RESEARCH
Mapping QTL, Selection Differentials, and the Effect of Rht-B1 under Organic and Conventionally Managed Systems in the Attila × CDC Go Spring Wheat Mapping Population M. Asif, R.-C. Yang, A. Navabi, M. Iqbal, A. Kamran, E. P. Lara, H. Randhawa, C. Pozniak, and D. Spaner*
ABSTRACT A randomly derived recombinant inbred line (RIL) population (n = 163) from a cross between CIMMYT spring wheat ‘Attila’ and the Canadian ‘CDC Go’ was used to map quantitative trait loci (QTL) affecting various agronomic and quality traits. The experiment was also designed to investigate the feasibility of organic wheat breeding by determining selection differentials and the effect of Rht-B1 in paired organic and conventional management systems. Heritability estimates differed between systems for five of nine traits measured; including grain yield, number of tillers, plant height, kernel weight, and grain protein content. Direct selection in each management system resulted in 50% or fewer selected individuals in common between the two systems, for eight of the nine (except for flowering time) studied traits. Most QTL were specific to either the organic or the conventional management system. However, consistent QTL for grain yield, grain volume weight, kernel weight, and days to flowering were mapped in both systems on chromosomes 6A, 1B, 3A, and 5B, respectively. The effect of Rht-B1 was more pronounced in organic systems, where RILs carrying the wild-type allele were taller, produced more grain yield with higher grain protein content, and suppressed weed biomass to a greater extent than those carrying dwarfing alleles. Results of the present study suggest that differences exist between the two management systems for QTL effects. Indirect selection of superior genotypes from one system to another will not result in the advancement of the best possible genotypes. Therefore, selection of spring wheat cultivars for organic systems should be conducted on organically managed lands.
crop science, vol. 55, may– june 2015
M. Asif, R.-C. Yang, M. Iqbal, A. Kamran, E.P. Lara, and D. Spaner, Department of Agricultural, Food and Nutritional Science, Univ. of Alberta, Edmonton, AB T6G 2P5, Canada; R.-C. Yang, Alberta Agriculture and Rural Development, #307, J.G. O’Donoghue Building, 7000- 113 St. Edmonton, AB T6H 5T6, Canada; A. Navabi, Agriculture and Agri-Food Canada, c/o Dep. of Plant Agriculture, Univ. of Guelph, 50 Stone Road, Guelph, ON N1G 2W1, Canada; M. Iqbal, National Institute for Genomics and Advanced Biotechnology, National Agricultural Research Centre, Islamabad, Pakistan; A. Kamran; Seed Centre, Dep. of Botany, Univ. of the Punjab, Lahore, Pakistan; H. Randhawa, Agriculture and Agri-Food Canada, 5403-1st Avenue South, Lethbridge, AB T1J 4B1, Canada; and C. Pozniak, Crop Development Centre and Dep. of Plant Sciences, Univ. of Saskatchewan, 51 Campus Drive, Saskatoon, SK S7N 5A8, Canada. Received 28 Jan. 2014. Accepted 22 Dec. 2014. *Corresponding author (
[email protected]). Abbreviations: CWRS, Canada Western Red Spring; DH, doubled haploid; PAR, photosynthetically active radiation; PCR, polymerase chain reaction; QTL, quantitative trait loci; RIL, recombinant inbred line.
M
any wheat (Triticum aestivum L.) breeding objectives for organically managed systems (including grain yield, resistance to abiotic and biotic stresses, and baking quality) are similar to conventionally managed systems. It may be, however, necessary to test the expression of these traits under low input, very weedy organic conditions to maximize gain from selection in organic environments. A few traits relevant to high input farming may have negative effects on organic lands. The breeding for semidwarf wheat cultivars as a result of the green revolution has resulted in (i) reduced depth and size of root systems, (ii) increased reliance on N fertilizers to attain satisfactory protein content, (iii)
Published in Crop Sci. 55:1–14 (2015). doi: 10.2135/cropsci2014.01.0080 © Crop Science Society of America | 5585 Guilford Rd., Madison, WI 53711 USA All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Permission for printing and for reprinting the material contained herein has been obtained by the publisher. www.crops.org 1
lower N use efficiency, and (iv) decreased competitiveness against weeds (Zerner et al., 2008). These semidwarf wheat cultivars may not have stable yield performance in organic and low-input farming systems (Mason et al., 2007a; Reid et al., 2009b; Gooding et al., 2012). Competition from weeds is a major production constraint in organic systems and causes significant grain yield losses. In organic production systems, the lack of efficient and effective weed management strategies is considered the major constraint to improved grain production (Hiltbrunner et al., 2002; Gianessi and Reigner, 2007; Hiltbrunner et al., 2007; Wszelaki et al., 2007). Weed control in organically managed lands is challenging. Most synthetic herbicides are forbidden in organic agriculture (Kruidhof et al., 2008). Some herbicides have been approved but are costly, nonselective, and potentially harmful to crops (Knezevic, 2009). Alternatives to herbicides such as hand weeding is expensive (prices vary from $300 to 800 ha–1), time consuming, and laborious (Kruidhof et al., 2008). Hand weeding is only practiced on pedigreed farms or in highvalue horticultural crops, and is only feasible on a large scale if weed abundance is low (Frick, 1998). Breeding for improved competitive ability against weed pressure has the potential to minimize yield losses in both conventional and organic production systems. The presence of genetic variability is a prerequisite for improvement of any trait. Previous studies have reported the existence of considerable genetic variation for competitive ability among wheat cultivars. Wheat exhibited poorer competitive ability against annual ryegrass (Lolium rigidum Gaudin) than barley (Hordeum vulgare L.), oat (Avena sativa L.), canola (Brassica napus L.), rye (Secale cereale L.), and triticale (´Tritticosecale; Lemerle et al., 1995). In a study to determine competitive ability of 63 historical and modern wheat cultivars, wheat cultivars with high competitive ability suppressed weed biomass 5.7 times more than cultivars with the lowest weed suppression ability (Murphy et al., 2008). Mokhtari et al. (2002) studied the genetic basis of variation for tolerance against annual ryegrass (L. rigidum) in two F2:3 wheat populations derived from crosses between locally adapted Australian wheat cultivars with good and poor competitive abilities. They found significant genetic variation for competitive ability in both wheat populations. They reported heritability estimates of 0.57 and 0.27 for percentage yield loss (tolerance) due to competition on an entry mean basis, in late and early crosses, respectively. Studies on the relationship between agronomic and/ or morphological traits and their effect on competitive ability with weeds have been well documented. Traits that can contribute to competitive ability include early season vigor, plant height (Huel and Hucl, 1996), flowering, maturity (Mokhtari et al., 2002), light interception (Lemerle et al., 1994; Reid et al., 2009b), and tiller 2
production (Lemerle et al., 1996a; Lemerle et al., 1996b). Plant height has an effect on wheat grain yield and is also associated with competitive ability of wheat. Taller plants exhibit better competitive ability against weeds than shorter ones mainly due to better light interception that directly alters the photosynthetic activity of crop plants (Cudney et al., 1991; Gooding et al., 1993; Thomas et al., 1994; Cousens et al., 2003a; Cousens et al., 2003b; Mason et al., 2007b; Beres et al., 2010). Huel and Hucl (1996) reported a significant positive correlation of plant height with competitive ability. They observed that shorter cultivars suffered greater yield losses than taller cultivars in a weed competitive environment. The effect of height reducing Rht genes in wheat was investigated in an organic environment (Gooding et al., 1997). The authors found a greater weed (Alopecurus myosuroides Huds.) infestation in wheat genotypes with Rht genes and attributed it to the reduction in shading ability of short statured wheat plants that allowed an increased penetration of photosynthetically active radiation (PAR), allowing for the proliferation of weeds. To be competitive, short statured wheat genotypes must have erect leaves with higher leaf area to maximize light interception (Watson et al., 2006). Early season vigor is another important trait that contributes to competitive ability. Early season vigor is highly dependent on relative growth rate (i.e., ability to increase biomass per unit of time) of crop plants (Grime, 1979). To breed for organic management, plant traits that efficiently utilize resources during early growth stages and improved competitive ability are required. These traits include greater PAR, and increased biomass and tillering capacity. In a population derived from early flowering parents, Mokhtari et al. (2002) reported that the effectiveness of indirect selection based on total dry weight and number of heads per plants in monoculture was higher than direct selection. However, the traits evaluated expressed low heritability and were poorly correlated with weed tolerance in a population derived from late-flowering parents. Moreover, a positive correlation between time to anthesis and percentage yield loss in both populations suggested that earliness is linked to competitiveness and/or tolerance (Mokhtari et al., 2002). The environment influences the expression of various agronomic and/or morphological quantitatively inherited traits associated with competitive ability. Coleman et al. (2001) studied a population of 161 doubled haploid (DH) wheat lines derived from a cross between ‘Cranbrook’ and ‘Halbred’, and mapped QTL for various competitive ability traits including plant height, flag leaf size, anthesis, and size of first two leaves on similar positions at chromosomes 2B and 2D for 2 yr. Steege et al. (2005) reported 85 QTL controlling early season vigor on chromosomes 1D, 2D, 4D, 5D, and 7D of Aegilops tauschii Coss., the D genome donor of hexaploid wheat. Spielmeyer et al.
www.crops.org
crop science, vol. 55, may– june 2015
(2007) reported that early season vigor was associated with greater seedling leaf area and longer coleoptiles in wheat and identified a QTL on chromosome 6A that explained 6 and 14% of the phenotypic variation for seedling leaf width and coleoptile length, respectively. The SSR marker NW3106 was linked to the QTL, and has been suggested as a tool for marker assisted breeding for coleoptile length in early generations to improve early season vigor. Our research group has been studying the competitive ability of wheat to develop a spring wheat ideotype for organic environments on a 4 ha of land that has been organically managed in Edmonton, AB, Canada, since 1999 (Mason et al., 2006, 2007a, 2007b; Reid et al., 2009a, 2009b, 2011; Asif et al., 2012). Previous findings reported from this location provided a basis for the present study and the following objectives: to (i) identify differences in heritability and selection response between organic and conventional production systems, (ii) identify and map genomic regions (QTL) associated with various agronomic traits, (iii) compare and uncover system-specific QTL, and (iv) examine the effect of Rht-B1 on various traits in both systems to facilitate the development of competitive and highyielding wheat cultivars for organically managed lands.
MATERIALS AND METHODS Population Development
A population of RILs derived from a cross between the CIMMYT spring wheat cultivar Attila (CM85836-50Y-0M0Y-3M-0Y) and the Canadian spring wheat cultivar CDC Go was used in the present study. CDC Go is a hollow-stemmed, high-yielding, semidwarf bread wheat cultivar that expresses high grain protein content. It is a western Canadian variety with a market class designation of Canada Western Red Spring (CWRS). In 2010, it was grown on 5.9% of the wheat production area of the Canadian Prairies regions (Canadian Wheat Board, 2010). Attila is an awned, semidwarf bread wheat cultivar widely grown in Southeast Asia (Rosewarne et al., 2008). Reid et al. (2009b) reported that Attila was high yielding (5.34 t ha–1), semidwarf (84 cm), with average maturity for the regions tested (135 d). The original population consisted of 171 F6 derived F7 RILs, which were advanced to F6 using single seed descent. The RILs and two parents were planted in double head rows to multiply seed for experimental use as F6 derived F7 in 2007.
Phenotyping Field experiments were conducted at the Edmonton Research Station, University of Alberta, South Campus, Edmonton, AB, Canada (latitude: 53’34° N, longitude: 113’ 31° W, elevation: 723.3 m) during 2008 to 2010 on organically and conventionally managed lands. In all 3 yr, the trials were conducted on paired sites, one organically and one conventionally managed, located approximately 500 m apart on the same soil type and with the same weather conditions. Sowing of plots was performed on May 29th, 13th and 18th at the organically managed site and on May 29th, 13th and 17th at the conventionally managed site in 2008, 2009, and 2010, respectively. Experiments crop science, vol. 55, may– june 2015
were harvested on September 19th (organic) and September 29th (conventional) in 2008; September 23rd (organic) and September 9th (conventional) in 2009 and on September 23rd (organic) and September 18th (conventional) in 2010. The experiments were arranged in randomized incomplete blocks within each of three (2008 and 2009) and two (2010) blocks. The plots were seeded with 300 viable seeds m–2. The plot (5.4 m 2) consisted of six rows 4m in length with a row spacing of 22.5 cm between rows. The 3-yr crop rotation on the conventionally managed site was wheat-pea-canola, while on the organically managed site wheat followed a green manure rye plow-down. On the conventionally managed site, fertilizer (11–52–0 N–P2O5 –K 2O) was banded with seed at the time of planting at a rate of 36, 40, and 40 kg ha–1 in 2008, 2009, and 2010, respectively. The organically managed site did not receive any chemical fertilizer or herbicide. No compost was applied on the organically managed site. Soil nutrient levels of N, P, K, and S were 34, 60, 555, and 20 mg kg–1, respectively, in the organic systems. Thus, all nutrient levels were optimum except N, which was considered to be marginal. Soil pH, electric conductivity, and organic matter was 6.7 (neutral), 0.70 dS m–1 (good), and 10.8% (high), respectively. Land was tilled 2 wk before planting to prepare fields for planting in both systems. Organically managed fields were also tilled and harrowed 1 d before planting as a weed control measure. The soil at both sites is an Udic Boroll (Orthic Black Chernozem in Canadian system) with loam or clay loam texture, neutral pH (6.7 to 7.4), and high soil organic matter (6 to 11%), which is typical of central Alberta (AAFRD, 2004). Standard agronomic practices including fertilizer and herbicide applications were performed on the conventionally managed sites only during the growing seasons to obtain optimum crop stands consistent with conventional management systems. Phenotypic data collected included days to flowering, physiological maturity, plant height, grain volume weight, kernel weight, number of tillers, grain protein content, weed biomass, light capture, and grain yield. Days to flowering were recorded as the number of days from seeding to when 75% of the spikes in a plot had visible peduncles. Physiological maturity was recorded as the number of days from seeding to a point when 75% of the peduncle in the plot lost green color. Plant height was recorded on a plot basis at the completion of stem elongation, measured from the soil surface to the tip of spike, excluding awns. Grain volume weight was calculated by weighing a 1-pint (473 mL) subsample of plot yield. Numbers of tillers m–2 were calculated by counting fertile tillers from a randomly chosen 0.5 m length of two rows. Grain protein content was determined using near-infrared reflectance spectroscopy using a Monochromator NIR system model 6500 (NIRSystems, Inc., Silver Springs, MD). Weed biomass was recorded to measure suppressive ability only in organically managed plots. Samples were collected at physiological maturity and dried for 3 d at 50°C to determine dry weight. The harvested grain samples were dried at 60°C for ~24 h and grain yield on a plot basis was determined by weighing the clean seeds. Photosynthetically active radiation was recorded only in conventionally managed plots using LI-COR LI-191SA Line Quantum Sensor (LI-COR Bioscience, Lincoln, NE). The Line Quantum Sensor was held in the middle of a field plot at ground level and above the crop
www.crops.org 3
Table 1. Least squares means, range, and heritability estimates of various traits for parents and 163 recombinant inbred lines of Attila × CDC GO mapping population under conventional (Con) and organic (Org) management systems in Edmonton, AB, Canada, during 2008–2010.
Attila†
Variables Grain yield, t ha Tillers, m –2
–1
CDC Go†
Difference between parents‡
Population
Con§
Org§
Heritability, %†
Difference between Org systems† min.
max.
min.
max.
Con
SE
Org
SE
4.51** 2.74** 4.93* 3.81* –0.42 –1.07** 4.63 481 408 549 404 –68 4 513 71* 61* 73 72 –2 –11* 77 77* 75* 79* 77* –2** –2** 78
3.24 454 72 76
1.39** 59** 5** 2**
0.43 217 30 71
8.32 853 114 88
0.12 217 28 41
6.94 403 116 82
37** 32** 58* 18
3 4 3 3
18** 16** 43* 24
3 3 4 3
41** 13.4 50 5 58 95**
39 14.4 40 4 59 97
2** –1.2**
27 8.4
62 17.7
4 3
1 42 59 –0.01
5 81 127 0.98
50 17.9 360 5 80 122
27** 64**
0.5* 0 7**
22 7.2 10 2 42 72
8 76 38 55
2 2 4 3
44** 27** 83 7 70 47
4 3 5 3 3 4
Con
Org
Con
Plant height, cm Grain volume weight, kg hL–1 Kernel weight, g 39* 37* 42** Grain protein, % 12.3** 14.4** 14.4 Weed biomass, g m –2 70 Early season vigor 3 4 4 Days to flowering 60 61 57 Days to maturity 106** 99** 102** Light capture 0.62 0.66
Org
Con
–3 –2.1** – –1* 3** 4 –0.04
Org
–4* 1.0 20** –1* 3** 4
Con
41 13.2 3.5 59 104 0.57
* Significant at P < 0.05. ** Significant at P < 0.01. †
Statistical differences tested between management systems.
‡
Statistical differences tested between Attila and CDC Go.
§
min. and max. represents LS means recorded for a particular trait in any of the years.
canopy, with PAR recorded in µmol s–1 m–2. Proportion of captured light was calculated using the following formula: Light Capture = 1 – (PAR Below Canopy)/ (PAR Above Canopy).
Genotyping Genomic DNA of RILs along with parents was extracted from the leaves of 3-wk-old seedlings using DArT protocol (www.diversityarrays.com, verified 24 Feb. 2015). DNA concentration of each sample was measured using NanoDrop (ND-1000 Spectrophotometer, Thermo Scientific, Waltham, MA), adjusted to 100 ng/ mL and was sent to Diversity Arrays Technology Pty. Ltd., Yarralumla, Australia, for genotyping against 7000 cloned sequences. Approximately 579 markers were found to be polymorphic within the population. The DArT technology followed protocols similar to those illustrated by Akbari et al. (2006). It consists of reducing the complexity of the DNA samples by restricting samples with methylation-sensitive restriction endonucleases and annealing adapters to facilitate polymerase chain reaction (PCR) amplification. Amplified fragments are then labeled, and hybridized to a microarray of variable fragments which represents diversity within a species. Polymorphisms are assessed as standardized hybridization intensities. The parents, along with the population, were also screened for allelic variation at Rht-B1 and Rht-D1 loci using perfect DNA markers (Ellis et al., 2002).
traits was performed using PROC MIXED in SAS v.9.2 (SAS, 2008). Data were analyzed for individual years (2008, 2009, and 2010) in organically and conventionally managed sites separately and then combined over years. PROC MIXED of SAS was used for estimation of Least squares means (LS Means; Yang, 2010) where RILs were used as fixed effects, and years, replications, blocks within replications as random effects. The Estimate statement of SAS was used to estimate the significance of differences between LS Means presented in Table 1. Broad-sense heritabilities were estimated on plot basis as 2 2 2 2 H = sG2 / ( sG2 + sGE + se2 ) , where sG , sGE , and se are among RILs, RIL × environment, and error variances, respectively. Genetic correlations among the traits were calculated using a multivariate restricted maximum likelihood (REML) method in the MIXED procedure of SAS (Holland, 2006). The estimated genetic [ rˆg( xy ) ] and phenotypic correlations [ rˆp( xy ) ] between traits x and y are given as follows: rˆg( xy ) =
rˆP( xy ) =
sˆ G( xy ) 2 ˆ2 sˆ G( x ) × sG( y )
sˆ G( xy ) 2 sˆ 2P( x ) × sˆ P( y)
,
=
sˆ G( xy ) + sˆ GE( xy ) + sˆ e( xy ) 2 ˆ2 ˆ2 ˆ2 ˆ2 ˆ2 sˆ G( x ) + sGE( x ) + se( x ) . sG( y ) + sGE( y ) + se( y )
,
Statistical Analysis Phenotypic data were first tested for normality and homogeneity of the error variances using Kolmogorov-Smirnoff and Levene’s tests, respectively (Steel et al., 1997). Analysis of variance for all 4
where sˆ G( xy ) , s ˆ P( xy ) , s ˆ GE( xy ) , and sˆ e( xy ) are the estimated genetic, phenotypic, genotype ´ environment and error covariances, respectively, between the two traits x and y, while
www.crops.org
crop science, vol. 55, may– june 2015
Figure 1. Frequency distribution showing transgressive segregation for grain yield and plant height of 163 wheat recombinant inbred lines (RILs) derived from a cross between ‘Attila’ and ‘CDC Go’ in Edmonton, AB, Canada, during 2008–2010.
2 , sˆ 2P , sˆ GE , and sˆ e2 are estimated genetic, phenotypic, genotype ´ environment, and error variances calculated for both traits (Holland 2006). The genotypic correlation coefficients were then transformed as suggested by Fisher (1925): Z xy = éê ln (1 + rxy ) - ln (1 - rxy )ùú / 2 to test whether correlation ë û coefficients are significantly different from zero. The new variable, Zxy, was tested under the null hypothesis of no correlation with Z ¢ = Z xy / (1/ n - 3 ) where n is the number of RILs. 2 ˆG s
QTL Detection and Analysis Composite interval mapping was performed using QTL Cartographer (WinQTL) version v.2.5.010 (Wang et al., 2010) to detect/identify QTL influencing each trait using LS Means of each trait individually for each site (organic and conventional) for 3 yr (2008, 2009, and 2010) and then combined over years. The overall significance level for declaring a definitive QTL was obtained using a permutation test (Churchill and Doerge, 1994) wherein the analysis was replicated 1000 times on data sets generated by random reshuffling of the trait values in the original data. The chromosome scan or walk for detecting QTL was performed at an interval of 1 cM at a time for all traits under study. The position at which the peak of logarithm of odds (LOD) scores reached the maximum was used to declare a QTL location. The percentage of phenotypic variation explained by a QTL was calculated using the coefficient of determination (R 2) based on the marker that is closest and thus linked to the QTL. The QTL identified in the study were named according to the guidelines illustrated in the catalogue of gene symbols for wheat (McIntosh et al., 2003).
crop science, vol. 55, may– june 2015
RESULTS Phenotypic Evaluation On average, the parents (Attila and CDC Go) yielded more grain, having taller plants with higher grain volume and kernel weights under conventional than organic systems (Table 1). The parents also took longer to mature in conventional management. The grain protein content of Attila (14.4%) was higher than CDC Go (13.4%) in organically managed systems. Differences between the two parents were observed for plant height (11 cm), grain volume weight (2 kg hL–1), kernel weight (5 g), early season vigor (1), days to flowering (3), and weed biomass (20 g m–2) in organic systems; and for grain volume weight (2 kg hL–1), grain protein content (2.1%), early season vigor (1), and days to flowering (3) in conventional management systems. The RILs (n = 163) differed for all traits except days to flowering between the two management systems (Table 1). Transgressive segregation was observed in the population for all traits, including grain yield and plant height (Fig. 1) under both management systems. Heritability estimates were found to be different for grain yield (37 and 18%), tillers (32 and 16%), plant height (58 and 43%), kernel weight (27 and 44%), and grain protein content (64 and 27%) between conventional and organic management systems (Table 1). Spearman’s rank correlations between the two management systems were high for kernel weight (0.70), early season vigor (0.80), and days to maturity (0.74), moderate for tillers (0.51), grain volume weight (0.60), and days to flowering (0.54), with no correlation for grain yield, plant height, and grain protein content (Table 2). Direct selection in the conventional and organic management system
www.crops.org 5
Table 2. Spearman’s rank correlations (rs) along with the number of recombinant inbred lines (RILs) common at three selection intensities (5, 10, and 15%) for 9 traits in the Attila × CDC Go population. Line selected in common Variables Grain yield, t ha Tillers, m –2
–1
Plant height, cm Grain volume weight, kg hL–1 Kernel weight, g Grain protein, % Early season vigor Days to flowering Days to maturity
5%†
10%†
15%†
rs
9‡
17‡
26‡
0.20 0.51 0.10 0.60 0.70 0.24 0.80 0.54 0.74
3 2 4 3 1 4 1 5 2
5 3 9 6 5 8 4 13 5
13 7 13 11 10 12 6 18 12
†
Selection intensity applied within each management system.
‡
Maximum number of lines selected from the experimental population of 163 RILs at the given selection intensity.
(up to a 15% selection intensity) resulted in 50% or less RILs selected in common (except for flowering time) for the nine studied traits. For instance, if the top-yielding nine (at 5% selection intensity), 17 (at 10% selection intensity), and 26 (at 15% selection intensity) RILs of the population were selected from each management system, only half or less (3, 5, and 13, respectively) were in common when selection was based on both systems (Table 2; Fig. 2). Phenotypic and genotypic correlation coefficients among different traits were estimated for each management system separately (Tables 3 and 4). Grain yield was moderately correlated with plant height (rg = 0.44) and days to maturity (rg = 0.49) in conventionally managed systems. There were negative genotypic and phenotypic correlations between grain yield and grain protein content in both management systems. Genotypic correlations of early season vigor with both days to flowering and maturity were negative in both management systems. Plant height
Figure 2. Observed rank changes in the top 10% recombinant inbred lines (RILs; represents 17 lines) ranked under conventional (C) and organic (O) management systems for nine different traits. The RILs were ranked according to the desired direction of selection (e.g., Rank 1 for grain yield was the highest-yielding RIL).
6
www.crops.org
crop science, vol. 55, may– june 2015
Table 3. Genetic and phenotypic correlations estimated under conventional management systems in Attila × CDC Go population in Edmonton, AB, Canada, during 2008–2010.†
Variable Grain yield, t ha –1 Tillers, m –2 Plant height, cm Grain volume weight, kg hL–1 Kernel weight Grain protein, % Early season vigor Days to flowering Days to maturity
Test weight
Kernel weight
Grain protein
Early season vigor
Days to flowering
Days to maturity
Light
Grain yield
Tillers
Plant height
----0.28 0.24 –
0.31 ----– –
0.44 –0.74 ----–
0.52 – – -----
0.39 – – –
–0.63 – – –
–‡ – – 0.52
0.31 0.35 0.41 –0.26
0.49 – – –
– – 0.57 –
0.29 –0.47 0.35 – 0.28
– –0.37 – – –
– – – 0.32 –
– – – – –
----– 0.29 – –
0.34 -----
0.54 0.29 ----– –0.27
–0.28 –0.47 –0.53 ----0.6
– –0.55 –0.64 0.91 -----
0.58 – – – –0.27
–0.35 –0.31
†
Values above and below the diagonal represent genotypic and phenotypic correlation coefficients, respectively.
‡
Correlation coefficient not different from zero (P > 0.05).
Table 4. Genetic and phenotypic correlations under organic management systems in Attila × CDC Go population in Edmonton, AB, Canada, during 2008–2010.†
Variable Grain yield, t ha –1 Tillers, m –2 Plant height, cm Grain volume weight, kg hL–1 Kernel weight Grain protein, % Early season vigor Days to flowering Days to maturity Weed biomass
Test weight
Kernel weight
Grain protein
Early season vigor
Days to flowering
Days to maturity
Weed biomass
Grain yield
Tillers
Plant height
----0.38 0.27
–‡ ----–0.31 –
0.46 –0.39 ----–
0.35 – 0.42 -----
– – 0.26 –
–0.75 – – –
0.36 – – 0.29
0.26 – 0.36 –
0.26 – 0.33 –
–0.64 – –0.98 –0.57
– –0.39 – – – –
– – – – – –
– – – 0.29 0.27 –
0.28 – – – – –
----– – – – –
0.26 ----– – – –
–0.37 – ----– – –
– – –0.63 ----0.66 –
– – –0.53 0.92 ----–
– – – – – -----
†
Values above and below the diagonal represent genotypic and phenotypic correlation coefficients, respectively.
‡
Correlation coefficient not different from zero (P > 0.05).
and number of tillers were negatively correlated in conventional (r = –0.74) and organic (r = –0.39) management systems. Weed biomass and plant height (r = –0.98), grain yield (r = –0.64), and grain volume weight (r = –0.57) were significantly and negatively correlated in organically managed land. Significant positive genotypic correlation between days to flowering and maturity occurred in both conventional (r = 0.91) and organic (r = 0.92) management systems. Light capture (measured only in the conventional system) was positively correlated (genotypically) with plant height (r = 0.57) and kernel weight (r = 0.58) and negatively with days to maturity (r = –0.27).
Marker Analysis, Map Construction, and QTL Detection In total, 579 polymorphic DArT markers (of 7000 clones) were evaluated in the Attila ´ CDC Go population. The linkage map was constructed using DArT markers (www. diversityarrays.com). Initially, 718 markers were found to be polymorphic; 139 DArt markers were noninformative. The map spanned a total distance of 2045 cM and covered 19 wheat chromosomes with an average distance of 3.53 crop science, vol. 55, may– june 2015
cM between markers. Genotyping results uncovered only one and six polymorphic markers on chromosomes 4D and 5D, respectively, which were noninformative. Previous studies have reported low or no polymorphism for DArT markers on the D genome of hexaploid wheat (Kamran et al., 2013; Cui et al., 2014). Overall, we found and mapped various QTL for grain yield (chromosome 6A), number of tillers (chromosome 4A), plant height (chromosome 4B), grain volume weight (on chromosomes 1A, 1B, 5B), kernel weight (on chromosomes 1B, 3A, 4A, 6A and 7A), grain protein (on chromosomes 3B, 5B, 6A, and 6B), early season vigor (chromosome 3B), days to flowering (chromosome 5B), days to maturity (on chromosomes 2B, 4B, and 5B) and light capture (on chromosome 6B) in the six management ´ year environment combinations (Table 5). Among these QTL, six were mapped consistently in both management systems over the 3 yr, on chromosomes 6A, 1B, 3A, and 5B, for grain yield, grain volume weight, kernel weight, and days to flowering, respectively (Table 6).
www.crops.org 7
Table 5. Summary of quantitative trait loci (QTL) identified for 10 different traits in the Attila ´ CDC Go mapping population in Edmonton, AB, Canada, during 2008–2010. Distance from closest marker
S. No.
Trait
QTL†
System†
Years
Chromosome
Closest locus
Map position
LOD‡
R2
Additive effect§
1
Grain yield
Qyld.dms-6A
Org
2010
6A
wPt-1375
93.6
3.2
7.0
0.27
–0.31
Qyld.dms-6A
Org
2009
6A
wPt-741026
91.3
3.7
9.0
1.03
–0.22
Qyld.dms-6A
Con
Combined
6A
wPt-741026
83.4
4.0
17.0
8.93
–0.32
Qyld.dms-6A
Con
2010
6A
wPt-1375
93.9
3.6
7.9
0.03
–0.24
Qyld.dms-6A
Con
2008
6A
wPt-741026
82.3
4.6
22.2
10.03
–0.50
1.00
%
2
Tillers
QTil.dms-4A
Con
2009
4A
wPt-2903
84.4
3.2
7.0
3
Plant height
QHt.dms-4B
Org
Combined
4B
Rht-B1
213.8
8.5
18.1
–3.56
QHt.dms-4B
Org
2010
4B
Rht-B1
213.8
6.0
17.5
–4.88
QHt.dms-4B
Org
2009
4B
Rht-B1
213.8
5.7
13.2
–4.27
QHt.dms-4B
Org
2008
4B
Rht-B1
213.8
7.3
14.4
–5.00
QHt.dms-4B
Con
Combined
4B
Rht-B1
213.8
9.0
19.2
–3.81
QHt.dms-4B
Con
2010
4B
Rht-B1
213.8
10.8
21.6
–4.15
QHt.dms-4B
Con
2009
4B
Rht-B1
213.8
4.8
11.0
–2.53
4
5
6
Kernel weight
Grain protein
QHt.dms-4B
Con
2008
4B
Rht-B1
213.8
8.61
17.7
QTwt.dms-1B
Org
Combined
1B
wPt-5279
79.4
3.7
8.3
0.10
–4.12 0.43
QTwt.dms-1B
Org
2008
1B
wPt-5279
79.3
3.4
7.4
0.10
0.53
QTwt.dms-1A
Con
Combined
1A
wPt-7339
133.7
4.6
10.9
0.60
0.41
QTwt.dms-1A
Con
2010
1A
wPt-7339
133.7
3.8
8.7
0.60
0.47
QTwt.dms-1B
Con
2009
1B
wPt-5279
79.4
4.2
10.2
0.10
0.41
QTwt.dms-5B
Con
2008
5B
wPt-3569
52.0
4.1
29.9
1.23
–0.25
QGwt.dms-4A
Org
Combined
4A
wPt-3398
86.1
4.3
11.2
1.50
1.11
QGwt.dms-6A.1
Org
Combined
6A
wPt-733151
38.1
3.2
7.5
7.79
0.92
QGwt.dms-1B
Org
2010
1B
wPt-1248
36.3
6.7
13.9
4.63
1.94
QGwt.dms-6A
Org
2010
6A
wPt-733151
31.1
3.8
8.0
0.79
1.37
QGwt.dms-3A
Org
2008
3A
wPt-8593
87.5
4.9
10.3
2.10
–1.16
QGwt.dms-3A
Con
2009
3A
wPt-8593
87.5
3.9
7.7
2.10
–0.77
QGwt.dms-4A
Con
2009
4A
wPt-3398
86.1
4.9
12.4
1.50
0.98
QGwt.dms-7A
Con
2009
7A
wPt-0514
78.3
4.5
12.5
3.20
–1.00
QGwt.dms-3A
Con
2008
3A
wPt-8593
87.4
4.1
9.6
2.30
–1.43
QGpc.dms-6A
Org
Combined
6A
wPt-7127
29.6
4.0
10.9
2.69
–0.49
QGpc.dms-6A.1
Org
2009
6A
wPt-729904
11.2
5.0
15.8
2.74
0.67
QGpc.dms-6A.2
Org
2009
6A
wPt-7127
29.6
4.0
10.2
2.69
–0.63
QGpc.dms-6B
Org
2009
6B
wPt-9971
86.3
4.3
9.2
1.00
0.51
QGpc.dms-5B
Con
2009
5B
wPt-4091
150.6
3.7
8.0
1.20
0.29
QGpc.dms-3B
Con
2008
3B
wPt-0751
111.9
4.2
9.3
7.00
0.28
7
Early season vigor
Qesv.dms.3B
Org
2008
3B
wPt-741750
25.1
3.5
9.7
0.70
–0.17
8
Days to flowering
Qflt.dms-5B
Org
2010
5B
wPt-3569
52.0
4.2
14.0
1.23
–2.12
Qflt.dms-5B
Org
2008
5B
wPt-666939
54.7
4.8
32.8
1.03
–2.67
Qflt.dms-5B
Con
2010
5B
wPt-3569
49.0
3.5
21.2
4.23
2.20
Qflt.dms-5B
Con
2008
5B
wPt-666939
54.7
3.6
32.8
1.03
–3.18
Qmat.dms-5B
Org
2009
5B
wPt-3569
50.0
3.5
4.9
3.23
–2.78
Qmat.dms-5B
Org
2008
5B
wPt-666939
53.7
5.9
15.8
0.30
–6.66 –3.43
9
10 †
Grain volume weight
2.91
Days to maturity
Light capture
Qmat.dms-4B
Con
2010
4B
wPt-6209
186.5
4.0
8.7
3.20
Qmat.dms-2B
Con
2009
2B
wPt-4199
77.4
4.4
9.1
2.20
0.68
Qlig.dms-6B
Con
2009
6B
wPt-744302
44.4
4.1
8.9
3.58
–0.03
Stable QTL over organic (Org) and conventional (Con) management systems are underlined.
‡
LOD, logarithm of odds.
§
Positive values indicate that Attila alleles increase the corresponding trait, and, conversely, negative values indicate that Attila alleles decrease it.
8
www.crops.org
crop science, vol. 55, may– june 2015
Table 6. Summary of consistent quantitative trait loci (QTL) identified for four different traits in the Attila × CDC Go mapping population in Edmonton, AB, Canada, during 2008–2010. Year
QTL
2010 2008 and 2009 2008 and 2009 2008 2008 2010
Qyld.dms-6A Qyld.dms-6A QTwt.dms-1B QGwt.dms-3A Qflt.dms-5B Qflt.dms-5B
wPt-1375 wPt-741026 wPt-5279 wPt-8593 wPt-666939 wPt-3569
Trait Grain yield Grain volume weight Kernel weight Days to flowering
Additive effect
R2
Closet locus
Organic
Conventional
Organic
Conventional
7.0 17.9 7.4 10.3 32.8 14.0
7.9 22.2 10.2 9.6 32.8 21.2
–0.31 –0.44 0.53 –1.16 –2.67 –2.12
–0.24 –0.50 0.41 –1.43 –3.18 2.20
Table 7. Effect of Rht-B1 (wild type and mutant) on various traits in the Attila × CDC Go population grown under conventional and organic management systems in Edmonton, AB, Canada, during 2008–2010. Conventional† Trait Grain yield, t ha –1 Tillers, m –2 Plant height, cm Grain volume weight, kg hL–1 Kernel weight, g Grain protein, % Weed biomass, g m –2 Early season vigor Days to flowering Days to maturity Light capture
Wild type
Mutant
4.84 111 76** 77 41 12.9
4.61 112 66** 78 41 12.9
3.7 53 96 0.5
3.5 52 96 0.47
Organic Wild type 3.73* 99 70** 76 39 14* 5.2** 4.01 53 91
Difference Mutant 2.90* 97 65** 76 39 12.8* 8.2** 4.36 52 90
Wild type
Mutant
1.11** 12** 6** 1** 2** –1.1**
1.71** 15** 1* 2** 2** 0.1
–0.31 0 5**
–0.86 0 6**
* Significant at P = 0.05. ** Significant at P = 0.01. †
The effect of Rht-B1 was estimated by comparing LS means of lines carrying wild type and mutant alleles in each management system.
DISCUSSION
Effect of Rht-B1 in Conventional and Organic Management Systems After detecting a consistent QTL (QHt.dms-4B) for plant height over the six management systems and year combinations, the parents along with population was screened for the Rht-B1 locus on chromosome 4B using PCR-based markers reported by Ellis et al. (2002). CDC Go carries the Rht-B1b dwarfing allele, and the population was found segregating for this locus. The Rht-B1 locus was mapped at 213.8 cM. It was 26.6 cM away from the closest DArT (wPt733038) marker. In the conventional management system, RILs carrying the wild-type (tall) and mutant (shorter) allele did not differ for all traits except plant height (Table 7). The RILs with the wild-type allele produced more grain yield (3.73 t ha–1), had higher grain protein content (14%), suppressed weed biomass (5.2 g m–2), and were taller (70 cm) than mutant types in the organic management system (Table 7). Wild-type grain yield was reduced less in the organic system (1.11 t ha–1) than mutant types (1.71 t ha–1). A similar trend occurred for number of tillers, plant height, and grain volume weight. The RILs with the wild-type allele had 1.1% higher grain protein content and matured 5 d earlier in the organic system (Table 7).
crop science, vol. 55, may– june 2015
The Attila × CDC Go mapping population was studied in organic and conventional management systems to investigate selection differences among RILs, map genomic regions controlling agronomic and quality traits, and investigate the effect of the Rht-B1 locus on other traits. Here, we report four main findings: (i) heritability estimates differed between the two systems for five of the nine measured traits, (ii) direct selection in each management system resulted in fewer lines selected in common for nine traits, (iii) QTL were found for grain yield, number of tillers, plant height, grain volume weight, kernel weight, grain protein content, early season vigor, days to flowering, days to maturity and light capture in the six management system × year combinations, and six consistent QTL were mapped on chromosomes 6A, 1B, 3A, and 5B for grain yield, grain volume weight, kernel weight, and days to flowering, respectively, and iv) the effect of Rht-B1 was more pronounced on grain yield, tiller number, grain volume weight, and kernel weight in the organic management system. Genotypic correlation between traits is due to pleiotropy and/or genetic linkage. It depicts the direction and magnitude of correlated response to selection and the relative efficiency of indirect selection (Holland, 2006). In the case of highly correlated traits, plant breeders may directly
www.crops.org 9
select for traits with high heritability to maximize genetic gain in a segregating population. In our study, heritability and correlation estimates differed between management systems. Heritability was higher in the conventional management system for grain yield, number of tillers, plant height and grain protein content. Kernel weight was greater in the organic system. Heritability estimates were high for days to flowering in both systems. It has also been previously reported that indirect selection, that is, selection in an environment different from the target environment, can lead to higher genetic gains for highly heritable traits than direct selection (Hill et al., 1998). Thus, direct selection is more appropriate for low heritability traits. Direct selection in each management system resulted in few RILs in common at 5% selection intensity. Grain yield, number of tillers, plant height, and early season vigor had only 3, 2, 4, and 1 RIL in common at 5% selection intensity. Weed biomass, an indicator of competitive ability, exhibited strong negative genotypic correlations with plant height, grain yield, and grain volume weight in the organic system. This indicated that taller plants suppressed weeds, resulting in lower weed biomass. Shorter plants allowed weeds to grow more. Grain yield decreased in the presence of high weed populations (high biomass) and vice versa. Higher weed population also decreased grain volume weight in the organic system. Spearman’s rank correlation between systems for all traits except early season vigor were below 0.80. We observed low to moderate genetic correlations for complex traits like grain yield between organic and conventional management system over 3 yr. These results suggest that the two systems were quite different for these traits. Thus, indirect selection in conventionally managed systems would be less efficient than direct selection in organic systems to identify superior genotypes. In a study to evaluate 800 barley (Hordeum vulgare L.) breeding lines in low and high yielding environments, Ceccarelli and Grando (1991) reported the best lines selected in low yielding environments outperformed superior lines selected in high yielding environments. In a similar study of a large number of German variety trials, substantial differences were found in the varietal rankings that were grown under high input, low input, and organic conditions (Baresel and Reents, 2006). Mason et al. (2007b) tested 27 spring wheat cultivars in organic and conventional management systems in western Canada and reported 63% more grain yield in the conventional than the organic system. They further reported that a competitive spring wheat ideotype for organically managed lands would be tall, early maturing, with fast early-season growth, and elevated fertile tillers. Few cereal breeders select directly on organic lands. Nevertheless, the Association of Biodynamic Plant Breeders have made direct selection on organic lands and released 12 cultivars exclusively for organic farming. These cultivars 10
were mostly taller, had low harvest indices, and higher grain protein content than cultivars released for conventional farming (Wolfe et al., 2008). Similar findings of direct selection under low input farming have been reported in other cereals like corn (Zea mays L.; Burger et al., 2008), barley (Ryan et al., 2008), oat (Atlin and Frey, 1990), and rice (Oryza sativa L.; Mandal et al., 2010). On the basis of our findings, we conclude that indirect selection of spring wheat genotypes on conventionally managed systems may not be useful for organically managed systems. Therefore, wheat breeding programs for organic systems should select promising genotypes directly on organic lands. We used the Attila × CDC Go population for genome wide QTL analyses to uncover putative QTL for traits conferring competitive ability. The grain yield QTL identified in this study are positioned between 82.3 and 93.9 cM on chromosome 6A and are most likely the same QTL previously reported. Heidari et al. (2011) mapped three grain yield QTL on chromosome 6A that explained up to 20.9% phenotypic variation for grain yield in wheat. In a similar study, Baenziger et al. (2011) developed a full set of chromosome substitution lines using two historically important wheat cultivars, Wichita and Cheyenne. They reported a grain yield advantage of 19 and 14% when Wichita chromosomes 3A and 6A were substituted for Cheyenne chromosomes, respectively. Reciprocal substitution lines showed an opposite trend (e.g., grain yield reduction of 17 and 23% for 3A and 6A chromosomes, respectively) leading to the conclusion that these chromosomes contain important genomic regions influencing grain yield in wheat. In our study, the plant height gene (Rht-B1) was mapped at 213.8 cM, and was associated with the Rht-B1 locus with CDC Go contributing the Rht-B1b (dwarfing) allele. The Rht-B1 locus was 26.6 cM away from the closest DArT (wPt-733038) marker. Attila is also a semidwarf wheat, but we did not find allelic variation between the two parents, and no segregation in the mapping population at a second major dwarfing locus (Rht-D1) based on the perfect DNA markers for this locus (Ellis et al., 2002). This indicated that Attila carries another dwarfing gene that we were not able to identify. The Rht-B1b (dwarf ) did not alter traits other than plant height in the conventional management system. Genotypes carrying Rht-B1b did have reduced plant height, lower grain yield, and higher weed biomass in plots where grown under organic management. Therefore, Rht-B1 interacted with the management (cropping) system and exhibited negative effects on wheat grain yield and competitive ability in organic systems. Several studies report that wheat cultivars carrying dwarfing alleles on the Rht-B1 and Rht-D1 loci exhibit shorter coleoptiles, lower rates of dry matter accumulation, and have a reduced leaf elongation rate when grown in hot and dry (stressed) environments of North America and Australia (Bai et al., 2004; Botwright et al., 2001;
www.crops.org
crop science, vol. 55, may– june 2015
Ellis et al., 2004; Richards, 1992). It has been suggested that Rht-B1b does not have as great an effect as Rht-D1b on the associated traits discussed above (Addisu et al., 2010). Tiller numbers, early season vigor, and light capture are considered important traits in conferring competitive ability in spring wheat (Mason et al., 2007b). We found QTL for each of these traits on chromosomes 4A, 3B, and 6B, respectively. The early season vigor QTL, Qesv. dms.3B, identified in organic management systems in our study, may be the previously reported early vigor QTL that was mapped on 3BL in the RAC875 × Kukri DH mapping population tested under drought and heat stress conditions (Bennett et al., 2012; Bonneau et al., 2013). To our knowledge, the QTL identified for number of tillers and light capture in the present study have not been previously reported. In our study, grain volume weight and kernel weight differed between the two parents and the population in conventional and organic management systems. Grain volume weight is of special interest to wheat millers due to its positive correlation with flour yield, whereas kernel weight is associated with grain yield. Grain volume weight QTL have been previously reported on chromosomes 1A, 1B, 1D, 2D, 3B, 3D, 4A, 4D, 5A, 5D, 6B, and 7A (Elouafi and Nachit, 2004; Huang et al., 2006; McCartney et al., 2005; Narasimhamoorthy et al., 2006; Zhang et al., 2008). We mapped a consistent QTL for grain volume weight on chromosome 1B at the same position (79.4 cM) in both systems. This QTL was linked to the DArT marker wpt5279. It accounted for 7.4% of the variation in the organic system and 10.2% of the phenotypic variation in the conventional management system for grain volume weight. The additive effect of this QTL was slightly more pronounced in the organic (0.53) than the conventional (0.41) system. The QTL QTwt.dms-1A, linked to marker wpt7339, was also detected under conventional management system. A grain volume weight QTL (QTwt.dms-5B) localized to chromosome 5B in the conventional management system during 2008 explained 29.9% of the variation for the trait, and to our knowledge has not been previously reported. We mapped kernel weight QTL on chromosomes 6A and 1B only in the organic system, and another on 7A only in the conventional management system. A grain volume weight QTL, QGwt.dms-3A linked to wPt8593, was mapped on chromosome 3A across management systems during 2008. The additive effect and phenotypic variation explained by this QTL did not differ between the two systems. Grain protein content is one of the most important quantitatively inherited traits and is an important consideration in the development of wheat cultivars in the CWRS class of Canadian wheat. Grain protein content is negatively correlated with grain yield and, therefore, poses a serious challenge to wheat breeders (Steiger et al., 1996). Grain protein content is also influenced by the rate and time of crop science, vol. 55, may– june 2015
N application and moisture availability in the soil. In this study, we mapped four QTL for grain protein content during 2009, the year with the least rainfall. Three QTL identified during 2009 had a positive additive effect on grain protein content. The QTL, QGpc.dms-6A.1 and QGpc. dms-6A.2 mapped on chromosome 6A during 2009 in the organic management system, collectively explained 26% of the phenotypic variation for grain protein content. Several attempts were previously made to identify minor and major QTL for grain protein content (Blanco et al., 2006; Groos et al., 2003; Huang et al., 2006; Kunert et al., 2007; Prasad et al., 2003; Sourdille et al., 2003). Most of these QTL were not consistent across environments due to QTL ´ QTL epistatic and/or QTL ´ environment interactions. In our study, grain protein content QTL on chromosomes 3B, 5B, 6A, and 6B were also mapped only in 1 yr and/or management system, most probably due to QTL by environment and/or management interaction for grain protein content. Earliness is a desirable trait in western Canada where wheat breeding programs generally exist in short growing season environments (Randhawa et al., 2013). We mapped one major QTL for days to flowering and three for days to maturity in this study. Most of the QTL were mapped on Chromosome 5B at 49 to 54.7 cM positions. The days to flowering QTL mapped during 2008 in organic and conventional management system explained 32.8% phenotypic variation in days to flowering. We mapped a days to maturity QTL, Qmat.dms-5B, to a similar region (53.7 cM) during 2008 in the organic system affecting maturity by 6.6 d. These results suggest that chromosome 5B contains an important genomic region (49 to 54.7 cM) influencing flowering and maturity time in wheat. The additive effects of flowering and maturity time QTL identified in this study ranged from –6.66 to 2.2 d. The flowering time QTL on chromosome 5B behaved in a similar manner during 2008 in both management systems and its earliness allele reduced flowering time in both systems (–2.67 and –3.18 d in organic and conventional systems, respectively). On the contrary, flowering time QTL on 5B chromosome linked to wPt-3569 during 2010 accelerated flowering by 2.12 d in the organic system but delayed flowering by 2.2 d in the conventional system. The phenotypic variation explained by this QTL and its map position also varied between the two systems. These may or may not be the same QTL. The difference between the map positions may be due to the fact that map positions are point estimates that should have sampling errors, which are not given by mapping software. Therefore, the difference may not be statistically different. Various studies have reported genotype ´ environment interaction for various agronomic and quality traits in spring wheat. This implies that some genes act differently in different environments. These environmentally dependent gene and QTL effects can be of special interest to wheat breeders. The detection and mapping of such
www.crops.org 11
genes and QTL may have practical implications for wheat breeding programs aiming to develop cultivars specifically for organically managed lands. In the present study, most of the detected QTL were specific to either the conventional or the organic management system with respect to their additive effects and phenotypic variation, suggesting that QTL express differently in different environments. Therefore, spring wheat breeders should test and select genotypes under environmental conditions which are comparable with target environments.
CONCLUSIONS Results of this study suggest that selection, heritability, and correlation differences exist between management systems for various agronomic and quality traits. Under conventional management, no differences except plant height were found between RILs with wild and mutant type alleles of the locus Rht-B1; however, in organic systems, genotypes carrying the Rht-B1b (mutant) had reduced plant height, higher weed biomass, and lower grain yield than in conventional systems. We found various QTL for agronomic and quality traits. The consistent QTL detected across management systems for grain yield, grain volume weight, kernel weight, and days to flowering on chromosomes 6A, 1B, 3A, and 5B, respectively, can be further used in marker assisted breeding through fine mapping the specific regions.
Supplemental Information Available Supplemental material is available with the online version of this article. Acknowledgments The authors would like to acknowledge Klaus Strenzke, Kelley Dunfield, Glen Hawkins, Lisa Raatz, Fabiana Dias, Alex Pswarayi, Joe Back, Ivan Adamyk, Henry Song, Graham Collier, Hua Chen, Neshat Pazooki, and Rachelle Rimmer for technical assistance. This research was supported by grants to the University of Alberta wheat breeding program from the Alberta Crop Industry Development Fund, Organic Agriculture Cluster grant from Agriculture and Agri-Food Canada, Western Grains Research Foundation Endowment Fund, and an NSERC Collaborative Grant to D. Spaner. This work was conducted in part within the project “Canadian Triticum Advancement Through Genomics (CTAG).” We would like to acknowledge CTAG funding provided by the Saskatchewan Ministry of Agriculture, Western Grains Research Foundation, Agriculture and Agri-Food Canada, Genome Canada, Genome Prairie, Genome Alberta and Alberta Innovates. The study was also supported by Canadian Wheat Board fellowships (CWB) to the first (Asif ) and fifth (Atif ) author and an NSERC Discovery grant to D. Spaner.
References AAFRD. 2004. Agron JourSoil group map of Alberta. Available at www1.agric.gov.ab.ca/soils/soils.nsf/soilgroupmap?readform (accessed 8 Dec. 2013; verified 17 Feb. 2015). Alberta Agriculture,
12
Food, and Rural Development (AAFRD), Lacombe, AB, Canada. Addisu, M., J.W. Snape, J.R. Simmonds, and M.J. Gooding. 2010. Effects of reduced height (Rht) and photoperiod insensitivity (Ppd) alleles on yield of wheat in contrasting production systems. Euphytica 172:169–181. doi:10.1007/s10681-009-0025-2 Akbari, M., P. Wenzl, V. Caig, J. Carling, L. Xia, S. Yang, G. Uszynski, V. Mohler, A. Lehmensiek, H. Kuchel, M. Hayden, N. Howes, P. Sharp, P. Vaughan, B. Rathmell, E. Huttner, and A. Kilian. 2006. Diversity arrays technology (DArT) for high-throughput profiling of the hexaploid wheat genome. Theor. Appl. Genet. 113:1409–1420. doi:10.1007/s00122-006-0365-4 Asif, M., H.S. Randhawa, R.C. Yang, A. Navabi, and D. Spaner. 2012. Agronomic evaluation of a wheat breeding population in weed competitive organic and weed free conventional management systems. Can. J. Plant Sci. 92:608–609. Atlin, G.N., and K.J. Frey. 1990. Selecting oat lines for yield in lowproductivity environments. Crop Sci. 30:556–561. doi:10.2135/ cropsci1990.0011183X003000030017x Baenziger, P.S., I. Dweikat, K. Gill, K. Eskridge, T. Berke, M. Shah, B.T. Campbell, M.L. Ali, N. Mengistu, A. Mahmood, A. Auvuchanon, Y. Yen, S. Rustgi, B. Moreno-Sevilla, A. Mujeeb-Kazi, and M.R. Morris. 2011. Understanding grain yield: Its a journey, not a destination. Czech J. Genet. Plant 47:S77–S84. Bai, G.H., M.K. Das, B.F. Carver, X.Y. Xu, and E.G. Krenzer. 2004. Covariation for microsatellite marker alleles associated with Rht8 and coleoptile length in winter wheat. Crop Sci. 44:1187–1194. doi:10.2135/cropsci2004.1187 Baresel, J.P., and H.J. Reents. 2006. Observations on long-term wheat variety trials under organic and conventional conditions in Germany. In: H. Østergard and L. Fontaine, editors, Proceedings of the COST SUSVAR workshop on cereal crop diversity: Implications for production and products. ITAB Press, Paris. Bennett, D., M. Reynolds, D. Mullan, A. Izanloo, H. Kuchel, P. Langridge, and T. Schnurbusch. 2012. Detection of two major grain yield QTL in bread wheat (Triticum aestivum L.) under heat, drought and high yield potential environments. Theor. Appl. Genet. 125:1473–1485. doi:10.1007/s00122-012-1927-2 Beres, B.L., K.N. Harker, G.W. Clayton, R.E. Blackshaw, and R.J. Graf. 2010. Weed competitive ability of spring and winter cereals in the Northern Great Plains. Weed Technol. 24:108–116. doi:10.1614/WT-D-09-00036.1 Blanco, A., R. Simeone, and A. Gadaleta. 2006. Detection of QTLs for grain protein content in durum wheat. Theor. Appl. Genet. 113:563–565. doi:10.1007/s00122-006-0313-3 Bonneau, J., J. Taylor, B. Parent, D. Bennett, M. Reynolds, C. Feuillet, P. Langridge, and D. Mather. 2013. Multi-environment analysis and improved mapping of a yield-related QTL on chromosome 3B of wheat. Theor. Appl. Genet. 126:747–761. doi:10.1007/ s00122-012-2015-3 Botwright, T., G. Rebetzke, T. Condon, and R. Richards. 2001. The effect of rht genotype and temperature on coleoptile growth and dry matter partitioning in young wheat seedlings. Aust. J. Plant Physiol. 28:417–423 10.1071/PP01010. Burger, H., M. Schloen, W. Schmidt, and H.H. Geiger. 2008. Quantitative genetic studies on breeding maize for adaptation to organic farming. Euphytica 163:501–510. doi:10.1007/s10681-008-9723-4 Canadian Wheat Board. 2010. Variety survey report. CWB, Winnipeg, MB, Canada. Ceccarelli, S., and S. Grando. 1991. Environment of selection and type of germplasm in barley breeding for low-yielding conditions. Euphytica 57:207–219. doi:10.1007/BF00039667 Churchill, G.A., and R.W. Doerge. 1994. Empirical threshold values for quantitative trait mapping. Genetics 138:963–971.
www.crops.org
crop science, vol. 55, may– june 2015
Coleman, R.D., G.S. Gill, and G.J. Rebetzke. 2001. Identification of quantitative trait loci for traits conferring weed competitiveness in wheat (Triticum aestivum L.). Aust. J. Agric. Res. 52:1235–1246. doi:10.1071/AR01055 Cousens, R.D., A.G. Barnett, and G.C. Barry. 2003a. Dynamics of competition between wheat and oat: I. Effects of changing the timing of phenological events. Agron. J. 95:1295–1304. doi:10.2134/ agronj2003.1295 Cousens, R.D., G.J. Rebetzke, and A.G. Barnett. 2003b. Dynamics of competition between wheat and oat: II. Effects of dwarfing genes. Agron. J. 95:1305–1313. doi:10.2134/agronj2003.1305 Cudney, D.W., L.S. Jordan, and A.E. Hall. 1991. Effect of wild oat (Avena fatua) infestations on light interception and growth-rate of wheat (Triticum aestivum). Weed Sci. 39:175–179. Cui, F., C. Zhao, A. Ding, J. Li, L. Wang, X. Li, Y. Bao, J. Li, and H. Wang. 2014. Construction of an integrative linkage map and QTL mapping of grain yield-related traits using three related wheat RIL populations. Theor. Appl. Genet. 127:659–675. doi:10.1007/ s00122-013-2249-8 Ellis, M.H., G.J. Rebetzke, P. Chandler, D. Bonnett, W. Spielmeyer, and R.A. Richards. 2004. The effect of different height reducing genes on the early growth of wheat. Funct. Plant Biol. 31:583–589. doi:10.1071/FP03207 Ellis, M., W. Spielmeyer, K. Gale, G. Rebetzke, and R. Richards. 2002. “Perfect” markers for the Rht-B1b and Rht-D1b dwarfing genes in wheat. Theor. Appl. Genet. 105:1038–1042. doi:10.1007/ s00122-002-1048-4 Elouafi, I., and M.M. Nachit. 2004. A genetic linkage map of the Durum × Triticum dicoccoides backcross population based on SSRs and AFLP markers, and QTL analysis for milling traits. Theor. Appl. Genet. 108:401–413. doi:10.1007/s00122-003-1440-8 Fisher, R.A. 1925. Theory of statistical estimation. Proc. Camb. Philos. Soc. 22:700–725. doi:10.1017/S0305004100009580 Frick, B. 1998. Weed management for organic producers: Literature Search. Available at http://saskorganic.com/sites/saskorganic. com/files/Weed_Management_for_Organic_producers.pdf (verified 17 Feb. 2015). Agricultural and Agri-Food Canada, Saskatoon Research Centre, Saskatoon, SK, Canada. Gianessi, L.P., and N.P. Reigner. 2007. The value of herbicides in U.S. crop production. Weed Technol. 21:559–566. doi:10.1614/ WT-06-130.1 Gooding, M.J., M. Addisu, R.K. Uppal, J.W. Snape, and H.E. Jones. 2012. Effect of wheat dwarfing genes on nitrogen-use efficiency. J. Agric. Sci. 150:3–22. doi:10.1017/S0021859611000414 Gooding, M.J., N.D. Cosser, A.J. Thompson, W.P. Davies, and R.J. Froud-Williams. 1997. The effect of cultivar and Rht genes on the competitive ability, yield, and bread-making qualities of organically grown wheat. Proc. of the 3rd ENOF workshop: Resource use in organic farming. Ancona, Italy. p. 113–121. Gooding, M.J., A.J. Thompson, and W.P. Davies. 1993. Interception of photosynthetically active radiation, competitive ability and yield of organically grown wheat varieties. Aspects Appl. Biol. 34:355–362. Grime, J.P. 1979. Plant strategy and vegetation processes. Wiley, Chichester, UK. Groos, C., N. Robert, E. Bervas, and G. Charmet. 2003. Genetic analysis of grain protein-content, grain yield and thousand-kernel weight in bread wheat. Theor. Appl. Genet. 106:1032–1040. doi:10.1007/s00122-002-1111-1 Heidari, B., B.E. Sayed-Tabatabaei, G. Saeidi, M. Kearsey, and K. Suenaga. 2011. Mapping QTL for grain yield, yield components, and spike features in a doubled haploid population of bread wheat. Genome 54:517–527. doi:10.1139/g11-017 crop science, vol. 55, may– june 2015
Hill, J., H.C. Becker, and P.M.A. Tigerstedt. 1998. Quantitative and ecological aspects of plant breeding. Chapman and Hall, London. Hiltbrunner, J., L. Bloch, W. Richner, and B. Streit. 2002. Development of weed populations with no-tillage of winter wheat into living mulches. J. Plant Dis. Prot. Special Issue XVIII:581–587. Hiltbrunner, J., M. Liedgens, L. Bloch, P. Stamp, and B. Streit. 2007. Legume cover crops as living mulches for winter wheat: Components of biomass and the control of weeds. Eur. J. Agron. 26:21–29. doi:10.1016/j.eja.2006.08.002 Holland, J.B. 2006. Estimating genotypic correlations and their standard errors using multivariate restricted maximum likelihood estimation with SAS Proc MIXED. Crop Sci. 46:642–654. doi:10.2135/cropsci2005.0191 Huang, X.Q., S. Cloutier, L. Lycar, N. Radovanovic, D.G. Humphreys, J.S. Noll, D.J. Somers, and P.D. Brown. 2006. Molecular detection of QTLs for agronomic and quality traits in a doubled haploid population derived from two Canadian wheats (Triticum aestivum L.). Theor. Appl. Genet. 113:753–766. doi:10.1007/ s00122-006-0346-7 Huel, D.G., and P. Hucl. 1996. Genotypic variation for competitive ability in spring wheat. Plant Breed. 115:325–329. doi:10.1111/j.1439-0523.1996.tb00927.x Kamran, A., M. Iqbal, A. Navabi, H. Randhawa, C. Pozniak, and D. Spaner. 2013. Earliness per se QTLs and their interaction with the photoperiod insensitive allele Ppd-D1a in the Cutler × AC Barrie spring wheat population. Theor. Appl. Genet. 126:1965–1976. doi:10.1007/s00122-013-2110-0 Knezevic, S.Z. 2009. Flaming: A new weed control tool in organic crops. Crop Watch, 19 June 2009. Available at http://cropwatch.unl.edu/archive/-/asset_publisher/VHeSpfv0Agju/content/886969 (verified 17 Feb. 2015). University of Nebraska-Lincoln Extension. Kruidhof, H.M., L. Bastiaans, and M.J. Kropff. 2008. Ecological weed management by cover cropping: Effects on weed growth in autumn and weed establishment in spring. Weed Res. 48:492– 502. doi:10.1111/j.1365-3180.2008.00665.x Kunert, A., A. Naz, O. Dedeck, K. Pillen, and J. Léon. 2007. ABQTL analysis in winter wheat: I. Synthetic hexaploid wheat (T. turgidum ssp. dicoccoides ´ T. tauschii) as a source of favourable alleles for milling and baking quality traits. Theor. Appl. Genet. 115:683–695. doi:10.1007/s00122-007-0600-7 Lemerle, D., R. Cousens, B. Verbeek, and J.A. Fisher. 1994. The potential to breed more competitive wheat cultivars. Proc. of the 7th Assembly of the Wheat Breeding Society of Australia. Wheat Breeding Society, Adelaide, Australia. p. 447–450. Lemerle, D., B. Verbeek, and N. Coombes. 1995. Losses in grain yield of winter crops from Lolium rigidum competition depend on crop species, cultivar and season. Weed Res. 35:503–509. doi:10.1111/j.1365-3180.1995.tb01648.x Lemerle, D., B. Verbeek, R.D. Cousens, and N.E. Coombes. 1996a. The potential for selecting wheat varieties strongly competitive against weeds. Weed Res. 36:505–513. doi:10.1111/j.1365-3180.1996. tb01679.x Lemerle, D., B. Verbeek, and P. Martin. 1996b. Breeding wheat cultivars more competitive against weeds. In: H. Brown et al., editors, 2nd Int. Weed Control Congress—Proceedings. 1–4:1323–1324. Mandal, N.P., P.K. Sinha, M. Variar, V.D. Shukla, P. Perraju, A. Mehta, A.R. Pathak, J.L. Dwivedi, S.P.S. Rathi, S. Bhandarkar, B.N. Singh, D.N. Singh, S. Panda, N.C. Mishra, Y.V. Singh, R. Pandya, M.K. Singh, R.B.S. Sanger, J.C. Bhatt, R.K. Sharma, A. Raman, A. Kumar, and G. Atlin. 2010. Implications of genotype ´ input interactions in breeding superior genotypes for favorable
www.crops.org 13
and unfavorable rainfed upland environments. Field Crops Res. 118:135–144. doi:10.1016/j.fcr.2010.05.004 Mason, H., B. Frick, J. O’Donovan, A. Navabi, and D. Spaner. 2006. Competitive ability of historical and modern Canadian western red spring (CWRS) wheat varieties under conventional and organic management. Can. J. Plant Sci. 86:191–191. doi:10.4141/P05-051 Mason, H., A. Navabi, B. Frick, J. O’Donovan, and D. Spaner. 2007a. Cultivar and seeding rate effects on the competitive ability of spring cereals grown under organic production in northern Canada. Agron. J. 99:1199–1207. doi:10.2134/agronj2006.0262 Mason, H.E., A. Navabi, B.L. Frick, J.T. O’Donovan, and D.M. Spaner. 2007b. The weed-competitive ability of Canada western red spring wheat cultivars grown under organic management. Crop Sci. 47:1167–1176. doi:10.2135/cropsci2006.09.0566 McCartney, C.A., D.J. Somers, D.G. Humphreys, O. Lukow, N. Ames, J. Noll, S. Cloutier, and B.D. McCallum. 2005. Mapping quantitative trait loci controlling agronomic traits in the spring wheat cross RL4452 ´ AC Domain. Genome 48:870–883. doi:10.1139/g05-055 McIntosh, R.A., K.M. Devos, J. Dubcovsky, C.F. Morris, and W.J. Rogers. 2003. Catalogue of gene symbols for wheat. Available at http://wheat.pw.usda.gov/ggpages/wgc/2003/GeneSymbol.html (accessed 6 June 2012; verified 17 Feb. 2015). GrainGenes, ARSUSDA, Washington, D.C. Mokhtari, S., N.W. Galwey, R.D. Cousens, and N. Thurling. 2002. The genetic basis of variation among wheat F3 lines in tolerance to competition by ryegrass (Lolium rigidum). Euphytica 124:355–364. doi:10.1023/A:1015752021568 Murphy, K.M., J.C. Dawson, and S.S. Jones. 2008. Relationship among phenotypic growth traits, yield and weed suppression in spring wheat landraces and modern cultivars. Field Crops Res. 105:107–115. doi:10.1016/j.fcr.2007.08.004 Narasimhamoorthy, B., B.S. Gill, A.K. Fritz, J.C. Nelson, and G.L. Brown-Guedira. 2006. Advanced backcross QTL analysis of a hard winter wheat ´ synthetic wheat population. Theor. Appl. Genet. 112:787–796. doi:10.1007/s00122-005-0159-0 Prasad, M., N. Kumar, P. Kulwal, M. Röder, H. Balyan, H. Dhaliwal, and P. Gupta. 2003. QTL analysis for grain protein content using SSR markers and validation studies using NILs in bread wheat. Theor. Appl. Genet. 106:659–667. doi:10.1007/s00122-002-1114-y. Randhawa, H.S., M. Asif, C. Pozniak, J.M. Clarke, R.J. Graf, S.L. Fox, D.G. Humphreys, R.E. Knox, R.M. DePauw, A.K. Singh, R.D. Cuthbert, P. Hucl, and D. Spaner. 2013. Application of molecular markers to wheat breeding in Canada. Plant Breed. 132:458–471. doi:10.1111/pbr.12057 Reid, T.A., A. Navabi, J.C. Cahill, D. Salmon, and D. Spaner. 2009a. A genetic analysis of weed competitive ability in spring wheat. Can. J. Plant Sci. 89:591–599 10.4141/CJPS08105. doi:10.4141/CJPS08105 Reid, T.A., R.C. Yang, D.F. Salmon, A. Navabi, and D. Spaner. 2011. Realized gains from selection for spring wheat grain yield are different in conventional and organically managed systems. Euphytica 177:253–266. doi:10.1007/s10681-010-0257-1 Reid, T.A., R.C. Yang, D.F. Salmon, and D. Spaner. 2009b. Should spring wheat breeding for organically managed systems be conducted on organically managed land? Euphytica 169:239–252. doi:10.1007/s10681-009-9949-9 Richards, R.A. 1992. The effect of dwarfing genes in spring wheat in dry environments. 1. Agronomic characteristics. Aust. J. Agric. Res. 43:517–527. doi:10.1071/AR9920517 Rosewarne, G.M., R.P. Singh, J. Huerta-Espino, and G.J. Rebetzke. 2008. Quantitative trait loci for slow-rusting resistance in wheat to leaf rust and stripe rust identified with multi-environment
14
analysis. Theor. Appl. Genet. 116:1027–1034. doi:10.1007/ s00122-008-0736-0 Ryan, J., S. Masri, S. Ceccarelli, S. Grando, and H. Ibrikci. 2008. Differential responses of barley landraces and improved barley cultivars to nitrogen-phosphorus fertilizer. J. Plant Nutr. 31:381–393. doi:10.1080/01904160801894939 SAS. 2008. Release 9.2. SAS Institute Inc., Cary, NC. Sourdille, P., T. Cadalen, H. Guyomarc’h, J.W. Snape, M.R. Perretant, G. Charmet, C. Boeuf, S. Bernard, and M. Bernard. 2003. An update of the Courtot x Chinese Spring intervarietal molecular marker linkage map for the QTL detection of agronomic traits in wheat. Theor. Appl. Genet. 106:530–538 10.1007/s00122–002–1044–8. Spielmeyer, W., J. Hyles, P. Joaquim, F. Azanza, D. Bonnett, M.E. Ellis, C. Moore, and R.A. Richards. 2007. A QTL on chromosome 6A in bread wheat (Triticum aestivum) is associated with longer coleoptiles, greater seedling vigour and final plant height. Theor. Appl. Genet. 115:59–66. doi:10.1007/s00122-007-0540-2 Steege, T.M.W., F.M. den Ouden, H. Lambers, P. Stam, and A.J.M. Peeters. 2005. Genetic and physiological architecture of early Vigor in Aegilops tauschii, the D-genome donor of hexaploid wheat. A quantitative trait loci analysis. Plant Physiol. 139:1078–1094. doi:10.1104/pp.105.063263 Steel, R.G.D., J.H. Torrie, and D.A. Dickey. 1997. Principles and procedures of statistics: A biometrical approach. McGraw-Hill Book Co., NY. Steiger, D.K., E.M. Elias, and R.G. Cantrell. 1996. Evaluation of lines derived from wild emmer chromosome substitutions. 1. Quality traits. Crop Sci. 36:223–227. doi:10.2135/cropsci1996.0011183X0 03600020001x Thomas, J.B., G.B. Schaalje, and M.N. Grant. 1994. Height, competition and yield potential in winter wheat. Euphytica 74:9–17. doi:10.1007/BF00033761 Wang, S., C.J. Basten, and Z.B. Zeng. 2010. Windows QTL cartographer 2.5. Available at http://statgen.ncsu.edu/qtlcart/WQTLCart. htm (verified 17 Feb. 2015). Department of Statistics, North Carolina State University, Raleigh. Watson, P.R., D.A. Derksen, and R.C. Van Acker. 2006. The ability of 29 barley cultivars to compete and withstand competition. Weed Sci. 54:783–792. doi:10.1614/WS-05-020R3.1 Wolfe, M.S., J.P. Baresel, D. Desclaux, I. Goldringer, S. Hoad, G. Kovacs, F. Löschenberger, T. Miedaner, H. Østergård, and E.T. Lammerts van Bueren. 2008. Developments in breeding cereals for organic agriculture. Euphytica 163:323–346. doi:10.1007/ s10681-008-9690-9 Wszelaki, A.L., D.J. Doohan, and A. Alexandrou. 2007. Weed control and crop quality in cabbage (Brassica oleracea (capitata group)) and tomato (Lycopersicon lycopersicum) using a propane flamer. Crop Prot. 26:134–144. doi:10.1016/j.cropro.2006.04.012 Yang, R.C. 2010. Towards understanding and use of mixed-model analysis of agricultural experiments. Can. J. Plant Sci. 90:605–627. doi:10.4141/CJPS10049 Zerner, M.C., G.S. Gill, and R.K. Vandeleur. 2008. Effect of height on the competitive ability of wheat with oats. Agron. J. 100:1729– 1734. doi:10.2134/agronj2008.0068 Zhang, W., S. Chao, F. Manthey, O. Chicaiza, J.C. Brevis, V. Echenique, and J. Dubcovsky. 2008. QTL analysis of pasta quality using a composite microsatellite and SNP map of durum wheat. Theor. Appl. Genet. 117:1361–1377. doi:10.1007/s00122-008-0869-1
www.crops.org
crop science, vol. 55, may– june 2015
