Draft - tspace.library.utoronto.ca · Draft 2 19 20 Abstract 21 As a primitive hexaploid wheat resource distributed only in Tibet, Tibetan semi-wild 22 wheat (Triticum aestivum ssp

Draft

Identification of quantitative trait loci for seedling root

traits from Tibetan semi-wild wheat (Triticum aestivum ssp. tibetanum)

Journal: Genome

Manuscript ID gen-2017-0097.R1

Manuscript Type: Article

Date Submitted by the Author: 02-Jul-2017

Complete List of Authors: Ma, Jian; Sichuan Agricultural University, Triticeae Research Institute

Luo, Wei; Sichuan Agricultural University, Triticeae Research Institute Zhang, Han; Sichuan Agricultural University, Triticeae Research Institute Zhou, Xiao-Hong; Sichuan Agricultural University, Triticeae Research Institute Qin, Na-Na; Sichuan Agricultural University, Triticeae Research Institute Wei, Yu-Ming; Sichuan Agricultural University, Triticeae Research Institute Liu, Yaxi; Sichuan Agricultural University, Triticeae Research Institute Jiang, Qian-Tao; Sichuan Agricultural University, Triticeae Research Institute Chen, Guo-Yue; Sichuan Agricultural University, Triticeae Research Institute Zheng, You-Liang; Sichuan Agricultural University, Triticeae Research

Institute Lan, Xiu-Jin; Sichuan Agricultural University, Triticeae Research Institute

Is the invited manuscript for consideration in a Special

Issue? : This submission is not invited

Keyword: Tibetan semi-wild wheat, recombinant inbred line, root trait, quantitative trait locus, correlation

https://mc06.manuscriptcentral.com/genome-pubs

Genome

Draft

1

Identification of quantitative trait loci for seedling root traits from 1

Tibetan semi-wild wheat (Triticum aestivum ssp. tibetanum) 2

Jian Ma#, Wei Luo

#, Han Zhang

#, Xiao-Hong Zhou, Na-Na Qin, Yu-Ming Wei, Ya-Xi 3

Liu, Qian-Tao Jiang, Guo-Yue Chen, You-Liang Zheng, Xiu-Jin Lan* 4

5

Triticeae Research Institute, Sichuan Agricultural University, 211 Huimin Road, 6

Wenjiang, Chengdu, Sichuan, 611130, China. 7

8

# Contributed equally to this work. 9

*Corresponding author 10

Xiu-Jin Lan: [email protected]. Tel: +86 28 82650350. Fax: +86 28 82650350. 11

The e-mail addresses of the other co-authors: [email protected] (Jian Ma), 12

[email protected] (Wei Luo), [email protected] (Han Zhang), 13

[email protected] (Xiao-Hong Zhou), [email protected] (Na-Na Qin), 14

[email protected] (Yu-Ming Wei), [email protected] (Ya-Xi Liu), 15

[email protected] (Qian-Tao Jiang ), [email protected] (Guo-Yue 16

Chen), [email protected] (You-Liang Zheng). 17

18

Page 1 of 20


Genome

Draft

2

19

Abstract 20

As a primitive hexaploid wheat resource distributed only in Tibet, Tibetan semi-wild 21

wheat (Triticum aestivum ssp. tibetanum Shao) possesses unique characteristics that 22

could be exploited in wheat breeding program. A good root system could offer a stable 23

platform for the above-ground components. Aiming at detecting possible excellent 24

locus for root traits from Tibetan semi-wild wheat, we here identified QTLs for root 25

traits using a recombinant inbred line population derived from a cross between 26

Tibetan semi-wild wheat Q1028 and Zhengmai 9023. A total of 15 QTLs on 8 27

chromosomes were detected, including four major QTLs, QMrl.sau-7B, QTrl.sau-4B, 28

QAd.sau-7A and QSa.sau-4B. The phenotypic variation explained by each of these 29

QTLs ranges from 5.67 to 16.68 %. Positive alleles of 6 QTLs were derived from 30

Q1028. Several novel QTLs for root traits were identified. In addition, significant 31

correlations were detected amongst root traits and agronomic traits. Taken together, 32

the results suggested that Tibetan semi-wild wheat and the newly identified novel 33

QTLs could be useful in future breeding programs. 34

Keywords Tibetan semi-wild wheat, recombinant inbred line, root trait, quantitative 35

trait locus, correlation 36

37

Page 2 of 20


Genome

Draft

3

Introduction 38

Common wheat (Triticum aestivum L., 2n=6x=42, AABBDD) is one of the most 39

important food crops in the world. With the increase of population, improving yield of 40

wheat has become an important target of breeding work. However, the current annual 41

rate of increase in wheat productivity is considerably lower than the demanded by 42

global wheat consumption (Reynolds et al. 2012). 43

As an important organ to absorb water and minerals, a root system plays a major role 44

in the growth and development and even yield in crops (Moudal and Kour 2004; 45

Partha et al. 2004; Rogers and Benfey 2015).Compared with root traits, the 46

above-ground components including plant height and harvest index have always been 47

focused on in wheat improvement programs, mainly due to the difficulties related to 48

root recovery and evaluating root traits in situ. Another two limitations, i.e. broad 49

phenotypic plasticity of root traits in response to changes in soil conditions and lack 50

of high-throughput and cost-effective screening techniques, make root studies highly 51

challenging as well (Fitter 2002; Manschadi et al. 2008; Poorter and Nagel 2000). 52

Despite of these restricted conditions, some appreciable progress has been made to 53

date on identification of QTLs for important root traits (Bai et al. 2013; Ehdaie et al. 54

2016; Kabir et al. 2015; Liu et al. 2013; Ren et al. 2012). For instance, Maccaferri et 55

al. (2016) detected several QTLs for root length, root number and root growth angle 56

in the whole genome of tetraploid wheat except 4A and 5B, Ayalew et al. (2017) 57

located some QTLs for root length on 1A, 1B, 2D, 3A, 5A, 6A and 7B in hexaploid 58

wheat. Furthermore, quite a few of QTLs for root diameter, surface area and root 59

volume have been reported by Petrarulo et al. (2015) and Shang et al. (2016). 60

Tibetan semi-wild wheat (T. aestivum ssp. tibetanum) is a hexaploid wheat found only 61

in Tibet (Shao et al. 1980), with unique and primitive characteristics such as hulled 62

glumes and brittle rachis (Jiang et al. 2014). It is characterized by numerous desirable 63

characteristics, such as tolerance to nutrition deficiency and strong seed dormancy 64

which can be introduced by crossing with common wheat breeding (Jiang et al. 2015; 65

Sun et al. 1998). Thus, identification and the utilization of desirable genes from 66

Tibetan semi-wild wheat could be important to develop modern wheat cultivars. 67

Most recently, we identified QTLs controlling seed dormancy (Jiang et al. 2015), 68

brittle rachis and threshing ability (Jiang et al. 2014), and important agronomic traits 69

(Luo et al. 2016) of T. aestivum ssp. tibetanum accession Q1028 collected in our 70

institute. In addition, a recent study reported a few QTLs controlling root traits using 71

RILs derived from common wheat and Tibetan semi-wild wheat (Kabir et al. 2015), 72

broadening our knowledge on Tibetan semi-wild wheat for breeding as well. It is 73

documented that more than 140 accessions of Tibetan semi-wild wheat with high 74

genetic diversity were collected (Liu 2010). As part of our systematic work to exploit 75

breeding potentialities of Tibetan semi-wild wheat in cultivar development, we, here, 76

report on QTLs controlling root traits of Tibetan semi-wild wheat accession Q1028 77

Page 3 of 20


Genome

Draft

4

different from that used in Kabir et al. (2015). 78

Materials and methods 79

Plant material 80

In this study, a total of 186 RILs derived from a cross between Tibetan semi-wild 81

wheat accession Q1028 (female parent) and a wheat cultivar Zhengmai 9023 (male 82

parent, ZM9023) were used to investigate root traits. Q1028 shows a longer root 83

system than ZM9023 (Fig. 1). Furthermore, Q1028 has longer flowering date, taller 84

plant height, higher spikelet number per spike and larger tiller number than ZM9023. 85

While ZM9023 has higher thousand-grain weight than Q1028 (Luo et al. 2016). 86

Hydroponic culture 87

The commonly used hydroponic culture system (Kabir et al. 2015; Ren et al. 2012) 88

was employed to study the root morphological parameters of wheat seedlings in the 89

present study. The experiments were conducted in greenhouse of the Triticeae 90

Research Institute of Sichuan Agricultural University. The settings for the greenhouse 91

were: 20 (±4) °C day/night temperature and 65/85 % day/night relative humidity, and 92

a 16-h photoperiod. 93

The experiment was repeated for three times to increase the credibility of root trait 94

measurements. Each replicated experiment was performed in a completely 95

randomized design. For each replicated experiment, twenty uniformly sized seeds 96

from each line were surface-sterilized by soaking in 10 % sodium hypochlorite for 5 97

min and then, rinsed three times with deionized water. Seeds were then germinated on 98

three layers of filter paper saturated with water in petri dishes at room temperature for 99

a week. After removing residual endosperm materials, four equal sized germinated 100

seeds of each line having shoots of about 1 cm were transferred to plastic tray. The 101

plastic trays (50 cm × 40 cm × 30 cm) contain Hoagland nutrient solution to supply 102

mineral nutrition and air pump to supply oxygen (An et al. 2006). The nutrient 103

solution was replaced twice in a week. 104

Phenotypic data 105

Root and shoot tissues were collected separately four weeks after transplanting for 106

measuring root traits. Maximum root length (MRL, the length between the roots base 107

and the tip of the longest root, cm/plant) was measured using a ruler. The collected 108

roots were scanned by Epson Expression 10000 XL. The output root imagines were 109

analyzed by Win-RHIZO developed by Regent Instruments Canada Inc. (Ottawa, ON, 110

Canada) for the following traits: total root length (TRL, the sum of the length for all 111

the roots including primary and lateral roots from a single measured plant, cm/plant), 112

root surface area (SA, the area of all the roots from a single measured plant, 113

Page 4 of 20


Genome

Draft

5

cm2/plant ), root average diameter (AD, the average diameter of all the roots from a 114

single measured plant, cm/plant), root volume (RV, the total size for all the roots from 115

a single measured plant, cm3/plant) and number of root tips (RTN, the total tips of all 116

the roots from a single measured plant, n/plant). Root dry weight (RDW, g/plant)) and 117

shoot dry weight (SDW, g/plant) were measured by electronic scale after the fresh 118

roots and shoot had been dried to a constant weight at 100 °C. The root/shoot ratio 119

(RSR, RDW/SDW) was calculated. Five agronomic traits investigated in our previous 120

study (Luo et al. 2016) were employed for correlation analysis, including flowering 121

date (FD), plant height (PH), tiller number (TN), spikelet number per spike (SNS) and 122

1000 grains weight (TGW). The best linear predictors (BLUPs) were calculated for 123

the measured root traits (from three repeated experiments) and agronomic traits (from 124

different growth seasons) and were further used for the following analysis. Frequency 125

distribution and the correlation between traits were calculated using SPSS version 126

20.0 for Windows (SPSS Inc., Chicago, IL). The estimated broad-sense heritability of 127

the investigated traits was calculated as follows: h2

= σ2

G/(σ2

G + σ2

e /r), where σ2

G is the 128

genetic variance, σ2

e is the residual variance, and r is the number of replicates per 129

genotype. Student’s t-test (P < 0.05) calculated by SPSS was used to evaluate the 130

differences of the investigated traits between the parents. 131

QTL mapping 132

The BLUPs of 12 seedlings of each RIL from three replicates was used to detect root 133

QTLs. A whole-genome genetic map constructed previously (Jiang et al. 2014) was 134

adopted for QTL mapping. The genetic map was constructed using a population of 135

186 RILs with 564 DArT makers and 117 SSR markers. These markers were 136

distributed in 22 linkage groups and covered a total genetic distance of 2, 727 cM 137

(Jiang et al. 2014). The relative positions of contiguous markers were tested through 138

Kosambi mapping function. QTL analysis was initially conducted using QTL 139

IciMapping version 3.2 which was followed by inclusive composite interval mapping 140

(ICIM) (Wang et al. 2012). Thereafter, single marker analysis (SMA) was applied to 141

ascertain the QTLs’ authenticity. LOD ≥ 2.5 was used to determine the presence of a 142

QTL (Lin et al. 1996). QTLs were named according to the rules of International Rules 143

of Genetic Nomenclature (http://wheat.pw.usda.gov/ggpages/wgc/98/Intro.htm). 144

Results 145

Phenotypic analysis 146

The phenotypes values for each trait varied among different lines (Table 1; Fig. 1). 147

Significant differences between parents were detected for MRL, TRL, SA, AD, RV, 148

MRL and RTN (Table 1). For example ZM9023 exhibits higher values of TRL, RSA, 149

RV and RTN than Q1028. However, Q1028 shows longer MRL and larger AD than 150

ZM9023 (Table 1). 151

Page 5 of 20


Genome

Draft

6

The transgressive segregation in the RIL population were observed for all traits (Fig. 152

2). The frequency distributions of the investigated traits showed continuous variation 153

with approximately normal distributions indicating that the traits are quantitative in 154

nature and are suitable for QTL analysis. The heritability value of RSR was the lowest 155

in this study (only 5%) (Table 1). Furthermore, a concentrated heritability values 156

ranged from 62 % to 85 % indicating that the other 8 traits are stable and mainly 157

controlled by genetic factors (Table 1). 158

Correlation between root and agronomic traits 159

The phenotypic correlations among the investigated root and agronomic traits are 160

shown in Table 2. Significant correlations were detected amongst most of the traits. 161

The correlation coefficients ranged from 0.008 to 0.988 (Table 2). FD was 162

significantly correlated with SDW and MRL as well as all other agronomic traits with 163

correlation coefficients ranging from 0.144 to 0.689 (P < 0.05). Significant positive 164

correlations were also detected between PH and 6 seeding traits including RDW, SDW, 165

TRL, SA, RV and RTN with the coefficients ranging from 0.255 to 0.329 (P < 0.01). 166

TN was significantly negatively correlated with TGW and AD (P < 0.01) but 167

positively correlated with SNS, MRL and RTN (P < 0.05 or 0.01). SNS showed 168

significantly negative correlation with TGW and significantly positive correlation 169

with RDW, SDW and MTL. TGW was significantly correlated with MRL, TRL, SA 170

and RV (P < 0.01). 171

Significant correlations were also detected among seeding traits (Table 2). 172

Significantly positive correlations were detected between RDW and all of the other 173

seeding traits except AD, and the coefficients ranged from 0.297 to 0.852 (P < 0.01). 174

Unsurprisingly, SDW was significantly negatively related to RSR (r = -0.186), but it 175

was significantly positively correlated with several seeding traits including RDW, 176

MRL, TRL, SA, RV and RTN (0.295 ≤ r ≤ 0.847). MRL was significantly correlated 177

with TRL, SA, RV and RTN (P < 0.01). TRL showed highly significant correlations 178

with PH, TGW and all the seeding traits except RSR (P < 0.01). AD was significantly 179

negatively correlated with TN, TRL and RTN (P < 0.01). SA was significantly 180

correlated with PH, TGW and all of the seeding traits (0.230 ≤ r ≤ 0.988, P < 0.01) 181

except AD and RSR. RV was significantly correlated with all of the seeding traits 182

(0.204 ≤ r ≤ 0.948) except RSR. RTN was significantly correlated with PH, TN and 183

all the seeding traits except RSR (Table 2). 184

QTL detection 185

In the present study, a total of 15 QTLs for seven root traits were identified with a 186

single QTL explaining 5.67–16.68 % of phenotypic variation observed (Table 3, Fig. 187

3). The LOD values ranged from 2.54 to 7.75 (Table 3). The 15 QTLs were 188

distributed on 8 chromosomes and they were 1B, 3B, 3D, 4B (4 QTLs), 5B, 6B, 7A (5 189

QTLs) and 7B. The additive effects of 6 QTLs were positive with Q1028 increasing 190

Page 6 of 20


Genome

Draft

7

the effects of these QTLs, while the remaining 9 were negative with ZM9023 191

increasing the effects. 192

Four QTLs for MRL were detected and they were mapped on chromosomes 3B, 6B, 193

7A, and 7B (Fig. 3). The phenotypic variation explained by these QTLs ranged from 194

6.18 to 16.68 % (Table 3). The positive allele of QMrl.sau-7B was contributed by 195

ZM9023 and those of the remaining QTLs were contributed by Q1028. The QTL, 196

QMrl.sau-7B, was a major QTL which explained 16.68 % of variation. 197

Two QTLs for TRL, QTrl.sau-4B and QTrl.sau-7A, were detected explaining 12.56 % 198

and 5.67 % of the phenotypic variation, respectively. The positive alleles of these two 199

QTLs were contributed by ZM9023. 200

For AD, a major (QAd.sau-7A) and two minor QTLs (QAd.sau-3D and QAd.sau-5B) 201

with 12.18 %, 8.74 %, and 5.83% phenotypic variation explained were identified on 202

7A, 3D, and 5B, respectively (Fig. 3, Table 3). The positive alleles of QAd.sau-3D 203

and QAd.sau-7A were contributed by Q1028 and that of QAd.sau-5B was contributed 204

by ZM9023. 205

One QTL only for RV was detected and it was mapped on chromosome 4B (Fig. 3). It 206

explained 8.79 % of the phenotypic variation and the positive allele was from 207

ZM9023 (Table 3). 208

A major QTL for SA was also detected on 4B (Fig. 3). It explained 10.61 % of the 209

phenotypic variation and the positive allele was from ZM9023 (Table 3). 210

Two QTLs for RTN were detected on 4B and 7A (Fig. 3). QRtn.sau-4B and 211

QRtn.sau-7A explained 8.00 % and 8.61 % of the phenotypic variation, respectively. 212

The positive alleles of these QTLs were all from ZM9023 (Table 3). 213

Two QTLs for RSR were located on 1B and 7A. QRsr.sau-1B explained 6.80 % of the 214

phenotypic variation and QRsr.sau-7A explained 5.73 % of that (Table 3). 215

Furthermore, the positive alleles of the two QTLs were from ZM9023 and Q1028, 216

respectively. 217

Discussion 218

A good root system could offer a stable platform for the above-ground components. 219

To explore desirable genes for root traits of wheat, we investigated several root traits 220

of a RIL population derived from the Tibetan semi-wild wheat Q1028 and a wheat 221

cultivar ZM9023. ZM9023 showed clear advantages in six out of the eleven assessed 222

seeding traits, indicating its breeding potential as a national wheat cultivar. A total of 223

15 QTLs on 8 chromosomes were detected. The phenotypic variation explained (PVE) 224

by each of these QTLs ranged from 5.67 to 16.68 % and 4 were considered as major 225

Page 7 of 20


Genome

Draft

8

QTLs with a higher than 10 % of PVE. Positive alleles of 6 QTLs were derived from 226

Q1028. Significant correlations were detected amongst most of root traits and several 227

agronomic traits. As the expression of QTL is easily affected by environment factors, 228

the BLUP model was adopted to correct the environmental effects aiming at 229

increasing the reliability and facticity of the detected QTLs (Doligez et al. 2013). 230

Possible potential of Tibetan semi-wild wheat in breeding program 231

The values of MRL and AD for Tibet semi-wild wheat Q1028 were significantly 232

higher than those for ZM9023. Reportedly, a deep root system could facilitate the 233

plant to extract water stored in deep soil layers to avoid drought stress (Boyer 1996; 234

Ludlow and Muchow 1990). It is also reported that increased root diameter was 235

related to drought tolerance in rice (Oryza sativa L.) because thicker roots are 236

characterized by large xylem vessels with improved axial conductance and are more 237

efficient in penetrating deep soil layers to extract water (Clark et al. 2008; Fukai and 238

Cooper 1995). Given that Tibetan semi-wild wheat was found only in Tibet, a 239

relatively dry growth environment (Shao et al. 1980), it is thus understandable that 240

Tibetan semi-wild wheat has adapted by increasing rooting depth and root diameter to 241

capture water and nutrition from the deeper layers of soil to avoid drought stress. 242

Therefore, Tibetan semi-wild wheat could be utilized as an excellent genetic resource 243

in breeding program aiming at improving tolerance to drought. 244

It was documented that more than 140 accessions of Tibetan semi-wild wheat with 245

high genetic diversity were collected (Liu 2010). Exploring various genotypes of 246

Tibetan semi-wild wheat with desirable genes could facilitate ongoing of wheat 247

improvement. For example, an accession of Tibetan semi-wild wheat (i.e. Zang 1817) 248

had been used to construct a RIL population and detected QTLs controlling root traits 249

(Kabir et al. 2015). Aiming at identifying more desirable genes from different 250

genotypes of Tibetan semi-wild wheat, we created a RIL population derived from 251

Tibetan semi-wild wheat Q1028 and a cultivar ZM9023 for detecting QTLs 252

responsible for root traits. Genotypes of Tibetan semi-wild wheat (Q1028 and Zang 253

1817) are different in phenotypes including plant height (Q1028:139.0cm; Zang 1817: 254

96.7cm), spike length (Q1028:12.7cm; Zang 1817: 11.2cm), thousand grain weight 255

(Q1028:21.3g; Zang 1817: 26.7g) (Liu et al. 2014; Luo et al. 2016). For the traits 256

determined in both study, i.e. MRL, TRL, (R)SA, AD (RVOL), RTN (TIP), some 257

novel QTLs were detected in the present study, such as QTLs for MRL, TRL and AD 258

on 7A. Of these newly identified QTLs, positive alleles of QMrl.sau-7A and 259

QAd.sau-7A are derived from Q1028. In addition, it was notable that transgressive 260

segregation was detected in all of the nine traits, and some lines showed better 261

phenotype than their parents (Fig. 2). The lines which expressed transgressive 262

segregation in multiple traits were also propitious to be applied in breeding 263

programs. Taken together, the results reported here combined with those from Kabir et 264

al. (2015) suggested that Tibetan semi-wild wheat can be used as a valuable gene 265

resource in breeding wheat. 266

Page 8 of 20


Genome

Draft

9

Comparison of QTLs detected in this study with those reported previously 267

Numerous QTLs for root traits have been detected by using various populations 268

(Ayalew et al. 2017; Bai et al. 2013; Ibrahim et al. 2012; Kabir et al. 2015; Liu et al. 269

2013; Maccaferri et al. 2016; Ren et al. 2012). Both the present study and that 270

reported by Kabir et al. (2015) detected QTLs for MRL on 3B and 7B. Ibrahim et al. 271

(2012) detected QTLs for TRL on 7A as this study. As the markers used for QTL 272

mapping were different, the detailed comparison of the markers flanking a given QTL 273

is not realistic. In this study, we also detected several QTLs that have not been 274

detected previously to our knowledge. For example, QMrl.sau-7A and the QTLs for 275

AD on 3D and 5B were newly identified in the present study. The fact that the QTLs 276

for MRL, TRL, AD, RTN and RSR were identified in 7A suggests that chromosome 277

7A could play a major role in growth and development of root and its function 278

performances. 279

Co-mapped and closely linked QTLs controlling different root traits 280

QTL clusters have been reported in previous studies (Kabir et al. 2015; Luo et al. 281

2016). The co-located QTLs for different traits were detected in specific chromosomal 282

intervals in the present study. Two QTL clusters distributed on chromosomes 4B and 283

7A were identified (Fig. 3). Co-localization of QTLs indicates the close association of 284

the traits. The loci for TRL, SA, RV and RTN on 4B are most likely controlled by a 285

same gene, given that the positive alleles of these loci came from the same parent 286

ZM9023. The cluster on 7A consisting of three QTLs for AD, RTN, and TRL mapped 287

between wPt-4515 and Xgpw3220 was identified. The positive alleles for 288

QTrl.sau-7A and QRtn.sau-7A were contributed by male parent ZM9023 and the 289

positive allele of QAd.sau-7A was contributed by female parent Q1028. The 290

resolution of the genetic map constructed in this study should be further improved to 291

determine whether these traits are controlled by the two different alleles of the same 292

locus. 293

The relationships among root and agronomic traits 294

In this study, SA showed significantly positive correlation with PH, TGW, SDW and 295

other root traits except AD and RSR (Table 2). Water and nutrient absorption from the 296

soil is directly proportional to contact areas between soil and roots (Caassen and 297

Barber 1976; Yoshida and Hasegawa 1982). It was reported that the absorptive 298

capacity of wheat roots increases with SA (Narayanan et al. 2014). The increased 299

intakes through increased SA might help plant to improve the above-ground biomass 300

production including SDW and TGW. As TRL, RV and RTN were significantly 301

positively correlated with SA (r ≥ 0.805, P < 0.01), it is thus understandable that these 302

three root traits were also significantly positively correlated with SDW. 303

A positive correlation between SDW and RDW (r = 0.833, P < 0.01; Table 2) was 304

Page 9 of 20


Genome

Draft

10

detected in the present study, being consistent with previous studies in wheat 305

(Narayanan et al. 2014), pea (McPhee 2005) and chickpea (Serraj et al. 2004). As 306

discussed above, the raised nutrients absorbed through increased root surface might 307

help to improve the biomass of SDW. In turn, the surplus photosynthate of the 308

increased shoot growth might be transported to roots and thus increase RDW 309

(Narayanan et al. 2014). And the mutual-promotion pattern might explain the 310

significant correlations between root traits and agronomic traits including PH, TN, 311

SNS and TGW. 312

Our previous study detected a QTL for TN on 6B where the MRL QTL QMrl.sau-6B 313

was located (Luo et al. 2016). In the present study, TN showed significant correlation 314

with MRL and RTN. Furthermore, significant correlations between PH and RDW, 315

TRL, SA, RV and RTN (P < 0.01) were also detected. We thus suggest that some root 316

and yield-related traits could be controlled by a single gene. For example, it was 317

reported that there were genes or QTLs likely controlling both plant height and root 318

traits (Bai et al. 2013; Ehdaie et al. 2016; Kabir et al. 2015). This might be another 319

reason that root traits were significantly related to agronomic traits (such as SNS and 320

TGW). 321

FD showed significant correlation with MRL. It is known that a plant with longer 322

growth period need to absorb more water and nutrients from deeper soil layers to 323

complete the period from vegetative to reproductive stage (Salvador R and Pearce R 324

1995). Therefore, longer roots could undoubtedly benefit the absorption and 325

utilization of the water and nutrients and thus help the growth and development of a 326

plant. The fact that Q1028 showed a longer growth stage (Luo et al. 2016) and root 327

length than ZM9023 may further explain the significant correlation between root traits 328

and FD in this study. 329

330

Acknowledgements 331

This work was supported by the projects from the National Natural Science 332

Foundation of China (31570335 and 31601292), the Applied Basic Research 333

(2016JY0010) and International Science and Technology Cooperation and Exchanges 334

(2017HH0076) Programs of Science and Technology Department of Sichuan 335

Province. 336

References 337

An, D., Su, J., Liu, Q., Zhu, Y., Tong, Y., Li, J., et al. 2006. Mapping QTLs for 338

nitrogen uptake in relation to the early growth of wheat (Triticum aestivum L.). Plant 339

Soil 284(1): 73-84. doi:10.1007/s11104-006-0030-3. 340

Ayalew, H., Liu, H., and Yan, G. 2017. Identification and validation of root length 341

Page 10 of 20


Genome

Draft

11

QTLs for water stress resistance in hexaploid wheat (Titicum aestivum L.). Euphytica 342

213(6): 126. doi:10.1007/s10681-017-1914-4. 343

Bai, C., Liang, Y., and Hawkesford, M.J. 2013. Identification of QTLs associated with 344

seedling root traits and their correlation with plant height in wheat. J. Exp. Bot. 64(6): 345

1745–1753. doi:10.1093/jxb/ert041. 346

Boyer, J.S. 1996. Advances in drought tolerance in plants. Adv. Agron. 56(08): 347

187-218. 348

Caassen, N., and Barber, S. 1976. Simulation model for nutrient uptake from soil by a 349

growing plant root system. Agron. J. 68(6): 961-964. 350

doi:10.2134/agronj1976.00021962006800060030x. 351

Clark, L.J., Price, A.H., Steele, K.A., and Whalley, W.R. 2008. Evidence from 352

near-isogenic lines that root penetration increases with root diameter and bending 353

stiffness in rice. Funct Plant Biol. 35(11): 1163-1171. doi:10.1071/FP08132. 354

Doligez, A., Bertrand, Y., Farnos, M., Grolier, M., Romieu, C., Esnault, F., et al. 2013. 355

New stable QTLs for berry weight do not colocalize with QTLs for seed traits in 356

cultivated grapevine (Vitis vinifera L.). BMC Plant Biology 13(1): 217. 357

doi:10.1186/1471-2229-13-217. 358

Ehdaie, B., Mohammadi, S.A., and Nouraein, M. 2016. QTLs for root traits at 359

mid-tillering and for root and shoot traits at maturity in a RIL population of spring 360

bread wheat grown under well-watered conditions. Euphytica 211(1): 17-38. 361

doi:10.1007/s10681-016-1670-x. 362

Fitter, A. 2002. Characteristics and functions of root systems. In: Waisel Y, Eshel A, 363

Kafkafi U, editors. Plant roots: the hidden half. Marcel Dekker, New York, USA.: 364

249-259. 365

Fukai, S., and Cooper, M. 1995. Development of drought-resistant cultivars using 366

physiomorphological traits in rice. Field Crops Res. 40(2): 67-86. 367

doi:10.1016/0378-4290(94)00096-U. 368

Ibrahim, S.E., Schubert, A., Pillen, K., and Léon, J. 2012. QTL analysis of drought 369

tolerance for seedling root morphological traits in an advanced backcross population 370

of spring wheat. Int. J. Agrisci. 2(7): 619-629. 371

Jiang, Y.F., Lan, X.J., Luo, W., Kong, X.C., Qi, P.F., Wang, J.R., et al. 2014. 372

Genome-wide quantitative trait locus mapping identifies multiple major loci for brittle 373

rachis and threshability in Tibetan semi-wild wheat (Triticum aestivum ssp. tibetanum 374

Shao). Plos One 9(12): e114066-e114066. doi:10.1371/journal.pone.0114066. 375

Jiang, Y.F., Wang, J.R., Luo, W., Wei, Y.M., Qi, P.F., Liu, Y.X., et al. 2015. 376

Quantitative trait locus mapping for seed dormancy in different post-ripening stages in 377

a Tibetan semi-wild wheat ( Triticum aestivum ssp . tibetanum Shao). Euphytica 378

203(3): 557-567. doi:10.1007/s10681-014-1266-2. 379

Kabir, M.R., Liu, G., Guan, P., Wang, F., Khan, A.A., Ni, Z., et al . 2015. Mapping 380

QTLs associated with root traits using two different populations in wheat (Triticum 381

aestivum L.). Euphytica 206(1): 175-190. doi: 10.1007/s10681-015-1495-z. 382

Lin, H., Qian, H., Zhuang, J., Lu, J., Min, S., Xiong, Z., et al. 1996. RFLP mapping of 383

QTLs for yield and related characters in rice (Oryza sativa L.). Theor. Appl. Genet. 384

Page 11 of 20


Genome

Draft

12

92(8): 920-927. doi: 10.1007/BF00224031. 385

Liu, G., Jia, L., Lu, L., Qin, D., Zhang, J., Guan, P.,et al. 2014. Mapping QTLs of 386

yield-related traits using RIL population derived from common wheat and Tibetan 387

semi-wild wheat. Theor. Appl. Gene. 127(11): 2415-2432. doi: 388

10.1007/s00122-014-2387-7. 389

Liu, S. 2010. Genetic diversity analysis of Triticum aestivum ssp. tibetanum Shao. 390

M.S. Thesis. Northwest Agriculture and Forestry Univ., Yangling, China. 391

Liu, X., Li, R., Chang, X., and Jing, R. 2013. Mapping QTLs for seedling root traits in 392

a doubled haploid wheat population under different water regimes. Euphytica 189(1): 393

51-66. doi: 10.1007/s10681-012-0690-4. 394

Ludlow, M.M., and Muchow, R.C. 1990. A Critical Evaluation of Traits for Improving 395

Crop Yields in Water-Limited Environments 1. Adv. Agron. 43: 107-153. doi: 396

10.1016/S0065-2113(08)60477-0. 397

Luo, W., Ma, J., Zhou, X.-H., Sun, M., Kong, X.-C., Wei, Y.-M.,et al. 2016. 398

Identification of Quantitative Trait Loci Controlling Agronomic Traits Indicates 399

Breeding Potential of Tibetan Semiwild Wheat (ssp.). Crop Sci. 56(5): 2410-2420. 400

doi:10.2135/cropsci2015.11.0700. 401

Maccaferri, M., El-Feki, W., Nazemi, G., Salvi, S., Canè, M.A., Colalongo, M.C., et al. 402

2016. Prioritizing quantitative trait loci for root system architecture in tetraploid 403

wheat. J. Exp. Bot. 67(4): 1161-1178. doi: 10.1093/jxb/erw039. 404

Manschadi, A.M., Hammer, G.L., Christopher, J.T., and Devoil, P. 2008. Genotypic 405

variation in seedling root architectural traits and implications for drought adaptation in 406

wheat (Triticum aestivum L.). Plant Soil 303(1-2): 115-129. doi: 407

10.1007/s11104-007-9492-1. 408

McPhee, K. 2005. Variation for seedling root architecture in the core collection of pea 409

germplasm. Crop Sci. 45(5): 1758-1763. doi:10.2135/cropsci2004.0544. 410

Moudal, S., and Kour, K. 2004. Genetic variability and correlation coefficients of 411

some root characteristics and yield components in bread wheat (Triticum aestivum L.) 412

under rainfed condition. Environ. Ecol. 22: 646-648. 413

Narayanan, S., Mohan, A., Gill, K.S., and Prasad, P.V.V. 2014. Variability of Root 414

Traits in Spring Wheat Germplasm. Plos One 9(6): e100317. 415

doi:10.1371/journal.pone.0100317. 416

Partha, D., Ali, M., and Sarkar, H. 2004. Genetical studies on roots in bread wheat. J 417

Interacademicia 8(2): 166-168. 418

Petrarulo, M., Marone, D., Ferragonio, P., Cattivelli, L., Rubiales, D., De Vita, et al. 419

Genetic analysis of root morphological traits in wheat. Mol. Genet. Genomics 290(3): 420

785-806. doi:10.1007/s0043. 421

Poorter, H., and Nagel, O. 2000. The role of biomass allocation in the growth 422

response of plants to different levels of light, CO2, nutrients and water: a quantitative 423

review. Funct. Plant Biol. 27(6): 1191-1191. doi:10.1071/PP99173_CO. 424

Ren, Y., He, X., Liu, D., Li, J., Zhao, X., Li, B., et al. 2012. Major quantitative trait 425

loci for seminal root morphology of wheat seedlings. Mol. Breed. 30(1): 139-148. 426

doi:10.1007/s11032-011-9605-7. 427

Page 12 of 20


Genome

Draft

13

Reynolds, M., Foulkes, J., Furbank, R., Griffiths, S., King, J., Murchie, E., et al. 2012. 428

Achieving yield gains in wheat. Plant, Cell Environ. 35(10): 1799-1823. 429

doi:10.1111/j.1365-3040.2012.02588.x. 430

Rogers, E.D., and Benfey, P.N. 2015. Regulation of plant root system architecture: 431

implications for crop advancement. Curr. Opin. Biotechnol. 32: 93-98. doi: 432

10.1016/j.copbio.2014.11.015. 433

Salvador R, J., and Pearce R, B. 1995. Proposed standard system of nomenclature for 434

maize grain filling events and concepts. Maydica 40(2): 141-146. 435

Serraj, R., Krishnamurthy, L., Kashiwagi, J., Kumar, J., Chandra, S., and Crouch, J. 436

2004. Variation in root traits of chickpea (Cicer arietinum L.) grown under terminal 437

drought. Field Crops Res. 88(2): 115-127. doi:10.1016/j.fcr.2003.12.001. 438

Shang, L., Cai, S., Ma, L., Wang, Y., Abduweli, A., Wang, M., et al. 2016. Seedling 439

root QTLs analysis on dynamic development and upon nitrogen deficiency stress in 440

Upland cotton. Euphytica 207(3): 645-663. doi: 10.1007/s10681-015-1564-3. 441

Shao, Q.Q., Li, C.S., and Basang, C.R. 1980. Semiwild wheat from Tibet. Acta Genet. 442

Sinica 7: 149-156. 443

Sun, Q., Ni, Z., Liu, Z., Gao, J., and Huang, T. 1998. Genetic relationships and 444

diversity among Tibetan wheat, common wheat and European spelt wheat revealed by 445

RAPD markers. Euphytica 99(3): 205-211. doi: 10.1023/A:1018316129246. 446

Wang, J., Li, H., Zhang, L., and Meng, L. 2012. QTL IciMapping version 3.2. 447

Quantitative Genetics Group Inst. of Crop Science, Chinese Academy of Agricultural 448

Sciences, Beijing. 449

Yoshida, S., and Hasegawa, S. 1982. The rice root system: its development and 450

function. IRRI, Los Bafios, Philippines. pp. 97-114. 451

452

453

Page 13 of 20


Genome

Draft

14

Table 1 Phenotypic variation and heritability (H2) of investigated traits for RIL 454

population. The data of the parents were the average of the three repetitions, and that 455

for lines were BLUP values. 456

457

Trait†

Parent RIL Population

Q1028±SD ZM9023±SD Mean Range H2

RDW (g) 0.051±0.012 0.053±0.010 0.04 0.03-0.06 0.62

SDW (g) 0.19±0.01 0.21±0.03 0.16 0.08-0.26 0.85

MRL (cm) 44.99±6.00** 36.17±6.43 37.71 27.37-47.72 0.68

TRL (cm) 645.33±118.88** 989.97±86.55 653.08 387.34-1079.33 0.80

AD (cm) 0.33±0.022** 0.31±0.018 0.32 0.30-0.34 0.64

SA (cm2) 67.60±14.90** 97.21±12.37 65.07 39.58-103.52 0.78

RV (cm3) 0.56±0.14** 0.76±0.13 0.52 0.33-0.80 0.75

RTN (n/plant) 1235.83±234.04** 2614.24±397.89 1466.06 788.27-2755.87 0.78

RSR 0.27±0.05 0.25±0.01 0.27 0.259-0.274 0.05

* Significant at the 0.05 probability level. 458

** Significant at the 0.01 probability level. 459

† RDW, Root dry weight; SDW, Shoot dry weight; MRL, Maximum root length; TRL, Total root length; AD, Root 460

average diameter; SA, Root surface area; RV, Root volume; RTN, Number of root tips; RSR, root/shoot ratio. 461

462

Page 14 of 20


Genome

Draft

15

Table 2 Correlation coefficients between investigated traits in the RIL population. 463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

479

480

* Correlation is significant at the 0.05 level 481

** Correlation is significant at the 0.01 level 482

The BLUP values for a given trait were used for correlation analysis. FD, Flowering date; PH, Plant height; TN, Tiller number; SNS, Spikelet number per spike; TGW, Thousand grains weight; 483

RDW, Root dry weight; SDW, Shoot dry weight; MRL, Maximum root length; TRL, Total root length; AD, Root average diameter; SA, Root surface area; RV, Root volume; RTN, Number of 484

root tips; RSR, root/shoot ratio. 485

PH TN SNS TGW RDW SDW MRL TRL AD SA RV RTN RSR

FD 0.298** 0.257** 0.508** -0.689** 0.088 0.144* 0.183* -0.048 -0.050 -0.053 -0.057 0.030 -0.095

PH 0.109 0.272** -0.040 0.326** 0.329** 0.103 0.329** -0.122 0.323** 0.308** 0.255** -0.008

TN 0.160* -0.220** 0.073 0.084 0.197** 0.041 -0.216** 0.014 -0.016 0.150* -0.012

SNS -0.391** 0.157* 0.194** 0.193** 0.075 0.087 0.091 0.104 0.124 -0.032

TGW 0.028 0.043 -0.211** 0.248** -0.053 0.243** 0.232** 0.120 0.014

RDW 0.833** 0.297** 0.784** 0.088 0.828** 0.852** 0.578** 0.358*

SDW 0.295** 0.807** 0.029 0.838** 0.847** 0.677** -0.186*

MRL 0.301** -0.003 0.313** 0.316** 0.409** 0.041

TRL -0.276** 0.988** 0.948** 0.855** 0.053

AD -0.127 0.035 -0.460** 0.085

SA 0.986** 0.805** 0.073

RV 0.731** 0.094

RTN -0.081

Page 15 of 20


Genome

Draft

16

486

Table 3 QTLs controlling maximum root length (MRL), total root length (TRL), root 487

average diameter (AD), root surface area (SA), root volume (RV), number of root tips 488

(RTN) and root/shoot ratio (RSR) detected in the RIL population of wheat derived 489

from the Q1028 × ZM9023 cross. 490

491

Trait QTL Chromosome Interval* LOD PVE (%)† Add.

‡

MRL QMrl.sau-3B 3BS Xgpw7452-wPt-6047 4.43 9.85 1.12

QMrl.sau-6B 6BS wPt-2564-wPt-8183 3.86 6.94 0.95

QMrl.sau-7A 7AS wPt-744715-Xgpw3208 3.20 6.18 0.89

QMrl.sau-7B 7B§ Xgpw4444-Xgpw3111 7.75 16.68 -1.46

TRL QTrl.sau-4B 4BL Xbarc1096-Xwmc47 4.51 12.64 -42.89

QTrl.sau-7A 7AS wPt-4515-Xgpw3220 2.55 5.67 -28.77

AD QAd.sau-3D 3DS Xcfd70-Xgpw5104 3.99 8.74 0.002

QAd.sau-5B 5BL wPt-1304-tPt-1253 2.93 5.83 -0.002

QAd.sau-7A 7AS wPt-4515-Xgpw3220 5.83 12.18 0.003

SA QSa.sau-4B 4BL Xbarc1096-Xwmc47 3.98 10.61 -3.70

RV QRv.sau-4B 4BL Xbarc1096-Xwmc47 3.28 8.79 -0.03

RTN QRtn.sau-4B 4BL Xbarc1096-Xwmc47 2.83 8.00 -97.82

QRtn.sau-7A 7AS wPt-4515-Xgpw3220 3.82 8.61 -101.63

RSR QRsr.sau-1B 1BS wPt-1876-wPt-3819 2.76 6.80 -0.0007

QRsr.sau-7A 7AS tPt-1755-rPt-0950 2.54 5.73 0.0007

* The left and right flanking markers of the interval of the logarithm of odds (LOD) 492

peak value for a given QTL. 493

† Percentage of the phenotypic variation explained. 494

‡ Additive effect of a QTL. Positive values of additive effect indicate that alleles from 495

Q1028 in the direction of increasing the traits scores, and the negative values indicate 496

that alleles from ZM9023 in the direction of increasing the score. 497

§ The arm location of QMrl.sau-7B is difficult to analysis. Xgpw4444 and Xgpw3111 498

were belonged to 7BS and 7BL, respectively. 499

500

Page 16 of 20


Genome

Draft

17

Figure legends 501

Fig. 1 Comparison of root systems for ZM9023 and Q1028. Scale represents 10 cm. 502

Fig. 2 Frequency distributions of the root traits from the RIL population. Data was 503

based on BLUP values of the investigated experiments. The black arrow indicates the 504

value of Q1028 and the white arrow indicates ZM9023. 505

Fig. 3 QTLs for the root traits of the RIL population. Abbreviations for traits are: 506

MRL, Maximum root length; TRL, Total root length; AD, root average diameter; SA, 507

Root surface area; RV, root volume; RSR, root/shoot ratio; RTN, number. 508

Page 17 of 20


Genome

Draft

Fig. 1 Comparison of root systems for ZM9023 and Q1028. Scale represents 10 cm.

122x279mm (300 x 300 DPI)

Page 18 of 20


Genome

Draft

Fig. 2 Frequency distributions of the root traits from the RIL population. Data was based on BLUP values of the investigated experiments. The black arrow indicates the value of Q1028 and the white arrow indicates

ZM9023.

73x63mm (300 x 300 DPI)

Page 19 of 20


Genome

Draft

Fig. 3 QTLs for the root traits of the RIL population. Abbreviations for traits are: MRL, Maximum root length; TRL, Total root length; AD, root average diameter; SA, Root surface area; RV, root volume; RSR, root/shoot

ratio; RTN, number.

76x85mm (600 x 600 DPI)

Page 20 of 20


Genome

Documents

Draft - tspace.library.utoronto.ca · Draft 2 19 20 Abstract 21 As a primitive hexaploid wheat resource distributed only in Tibet, Tibetan semi-wild 22 wheat (Triticum aestivum ssp