Introduction
Maize (Zea mays L.) root system plays a critical role in plant
anchorage and acquisition of water and nutrients (Song et al. 2016). Changes in the root architecture of
maize strongly affect the yield (Hammer et al.
2009). Previous studies have shown that high-yielding maize varieties
displayed propitious root architecture that can provide strong supplies of
water and nutrients, resulting in increased grain yields (Hammer et al. 2009; Abdel-Ghani et al.
2013). However, notwithstanding their notorious involvement in plant
development, agricultural performance, and competition in the wild, roots in
plant genetics remain an under-explored frontier. The visualization and
measurement of root structures and their growth are much challenging than the
characterization of aboveground parts of the plant and is often simply avoided (Bray and Topp 2018). The majority of maize
root investigations have focused on their anatomy, physiology, growth,
development, and soil interactions but far less is known about the genetics
that regulates root quantitative traits (Bray
and Topp 2018). To date, quantitative variation has been proposed for
root architecture to promote trait optimization in various environments through
a systematic "fine-tuning" of many loci (Gifford et al. 2013; Rosas et al. 2013).
Nevertheless, this concept has not been well tested in crops, mainly because
few genes that regulate root traits have been identified. The main challenge in
the genetic analysis of root characteristics is the need for reliable and
high-throughput phenotypic assessment methods that can provide a proxy for
field performance since measuring root characteristics under open-field
conditions can be extremely challenging (Iannucci
et al. 2017). This is especially the case for genetic studies
involving very large samples. Therefore, different growth strategies are being
used under laboratory, greenhouse, as well as field conditions, and soil-less
growth media is the most commonly-used growth system (Wasaya et al. 2018). Plant growing methods that nearly
mimic the soil media have been reported to be more stable in mineral elements
and environmental factors, easier to operate, and more appropriate for root
morphological traits phenotyping in maize (Ju et
al. 2018). Therefore, to offer a better and robust tool for plant
behavior prediction under field conditions, various experimental growing
systems with soil-based substrates have been adopted (Zhu et al. 2005; Laperche et al. 2006; Ren et al.
2012; Liu et al. 2013). The root phenotypic data acquisition is
now becoming cheaper, quicker, and more effective, mainly due to rapid progress
in digitally automatic image analysis (Galkovskyi
et al. 2012; Pierret et al. 2013; Pace et al. 2014; Das et
al. 2015; Rellán-Álvarez et al. 2015; Symonova et al. 2015).
Today, image processing approaches have been widely used as reliable and less
time-consuming root phenotyping techniques and have become accessible through
various softwares. Recently, several studies have attempted to examine the
genetic basis of maize root system architectures in both field (Zaidi et al. 2016; Gu et al. 2017;
Zhang et al. 2018) and controlled conditions (Pace et al. 2015a, b; Zurek et al.
2015; Song et al. 2016; Liu et al. 2017; Sanchez et al.
2018). However, most of those studies used relatively short time-point
intervals and low-density marker genetic maps, resulting in large inter-marker
intervals (Song et al. 2016). The
full realization of the importance of root architecture for crop improvement
will require, therefore, a more comprehensive understanding of the particular
genetic loci involved in the variability of quantitative root traits (Price et al. 2007; Hochholdinger and Tuberosa
2009; Lynch 2013). Thus, we hypothesized that longer time-point
intervals combined with high-density Single Nucleotide Polymorphism (SNP)
bin-map would more efficiently and accurately estimate Quantitative Trait Loci
(QTLs).
The present study aimed to (i) construct a high-density GBS-SNPs genetic
bin-map for a RIL population obtained from a cross between P014 × E1312; (ii)
map novel potential chromosomal regions underlying maize root morphological
traits at 5 and 15 days after germination; and (iii) detect candidate genes
associated with root related-traits.
Materials and Methods
Genetic material and growth
conditions
The two parental maize lines used in this study, P014, E1312, and their
179 offspring were used as genetic material for the analysis of root
characteristics. The RIL family comprising 179 inbred lines was developed by a
cross between the maternal line P014 and the male parent E1312 and continuous
inbreeding for 9 generations. Genetically pure seeds were obtained from the
Jilin Agricultural University's experimental field in 2018.
Plant materials and growth
conditions
The experiments were laid out using a completely randomized block (CRD)
in growth chambers under a photoperiod of 14/10 h at a temperature of 25/22°C
and relative humidity of 70/80% (light/darkness). The light intensity was set
at 200 μmol photons m−2
s−1. Two independent growth chamber trials were
completed in April and July 2019. Seedlings were planted in polyvinyl chloride
(PVC) pipes sealed at the bottom with 4 manually pierced holes for drainage.
The dimensions of the PVC pipe were 26 cm and 9 cm in height and bottom
diameter. The PVC pots were filled with a mixture of
sandy soil and vermiculite (2:1 ratio). For data collection, each pot
consisted of three (3) seedlings was considered an experimental unit.
Root phenotyping
At each indicated time-point (5 and 15dag), seedlings were taken off
from the growth chamber and harvested for root traits analysis. To eliminate
soil residuals, the samples were carefully removed from the pots and washed
concisely with water. In each trial, 3 seedlings per line were examined
simultaneously at each specified time-point. All the collected traits are
listed in Table 1. To generate high-resolution images, each root sample was put
on a scanner (Perfection V800 Epson, resolution of 12800 dots per inch (dpi:
5039.37 dots per cm)) and using DJ-GXG02 software (www.
Dianjiangtech.com), the root images were then processed. If root
scanning could not be completed in a single day, seedlings were preserved by
submerging the roots in 30% ethanol and storing them in a cold room (about 4°C)
to prevent further growth (Sanchez et al.
2018). Root and shoot dry weights were measured after drying in an oven
dryer set at 75°C for at least 48 h until constant weight. For that purpose, an
electronic scale was used. A total of 10 traits including five (4) manually
measured (root dry weight, shoot dry weight, root to shoot dry weight ratio,
total plant biomass) and five (5) electronically recorded via DJ-GXG02 (total root length, surface area, projected area,
average root diameter, and root tips) were collected (Table 1).
Root traits statistical analysis
All statistical
analyses were performed using Minitab17 software (Minitab Inc., State College,
P.A., U.S.A.). For each root related trait in the two parental lines and RIL
population, descriptive statistics including mean, standard deviation, maximum
and minimum were calculated. Analysis of variance (ANOVA) using the Fit General
Linear Model was carried out to compare the two parental maize inbred lines for
the different morphological traits assessed at the respective time-points. As Table 1: Collected root related traits
initials and illustrations
|
Trait Name |
Abbreviations |
Trait description |
|
Root dry weight |
RDW |
Total root dry weight of the plant in gram |
|
Shoot dry weight |
SDW |
Total shoot dry weight the plant in gram |
|
Root dry weight/Shoot
dry weight |
RDW/SDW |
Root to shoot dry weight ratio in gram |
|
Total plant biomass |
TPB |
Total root dry weight + Total shoot dry weight
in gram |
|
Total root length |
TRL |
Cumulative length of all the roots in cm |
|
Projected area |
PRA |
Whole root system projected area in cm2 |
|
Surface area |
SUA |
|
|
Average root diameter |
ARD |
The average diameter of the entire root system
mm |
|
Root tips |
ROT |
Total number of all the root tips |
well, broad-sense heritability (h2) was
determined on an entry mean basis as described by Pace et al. (2015a, b). Normal distribution and Pearson
coefficients of correlation among traits in the RIL family were also generated.
Genotyping, QTL mapping, and
linkage map analysis
The RIL population was genotyped by sequencing (GBS) (www.broadinstitute.org/gatk), and MSTmap software (Wu et al.
2008) was used for linkage analysis
of marker data. Version 4.1 of QTL IciMapping software was used to perform QTL
mapping (Wang et al. 2016). With
reference to Meng et al. (2015)
and based on genotypic data of the 179 RILs with a final total of 4235
high-quality SNP markers spanned on all chromosomes, linkage map analysis was
undertaken using automatic parameter settings with the required data. Using the
Kosambi mapping function, recombination frequencies and pairwise distance were
converted in centimorgans (cM). The relative segregation ratio of each marker
and its relative deviation from the expected ratio were determined via the squared chi test. Briefly, the
markers were grouped at a LOD of ≥ 3.0, ordered, rippled then outputted
to construct the linkage map which covered a total length of 1514.57 cM
distributed on the 10 linkage groups. QTL analysis was performed following inclusive composite interval mapping for additive QTL (ICIM-ADD),
and a 1000 permutation test at 95% confidence level was used to define QTL
logarithm of odds (LOD) scores threshold. The walking speed was 1.0 cM and the
size of the windows was 5.0 cM. A LOD threshold peak score value of ≥ 2.5
was set to declare a significant QTL, which is commonly used in maize QTL
mapping (Li et al. 2014; Song et al.
2016). QTL additive effects and phenotypic variance explained (PVE) were
also analyzed.
Gene annotation and expression analysis
With
reference to the B73 genome, MaizeGDB (http://www.maizegdb.org/) and Gramene (https://www.gramene.org/) databases were used for
predicting functional annotations of the candidate genes. Two candidate genes GRMZM2G109056 and
GRMZM2G053458, with known functions in maize root growth (Stelpflug et al. 2016; Hoopes et
al. 2019) were tested for expression analysis via quantitative real-time PCR (qRT-PCR)
using primary root samples from the two
parental accessions ( P014 and E1312) at the two evaluated
time-points (5dag and 15dag). The total
RNA was mined using an RNA kit (TIANGEN, China), and cDNA was synthesized using a standard protocol
based on the Prime ScriptTM RT Reagent Kit (TaKaRa, Japan). qRT-PCR primers were
designed using Primer Premier 5.0 software, and Leunig (GenBank accession: GRMZM2G425377_T01) was selected as the
reference gene (Manoli et al. 2012).
All primer sequences were presented in Table 2. qRT-PCR trials in triplicates
were carried out in a system of 20 μL, consisting of 2 μL
cDNA, 1 μL of every single primer (μM), and 10 μL
qPCR Master mix using a Stratagene Mx3000P instrument. The relative level of
expression of the two indicated genes was determined using 2-∆∆CT method (Livak and Schmittgen 2001).
Results
Root morphological traits
phenotypic variability
The
seedlings were developed in a 2:1 ratio mixture of sandy soil and vermiculite
in a growth chamber for root trait measurements. The results related to
morphological performances of root and shoot traits display large variations
both in parents and their offspring (Table 3). Apparent differences about roots
between the two parents of the 179 RILs, P014 and E1312,
were observed at 5dag and 15dag. P014 possessed a longer root system than E1312
from 5dag. At 15dag, P014 compare to E1312 showed a ticker root system with a
larger and longer number of lateral roots present on the primary and seminal
roots. The inbred line P014 was therefore superior in phenotype because of its
larger number of roots especially at the Table 2: Oligonucleotide primers used for qRT-PCR
|
GRMZM2G109056-forward |
5’ATGTTCTGGCACGGGGTCGCGGA3’ |
|
GRMZM2G109056-reverse |
5’CGAGGAGCGAGGCGTTGAAGTCG3’ |
|
GRMZM2G053458-forward |
5’ATGCCAGACCACGGGCACGGGGT 3’ |
|
GRMZM2G053458-reverse |
5’TGGAACTGCCCGGACGGACACCC3’ |
|
Leunig-forward |
5’TCCAGTGCTACAGGGAAGGT3’ |
|
Leunig-reverse |
5’GTTAGTTCTTGAGCCCACGC3’ |
Table 3: Root
parameters performances in both parental and RIL lines
a Time points for root traits assessment: 5 and 15 dag
(days after germination). b Probability level of significance via student-test with, ns: no
significant difference; * : significantly different at P < 0.05; ** : significantly
different at P < 0.01 and ***
: significantly different at P < 0.001.
c Heritability estimates (broad sense). Root dry weight (RDW), shoot dry weight (SDW), total
plant biomass (TPB), root to shoot dry weight (RDW/SDW), total root length
(TRL), projected area (PRA), surface area (SUA), average root diameter (ARD),
and root tips (ROT)
later time point (15 dag). We also evaluated the architectural
differences among the two root systems of the parental lines about the nine
measured traits. Results from the study of the parental phenotypic data showed
gradual increases in eight of the nine root and shoot morphological traits
across the two time-points. In contrast, among the other traits, ARD and
RDW/SDW, progressively dropped, due likely to the continual emergence of new
leaves and roots by time. This result shows the instantaneous nature of the
development of the two parental root systems overtime which confirms the
pertinence of the two experimental time-points selected for root traits
assessment in this study. Analysis of variance related to the morphological
differences between the root systems of the two parental lines reveals highly
significant variations (P < 0.05; P < 0.01; P <
0.001) in the measured root related traits from the parental inbred lines
(Table 2). Comparatively, P014 showed higher values for, RDW,
TPB, TRL, PRA, SUA, and ROT than E1312 across both time-points (Table 3). This
supports the pertinence as well as the relevance of using the two parental
lines in this current study. For illustration, at 5dag, RDW, TPB, TRL, PRA, SUA and ROT in P014 versus E1312
were 0.04 g, 0.08 g, 70.45 cm, 9.25 cm2,
28.19 cm2, and 381; 0.01 g, 0.04 g, 36.86 cm, 5.12 cm2,
16.09 cm2, and 241, respectively (Table 3).
At 15 dag, P014 performed for the traits RDW, TPB, TRL, PRA, SUA, and ROT, 0.06
g, 0.17 g, 216.63 cm, 22.68 cm2, 72.96 cm2 and 1175.33,
respectively, while E1312 presented respectively 0.03 g, 0.13 g, 107.86 cm,
11.31 cm2, 35.71 cm2 and 641.67 for the same traits
(Table 3). Given together, these findings indicate that the two parents of the
RIL population exhibited substantial differences in most of the evaluated root
and shoot features at both time-points.
Subsequently, we
examined the distribution of the root related-traits in the 179 RILs. A large spectrum of variation was detected
for each trait (Table 3). The heritability values ranged between 7.6 and 71.9%,
and the highest value was observed from ROT with 71.9%, while SUA and PRA with
7.6% showed the lowest heritability (Table 3). For
most of the traits evaluated, heritability estimates were moderate (> 41%).
Similar correlation patterns were detected at all two time-points. At 15 dag, a
greater extent of significant correlations was observed comparing to 5 dag (Table 4). In terms of comparison, the
correlation coefficients between RDW, TPB, TRL, PRA, SUA, and ROT were similar
at 5 dag and 15 dag, but more strongly in 15 days old seedlings (Table 4). In
regards of all traits, SUA and PRA exhibited the strongest significant
correlations (r=
0.990) while RDW and SHL showed the lowest ones both at 5 dag (r = 0.157) and 15 dag (r =
0.194). TRL is correlated with all other traits at both time-points except ARD
and RDW/SDW at 5 dag and ARD at 15 dag (Table 4).
For example, strong to Table 4: Correlation coefficients of Pearson for root and shoot related traits in
the 179 RILs at each time-point
|
SDW |
RDW |
TPB |
RDW/SDW |
TRL |
PRA |
SUA |
ARD |
ROT |
|
|
SDW |
1 |
0.138 |
0.635** |
-0.296** |
0.489** |
0.292** |
0.292** |
-0.127 |
0.401** |
|
RDW |
0.675** |
1 |
0.838** |
0.738** |
0.312** |
0.215** |
0.214** |
0.159* |
0.287** |
|
TPB |
0.959** |
0.853** |
1 |
0.416** |
0.509** |
0.326** |
0.326** |
0.045 |
0.441** |
|
RDW/SDW |
-0.018 |
0.628** |
0.226** |
1 |
-0.010 |
0.010 |
0.010 |
0.200** |
0.062 |
|
TRL |
0.594** |
0.761** |
0.707** |
0.380** |
1 |
0.770** |
0.771** |
-0.092 |
0.834** |
|
PRA |
0.652** |
0.766** |
0.749** |
0.322** |
0.945** |
1 |
0.990** |
0.292** |
0.628** |
|
SUA |
0.652** |
0.766** |
0.749** |
0.322** |
0.945** |
0.990** |
1 |
0.292** |
0.628** |
|
ARD |
0.242** |
0.132 |
0.218** |
-0.109 |
0.078 |
0.350** |
0.350** |
1 |
0.063 |
|
ROT |
0.364** |
0.600** |
0.485** |
0.381** |
0.883** |
0.822** |
0.822** |
0.076 |
1 |
Correlation coefficients at 5 dag
(upper right) and 15 dag (lower left). The symbol * and ** indicate
respectively, significance at P <
0.05 and at P < 0.01. Root dry
weight (RDW), shoot dry weight (SDW), total plant biomass (TPB), root to shoot
dry weight (RDW/SDW), total root length (TRL), projected area (PRA), surface
area (SUA), average root diameter (ARD), and root tips (ROT)
%20904-914_files/image002.png)
Fig. 1: P014 × E1312 RIL high-density
GBS-SNPs genetic bin-map (A) and
consistency map (B)
moderate significant correlations at 15 dag were
obtained among TRL and PRA (r = 0.945), TRL and SUA (r =
0.945), TRL and ROT (r = 0.883),
TRL and RDW (r = 0.761), and TRL and
TPB (r = 0.707) (Table
3; P < 0.01). Interestingly,
at 15dag RDW is significantly (P <
0.05; P < 0.01) linked positively
to all traits excluding ARD (Table 4). The high correlation between ROT and RDW
at 15dag (r = 0.600) confirmed the
predominant contribution of root tip number to RDW, TPB and TRL. ARD is weakly
correlated (r = 0.218–0.350) with
only SDW, TPB, SUA, and PRA, in seedlings at 15dag (Table 4).
High-quality genetic linkage map
With reference to the B73 genome,
an initial total of 7275986 SNPs was obtained using GATK-samtools
software (www.broadinstitute.org/gatk).
After screening and filtering out the low-quality SNPs according to the
following criteria: (i) SNPs sequencing depth for parents < 6 fold, (ii)
SNPs coverage for offsprings’ number <60%, and (iii) SNPs displaying a
highly significant (P < 0.01)
segregation ratio, a panel of 63990 SNPs were obtained for linkage mapping.
With the removal of unlinked and co-segregated SNP markers, we established the
bin-map with 4235 high-quality SNP markers spanned along 10 chromosomes
(linkage groups). The generated map covered a total genetic length of 1514.57
cM, and no evident issues with the map were detected (Table
5, Fig. 1). The average number of markers per linkage group and the
distance between successive markers were 423.5 and 0.38 cM, respectively (Table
5). Chromosome 4
contained the highest number of markers with 635 markers spanning 154.16 cM
(the highest density) while chromosome 8 contained the lowest number with 285
markers spanning 124.63 cM. The lowest marker density consisted of 389 markers
spanning 185.41 cM on chromosome 1 (Table 5). Among the 10 chromosomes, the
most extented genetic length was 209.19 cM containing 477 bin markers for
chromosome 2. In contrast, chromosome 6, with 362 bin markers covering 123.82
cM, corresponds to the shortest genetic length covered in this map (Fig. 1,
Table 5). The genetic gap length varied from 2.5 cM (on chromosome 6) to 8.22
cM (on chromosome7), and five genetic gaps varied from 5 cM to 9 cM while the
other five ranged from 2.5 cM to 4.3 cM (Table 5).
QTLs and
genes associated with root related traits at the seedling stage
%20904-914_files/image004.png)
Fig. 2: Maize GBS-SNPs map showing QTLs
for root dry weight (RDW), shoot dry weight (SDW), total plant biomass(TPB),
root to shoot dry weight (RDW/SDW), total length of roots (TRL), projected area
(PRA), root surface area (SUA) and root tips (ROT) at 5 and 15 days after
germination in the P014 × E1312 RIL population. Each trait is symbolized by one
specific color and the location is relatively proportional to that of the
linkage map length expressed in cM
%20904-914_files/image006.png)
Fig. 3: Relative expression levels of
two candidate genes (GRMZM2G109056 and GRMZM2G053458) in the primary roots
(sampled at 5dag and 15dag) of the two parental accessions E1312 (in blue) and
P014 (in red) tested via qRT-PCR. * and ** show the significance at P < 0.05 and P < 0.01, respectively
The whole maize genome was scanned for root QTLs
via inclusive composite interval
mapping for additive QTL (ICIM-ADD) with LOD ≥ 2.5 as the threshold. A
total of fourteen (14) substantial QTLs with a phenotypic variance explained,
ranging from 1.78 to 16.05% were obtained across the two time-points (Fig. 2;
Table 6). The detected QTLs were allocated to all chromosomes except
chromosomes 3, 6, and 9, with no significant QTL detected. Chromosome 4
contained the highest number of QTLs, with 4 QTLs detected (Fig. 2; Table 6).
Other chromosomes contained between 1 and 3 QTLs. Different QTL numbers were
identified for each time-point, eleven and three at 5 and 15 dag, respectively
(Table 6). When examining the number of QTL inheriting parental favorable
alleles per time-point, as expected, those alleles did not segregate from the
parents uniformly (Table 6). The alleles involved in increasing root morphological characteristics at
ten chromosomal positions belonged to the maternal inbred P014. At the same
time, the paternal line E1312 contributed to four underlying loci,therefore, the imperative implication of the two
parents in root features discrimination. Subsequently, QTL clusters were
identified at both time-points. The four QTLs detected on chromosome 4 (qPRA5d-4-1,
qSUA5d-4-1, qSDW15d-4-1, and qTPB15d-4-1)
overlapped and spanned one genetic region, at 90.5–92.5cM interval (Fig. 2,
Table 6). The gene model GRMZM2G068506 predicted to encode a glucose-1-phosphate
adenylyltransferase was found within this chromosomal
region as associated with these four roots above traits (Table 7). All QTLs
detected on chromosome 1 are related to RDW (qRDW5d-1-1, qRDW5d-1-2,
and qRDW5d-1-3) and spanned the 0–35.5cM chromosomal region with QTL
(qRDW5d-1-2, LOD = 19.39, PVE = 16.05%) as the most significant QTL
detected at 5dag (Table 6). Interestingly, this region harbored five candidate
genes viz., GRMZM2G109056,
GRMZM2G123159, GRMZM2G153476, GRMZM5G874478, and GRMZM2G053458 encoding for a
lipoxygenase4, a silencing gene B102, a 30S ribosomal protein S13
chloroplastic-like, a glycine-rich RNA-binding protein 8, and a ferredoxin 3
protein, respectively (Table 7). At 5 dag, the QTL on chromosome 10 (qTRL5d-10-1,
LOD= 3.46, PVE= 8.5%) presented the largest additive effect (Table 6). This QTL
inherited the favorable alleles from E1312 line. It harbored two candidate
genes GRMZM2G116542 and GRMZM2G016477 predicted to encode a putative Spc97 /
Spc98 family of spindle pole body (SBP) component and a putative leucine-rich
repeat receptor-like protein kinase, respectively (Table 7). At 15dag, the QTL
located on chromosome 7 and involved in TRL (qTRL15d-7-1, LOD =
62.6, PVE = 1.78%) showed the highest effect in increasing root related-traits
(Table 6). This QTL inherited the favorable alleles from P014 line.
Collectively, four QTL clusters on chromosome 4
(qPRA5d-4-1, qSUA5d-4-1, qSDW15d-4-1, and qTPB15d-4-1)
and two closely located QTLs on chromosome 1 (qRDW5d-1-2, qRDW5d-1-3)
were detected in this study (Fig. 2, Table 6). The remaining QTLs were
distributed in different chromosomal regions. Further analysis of the QTL
results indicated that each co-located QTL position's root characteristics were
strongly linked to one another. The details of all candidate genes (27)
associated with the different QTLs and the functional annotations are presented
in Table 7.
Gene expression in parental
lines
The
expression analysis results related to the two evaluated candidate genes
GRMZM2G109056 and GRMZM2G053458 showed significant differences in gene
expression levels between the two parental materials in primary roots at both 5
dag and 15 dag (P < 0.05, P < 0.01, Fig. 3).
As shown in Figure 3, the two genes showed higher expression level for P014
than E1312 at both time-points (5 dag and 15 dag), which supports the apparent
significant phenotypic differences observed among the two parental lines (Table
3) as well as the regulating role of the evaluated genes in root growth. Thus, at
5 dag, GRMZM2G109056 and GRMZM2G053458 exhibited approximately 2.5-fold and
1.8-fold higher expression in the P014 line compared to E1312, respectively.
Similarly, at 15 dag, GRMZM2G109056 and GRMZM2G053458 showed 2.2-fold higher
expression in P014 compared to E1312 (Fig. 3).
Discussion
Table 5: Properties of the high-density linkage map generated
from P014 × E1312
|
Linkage Group ID |
Total Marker |
Total Distance (cM) |
Average Distance (cM) |
Max Gap (cM) |
Gap < 5 cM (%) |
|
1 |
389 |
185.41 |
0.48 |
4.10 |
3.11 |
|
2 |
477 |
209.19 |
0.44 |
4.30 |
0.95 |
|
3 |
453 |
150.54 |
0.33 |
5.62 |
5.18 |
|
4 |
688 |
154.16 |
0.22 |
3.02 |
1.97 |
|
5 |
329 |
147.19 |
0.45 |
6.13 |
5.51 |
|
6 |
362 |
123.82 |
0.34 |
2.50 |
3.11 |
|
7 |
635 |
155.62 |
0.25 |
8.22 |
3.02 |
|
8 |
285 |
124.63 |
0.44 |
4.00 |
2.52 |
|
9 |
321 |
124.29 |
0.39 |
5.58 |
5.23 |
|
10 |
296 |
139.12 |
0.47 |
5.53 |
16.44 |
|
Total |
4235 |
1514.57 |
0.38 |
8.22 |
4.70 |
Table 6: Summary of root and shoot traits QTLs detected in P014 × E1312 RIL population
|
Chr |
Bin |
Peak
Pos (cM) |
Marker
interval |
Genetic
interval (cM) |
LOD
score |
PVEb (%) |
Addc eff. |
|
|
5 dag |
|
|
|
|
|
|
|
|
|
qSDW5d-8-1 |
8 |
8.09 |
17 |
snp53934-snp53932 |
16.5-19.5 |
3.01 |
6.18 |
0.01 |
|
qRDW5d-1-1 |
1 |
1.11 |
1 |
snp9142-snp9174 |
0-1.5 |
8.17 |
5.95 |
-0.01 |
|
qRDW5d-1-2 |
1 |
1.09 |
30 |
snp8010-snp8064 |
29.5-30.5 |
19.39 |
16.05 |
0.01 |
|
qRDW5d-1-3 |
1 |
1.09 |
35 |
snp7954-snp7955 |
34.5-35.5 |
10.19 |
7.71 |
-0.01 |
|
qRDW5d-5-1 |
5 |
5.03 |
110 |
snp32110-snp32111 |
107.5-110.5 |
2.90 |
1.89 |
0.01 |
|
qTPB5d-7-1 |
7 |
7.04-05 |
34 |
snp48526-snp48153 |
31.5-38.5 |
2.61 |
6.26 |
0.01 |
|
qRDW/SDW5d-5-1 |
5 |
5.04-05 |
62 |
snp36024-snp35632 |
60.5-62.5 |
3.38 |
8.37 |
-0.22 |
|
qTRL5d-10-1 |
10 |
10.05-06 |
51 |
snp62466-snp62578 |
50.5-51.5 |
3.46 |
8.50 |
-7.48 |
|
4 |
4.05 |
92 |
snp25434-snp25474 |
90.5-92.5 |
2.63 |
6.57 |
0.83 |
|
|
4 |
4.05 |
92 |
snp25434-snp25474 |
90.5-92.5 |
2.63 |
6.58 |
2.61 |
|
|
qROT5d-2-1 |
2 |
2.09 |
23 |
snp16526-snp16488 |
22.5-23.5 |
2.54 |
8.30 |
39.54 |
|
15 dag |
|
|||||||
|
4 |
4.05 |
91 |
snp25452-snp25434 |
90.5-92.5 |
3.28 |
8.22 |
0.02 |
|
|
qTPB15d-4-1 |
4 |
4.05 |
91 |
snp25452-snp25434 |
90.5-92.5 |
3.59 |
8.91 |
0.02 |
|
qTRL15d-7-1 |
7 |
7.06 |
4 |
snp48875-snp48869 |
3.5-4.5 |
62.60 |
1.78 |
123.40 |
a The
identified QTLs; the name contains trait initials, seedling growing time-point,
and the number of correspondent chromosome; b The percentages of phenotypic
variations explained for every single QTL; c The QTL additive effect (positive values
indicate that P014 provides increased alleles and negative ones indicate that
E1312 alleles increased the trait)
Table 7: Candidate genes within the QTL
regions and functional annotations
|
QTL |
Trait |
Chr |
Candidate gene |
Functional annotation |
|
qRDW5d-1-2; qRDW5d-1-3 |
RDW |
1 1 |
GRMZM2G109056 GRMZM2G123159 |
lipoxygenase4 silencing gene B 102 |
|
|
|
1 |
GRMZM2G153476 |
30S ribosomal protein S13, chloroplastic-like |
|
qRDW5d-1-1 |
RDW |
1 |
GRMZM5G874478 |
glycine-rich RNA-binding protein 8 |
|
|
|
1 |
GRMZM2G053458 |
ferredoxin 3 |
|
qROT5d-2-1 |
ROT |
2 |
GRMZM5G841015 |
pollen-specific leucine-rich repeat extensin-like protein 3 |
|
qPRA5d-4-1; qSUA5d-4-1 qSDW15d-4-1; qTPB15d-4-1 |
PRA; SUA SDW; TPB |
4 4 |
GRMZM2G068506 GRMZM2G068506 |
Glucose-1-phosphate adenylyltransferase |
|
qRDW/SDW5d-5-1 |
RDW/SDW |
5 |
GRMZM2G139300 |
cell wall invertase 1 |
|
|
|
5 |
GRMZM2G118737 |
Alkaline/neutral invertase CINV2 |
|
|
|
5 |
GRMZM2G326111 |
peptidyl-prolyl cis-trans isomerase-like |
|
|
|
5 |
GRMZM2G032628 |
amylose extender 1 |
|
|
|
5 |
GRMZM2G143008 |
acetolactate synthase 1 |
|
|
|
5 |
GRMZM2G176358 |
histone H3.3 |
|
|
|
5 |
GRMZM2G074017 |
ATPase inhibitor |
|
|
|
5 |
GRMZM2G114241 |
pentatricopeptide repeat-containing protein
At2g29760, chloroplastic |
|
|
|
5 |
GRMZM2G113453 |
calmodulin binding protein1 |
|
|
|
5 |
GRMZM2G337749 |
DDT domain-containing protein |
|
|
|
5 |
GRMZM2G064255 |
glutathione transferase17 |
|
qTPB5d-7-1 |
TPB |
7 |
GRMZM2G027217 |
peroxidase 2-like |
|
|
|
7 |
GRMZM2G054136 |
ribosomal proteinS6 |
|
|
|
7 |
GRMZM2G025992 |
Superoxide dismutase |
|
|
|
7 |
GRMZM2G056407 |
fused leaves1 |
|
|
|
7 |
GRMZM2G015703 |
putative protein kinase superfamily protein |
|
|
|
7 |
GRMZM2G307119 |
branched silkless
1 |
|
qTRL5d-10-1 |
TRL |
7 10 10 |
GRMZM2G427815 GRMZM2G116542 GRMZM2G016477 |
peroxidase3 Spc97 / Spc98 family of spindle pole body
(SBP) component putative leucine-rich repeat receptor-like
protein kinase family protein |
Wide ranges of variations in terms of root and shoot biomass (RDW, SDW,
TPB) amassed were observed between P014 and E1312 at both time-points
investigated. The female parent P014 achieved significantly more root tip
number (ROT) with higher total root length (TRL). The decrease in ARD may be
attributed to the formation over time of new and fresh roots. Besides, a large
spectrum of phenotypic variation was observed from the offspring for most of
the evaluated root morphological traits (Table 3). Considerable phenotypic
variations for root and shoot traits in different maize panels have been
previously reported (Pace et al. 2014;
Pace et al. 2015a, b; Abdel et al. 2015; Pace et al. 2015a,
b; Song et al. 2016; Ju et al. 2018; Moussa et al. 2018).
Broad sense heritability ranged from 7.6 to 71.9% (Table 3). Similar
heritability ranges have been earlier recorded in similar studies regarding
maize root traits at various growth stages under field as well as growth
chamber conditions (Gu et al. 2017;
Sanchez et al. 2018). Due to the high plasticity and quantitative
nature of root growth, some root traits were more reproducible than others
under the same conditions. In the present study, similar (but with a greater
extent at 15 dag) correlations patterns were detected at all time-points.
Therefore, at 15dag, low to strong positive correlations (r between 0.194–0.959, P < 0.01) were found between root
(tips, length, dry weight, surface area, projected area) and shoot dry weight
(Table 4). As well, RDW and TRL showed significant positive correlation with
all other traits except ARD at 15dag. These findings were consistent with those
observed by several authors in similar studies (Song
et al. 2016; Hu et al. 2017; Ju et al. 2018).
Strong correlation between ROT and RDW confirmed the predominant contribution
of root tip number to RDW, TPB and TRL. Recent reports speculated that root
system made up of longer but lesser lateral roots are greatly indicated for
drought resistance, nitrogen assimilation from soils nitrogen-deficient, and
increasing final yields under biotic and abiotic constraints than root system
composed of shorter lateral roots (Zhan and
Lynch 2015; Zhan et al. 2015).
Recent evidence suggests that genetic map quality
significantly affects the accuracy of QTL localization (Zhou et al. 2016; Wang et al. 2018). Besides, high
marker density maps are more suitable to accurately identify more recombination
breakpoints and QTL locations in bi-parental mapping populations. Moreover, it
was previously reported that high-density maps could reliably predict potential
genes between two successive markers in narrow regions (Zhou et al. 2016). However, most of the root QTL studies in
crops have been carried out using relatively low-density marker genetic linkage
maps made up with simple sequence repeat (SSR), random amplified polymorphic
DNA (RAPD), and restriction fragment length polymorphism (RFLP), resulting
consequently in relatively scanty inter-marker intervals. Today, the advances
in next-generation sequencing (NGS) technology such as GBS have enabled higher
marker densities genetic maps via SNP
markers for improved QTLs identification in numerous maize panels for various
root morphological traits (Pace et al.
2015a, b; Burton et al. 2015; Zaidi et al. 2016; Sanchez et al.
2018). In this study, a high-quality bin-map composed of 4235 SNPs
spanning a genomic distance of 1514.57 cM with an average genetic marker
interval of 0.38 cM was constructed (Table 5, Fig. 1). Compared to previous
genetic linkage maps using common molecular markers (Mano et al. 2007) as well as GBS-SNPs markers (Burton et al. 2015) in nineteen maize
recently root-trait QTL studies (last summary) (Bray
and Topp 2018), in which the genetic marker intervals ranged from 0.6 to
17.2 cM, the resolution in our study is substantially enhanced. In relatively large populations, QTL mapping based on
GBS-SNPs has been lately proposed as an efficient and accurate way to detect
rapidly beneficial alleles in genetic resources (Zhou
et al. 2016). Mapping of important QTLs involved in root
morphological traits is crucial for root and shoot improvement via marker-assisted selection (MAS) and
cloning for potential candidate genes controlling root system architecture
growth and development in corn (Cai et al.
2012). The availability of fast and reliable root phenotyping techniques
through numerous digital imaging softwares and growth media, as well as diverse
controlled environment growing systems, offers an efficient and accurate method
for fine mapping and cloning root system architecture QTLs. Thus, a
considerable number of QTLs underlying root system development in various
bi-parental mapping populations have been recently reported (Zurek et al. 2015; Song et al. 2016; Gu
et al. 2017; Ju et al. 2018; Bray and Topp 2018). In the
present investigation, we detected 14 QTLs controlling the variations of the
nine seedling roots related traits evaluated across two different time-points
in the P014 × E1312 RIL mapping population. Each QTL explained relatively
between 1.78 and 16.05% of the overall phenotypic variation per trait. Song et al. (2016) recently detected 62
QTLs for 24 seedling root system architecture-related traits using a
recombinant inbred line population derived from two Chinese popular commercial
inbred lines and also reported small size effects with every single QTL
explained from 1.6 to 11.6% of the total phenotypic variation. Burton et al. (2014, 2015) reported similar ranges of phenotypic
variation explained, 0.44 to 13.5% and 4.7 to 12% using three famous maize US
inbred populations in two different studies for various anatomical and
architectural root traits, respectively. Compared to those different studies,
our findings exhibited somehow higher percentages of phenotypic variance
explained.
In this current study, root traits QTLs
overlapped or closely linked were localized on chromosomal bins 1.09, 4.05, and
10.05–10.06 (Table 6). An important number of root related trait QTLs
identified are in line with those reported in previous findings. Using the
results of fifteen QTL studies from nine different bi-parental mapping
populations, a meta-analysis study of QTLs recapitulated many putative QTL
clusters related to maize root (Hund et al.
2011). Interestingly, many clusters including Ax-5 (at bin 2.08–2.10),
Ax-10 (at bin 5.01–5.03), Rt-11
(at bin 5.03), Ax-11 (at bin 5.04), Ax-16 (at bin 7.04), Rt-14 (at bin 7.04–7.05), and Ax-20 (at bin 10.05–10.06), overlapped with
several QTLs viz., qROT5d-2-1 (at bin 2.09), qRDW5d-5-1
(at bin 5.03), qRDW/SDW5d-5-1 (at bin 5.04–5.05),
qTPB5d-7-1 (at bin 7.04–7.05), and qTRL5d-10-1 (at bin
10.05–10.06) detected in this current study. Besides, among the
twenty-four important MQTLs listed by Hund
et al. (2011), four MQTLs
namely Rt-11, Ax-11, Ax-16, and Rt-14 collocated with three of our detected
QTLs, qRDW5d-5-1 (at bin 5.03), qRDW/SDW5d-5-1
(at bin 5.04–5.05), and qTPB5d-7-1 (at bin 7.04–7.05)). QTL qRDW/SDW5d-5-1 located on
chromosome 5 was closely located near a QTL for total root dry weight to total
shoot dry weight ratio identified recently from a widely adapted Chinese
hybrid ZD958 (Zheng58 × Chang7-2) in bin 5.05 (Song
et al. 2016). QTL qRDW/SDW5d-5-1 located on Chromosome 5 and significantly
associated with root to shoot dry weight housed the gene models GRMZM2G326111
and GRMZM2G176358 highly expressed in the primary root at VE and V1 stages (Stelpflug et
al. 2016; Hoopes et al. 2019).
Based on B73, the absolute expression values for GRMZM2G326111 were 41882.26
for the primary roots in VE stage and 34480.2 during V1 developmental stage,
relative to 51527.2, which was the maximum expression level (Stelpflug et
al. 2016; Hoopes et al. 2019).
The expression values for GRMZM2G176358 were 3291.8 at VE stage and 4077.44 at
V1 stage, relative to 6319.12, as the maximum expression of GRMZM2G176358 (Stelpflug et
al. 2016; Hoopes et al. 2019).
QTL qTPB5d-7-1
associated with total plant biomass on chromosome 7 harbored the candidate gene
GRMZM2G054136 encoding a
ribosomal protein S6. Its absolute expression values were 28748.56 at VE and
24409.6 during V1, out of 35599.1, as the maximum expression level of
GRMZM2G054136 (Stelpflug et al. 2016; Hoopes et
al. 2019). The chromosomal bin 10.05–10.06 region was
previously revealed as rich in QTLs for maize root traits (Gu et al. 2017). In agreement with the involvement of the chromosomal bins 10.05–10.6
(Ax-20) and 2.08–2.10 (Ax-5) in the regulation of axile root number and length
reported by Hund et al. (2011), we indicate the role of these regions in the
control of total root tips number and length (qROT5d-2-1 and qTRL5d-10-1).
Three closely linked loci for root dry weight on bins
1.09 and 1.11 (qRDW5d-1-2,
qRDW5d-1-3, and qRDW5d-1-1) were found in linkage
disequilibrium with a QTL detected in a similar study (Gu et al. 2017). Interestingly, five genes GRMZM2G109056, GRMZM2G123159,
GRMZM2G153476, GRMZM5G874478, and GRMZM2G053458 were remarkably detected within
this genetic region. Gene
model GRMZM2G109056 codes for a lipoxygenase4 enzyme. Its absolute expression
levels were 12832.73 and 18179.96 for the primary root at VE and V1,
respectively, out of a maximum expression level of 21054.8 (Stelpflug et
al. 2016; Hoopes et al. 2019).
Gene model GRMZM2G053458 encodes a ferredoxin 3protein with absolute expression
levels of 13800.03 at VE and 27658.8 at V1, relative to 38759.4, which was the
maximum expression level for GRMZM2G053458 (Stelpflug et
al. 2016; Hoopes et al. 2019). This
putative region might be a special noteworthy locus for marker-assisted
breeding critical for further investigations. However, little overlap in QTL
detected between previous studies and our work was found for many loci in
chromosomal bins 2.09 (qROT5d-2-1),
4.05 (qPRA5d-4-1, qSUA5d-4-1,
qSDW15d-4-1, qTPB15d-4-1), 7.06 (qTRL15d-7-1), and 8.09 (qSDW5d-8-1); which may
therefore be considered novel. When
looking for QTLs overlapping previously recorded gene mutants involved in maize
roots development, QTL qRDW5d-1-3 detected for RDW in bin 1.09 at
the physical position 263,312,877-263,312,878 bp is positioned just a little
more 9 Mb from Rth1, a maize
roothairless gene encoding a homolog of sec3 involved in polar exocytosis (Wen et al. 2005). As well, Rth2,
another gene regulating maize root hair elongation (Wen and Schnable 1994) collocated with our detected QTL for root
to shoot dry weight (qRDW/SDW5d-5-1) in bin 5.04 on chromosome 5.
Quantitative real-time PCR was widely used to validate
gene expression with high accuracy and sensitivity (Bustin 2002; Bustin and Nolan 2004). In this study, two potential
genes were checked for expression levels using primary root samples from the
two parental maize accessions. At all considered time-points (5 dag, 15 dag),
the two evaluated genes GRMZM2G109056 and GRMZM2G053458 acted as positive regulators
for root growth with significantly higher expression levels in P014 as compared
to E1312. The genes being assessed could be of great value for future investigations.
Conclusion
This study provides a comprehensive analysis of the genetic
architecture of maize root related traits. We performed a QTL mapping using a high-density marker genetic map through 179
maize accessions genotyped by GBS-seq for root morphological traits at two
vegetative time-points (5 dag and 15 dag) under standard laboratory conditions.
We identified a total of 14 QTLs using the inclusive
composite interval mapping method. The detected QTLs showed individual phenotypic
contribution rates ranging from 1.78 to 16.05% and were
significantly associated with SDW (2), RDW (4), TPB (2), RDW/SDW (1), SUA (1),
PRA (1), ROT (1), and TRL (2). However, one major QTL
region on chromosome 4 was found to be associated with multiple traits, including
PRA, SUA, SDW, and TPB (qPRA5d-4-1, qSUA5d-4-1,
qSDW15d-4-1, and qTPB15d-4-1). Compared to several root QTL previous studies, QTLs located in
chromosomal bins 2.09 (qROT5d-2-1), 4.05 (qPRA5d-4-1,
qSUA5d-4-1, qSDW15d-4-1, qTPB15d-4-1), 7.06
(qTRL15d-7-1), and 8.09 (qSDW5d-8-1) were found to be
novel. Furthermore, two candidate genes GRMZM2G109056 and GRMZM2G053458,
overlapping with QTLs for root dry weight on chromosome 1, were detected and
verified for expression level. The evaluated genes were shown to serve as
positive regulators for root growth in the mapping parents. Upcoming research
will deeply explore these genomic regions to fully illuminate the genetic basis
regulating root traits in maize.
Acknowledgments
This research was funded by the Modern Crop Seed Industry development of
Jilin Province, China, (to Piwu Wang).
Author Contributions
AAM, PW, and JQ planned the experiments; AAM, AM, YJ, and QZ performed
the phenotyping and analyzed the data; AAM prepared the original draft; MGAE,
AS, and GA reviewed and edited the manuscript.
Conflict of Interest
There is no conflict of interest among the
authors and institutions where the research has been conducted
Data Availability Declaration
Data included in this paper are in the custody
of corresponding author and can be shared on request
References
Abdel-Ghani
AH, G Bharath, K Jordon, P Constantin, PJ Gonzalez, P Jenaro, R Matamoros, J
Pablo, S Martin, M Lee, T Lübberstedt, J Pace (2015). Association analysis of
genes involved in maize (Zea mays L.)
root development with seedling and agronomic traits under contrasting nitrogen
levels. Plant Mol Biol 88:133‒147
Abdel-Ghani
AH, B Kumar, J Reyes-Matamoros, PJ Gonzalez-Portilla, C Jansen, SMJ Pablo, M
Lee, T Lübberstedt (2013). Genotypic variation and relationships between
seedling and adult plant traits in maize (Zea
mays L.) inbred lines grown under contrasting nitrogen levels. Euphytica 189:123‒133
Bray AL,
CN Topp (2018). The Quantitative Genetic Control of Root Architecture in Maize.
Plant Cell Physiol 59:1919‒1930
Burton
AL, J Johnson, J Foerster, MT Hanlon, SM Kaeppler, JP Lynch, KM Brown (2015).
QTL mapping and phenotypic variation of root anatomical traits in maize (Zea mays L.). Theor Appl Genet 128:93‒106
Burton
AL, JM Johnson, JM Foerster, CN Hirsch, C Buell, MT Hanlon, SM Kaeppler, KM
Brown, JP Lynch (2014). QTL mapping and phenotypic variation for root
architectural traits in maize (Zea mays
L.). Theor Appl Genet 127:2293‒2311
Bustin
SA (2002). INVITED REVIEW Quantification of mRNA using real-time reverse
transcription PCR (RT-PCR): Trends and problems. J Mol Endocrinol 29:23‒39
Bustin SA, T Nolan (2004). Pitfalls of quantitative
real-time reverse-transcription polymerase chain reaction. J Biomol Technol 15:155‒166
Cai H, F
Chen, G Mi, F Zhang, HP Maurer, W Liu, JC Reif, L Yuan (2012). Mapping QTLs for
root system architecture of maize (Zea
mays L.) in the field at different developmental stages. Theor Appl Genet 125:1313‒1324
Das A, H
Schneider, J Burridge, AKM Ascanio, T Wojciechowski, CN Topp, JP Lynch, JS
Weitz, A Bucksch (2015). Digital imaging of root traits (DIRT): A
high-throughput computing and collaboration platform for field-based root
phenomics. Plant Meth 11:51-62
Galkovskyi T, Y Mileyko, A
Bucksch, B Moore, O Symonova, CA Price, CN Topp, AS Iyer-Pascuzzi, PR Zurek, S
Fang, J Harer, PN Benfey, JS Weitz (2012). GiA Roots:
Software for the high throughput analysis of plant root system architecture. BMC Plant Biol 12; Article 116
Gifford
ML, JA Banta, MS Katari, J Hulsmans, L Chen, D Ristova, D Tranchina, MD
Purugganan, GM Coruzzi, KD Birnbaum (2013). Plasticity regulators modulate
specific root traits in discrete nitrogen environments. PLoS Genet 9; Article e1003760
Gu D, X
Mei, T Yu, N Sun, D Xu, C Liu, Y Cai (2017). QTL identification for brace-root traits of maize in
different generations and environments. Crop
Sci 57:13‒21
Hammer
GL, Z Dong, G McLean, A Doherty, C Messina, J Schussler, C Zinselmeier, S
Paszkiewicz, M Cooper (2009). Can changes in canopy and/or root system
architecture explain historical maize yield trends in the US corn belt? Crop Sci 49:299‒312
Hochholdinger
F, R Tuberosa (2009). Genetic and genomic dissection of maize root development
and architecture. Curr Opin Plant Biol 12:172‒177
Hoopes
GM, JP Hamilton, JC Wood, E Esteban, A Pasha, B Vaillancourt, NJ Provart, CR
Buell (2019). An updated gene atlas for maize reveals organ-specific and stress-induced
genes. Plant J 97:1154‒1167
Hu S, DL
Sanchez, C Wang, AE Lipka, Y Yin, CAC Gardner, T Lübberstedt (2017).
Brassinosteroid and gibberellin control of seedling traits in maize (Zea mays L.). Plant Sci 263:132‒141
Hund A,
R Reimer, R Messmer (2011). A consensus map of QTLs controlling the root length
of maize. Plant Soil 344:143‒158
Iannucci
A, D Marone, MA Russo, PD Vita, V Miullo,
P Ferragonio, A Blanco, A Gadaleta, AM Mastrangelo (2017). Mapping QTL for root
and shoot morphological traits in a Durum Wheat × T. dicoccum segregating population at seedling stage. Intl J Genomics 2017; Article 6876393
Ju C, W
Zhang, Y Liu, Y Gao, X Wang, J Yan, X Yang (2018). Genetic analysis of seedling
root traits reveals the association of root traits with other agronomic traits
in maize. BMC Plant Biol 18; Article
117
Laperche
A, F Devienne-Barret, O Maury, JL Gouis,
B Ney (2006). A simplified conceptual model of carbon/nitrogen functioning for
QTL analysis of winter wheat adaptation to nitrogen deficiency. Theor Appl Genet 113:1131‒1146
Li F, H
Jia, L Liu, C Zhang, Z Liu, Z Zhang (2014). Quantitative trait loci mapping for
kernel row number using chromosome segment substitution lines in maize. Genet Mol Res 13:1707‒1716
Liu X, R
Li, X Chang, R Jing (2013). Mapping QTLs for seedling root traits in a doubled
haploid wheat population under different water regimes. Euphytica 189:51‒66
Liu Z, K
Gao, S Shan, R Gu, Z Wang, EJ Craft (2017). Comparative analysis of root traits
and the associated QTLs for maize seedlings grown in paper roll, hydroponics
and vermiculite culture system. Front
Plant Sci 8; Article 436
Livak
KJ, TD Schmittgen (2001). Analysis of relative gene expression data using
real-time quantitative PCR and the 2−ΔΔCT method. Methods 25:402‒408
Lynch JP
(2013). Steep, cheap, and deep: An ideotype to optimize water and N acquisition
by maize root systems. Ann Bot 112:347‒357
Mano Y,
F Omori, T Takamizo, B Kindiger, RM Bird, C Loaisiga, H Takahashi (2007). QTL
mapping of root aerenchyma formation in seedlings of a maize × rare teosinte Zea nicaraguensis Cross. Plant Soil 295:103‒113
Manoli
A, A Sturaro, S Trevisan, S Quaggiotti, A Nonis (2012). Evaluation of candidate
reference genes for qPCR in maize. J Plant
Physiol 169:807‒815
Meng L,
H Li, L Zhang, JJTCJ Wang (2015). QTL IciMapping: Integrated
software for genetic linkage map construction and quantitative trait locus
mapping in biparental populations. Crop J 3:269‒283
Moussa AA, V
Salako, DC Gbemavo, M Zaman-Allah, R Kakaď, Y Bakasso (2018). Performances
agro-morphologiques des variétés locales et améliorées de maďs au sud-ouest du
Niger. Afr Crop Sci J 26:157‒173
Pace J,
C Gardner, C Romay, B Ganapathysubramanian, T Lübberstedt (2015a). Genome-wide
association analysis of seedling root development in maize (Zea mays L.). BMC Genomics 16; Article 47
Pace J,
X Yu, L Thomas (2015b). Genomic prediction of seedling root length in maize (Zea mays L.). Plant J 83:903‒912
Pace J,
N Lee, HS Naik, B Ganapathysubramanian, T Lübberstedt (2014). Analysis of maize
(Zea mays L.) seedling roots with the
high-throughput image analysis tool ARIA (Automatic Root Image Analysis). PLoS One 9; Article e108255
Pierret A, S Gonkhamdee, C
Jourdan, JL Maeght (2013). IJ_Rhizo: An open-source
software to measure scanned images of root samples. Plant Soil 373:531‒539
Price A,
R Tuberosa, SD Dorlodot, B Forster (2007). Root system architecture: Opportunities
and constraints for genetic improvement of crops. Trends Plant Sci 12:474‒481
Rellán-Álvarez
R, G Lobet, H Lindner, P-L Pradier, J Sebastian, MC
Yee, Y Geng, C Trontin, T LaRue, A Schrager-Lavelle,
CR Haney, R Nieu, J Maloof, JP Vogel, JR Dinnenny (2015). GLO-Roots: An imaging
platform enabling multidimensional characterization of soil-grown root systems.
Elife 4;
Article e07597
Ren Y, X
He, D Liu, J Li, X Zhao, B Li, Y Tong, A Zhang, Z Li (2012). Major quantitative
trait loci for seminal root morphology of wheat seedlings. Mol Breed 30:139‒148
Rosas U, A
Cibrian-Jaramillo, D Ristova, JA Banta, ML Gifford, AH Fan, RW Zhou, GJ Kim, G
Krouk, KD Birnbaum (2013). Integration of responses within and across
Arabidopsis natural accessions uncovers loci controlling root systems
architecture. Proc Natl Acad Sci USA 110:15133‒15138
Sanchez DL, S Liu, R Ibrahim,
M Blanco, T Lübberstedt (2018). Genome-wide association studies of doubled
haploid exotic introgression lines for root system architecture traits in maize
(Zea mays L.). Plant Sci 268:30‒38
Song W,
B Wang, AL Hauck, X Dong, J Li, J Lai (2016). Genetic dissection of maize
seedling root system architecture traits using an ultra-high-density bin-map
and a recombinant inbred line population. J
Integr Plant Biol 58:266‒279
Stelpflug
SC, RS Sekhon, B Vaillancourt, CN Hirsch, CR Buell, N de-Leon, SM Kaeppler
(2016). An expanded maize gene expression atlas based on RNA sequencing and its
use to explore root development. Plant Genome
9; Article 0025
Symonova
O, CN Topp, H Edelsbrunner (2015). DynamicRoots: A
software platform for the reconstruction and analysis of growing plant roots. PLoS One 10; Article e0127657
Wang B, H
Liu, Z Liu, X Dong, J Guo, W Li, J Chen, C Gao, Y Zhu, X Zheng (2018).
Identification of minor effect QTLs for plant architecture related traits using
super high-density genotyping and large recombinant inbred population in maize
(Zea mays L.). BMC Plant Biol 18; Article 17
Wang J,
H Li, L Zhang, L Meng (2016). User’s
manual of QTL IciMapping. The Quantitative
Genetics Group, Institute of Crop Science, Chinese Academy of Agricultural
Sciences (CAAS), Beijing 100081, China, and Genetic Resources Program,
International Maize and Wheat Improvement Center (CIMMYT), Apdo Postal 6-641,
06600 Mexico, DF, Mexico
Wasaya A, X Zhang, Q Fang, Z Yan (2018).
Root phenotyping for drought
tolerance: A review. Agronomy 8;
Article 241
Wen T, F
Hochholdinger, M Sauer, W Bruce, PS Schnable (2005). The roothairless1 gene of
maize encodes a homolog of sec3, which is involved in polar exocytosis. Plant Physiol 138:1637‒1643
Wen T, PS Schnable (1994). Analyses of mutants of three genes that
influence root hair development in Zea
mays (Gramineae) suggest that root hairs are dispensable. Amer J Bot 81:833‒842
Wu Y, PR
Bhat, TJ Close, S Lonardi (2008). Efficient and accurate construction of
genetic linkage maps from the minimum spanning tree of a graph. PLoS Genet 4; Article e1000212
Zaidi P,
K Seetharam, G Krishna, L Krishnamurthy, S Gajanan, R Babu, M Zerka, M Vinayan,
B Vivek (2016). Genomic regions associated with root traits under drought
stress in tropical maize (Zea mays L.).
PLoS One 11; Article e0164340
Zhan A,
JP Lynch (2015). Reduced frequency of lateral root branching improves N capture
from low-N soils in maize. J Exp Bot 66:2055‒2065
Zhan A,
H Schneider, JP Lynch (2015). Reduced lateral root branching density improves
drought tolerance in maize. Plant Physiol
168:1603‒1615
Zhang K,
S Liu, W Li, S Liu, X Li, Y Fang, J Zhang, Y Wang, S Xu, J Zhang, J Song, Z Qi,
X Tian, Z Tian, W X Li, H Ning (2018). Identification of QTNs controlling seed
protein content in soybean using multi-locus genome-wide association studies. Front Plant Sci 9; Article 1690
Zhou Z, C Zhang, Y Zhou, Z
Hao, Z Wang, X Zeng, H Di, M Li, D Zhang, H Yong, S Zhang, J Weng, X Li (2016).
Genetic dissection of maize plant architecture with an ultra-high-density bin
map based on recombinant inbred lines. BMC
Genomics 17; Article 178
Zhu J, SM
Kaeppler, JP Lynch (2005). Mapping of QTLs for lateral root branching and length
in maize (Zea mays L). under
differential phosphorus supply. Theor Appl
Genet 111:688‒695
Zurek
PR, CN Topp, PN Benfey (2015). QTL mapping for maize root system traits. Plant Physiol 167:1487-1496