‘=’’, is placed after the parentheses to denote a
conditional QTL with an equal additive effect to that of the unconditional QTL
Mol Breeding (2015) 35:24 Page 11 of 16 24
123
Meanwhile, two major QTLs each for plant height
(data not shown) and for yield (Cui et al. 2014) were
detected in this interval. Breeding the variety with
desirable plant type and high yield potential, including
semidwarfing plant height, moderate flag leaf size, and
more kernel number, has been a major goal of
breeders. Considering that the alleles from KN9204
in this interval decreased plant height but increased the
FLW, KN, and yield, this interval should be of value in
further wheat breeding programs. qFlw-6B.2 was the
other major stable QTL for FLW, with J411-derived
alleles increasing FLW. Pyramiding elite alleles from
both the parents by further backcrossing and marker-
assisted breeding in wheat molecular breeding pro-
grams might be an optimal method by which to
improve ideotype and thus increase yield. Quarrie
et al. (2006) pyramided a yield QTL with QTL for
wider flag leaves and more flag leaf chlorophyll
content, which might improve wheat yield.
FLRTs: their influence on YRTs at the QTL level
The identification of co-localized QTLs conferring
multiple traits was used to dissect the genetic mech-
anism of pleiotropy and might be useful in improving
the efficiency of breeding for multiple elite traits.
Therefore, it is necessary to investigate the genetic
relationship among goal traits at each co-localized
locus. Numerous studies have shown that flag leaf size
is positively correlated with yield-related traits such as
KN, GY, and thousand kernel weight in cereal crops
(Cui et al. 2003; Mei et al. 2003; Khaliq et al. 2008;
Wang et al. 2011; Xue et al. 2013). In addition, QTL
clusters for FLRTs and YRTs have been reported in
rice (Wang et al. 2011). In this study, 12 genomic
regions that harbored QTL clusters for FLRTs and for
YRTs were identified. Multivariate conditional QTL
mapping was conducted to dissect the genetic effects
of FLRTs on the expression of QTLs for YRTs in these
12 regions.
All investigated YRTs in this study were positively
correlated with FLRTs (Table S3) at the phenotypic
level; thus, flag leaf size was supposed to improve the
expression of these clustered QTLs, altering yield
potential (Wang et al. 2011; Xue et al. 2013). When
conditioned on FLRTs, 60.87 % of the QTLs for
YRTs indeed showed reduced additive effects or were
even undetectable in the corresponding environments,
indicating that most of these QTLs for YRTs were
induced by the given FLRTs. However, 5.07 % of the
QTLs for YRTs showed enhanced additive effects
compared with those in the unconditional QTLs,
25
FLA
0
5
10
15
20
QQ qq
KN9204 J411 KJ126 KJ124
N=56 N=56
A B
Fig. 2 Allelic effect of qFla-5B on FLA (a) and flag leaves of the
parents (KN9204 and J411) and two KJ-RILs lines (KJ126 and
KJ124) (b) at 15 days after flowering. The group with QQ
representing the flanking markers of qFla-5B showed the KN9204-
genotype, indicating that it carries KN9204-derived alleles of
qFla-5B; the group with qq representing the flanking markers of
qFla-5B showed the J411-genotype, indicating that it carries J411-
derived alleles of qFla-5B. KJ126 was one member of the group
with KN9204-derived alleles of qFla-5B; KJ124 was one member
of the group with J411-derived alleles of qFla-5B
24 Page 12 of 16 Mol Breeding (2015) 35:24
123
indicating that FLRTs suppressed the expression of
these QTLs for YRTs. In addition, 34.06 % of the
QTLs for YRTs showed similar or equal additive
effects compared with those in the unconditional
QTLs, indicating that these QTLs for YRTs were
independent of FLRTs (Table 3). The above results
indicated that most but not all the QTLs for YRTs were
improved by FLRTs.
When conditioned on FLW, FLL, or FLA, 21.01,
15.94, or 23.91 % of the conditional QTLs for YRTs,
respectively, showed reduced additive effects or were
even undetectable compared with the results of the
unconditional QTL analysis in the corresponding
environments (Table 3). These results indicated that,
at the QTL level, FLA had the greatest contribution to
YRTs, followed by FLW and FLL, which was
consistent with the phenotypic correlation coefficients
among FLRTs and YRTs (Table S3). Regarding the
phenotypic and genetic association between FLA and
FLW, we should select individuals with wider flag
leaves to increase FLA and thus increase yield
potential in wheat breeding programs.
YRTs are complex traits and are sensitive to
environmental factors. This sensitivity to the environ-
ments was also observed in studies of the effect of
FLRTs that control the expression of the QTLs for
YRTs, which was similar to the cases reported by Cui
et al. (2013). For instance, when conditioned on FLA,
qSn-4B was induced in E6 and E8 but inhibited in E5.
Therefore, the loci for YRTs that showed stable
significance and consistent relationships with FLRTs
in all corresponding environments could be more
easily to be utilized in MAS.
FLRTs: their genetic basis of the response
to nitrogen supply
Genomic regions detected specifically under a nitro-
gen treatment are more probably involved in wheat
adaptation to nitrogen constraint (Laperche et al.
2007). In this study, 11 LN-specific QTLs (two
associated with FLW, seven associated with FLL,
and two associated with FLA) and 16 HN-specific
QTLs (seven associated with FLW, three associated
with FLL, and six associated with FLA) were detected,
respectively (Table 1). These regions harboring major
QTLs are valuable for further studying the effect of
nitrogen on flag leaf size. We identified a total of eight
environment-specific major QTLs herein. Among
these, qFlw-4B.1 was a major HN-specific QTL which
showed significance in multiple environments, indi-
cating that this interval has a higher possibility of
containing genes which are involved in response to the
sufficient nitrogen supply condition.
The varieties with ability to maintain desired leaf
size in low-input agricultural practices, particularly in
LN input management are of value in wheat breeding
programs. Among the LN-specific QTLs, qFlw-3B,
qFlw-4B.2,qFll-1B.1,qFll-1B.2,qFll-4A.2, and qFla-
1B.2 showed positive additive effects that increase
FRTs, indicating that alleles from KN9204 of these
QTLs were elite alleles (Table 1). qFll-2B.2,qFll-
4A.1,qFll-4B.2,qFll-5D, and qFla-4A showed neg-
ative additive effects, indicating that alleles from J411
of these QTLs increase FRTs (Table 1). Pyramiding
these elite alleles might be an optimal approach to
acquire desired leaf size under LN and thus facilitate
nitrogen use efficiency improvement in wheat molec-
ular breeding programs.
The major stable QTL can be used in marker-
assisted selection (MAS) designed to improve wheat
flag leaf size and yield potential. Except the environ-
ment-specific QTLs, 11 QTLs were identified under
both LN and HN treatments, including the three major
stable QTLs of qFlw-4B.3,qFlw-6B.2, and qFla-5B
(Table 1). This result further confirmed the expression
of these three QTLs was stable and insusceptible to
nitrogen supply. Thus, the close linkage markers of
these three major stable QTLs are of value in MAS.
Conserved QTLs across different mapping
populations
Several QTLs for FLRTs and YRTs reported herein
have been confirmed in previous studies. XwPt-1328
on chromosome 1B associated with FLL, FLA, and SN
in this study was also reported to be significantly
associated with grain yield by Crossa et al. (2007). The
SSR marker, Xwmc332, associated with qFll-2B.1,
was the flanking marker of the QTLs for grain yield
components in the Svevo 9Ciccio population
(Blanco et al. 2012). qFlw-2D.1 and qSn-2D on
chromosome 2D were flanked by the SSR marker
Xwmc601 in this study. This marker was reported to be
closely linked with QTLs for KN, thousand kernel
weight and days to heading (Cuthbert et al. 2008). The
STS marker Xmag3596 on chromosome 2D was
simultaneously associated with KW and FLW, which
Mol Breeding (2015) 35:24 Page 13 of 16 24
123
was also involved in the same QTL for KW found by
Cui et al. (2013). A major QTL for FLL and a stable
QTL for KN were flanked by Xwmc161 on chromo-
some 4A, which was reported to be associated with a
QTL for green leaf duration after heading and grain
yield (Naruoka et al. 2012). Xmag4087, one flanking
marker of two major stable QTLs of qFlw-4B.3 and
qKn-4B, has also been reported to be one flanking
marker of the QTL for grain number per spike detected
in four trials by Jia et al. (2013). The SSR marker
Xwmc487 was associated with the major QTL for
FLW (qFlw-6B.1) in this report, and it was also the
flanking marker of the QTL for sterile spikelet number
per spike identified by Li et al. (2007). The stable QTL
qSn-7D was clustered with qFlw-7D, with Xgdm67 as
one flanking marker, corresponding to a QTL cluster
for sterile SN and fertile SN (Li et al. 2007). Hu et al.
(2010) isolated three allelic rice mutants of nrl1 that
had reduced leaf width and semi-rolled leaves, indi-
cating that the cellulose synthase-like protein D4
(OsCslD4) encoded by NRL1 played an important role
in determining leaf morphogenesis and vegetative
development. In this study, qFlw-4B.1 was flanked by
the morphological marker Leaftype based on flag leaf
rolling type (curling for KN9204 vs. flat for J411), and
it was clustered with two major QTLs (qFll-4B.1 and
qSl-4B) in the interval Leaftype–Xme16em26–
Xwmc657 on chromosome 4B. In addition, a major
pleiotropic QTL for SL, KN, and spike number was
fine mapped by Deng et al. (2011) in this chromosomal
region. Wheat is allohexaploid, and the QTLs at the
similar position of homologous chromosomes might
associate with one gene family. Thus, the QTL clusters
of qFla-5B,qFLw-5B,qFll-5B, and qKn-5B on 5BL
might contain the homologous gene of the major locus
for FLW on chromosome 5AL detected by Xue et al.
(2013). The coincidence of QTL across different
mapping populations not only implies the reliability of
the QTL for FLRTs reported herein but also indicates
that FLRTs might affect YRTs consistently across
different mapping populations.
Conclusion
A total of 38 QTLs for FLRTs were identified in eight
environments via a KJ-RIL population. Of these,
qFlw-4B.3 and qFlw-6B.2 were two major stable
QTLs for FLW, and qFla-5B was one major stable
QTL for FLA. Phenotypic and QTL mapping analyses
indicated that FLW was the major contributor to flag
leaf size. Conditional QTL mapping analysis revealed
that most but not all of the QTLs for YRTs were
improved by FLRTs. At the QTL level, FLA made the
greatest contribution to the expression of QTLs for
YRTs, followed by FLW and FLL. This study
provides useful information for genetic improvement
of desirable plant morphological types and thus for
improving yield potential in wheat breeding programs.
Acknowledgments This researchwas supported by grantsfrom
the Ministry of Science and Technology of China (No. 2011AA
100103), Chinese Academy of Sciences (No. XDA08030107),
National Natural Science Foundation of China (No. 31471573),
and the Ministry of Agriculture of China (No. CARS-03-03B).
References
Abdelkhalik AF, Shishido R, Nomura K, Ikehashi H (2005)
QTL-based analysis of leaf senescence in an indica/
japonica hybrid in rice (Oryza sativa L.). Theor Appl
Genet 110:1226–1235
Blanco A, Mangini G, Giancaspro A, Giove S, Colasuonno P,
Simeone R, Signorile A, De Vita P, Mastrangelo AM,
Cattivelli L, Gadaleta A (2012) Relationships between
grain protein content and grain yield components through
quantitative trait locus analyses in a recombinant inbred
line population derived from two elite durum wheat culti-
vars. Mol Breed 30:79–92
Crossa J, Burgueno J, Dreisigacker S, Vargas M, Herrera-
Foessel SA, Lillemo M, Singh RP, Trethowan R, War-
burton M, Franco J, Reynolds M, Crouch HJ, Ortiz R
(2007) Association analysis of historical bread wheat
germplasm using additive genetic covariance of relatives
and population structure. Genetics 177:1889–1913
Cui KH, Peng SB, Xing YZ, Yu SB, Xu CG, Zhang Q (2003)
Molecular dissection of the genetic relationships of source,
sink and transport tissue with yield traits in rice. Theor
Appl Genet 106:649–658
Cui F, Li J, Ding AM, Zhao CH, Wang L, Wang XQ, Li SS, Bao
YG, Li XF, Feng DS, Kong LR, Wang HG (2011) Con-
ditional QTL mapping for plant height with respect to the
length of the spike and internode in two mapping popula-
tions of wheat. Theor Appl Genet 122:1517–1536
Cui F, Zhao CH, Li J, Ding AM, Li XF, Bao YG, Li JM, Ji J,
Wang HG (2013) Kernel weight per spike: what contrib-
utes to it at the individual QTL level? Mol Breed
31:265–278
Cui F, Fan XL, Zhao CH, Zhang W, Chen M, Ji J, Li JM (2014)
A novel genetic map of wheat: utility for mapping QTL for
yield under different nitrogen treatments. BMC Genet
15:57
Cuthbert JL, Somers DJ, Bru
ˆle
´-Babel AL, Brown PD, Crow GH
(2008) Molecular mapping of quantitative trait loci for
24 Page 14 of 16 Mol Breeding (2015) 35:24
123
yield and yield components in spring wheat (Triticum
aestivum L.). Theor Appl Genet 117:595–608
Deng SM, Wu XR, Wu YY, Zhou RH, Wang HG, Jia JZ, Liu SB
(2011) Characterization and precise mapping of a QTL
increasing spike number with pleiotropic effects in wheat.
Theor Appl Genet 122:281–289
Ding XP, Li XK, Xiong LZ (2011) Evaluation of near-isogenic
lines for drought resistance QTL and fine mapping of a
locus affecting flag leaf width, spikelet number, and root
volume in rice. Theor Appl Genet 123:815–826
Gladun I, Karpov E (1993) Distribution of assimilates from the
flag leaf of rice during the reproductive period of devel-
opment. Russia J Plant Physl 40:215–218
Guo LB, Xing YZ, Mei HW, Xu CG, Shi CH, Wu P, Luo LJ
(2005) Dissection of component QTL expression in yield
formation in rice. Plant Breed 124:127–132
Guo PG, Baum M, Varshney RK, Graner A, Grando S, Cec-
carelli S (2008) QTLs for chlorophyll and chlorophyll
fluorescence parameters in barley under post-flowering
drought. Euphytica 163:203–214
Hessler TG, Thomson MJ, Benscher D, Nachit MM, Sorrells
ME (2002) Association of a lipoxygenase locus, Lpx-B1,
with variation in lipoxygenase activity in durum wheat
seeds. Crop Sci 42:1695–1700
Horton P (2000) Prospects for crop improvement through the
genetic manipulation of photosynthesis: morphological
and biochemical aspects of light capture. J Exp Bot
51:475–485
Hu J, Zhu L, Zeng DL, Gao ZY, Guo LB, Fang YX, Zhang GH,
Dong GJ, Yan MX, Liu J, Qian Q (2010) Identification and
characterization of NARROW AND ROLLED LEAF 1,a
novel gene regulating leaf morphology and plant archi-
tecture in rice. Plant Mol Biol 73:283–292
Jia HY, Wan HS, Yang SH, Zhang ZZ, Kong ZX, Xue SL,
Zhang LX, Ma ZQ (2013) Genetic dissection of yield-
related traits in a recombinant inbred line population cre-
ated using a key breeding parent in China’s wheat breed-
ing. Theor Appl Genet 126:2123–2139
Khaliq I, Irshad A, Ahsan M (2008) Awns and flag leaf contri-
bution towards grain yield in spring wheat (Triticum aes-
tivum L.). Cereal Res Commun 36:65–76
Kobayashi S, Fukuta Y, Morita S, Sato T, Osaki M, Khush GS
(2003) Quantitative trait loci affecting flag leaf develop-
ment in rice (Oryza sativa L.). Breed Sci 53:255–262
Kumar U, Joshi AK, Kumar S, Chand R, Ro
¨der MS (2010)
Quantitative trait loci for resistance to spot blotch caused
by Bipolaris sorokiniana in wheat (T. aestivum L.) lines
‘Ning 8201’ and ‘Chirya 3’. Mol Breed 26:477–491
Laperche A, Brancourt-Hulmel M, Heumez E, Gardet O, Ha-
nocq E, Devienne-Barret F, Le Gouis J (2007) Using
genotype 9nitrogen interaction variables to evaluate the
QTL involved in wheat tolerance to nitrogen constraints.
Theor Appl Genet 326:399–415
Li SS, Jia JZ, Wei XY, Zhang XC, Li LZ, Chen HM, Fan YD,
Sun HY, Zhao XH, Lei TD, Xu YF, Jiang FS, Wang HG, Li
LH (2007) A intervarietal genetic map and QTL analysis
for yield traits in wheat. Mol Breed 20:167–178
Mei HW, Luo LJ, Ying CS, Wang YP, Yu XQ, Guo LB, Pat-
erson AH, Li ZK (2003) Gene actions of QTLs affecting
several agronomic traits resolved in a recombinant inbred
rice population and two testcross populations. Theor Appl
Genet 107:89–101
Mei HW, Li ZK, Shu QY, Guo LB, Wang YP, Yu XQ, Ying CS,
Luo LJ (2005) Gene actions of QTLs affecting several
agronomic traits resolved in a recombinant inbred rice
population and two backcross populations. Theor Appl
Genet 110:649–659
Naruoka Y, Sherman JD, Lanning SP, Blake NK, Martin JM,
Talbert LE (2012) Genetic analysis of green leaf duration
in spring wheat. Crop Sci 52:99–109
Quarrie SA, Quarrie SP, Radosevic R, Rancic D, Kaminska A,
Barnes JD, Leverington M, Ceoloni C, Dodig D (2006)
Dissecting a wheat QTL for yield present in a range of
environments: from the QTL to candidate genes. J Exp Bot
57:2627–2637
Sharma SN, Sain RS, Sharma RK (2003) The genetic control of
flag leaf length in normal and late sown durum wheat.
J Agric Sci 141:323–331
Tian F, Bradbury PJ, Brown PJ, Hung H, Sun Q, Flint-Garcia S,
Rocheford TR, McMullen MD, Holland JB, Buckler ES
(2011) Genome-wide association study of leaf architecture
in the maize nested association mapping population. Nat
Genet 43:159–162
Tsukaya H (2005) Leaf shape: genetic controls and environ-
mental factors. Int J Dev Biol 49:547–555
Tsukaya H (2006) Mechanism of leaf-shape determination.
Annu Rev Plant Biol 57:477–496
Verma V, Foulkes MJ, Worland AJ, Sylvester-Bradley R, Cal-
igari PDS, Snape JW (2004) Mapping quantitative trait loci
for flag leaf senescence as a yield determinant in winter
wheat under optimal and drought-stressed environments.
Euphytica 135:255–263
Wang P, Zhou GL, Yu HH, Yu SB (2011) Fine mapping a major
QTL for flag leaf size and yield-related traits in rice. Theor
Appl Genet 123:1319–1330
Wang P, Zhou GL, Cui KH, Li ZK, Yu SB (2012) Clustered
QTL for source leaf size and yield traits in rice (Oryza
sativa L.). Mol Breed 29:99–113
Wen YX, Zhu J (2005) Multivariable conditional analysis for
complex trait and its components. Acta Genet Sin
32:289–296
Xue DW, Chen MC, Zhou MX, Chen S, Mao Y, Zhang GP
(2008) QTL analysis of flag leaf in barley (Hordeum
vulgare L.) for morphological traits and chlorophyll con-
tent. J Zhejiang Univ Sci B 9:938–943
Xue SL, Xu F, Li GQ, Zhou Y, Lin MS, Gao ZX, Su XH, Xu
XW, Jiang G, Zhang S, Jia HY, Kong ZX, Zhang LX, Ma
ZQ (2013) Fine mapping TaFLW1, a major QTL control-
ling flag leaf width in bread wheat (Triticum aestivum L.).
Theor Appl Genet 126:1941–1949
Yue B, Xue WY, Luo LJ, Xing YZ (2008) Identification of
quantitative trait loci for four morphologic traits under
water stress in rice (Oryza sativa L.). J Genet Genomics
35:569–575
Zeng D, Hu J, Dong G, Liu J, Zeng L, Zhang G, Guo L, Zhou Y,
Qian Q (2009) Quantitative trait loci mapping of flag-leaf
ligule length in rice and alignment with ZmLG1 Gene.
J Integr Plant Biol 51:360–366
Zhao J, Becker HC, Zhang D, Zhang Y, Ecke W (2006) Con-
ditional QTL mapping of oil content in rapeseed with
Mol Breeding (2015) 35:24 Page 15 of 16 24
123
respect to protein content and traits related to plant
development and grain yield. Theor Appl Genet 113:33–38
Zhao XQ, Xu JL, Zhao M, Lafitte R, Zhu LH, Fu BY, Gao YM,
Li ZK (2008) QTLs affecting morph-physiological traits
related to drought tolerance detected in overlapping
introgression lines of rice (Oryza sativa L.). Plant Sci
174:618–625
24 Page 16 of 16 Mol Breeding (2015) 35:24
123
