Genetic mapping of QTL for maize leaf width combining RIL and IF2 populations
Genetic mapping of QTL for maize leaf width combining RIL and IF2 populations
Ruixiang Liu 0 1
Qingchang Meng 0 1
Fei Zheng 0 1
Lingjie Kong 0 1
Jianhua Yuan 0 1
Thomas LuÈ bberstedt 1
0 Institute of Food Crops, Provincial Key Laboratory of Agrobiology, Jiangsu Academy of Agricultural Sciences , Nanjing, Jiangsu Province , China , 2 Department of Agronomy, Iowa State University , Ames, Iowa State , United States of America
1 Editor: Zhiwu Zhang , Washington State Univeristy , UNITED STATES
Leaf width is an important component of plant architecture that affects light capture during photosynthesis and wind circulation under dense planting conditions. To improve understanding of the genetic mechanisms involved in leaf width at different positions, a comprehensive evaluation using the RIL (Recombinant Inbred Line) and IF2 (Immortalized F2) populations and a subsequent meta-analysis were performed. Forty-seven QTL associated with leaf width at different positions below the tassel were detected. The individual effects of QTL explained 3.5% to 17.0% of the observed phenotypic variation, and ten QTL explained over 10%. The initial QTL were integrated into eight mQTL (meta-QTL) through a metaanalysis. Our results suggested that leaf widths at different positions may be affected by several of the same mQTL and may also be regulated by many different mQTL. These results provide useful information for breeding high density tolerant inbred lines and hybrid cultivars, as well as for using marker-assisted selection for important mQTL.
Data Availability Statement: All relevant data are
within the paper and there are no Supporting
Funding: This work was financed by the National
Natural Science Foundation of China
(No.3171444), Jiangsu Academy of Agricultural
Science and Technology Innovation Fund (CX(13)
5003), and National Key Research and
Development Program of China
(2016YFD0101205). The funders had no role in
study design, data collection and analysis, decision
to publish, or preparation of the manuscript.
Over the past several decades, improvements in plant architecture greatly increased maize
grain yields [
]. All of leaf size and shape morphological traitsplay important role in
determining plant architecture. Leaf size, determinated by leaf width, leaf lengthand leaf area, is an
important component of leaf morphology and significantly influences the canopy morphology,
photosynthetic activity, and hence grain yield . Leaf width is an important component of
leaf size. Smaller and narrower leaf widths decrease shade effects on lower leaves and enhance
light capture for photosynthesis in dense plantings with a high leaf area index [
leaf size may decrease grain yield, owing to a decrease in photosynthetically active radiation
]. Therefore, understanding the genetic mechanisms of maize leaf width at
different position would not only address the radiation use efficiency in plant science but also
facilitate the breeding of tolerant to high planting density maize with an optimized leaf width.
Leaf development is initiated from the shoot apical meristem (SAM) [
]; then the leaf
polarity is established by three main axes, named proximal-distal (longitudinal), midvein-margin
(mediolateral) and adaxial-abaxial (dorsoventral); finally the leaf shape and size is determined
by a coordinated regulation of cell differentiation and expansion along these axes[
width is determined via differentiation along the mediolateral of founder cells in the peripheral
zone of the shoot apical meristem [
]. In maize, several key genes that affect the development
of the axes have been identified by using mutants, such as the narrow sheath ns1 and ns2
], the narrow and threadlike leaf phenotypes of lbl1 [
] and rgd2 [
]. These mutants
have helped to elucidate the molecular mechanisms of leaf-width development in maize.
In the past several decades, natural variation in maize leaf width has been determined by
using quantitative trait locus (QTL) mapping [3, 4, 11±14]. For example, Guo et al. [
identified 46 QTL associated with the width of the four consecutive leaves above the
uppermost ear in four RIL populations and in three environments. In addition, Yang et al. [
detected 83 QTL associated with the width of eight leaves below the tassel. A genome-wide
association study (GWAS) method has been used to detect variants at candidate loci and genes
responsible for leaf width. Tian et al. [
] have identified 34 QTL for leaf width through
NAM-GWAS (nested association mapping population). Additionally, Yang et al. [
found 18 SNPs associated with ear leaf width. Current research is often focus on the leaves
(one to three) near the ear, owing to their effect on grain yield; only two QTL mapping studies
on leaves at different positions have been published to date [
], and these studies have
shown inconsistent results regarding the QTL regions. This discrepancy may limit the
application of these QTL in marker-assisted selection. Therefore, further investigation of the genetic
basis of leaf width at different positions and the heterosis for leaf width is needed.
In this study, we conducted a comprehensive genetic dissection of leaf width to assess the
genetic architecture of seven consecutive leaves, by using a RIL and IF2 population derived
from a cross between the Chinese elite inbred lines S951 and Qi319. The major objective of
this study was to improve understanding of the genetic basis (including additive and
dominance effect) of leaf width at different positions and to determine how leaf width affects plant
architecture and adaptation to high density planting. In addition, we sought to identify and
fine map major leaf width QTL.
Materials and methods
Plant materials and leaf width collection
A total of 164 F9 RIL families derived from a cross between inbreds S-951 and Q319 was used
for leaf width analysis. The parents of this population were chosen on the basis of distinct
maize germplasm groups. S-951 is an inbred line derived from Chinese Stiff Stalk germplasm,
a heterotic group widely used in China, whereas Qi319 is an inbred line derived from Chinese
non-Stiff Stalk germplasm, also widely used in China. Similar to the procedure for generating
the previously described intercrossed F2 population [
], the 164 RILs were randomly
divided into two groups of 82 RILs. Single crosses were randomly conducted between the two
groups. Each RIL was used only once in each group of matings to generate crosses. This
procedure was repeated four times to produce 328 single crosses, forming the IF2 population. The
RIL and IF2 populations were planted in Dafeng, Liuhe in 2013, and Sanya in 2014. At three
year-location combinations, populations of RIL and IF2 were in neighboring blocks, each
planted in a randomized complete block design with three replications. Each plot was
singlerow, with 4 m long and 0.67 m between rows. The population density was 45,000 plants per
ha. Ten days after pollen shedding, five consecutive plants from the middle of each plot were
chosen to evaluate the top seven leaves' width (LW). The seven consecutive leaves below the
tassel were designated leaves one to seven and named L1W, L2W, L3W, L4W, L5W, L6W and
L7W, respectively. LW and trait value was determined according to the methods described by
Guo et al. .
2 / 13
Molecular markers and genetic linkage maps
The 164 RILs were genotyped with 209 polymorphic SSR markers that covered the whole
genome. The genotypes of each hybrid in the IF2 population were deduced from the genotype
of both its RIL parents. A linkage map was constructed by using the SSR genotypes of the 164
RILs. Molecular linkage maps for the RILs and IF2 population were constructed by using
] with a logarithm-of-odds (LOD) threshold of 3.0. The Kosambi mapping
function was used for calculating map distances [
Data analysis and QTL Mapping
The broad-sense heritability (H2) for each leaf width below the tassel was calculated according
to the method described by Guo et al. [
], as the genetic variance (VG) divided by the sum of
genetic variance (VG), G × E variance (VG×E) and error variance (Vε), based on the plot basis.
The calculation of best linear unbiased predictors (BLUPs) for each line in RILs and each
hybrid in IF2 was done with a mixed linear model that accounted for the effects of
environment, replication, genotype, and genotype by environment described by Guo et al. [
estimated genetic variance and error variance by analysis of variance were for BLUP
estimators. The BLUPs for each line and hybrid from three locations were used in QTL mapping
The mapping of QTL were performed with QTL ICIMapping software (http://www.
]. QTL were identified using the ICIM-ADD mapping method with the
default software parameters (PIN (probability in stepwise regression) = 0.001, step = 1.0 cM).
The threshold levels for declaring the existence of a QTL with an additive and/or dominance
effect were determined by performing 1,000 permutations on the data with a significance level
of P 0.05. Gene action was determined by the ratio of the absolute value of the estimated
dominance effect divided by the absolute value of the estimated additive effect following [
(additive 0±0.20, partial dominance 0.21±0.80, dominance 0.81±1.20, over dominance>1.20).
QTL integration and meta-analysis
The meta-analysis method was used to integrate the QTL information identified in the RIL
and IF2 population, and performed using BioMercator 4.2 [
]. The significant QTL model
for indicating the number of meta-QTL(mQTL) in each chromosome was determined by the
lowest Akaike information criterion (AIC) values as described by . The number of mQTL
that showed the best fit to the results in a given linkage group was determined on the basis of a
modified Akaike criterion [
]. The names of the mQTL were assigned according the
following nomenclature: "m" + "QLW" + "chromosome number" + "meta-QTL number" described
by Guo et al. [
The inbred line Qi319 showed decreased values for consecutive leaf-widths below the tassel, as
compared with those of the inbred line S-951. The values of L1W, L5W, L6W, and L7W were
markedly lower than those of S-951, with the exception of L2W, L3W, and L4W, which
presented no significant differences (Table 1). Among IF2 families, trait values were generally
higher than those of RILs. For most traits, phenotypic values were normally distributed, with
increased variation in both the RIL and IF2 populations. All leaf widths exhibited substantial
bidirectional transgressive segregation, thus supporting a polygenic quantitative genetic
control of these traits. Broad-sense heritabilities (h2b) for each leaf width ranged from 0.61 (L1W)
3 / 13
to 0.78 (LW4) in the RIL population and from 0.62 (L1W) to 0.80 (L4W) in the IF2 population,
thus indicating that genetic factors are determinants of these traits.
Phenotypic correlation coefficients among LW traits (Table 2) ranged from 0.22 to 0.90 in
the RIL and 0.50 to 0.92 in the IF2 population. Phenotypic correlations were significant and
positive, and closer phenotypic correlations existed between adjacent leaves than non-adjacent
leaves, and decreased with the distance of leaves [
Identification of major leaf width QTL in the RIL
Threshold values (P<0.05 significance level) were determined with 1000 permutations of the
leaf width data. In the RIL population, a total of 17 QTL were identified for the seven leaf
widths across chromosomes 1, 2, 5, 7, 8 and 9 (Table 3). The phenotypic variance of individual
QTL ranged from 6.5% (contributed by qL2WR1) to 17.0% (qL3WR5), with seven QTL
explaining more than 10% of the phenotypic variation. Among these QTL, three were
associated with L1W, three with L2W, two with L3W, two with L4W, two with L5W, three with
L6W and two with L7W. Fifteen positive alleles among the 17 QTL originated from S-951 and
contributed to increased leaf width values. The QTL qL2WR5, qL3WR5 and qL4WR5 were
detected within the marker interval umc1019-bnlg118 on chromosome 5 and explained more
than 10% of the total phenotypic variation. The QTL qL2WR7 and qL3WR7 were detected
4 / 13
aRepresent the leaves' width as described in material and methods.
bRepresent the chromosome number.
cThe genetc distance (centiMorgans, cM) of the QTL on the relevant chromosome in the genetic linkage map.
dThe threshold LOD values were determined with 1000-times permutations of the data.
eThe proportion of phenotypic variation explained by each QTL.
fThe estimated additive effect of the QTL.
within the marker interval mmc0411-umc1015 on chromosome 7 and explained 8.9% and 10.9%
of the observed phenotypic variation, whereas QTL qL4WR7 and qL5WR7 were identified near
the same marker interval phi008-mmc041 and explained 7.0% and 8.6% phenotypic variation,
respectively. The QTL qL1WR2 was detected within the marker interval bnlg125-umc2248 and
accounted for 9.8% of the total phenotypic variation, with a decreasing leaf width additive effect
of 0.15. The QTL qL5WR5 was located on chromosome 5 between umc1680 and umc1019 and
accounted for 13.4% of the total phenotypic variation.
Identification of major leaf width QTL in the IF2
A total of 30 QTL were identified for the seven leaf widths across the whole genome, except for
chromosome 4, 6 and 10, in the IF2 population (Table 4). Among these QTL, seven were
associated with L1W, six with L2W, six with L3W, four with L4W, three with L5W, two with L6W
and two with L7W. The phenotypic variance explained by individual QTL ranged from 3.5%
(qL2WF7) to 12.8% (qL2WF1.1), whereas four of the QTL accounted for more than 10% of
the phenotypic variation (Table 4). The analysis of the positive or negative effects of QTL
revealed that 60% (18/30) of the leaf width QTL were associated with an increase in leaf width.
The allelic effect distribution of different leaf widths was different. For L1W, 4 of 7 (57.1%)
QTL were associated with a decrease in leaf width. However, in contrast to L4W, all QTL
tended to increase leaf width. Three QTL, including qL2WF5, qL3WF5 and qL5WF5, were
detected on chromosome 5 within the marker interval umc1680 and umc1019 and showed
partial dominance. QTL qL1WF5 was detected in the marker interval umc1680-umc1019 on
5 / 13
chromosome 5 and had additive effects from parent S-951 and explained 5.6% of the total
phenotypic variance. The QTL qL4WF8 on chromosome 8 showed over-dominance and
accounted for 5.2% of the phenotypic variance. The QTL qL6WF2.2 was located on
chromosome 2, showed dominance, and accounted for 7.9% of phenotypic variance. The three QTL
qL1WF7.2, qL3WF7.2 and qL4WF7 explained 4.6%, 6.2% and 5.7% of the phenotypic
variance, respectively, and were associated with increased leaf width values, had additive effects,
and were detected in the marker interval mmc0411-umc1015 on chromosome 7. The QTL
qL1WF1.1, qL2WF1.1 and qL3WF1 accounted for 11.9%, 12.8% and 9.2% of the total
phenotypic variation, respectively, and were detected in the region between bnlg1055 and umc1009
on chromosome 1. L1W, L2W and L3W were controlled by one or two large-effect QTL in
addition to several small-effect QTL. L4W, L5W, L6W and L7W were controlled by
smalleffect QTL. These results suggested that the genetic architecture of L1W, L2W and L3W is
controlled by one or two large-effect QTL and a few small-effect QTL. The architecture of L4W,
L5W, L6W and L7W is complex and is controlled by many small-effect QTL.
6 / 13
Comparison of QTL positions between RIL and IF2
In the present study, a total of 47 QTL (17 in the RILs and 30 in the IF2 population) for the
seven leaf-widths were identified by QTL mapping, thus revealing that the IF2 population had
greater power to detect QTL. To evaluate the genetic overlap among different positions of leaf
widths in the RIL and IF2 population, the QTL marker intervals were compared. On the basis
of the same marker intervals, all QTL were distributed into 25 marker intervals in the maize
genome. Non-common marker interval regions controlled all seven leaf-width in different leaf
positions, althought the seven leaf-width were correlated highly. This finding was similar to
previous results [
]. Of these 25 regions, 14 (56%) were specific to different positions and leaf
widths. For example, five marker intervals (bnlg1297-mmc0111 and bnlg125-bnlg2248 on
chromosome 2, phi453121-bnlg1647 on chromosome 3, umc2334-umc1799 on chromosome
7 and bnlg619-umc1277 on chromosome 9) affected only L1W; the marker intervals
phi115phi121 on chromosome 8, bnlg1268-phi265454 on chromosome 1 and bnlg1176-bnlg1599 on
chromosome 8 were specific to L3W, L5W and L7W, respectively. Eleven (44%) regions
exhibited pleiotropic effects on leaf widths at different positions. The umc1680-umc1019 marker
interval found on chromosome 5 was responsible for L1W, L2W, L3W and L5W, whereas the
mmc0411-umc1015 interval on chromosome 7 was responsible for L1W, L2W, L3W and
L4W. There were ten regions that controlled adjacent leaf width, such as the marker interval
bnlg1055-umc1009 and umc2177-umc1378, which controlled L1W, L2W and L3W; the
phi008-mmc0411 regions, which controlled the L4W and L5W; and the bnlg1662-bnlg1863
and umc1551-umc1525 marker intervals, which were responsible for L6W and L7W. These
regions appeared to harbor pleiotropic locus affecting different positional widths of leaves.
These results may explain why adjacent leaf widths exhibited a higher phenotypic correlation.
To identify stable and consistent QTL between the RIL and IF2 populations, as well as to
discriminate between pleiotropic and linked QTL for the seven leaf widths, the initial QTL were
analyzed via meta-analysis. Eight mQTL were identified from the 47 initial QTL on the basis
of the variation in leaf width (Table 5). The eight mQTL were identified on six chromosomes:
one on chromosome 1, 5, 8 and 9, and two on chromosome 2 and 7. On average, one mQTL
included 5 initial QTL and ranged from 2 to 10 QTL (Fig 1). Importantly, 10 of the initial QTL
that showed an R2 > 10% in the RIL and IF2 populations were included in 4 of the mQTL,
Fig 1. Initial QTL and mQTL detected for the seven leaf widths. The ªChrº represents chromosome. The white bar and the bold bar represents the initial
QTL detected in IF2 and RIL population respectively. The bold segments in chromosome represent the region of mQTL.
including mQLW1.1, mQLW2.2, mQLW5 and mQLW7.2. mQLW5 included 10 QTL
associated with the widths of the first to sixth leaves in the RIL and IF2 populations and explained
5.6%-17.0% of the total phenotypic variation. mQLW7.2 comprised seven QTL associated
with the widths of the first to fifth leaves in the two populations and explained 4.6%-10.9% of
the phenotypic variation. mQLW1 included six QTL associated with the widths of the first to
fifth leaves in the two populations and explained 6.5%-12.8% of the phenotypic variation. Fine
mapping of these meta-QTL is a reliable strategy for QTL cloning, which is currently
underway in our laboratory. Importantly, the initial QTL included in mQLW7.1 was associated with
8 / 13
L1W, L2W, and L3W, while the initial QTL included in mQLW2.2 was associated with L6W
Association between leaf-width mQTL and known genes
A mutant gene is a logical candidate gene, if it alters the expression of a target trait and is
located within the QTL region associated with that trait [
]. Based on this, we investigated the
association between eleven mQTL that were observed in our study and known mutant genes
affecting leaf size in maize. The results showed that several mutant genes responsible for leaf
size corresponded to the mQTL region (Table 5). The lg4a [
] genes had map positions
neighboring mQLW8. mQLW2.1 was located close to the mrl1 gene, which played important role in
development of the mediolateral axis [
]. mQLW5 was located near the ZmPIN1b gene [
which mediates auxin transport, increases auxin concentrations in specific tissues and controls
gene expression. However, these mutant genes and the meta-QTL might be located too far to
provide effective validation. Isolation of the near-isogenic lines with enriched molecular
markers (which is underway for some of the mQTL) and fine mapping of these mQTL will help to
determine, whether these mutant candidate genes are the causal genetic variants of the mQTL.
Genetic structure of leaf width
One important maize breeding goal is to increase the stress tolerance for high-density planting
of new hybrids, which can be achieved by improving the plant type and canopy architecture
]. Leaf width is an important trait for high-density planting tolerance in maize breeding.
Therefore, unraveling the genetic mechanisms underlying leaf width is essential. Several QTL
mapping studies on maize leaf width have been published [
3, 4, 12, 18, 28, 35
], although they
are inconsistent in regard to the QTL regions. Therefore, further investigations addressing the
QTL that underlie the phenotypic variance in leaf width, especially in different leaf positions,
are required. In this study, we measured the widths of seven consecutive leaves below the
tassel, performed phenotypic data analyses, conducted QTL identification, and attempted to
investigate the genetic controls underlying leaf width in maize, by using RIL and IF2
populations. The results of the phenotypic analysis showed that the width of the seven leaves
displayed significantly positive correlations, especially for widths of adjacent leaves. The number
of significant QTL for different leaf widths at different positions ranged from 4 (L7W) to 10
(L1W) and the asymmetric and clustered distribution among genomic regions revealed the
complex nature of leaf width. Among 30 leaf-width QTL identified in IF2 populations, 12 QTL
(40%) showed additive gene action, 16 QTL (53.3%) showed partial dominance, and one each
showed overdominance and dominance. Collectively, these data indicate that the inheritance
of leaf width traits is controlled by a few major QTL and numerous minor QTL. Additive and
partial dominance effects play important roles in controlling leaf width.
Meta-QTL associated with different leaf widths at different positions
The leaf width of adjacent leaves of maize plants showed higher correlations than non-adjacent
leaves, thus indicating that the more common QTL can be expected in adjacent leaf widths. As
expected, the more common QTL (approximately 70.2% of total loci) were found in leaf
widths of adjacent leaves. The findings from several studies are in agreement with this result.
Fifty percent of the total loci were observed in adjacent leaf widths in one 253 RIL line derived
from a cross between B73 and SICAU1212 [
]. In multiple RIL populations, 48.6, 44.4, 47.5,
and 25% of the corresponding QTL were responsible for the widths of adjacent leaves [
9 / 13
However, only 6.5% were associated with the leaf width at all investigated positions in the four
connected RILs [
]. According to Hou et al. [
], 40% of the leaf width QTL control the leaf
widths in the first leaf above and below the primary ear.
Interestingly, the mQTL appeared to play distinct different effects on adult leaves at
different positions. The mQLW7.1 was required for only the first to third leaf widths and mQTL9
affected only the second and third leaf widths. In contrast, meta-QTL mQLW2.2 affected only
the sixth and seventh leaf widths. mQLW5 affected L1W to L6W and mQLW1 and mQLW7.2
affected L1W to L5W, which appeared to transition, bridge or switch between mQTL for L1W
and L2W and those for L6W and L7W (Table 5). The meta-QTL were composed of loci that
jointly affected adjacent leaf widths. This suggests that the architecture of adult leaf widths is
mediated by several key common regions and that most of the key genes or loci controlling
leaf width might be expressed in a few specific transition leaves (from juvenile to adult leaf
transition). Our data suggest that the width of the third and fourth leaf may form a connecting
link between the first to second and the sixth to seventh leaves. This possibility is consistent
with other studies. Guo et al. (2015) found that the width of the second leaf above the
uppermost ear is regulated by more common regions, as compared with the other three leaves,
regardless of the germplasm background [
]. Therefore, this leaf might be deemed the
transition leaf, which is controlled by more common loci [
]. The width of the fourth leaf shared the
highest proportion of loci with the other leaves and might have a transitional effect on other
]. Both leaf positions in the two studies are similar to the third and fourth leaves in our
study. The transcriptomes of the base and tip of developing leaves were different and the
expression of genes affecting lignin synthesis was distinct in the leaves of the mature and
immature leaves [
The QTL involved in leaf width have been found in previous QTL mapping reports [
11, 28, 30, 35, 37
], although the chromosomal regions for which QTL were located vary across
these studies. Base on the marker physical position, we compared the published QTL
controlling leaf width with the meta-QTL identified in this study. In the present study, six of the eight
meta-QTL were reported to be associated with leaf width in diverse populations across
different environments (Table 5). mQLW5 has been confirmed by several studies. Li et al. [
detected two leaf width QTL on chromosome 5, one of which is located on 197-202Mb and
explains more than 10.2% of the phenotypic variance. Yang et al. [
] found one QTL near
umc1852, which contributes 11.8% of the phenotypic variance and controls the first, second
and third leaf widths on chromosome 5 between the markers umc1822 and phi048. Ku et al.
] have also found one important QTL for leaf area on chromosome 5 between bnlg1287
and mmc0282 and have predicted YABBY15 to be the candidate gene. mQLW5 has also been
detected in an enlarged maize association panel [
]. Because the highest correlation occurred
between leaf area and leaf width, this region may include a common gene that controls both
traits. mQLW2.1, which resides on chromosome 2 between the markers mmc0111 and
bnlg2277, is near the lg1 gene and has been confirmed by Tian et al. [
] in the NAM (Nested
Association Mapping) population and further validated by Cai et al. [
]. Yang et al. [
found one QTL controlling the width of the sixth and seventh leaves below the tassel. These
QTL, which have been detected in different genetic backgrounds and environments, share a
high congruence, thus strongly supporting the candidacy of lg1 for mQLW2.1. The mQLW8.1
located in bin 8.3 has also been identified in a NAM-GWAS study [
]. In addition, one
common QTL controlling the leaf width has been mapped to the same region in the RIL derived
from B73 × SICAU1212 [
] and Yu82 × D132 .
The meta-QTL mQLW7.2 and mQLW9 did not overlap with the reported maize leaf width
QTL. Thus, our results identified not only stable and robust QTL validated by other studies
but also new QTL, thus further indicating that the genetic architecture of leaf width is
10 / 13
complicated and dominated by small but effective alleles. These newly reported QTL not only
provide new target genomic regions for further identification and characterization of genes
responsible for maize leaf width but also facilitate marker-assisted selection for maize plant
architecture improvements to develop hybrids that are better suited to high density planting.
Application of the QTL associated with leaf widths in maize breeding
Modern maize hybrids provide higher yields than those that were bred several decades ago,
primarily because they are adapted to high densities. Plant morphologies that enable efficient
light interception at high population densities may increase yield and production [
maize ideotype should exhibit vertically oriented leaves above the ear and horizontally oriented
leaves below the ear, such that more light can reach the ear leaves [
]. Furthermore, it is
necessary that the appropriate sizes of leaves offer maximized photosynthate without shading the
surrounding plant layers [
]. Relatively wider leaves at nodes that are closer to the ear plus
narrower leaves that are near the tassel may meet this requirement. However, significant
correlations between the different leaf widths at different positions make it difficult to increase the
leaf width below the primary ear without increasing the width of the three or four leaves under
the tassel. In contrast to the QTL mapping results for leaf widths in different positions, we
found a difference in the molecular basis of leaf width among differently positioned leaves. No
one QTL that affected all leaf widths was detected. We found two meta-QTL (mQLW7.1 and
mQLW9) that affected the widths of the two leaves below the tassel (L1W and L2W) and two
meta-QTL that affected the widths of the two leaves below the ear (L6W and L7W). These
effects were independent. Our results suggested that the molecular basis of different leaf widths
has similarities and differences. Hou et al [
] have also found differences in the molecular
basis of leaf width in the leaf above the ear compared with the leaf below the ear. Thus,
simultaneously decreasing the width of the two leaves below the tassel (L1W and L2W) and increasing
the width of the two leaves (L6W and L7W) near the ear may be possible by manipulating
these loci by MAS. The other meta-QTL that have distinct effects on the sizes of leaves at
different nodes may be used to alter the plant type further according to the specific application
]. For example, plants with wider leaves may increase the biomass and taste of forage maize.
Therefore, the results of this study may provide valuable information for breeding high density
tolerant inbred lines and for using marker-assisted selection for important mQTL.
Conceptualization: Qingchang Meng.
Data curation: Ruixiang Liu.
Investigation: Fei Zheng.
Methodology: Lingjie Kong.
Writing ± original draft: Ruixiang Liu.
Writing ± review & editing: Jianhua Yuan, Thomas LuÈbberstedt.
11 / 13
12 / 13
1. Li Y , Ma X , Wang T , Li Y , Liu C , Liu Z , et al. Increasing maize productivity in China by planting hybrids with germplasm that responds favorably to higher planting densities . Crop science . 2011 ; 51 ( 6 ): 2391 ± 400 .
2. Duvick DN . The contribution of breeding to yield advances in maize (Zea mays L.) . Advances in agronomy. 2005 ; 86 : 83 ± 145 .
3. Yang C , Tang D , Qu J , Zhang L , Zhang L , Chen Z , et al. Genetic mapping of QTL for the sizes of eight consecutive leaves below the tassel in maize (Zea mays L.) . Theoretical and Applied Genetics . 2016 ; 129 ( 11 ): 2191 ± 209 . https://doi.org/10.1007/s00122-016 -2767-2 PMID: 27550554
4. Guo S , Ku L , Qi J , Tian Z , Han T , Zhang L , et al. Genetic analysis and major quantitative trait locus mapping of leaf widths at different positions in multiple populations . PLoS One . 2015 ; 10 ( 3 ):e0119095. https://doi.org/10.1371/journal.pone. 0119095 PMID: 25756495
5. Stewart D , Costa C , Dwyer L , Smith D , Hamilton R , Ma B. Canopy structure, light interception, and photosynthesis in maize . Agronomy Journal . 2003 ; 95 ( 6 ): 1465 ± 74 .
6. Moon J , Hake S. How a leaf gets its shape. Current opinion in plant biology . 2011 ; 14 ( 1 ): 24 ± 30 . https:// doi.org/10.1016/j.pbi. 2010 . 08 .012 PMID: 20870452
7. Byrne ME . Networks in leaf development . Current opinion in plant biology . 2005 ; 8 ( 1 ): 59 ± 66 . https://doi. org/10.1016/j.pbi. 2004 . 11 .009 PMID: 15653401
8. Scanlon MJ . Developmental complexities of simple leaves. Current opinion in plant biology . 2000 ; 3 (1):31±6 . PMID: 10679452
9. Henderson DC , Muehlbauer GJ , Scanlon MJ . Radial leaves of the maize mutant ragged seedling2 retain dorsiventral anatomy . Developmental biology . 2005 ; 282 ( 2 ): 455 ± 66 . https://doi.org/10.1016/j. ydbio. 2005 . 03 .027 PMID: 15950610
10. Timmermans M , Schultes NP , Jankovsky JP , Nelson T. Leafbladeless1 is required for dorsoventrality of lateral organs in maize . Development . 1998 ; 125 ( 15 ): 2813 ± 23 . PMID: 9655804
11. Hou X , Liu Y , Xiao Q , Wei B , Zhang X , Gu Y , et al. Genetic analysis for canopy architecture in an F2:3 population derived from two-type foundation parents across multi-environments . Euphytica . 2015 ; 205 ( 2 ): 421 ± 40 . https://doi.org/10.1007/s10681-015-1401-8
12. Ku LX , Zhao WM , Zhang J , Wu LC , Wang CL , Wang PA , et al. Quantitative trait loci mapping of leaf angle and leaf orientation value in maize (Zea mays L.) . Theoretical and Applied Genetics . 2010 ; 121 ( 5 ): 951 ±9. https://doi.org/10.1007/s00122-010-1364-z PMID: 20526576
13. Cai H , Chu Q , Yuan L , Liu J , Chen X , Chen F , et al. Identification of quantitative trait loci for leaf area and chlorophyll content in maize (Zea mays) under low nitrogen and low phosphorus supply . Molecular breeding . 2012 ; 30 ( 1 ): 251 ± 66 .
14. Pelleschi S , Leonardi A , Rocher J-P , Cornic G , De Vienne D , Thevenot C , et al. Analysis of the relationships between growth, photosynthesis and carbohydrate metabolism using quantitative trait loci (QTLs) in young maize plants subjected to water deprivation . Molecular Breeding . 2006 ; 17 ( 1 ): 21 ± 39 .
15. Tian F , Bradbury PJ , Brown PJ , Hung H , Sun Q , Flint-Garcia S , et al. Genome-wide association study of leaf architecture in the maize nested association mapping population . Nature Genetics . 2011 ; 43 ( 2 ): 159 ± 62 . https://doi.org/10.1038/ng.746 PMID: 21217756
16. Yang N , Lu Y , Yang X , Huang J , Zhou Y , Ali F , et al. Genome wide association studies using a new nonparametric model reveal the genetic architecture of 17 agronomic traits in an enlarged maize association panel . PLoS Genetics . 2014 ; 10 ( 9 ):e1004573. https://doi.org/10.1371/journal.pgen. 1004573 PMID: 25211220
17. Hua J , Xing Y , Xu C , Sun X , Yu S , Zhang Q . Genetic dissection of an elite rice hybrid revealed that heterozygotes are not always advantageous for performance . Genetics . 2002 ; 162 ( 4 ): 1885 ± 95 . PMID: 12524357
18. Tang J , Yan J , Ma X , Teng W , Wu W , Dai J , et al. Dissection of the genetic basis of heterosis in an elite maize hybrid by QTL mapping in an immortalized F2 population . Theoretical and applied genetics . 2010 ; 120 ( 2 ): 333 ± 40 . https://doi.org/10.1007/s00122-009 -1213-0 PMID: 19936698
19. Meng L , Li H , Zhang L , Wang J. QTL IciMapping: Integrated software for genetic linkage map construction and quantitative trait locus mapping in biparental populations . The Crop Journal . 2015 .
20. Kosambi DD . The estimation of map distances from recombination values . Annals of Human Genetics . 1943 ; 12 ( 1 ): 172 ± 5 .
21. Tingting G , Ning Y , Hao T , Qingchun P , Xiaohong Y , Jihua T , et al. Genetic basis of grain yield heterosis in an "immortalized F2" maize population . Theoretical and applied genetics . 2014 ; 127 ( 10 ): 21492158 .
22. Wang J , editor QTL IciMapping: Integrated Software for Building Linkage Maps and Mapping Quantitative Trait Genes. Plant and Animal Genome XXI Conference; 2013: Plant and Animal Genome .
23. Stuber CW , Edwards MD , Wendel J . Molecular marker-facilitated investigations of quantitative trait loci in maize. II. Factors influencing yield and its component traits . Crop Science . 1987 ; 27 ( 4 ): 639 ± 48 .
24. Sosnowski O , Charcosset A , Joets J. BioMercator V3: an upgrade of genetic map compilation and quantitative trait loci meta-analysis algorithms . Bioinformatics . 2012 ; 28 ( 15 ): 2082 ± 3 . https://doi.org/10. 1093/bioinformatics/bts313 PMID: 22661647
25. Arcade A , Labourdette A , Chardon F , Falque M , Charcosset A , Joets J . BioMERCATOR version 2 .0. 2005 .
26. Akaike H. A new look at the statistical model identification . IEEE transactions on automatic control . 1974 ; 19 ( 6 ): 716 ± 23 .
27. Goffinet B , Gerber S. Quantitative trait loci: a meta-analysis . Genetics . 2000 ; 155 ( 1 ): 463 ± 73 . PMID: 10790417
28. Li C , Li Y , Shi Y , Song Y , Zhang D , Buckler ES , et al. Genetic control of the leaf angle and leaf orientation value as revealed by ultra-high density maps in three connected maize populations . PLoS One . 2015 ; 10 ( 3 ):e0121624. https://doi.org/10.1371/journal.pone. 0121624 PMID: 25807369
29. Carraro N , Forestan C , Canova S , Traas J , Varotto S. ZmPIN1a and ZmPIN1b encode two novel putative candidates for polar auxin transport and plant architecture determination of maize . Plant physiology . 2006 ; 142 ( 1 ): 254 ± 64 . https://doi.org/10.1104/pp. 106 .080119 PMID: 16844839
Wei X , Wang X , Guo S , Zhou J , Shi Y , Wang H , et al. Epistatic and QTL× environment interaction effects on leaf area-associated traits in maize . Plant Breeding . 2016 ; 135 ( 6 ): 671 ± 6 .
31. Moreno MA , Harper LC , Krueger RW , Dellaporta SL , Freeling M. liguleless1 encodes a nuclear-localized protein required for induction of ligules and auricles during maize leaf organogenesis . Genes & Development . 1997 ; 11 ( 5 ): 616 ± 28 .
32. Upadyayula N , da Silva HS , Bohn MO , Rocheford TR . Genetic and QTL analysis of maize tassel and ear inflorescence architecture . Theoretical and Applied Genetics . 2006 ; 112 ( 4 ): 592 ± 606 . https://doi. org/10.1007/s00122-005-0133 -x PMID : 16395569
33. Bauer P , Lubkowitz M , Tyers R , Nemoto K , Meeley RB , Goff SA , et al. Regulation and a conserved intron sequence of liguleless3/4 knox class-I homeobox genes in grasses . Planta . 2004 ; 219 ( 2 ): 359 ± 68 . https://doi.org/10.1007/s00425-004 -1233-6 PMID: 15034715
34. Ma D , Xie R , Niu X , Li S , Long H , Liu Y. Changes in the morphological traits of maize genotypes in China between the 1950s and 2000s . European Journal of Agronomy . 2014 ; 58 :1± 10 .
35. Ku L , Zhang J , Zhang JC , Guo S , Liu H , Zhao R , et al. Genetic dissection of leaf area by jointing two F2:3 populations in maize (Zea MaysL .). Plant Breeding . 2012 ; 131 ( 5 ): 591 ± 9 .
36. Alam MF , Khan MR , Nuruzzaman M , Parvez S , Swaraz AM , Alam I , et al. Genetic basis of heterosis and inbreeding depression in rice (Oryza sativa L.) . J Zhejiang Univ Sci . 2004 ; 5 ( 4 ): 406 ± 11 . PMID: 14994428
Wassom JJ . Quantitative Trait Loci for Leaf Angle, Leaf Width, Leaf Length, and Plant Height in a maize (Zea mays L) B73× Mo17 population . Maydica . 2013 ; 58 ( 3 ±4): 318 ± 21 .
38. Mock J , Pearce R. An ideotype of maize . Euphytica . 1975 ; 24 ( 3 ): 613 ± 23 .