1942 www.crops.org crop science, vol. 58, september–october 2018

RESEARCH

Significant proportions of the world’s population suffer from deficiencies of Fe, Zn, or both (FAO, 2016). This situation is

attributed to consumption of staple crops with low tissue mineral concentrations derived from crop genetic backgrounds and production in areas with low essential mineral availability (White and Broadley, 2009). As one of these staple crops, common wheat (Triticum aestivum L.) supplies 20% of the daily protein and calories for almost 4.5 billion people. Thus, it plays an important role in resolving human malnutrition (GCARD, 2012).

Although Fe fortification of wheat flour is widely recommended to relieve nutrient deficiencies, this fortification does not address the Zn issue. Furthermore, centralized fortification strategies in many African and Asian countries are heavily reliant on local food and distribution systems. In addition, rural and peri-urban residents often mill their own cereals instead of purchasing processed ingredients or ready-made cereal foods (Mildon et al., 2015). For these reasons, biofortification of crops through the development of new cultivars with an increased ability to acquire or partition Fe and other mineral

Biofortification of Hard Red Winter Wheat by Genes Conditioning Low Phytate and High

Grain Protein Concentration

Jorge P. Venegas, Robert A. Graybosch,* Brian Wienhold, Devin J. Rose, Brian M. Waters, P. Stephen Baenziger, Kent Eskridge, Guihua Bai, and Paul St. Amand

ABSTRACTRecombinant inbred lines (RILs) of winter wheat (Triticum aestivum L.) were used to determine whether the combination of low grain phytate (LPA) conditioned by lpa1-1 and Gpc-B1 (where GPC stands for grain protein content) alleles would simultaneously increase beneficial mineral concentrations and grain protein without pleiotropic effects on grain yield. Four different genotypes (LPA-GPC, LPA-wild type [WT], WT-GPC, or WT-WT) were used as treatments in field experiments in Nebraska. Genotypic effects on senescence, grain yield, grain volume weight, grain protein, Fe, Zn, and other mineral grain concentrations were determined. Low grain phytate alone and in combination with GPC increased dialyzed Zn, Ca, and Mn. Gpc-B1 had a slight effect on grain protein concentration in the tested genetic backgrounds and environ-ments. The combination of LPA and GPC did not lower grain yield, grain protein, or total grain Fe and Zn concentrations. However, the LPA-GPC combination significantly reduced grain volume weight. The LPA allele alone reduced grain protein concentration. Introgression of lpa1-1 alleles into adapted Great Plains winter wheat materials can improve dialyzed Zn, Ca, and Mn concentrations without reducing grain yield and, coupled with introgression of Gpc-B1, provide more nutritious wheat kernels.

J.P. Venegas, B.M. Waters, and P.S. Baenziger, Dep. of Agronomy and Horticulture, Univ. of Nebraska, Lincoln, NE 68583; R.A. Graybosch and B. Wienhold, USDA-ARS, Lincoln, NE 68583; K. Eskridge, Dep. of Statistics, Univ. of Nebraska, Lincoln, NE 68583; D.J. Rose, Dep. of Food Science and Agronomy and Horticulture, Univ. of Nebraska, Lincoln, NE 68583; G. Bai and P. St. Amand, USDA-ARS, Manhattan, KS, 66502. Received 13 Mar. 2018. Accepted 4 June 2018. *Corresponding author (bob.graybosch@ars.usda.gov). Assigned to Associate Editor Adam Heuberger.

Abbreviations: ANCOVA, analysis of covariance; DTPA, diethylenetriaminepentaacetic acid; G ´ E, genotype ´ environment; GPC, grain protein content; HIP, high inorganic phosphate; ICAP, inductively coupled argon cooled plasma spectrometer; LPA, low grain phytate; Pi, inorganic phosphate; RED, rapid equilibrium dialysis; RIL, recombinant inbred line; WT, wild type.

Published in Crop Sci. 58:1942–1953 (2018). doi: 10.2135/cropsci2018.03.0175 © Crop Science Society of America | 5585 Guilford Rd., Madison, WI 53711 USA All rights reserved.

Published July 26, 2018

crop science, vol. 58, september–october 2018 www.crops.org 1943

elements into the seed is advocated as a strategy to solve malnutrition in the developing world.

To make mineral elements of biofortified crops avail-able to humans, various strategies need to be pursued. Increasing the concentration of factors that stimulate the absorption of minerals by digestion, and reducing the concentration of antinutrients such as phytic acid or phytate (IP6), which interferes with mineral absorption, can increase nutrient bioavailability to humans (White and Broadley, 2009). Phytic acid sequesters inorganic phosphate (Pi) and divalent elements such as Ca, Fe, and Zn, making them less digestible in monogastric animals including humans. Additionally, monogastric livestock production (i.e., poultry, swine, and fish) produces waste containing high concentrations of phytate-phosphate, which contributes to water pollution (Raboy, 2007).

Natural genetic variation for wheat grain and flour Fe content is limited and heritability is low (Oury et al., 2006; Guttieri et al., 2015). In a study using historical French landraces, materials from worldwide germplasm collections, and elite breeding lines in different envi-ronments, genotypic effects for Fe were not significant, genotype ´ environment (G ´ E) interactions were very high, and Fe was poorly correlated to other mineral concentrations (Oury et al., 2006). Similar results were found by Guttieri et al. (2015) in a panel of 299 winter wheat genotypes evaluated in different environments. Consequently, breeding for high Fe concentration will probably “prove illusory” (Oury et al., 2006). It may be more advantageous if breeders attempt the introduction of genes from outside the primary gene pool of common wheat as a means of improving Fe concentration. Triticum turgidum L. var. dicoccoides, a wild wheat relative, has been identified as a possible source of genes for higher Fe concentrations (Chatzav et al., 2010). Gpc-B1, a T. turgidum var. dicoccoides-derived gene encoding a sequence-specific DNA-binding protein (NAM-B1), has been associated with increased grain protein and elevated Zn and Fe concentrations without negative effects on grain yield (Uauy et al., 2006b).

As in other grains, Fe and Zn bioavailability from wheat grain is inhibited by the presence of phytates (Hallberg, 1987). Therefore, attempts to elevate Fe and Zn concentrations should be made in low-phytate (LPA) genetic backgrounds. A LPA mutant in wheat (lpa1-1) is available (Guttieri et al., 2004), but the mutation has been associated with reduced grain yield (Raboy, 2009). However, Guttieri et al. (2006) determined that such effects in wheat were dependent on environmental yield potential, were not evident in lower yielding environ-ments, and were modified by the genetic background (Guttieri et al., 2006).

Breeding efforts to deploy LPA mutants in wheat in typically lower yielding Great Plains environments have

not been attempted, and little work on developing LPA wheat cultivars is known. In and of itself, deployment of lpa1-1 could be beneficial to human nutrition, as the presence of phytate in the gut reduces mineral absorption from all consumed foods (Raboy, 2009). Incorporation of Gpc-B1 in LPA backgrounds also has not been eval-uated. This high-protein and LPA combination could have additional beneficial effects of enhanced Fe and Zn bioavailability and reduce potential environmental impacts by lowering the P content of manure from monogastric animals fed with wheat. Released germplasm combining Gpc-B1 and LPA traits would be beneficial in wheat breeding programs with a grain biofortification goal.

For this study, winter wheat recombinant inbred lines (RILs) were used to determine whether combining LPA and Gpc-B1 (GPC) traits would increase grain protein and mineral element concentrations and dialyzability without pleiotropic effects on grain yield. Element dialyz-ability is a means of predicting bioavailability of mineral elements when more sophisticated direct absorption assays are impractical (Etcheverry et al., 2012). Our objectives were to assess the effects of LPA and GPC traits on (i) total and dialyzable grain mineral element concentrations and (ii) grain yield and quality traits such as grain protein concentration and grain volume weight, using RILs of a biparental population planted at four Nebraska locations across two growing seasons.

MATERIALS AND METHODSPlant MaterialsRecombinant inbred lines were developed from a single cross between ORACG:0019 (GPC donor parent) and AO2568WS-A-12-10 (LPA donor parent). ORACG:0019 was derived from OR943576/‘Langdon’ (DIC-6B)//*3 OR943576. OR943576 is a hard winter wheat breeding line developed by Oregon State University from the pedigree ‘Mildress’/CI14482//‘Yamhill’/‘Hyslop’/3/‘Rondezvous’(Morris et al., 2009). A02568WS-A-12-10 was derived from ‘Grandin’*3/Js-12lpa//’Boundary’. Js-12lpa is a soft white spring wheat with the LPA mutation. After the F4 generation, all the RILs were classified as either wild type (WT) or LPA using the high Pi (HIP) protocol (Guttieri et al., 2004), an indirect colorimetric phenotypic assay of grain phytate levels. Homozygous LPA and WT RILs also were scored as either WT or Gpc-B1 positive via DNA poly-merase chain reaction (PCR) using the sequence-tagged site (STS) marker HGPC/Yr36 (Distelfeld et al., 2006). Thirty-nine RILs were then classified as one of the following homozygous genotypes: LPA-GPC, LPA-WT, WT-GPC, or WT-WT; each genotype was used as a treatment in field experiments. At least seven RILs of each genotype were used. Seed of all 39 RILs has been deposited in the USDA-ARS National Small Grains Collection, Aberdeen, ID, under Plant Introduction numbers PI 682676 through PI 682714.

1944 www.crops.org crop science, vol. 58, september–october 2018

weight apparatus. Protein concentration was determined using a Perten DA7250 near-infrared reflectance spectrometer (Perten Instrument) calibrated to combustion analysis (LECO FP528, LECO Corporation).

Micronutrient, Inorganic Phosphate, and Bioavailability AnalysisApproximately 50 g of dried, whole kernels was weighed and a subsample was analyzed for grain micronutrient concentra-tion by Ward Laboratories. Grain samples were digested with 3 mL of concentrated HCl and 6 mL of concentrated HNO3 at 90°C for 105 min. Hydrogen peroxide (1 mL, 300 mL L−1) was added and samples were incubated at 140°C for 15 min. The same volume of H2O2 and temperature and incubation time procedure was repeated two more times. Samples were cooled and resuspended in 50 mL of 20 mL L−1 HNO3. Samples were filtered and analyzed using the ICAP instrument.

To determine grain Pi and grain bioaccessibility, ?1.5 g of whole kernels per sample (?45 seeds) was weighed and ground by impact using three 6.35-mm stainless steel balls per sample with a Geno/Grinder (SPEX SamplePrep). Three grinding repetitions at 1500 rpm for 3 min each were used to obtain homogeneous flour particles <100 mm. Forty milligrams per sample of the resulting flour was used to analyze Pi concentra-tions using the HIP protocol (Guttieri et al., 2004). Inorganic phosphate concentration data were only generated in 2016.

Potential grain bioaccessibility was estimated by dialysis using a modification of a previous method (Luten et al., 1996). The modified method used rapid equilibrium dialysis (RED) plates (Thermo Fisher Scientific) with a membrane pore size of 8K MWCO (molecular weight cutoff). This allowed high-throughput processing of samples. Twenty milligrams of flour was added into each sample chamber of a 48-well RED plate. One-hundred microliters of pepsin (Sigma, P7000) solution (100 mg mL−1 of 50 mM HCl) was added to the sample chamber. The RED plate was covered with sealing tape (ThermoScientific, 15036) and mixed at 125 rpm in an incubator at 37°C for 1 h. After this time, 258 mL of dialysis buffer (0.1 M NaHCO3) was added to each chamber of the 48-well RED plate, and the plate was mixed at 125 rpm in an incubator at 37°C for 55 min. The quantity of dialysis buffer added to the dialysis chamber amount was previously determined on a small subset of samples as the amount needed to reach optimal pH of 6.8. A pancreatin-bile solution was prepared by dissolving 0.4 g of pancreatin (Sigma, P7545) and 2.5 g of bile salts (Oxoid, LP0055) in 100 mL of 0.1 M NaHCO3. The solution was centrifuged at 4000 rpm for 1 min to remove clumpy threads that failed to dissolve. After centrifugation, 20 mL of the pancre-atin-bile solution was added into each sample chamber and the RED plate was mixed at 125 rpm in an incubator at 37°C for 2 h. Finally, the dialysis buffer was collected from the buffer chamber, and macro- and microelement (Li, B, Na, Mg, P, S, K, Ca, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, As, Se, Mo, and Cd) analysis was performed using the ICAP instrument (Agilent Technologies).

Statistical AnalysisAnalysis of variance was performed using PROC MIXED procedure in SAS 9.3 (SAS Institute, 2011). All variables were analyzed as a two-by-two factorial with four genotype

Field ExperimentsClassified F4:5 RILs of each genotype were planted at the University of Nebraska Experimental Field Stations in Havelock (40°51¢21.91¢¢ N, 96°36¢28.32¢¢ W) and Mead (41°9¢53.76¢¢ N, 96°24¢39.60¢¢ W) in eastern Nebraska and at Sidney (41°14¢40.33¢¢ N, 103°0¢11.18¢¢ W) and North Platte (41°2¢56.95¢¢ N, 100°44¢56.18¢¢ W) in western Nebraska in autumn 2015 and 2016, for harvest during 2016 and 2017 winter wheat production years. Growing conditions in eastern and western Nebraska were markedly different in precipitation. Eastern locations were more humid than western locations. The experiments were arranged in a randomized complete block design with three blocks, thirty-nine RILs, two parents, and nine Great Plains-adapted cultivars (‘Anton’, ‘Big Sky’, ‘Freeman’, ‘Intrada’, ‘Jerry’, ‘Millennium’, ‘Ruth’, ‘Settler CL’, and ‘Siouxland’) as controls. The experiments were sown in autumn and harvested in July the following year.

Agronomic ManagementThe fields were managed using the practices best suited for each location. In Mead, plots of each RIL were seeded at ?260 seeds m−2 in four 3.2-m rows spaced 30 cm apart. In Havelock, North Platte, and Sidney, plots of each genotype were seeded at ?210  seeds m−2 in five 3-m rows spaced 23 cm apart. Trials were grown under weed-free conditions. When necessary, the plots were maintained as disease free by fungicide applications. Fungicides metconazole {5-[(4-chloro-phenyl) methyl]-2,2-dimethyl-1-(1H-1,2,4-triazol-1-ylmethyl) cyclopentanol; Caramba, BASF} and pyraclostrobin (carbamic acid {2-[({[1-(4-chlorophenyl)-1H-pyrazol-3-yl]oxy}methyl)phenyl]methoxy-, methyl ester} plus metconazole; Twinline, BASF) were applied alternately during growing, flowering, and grain filling. Virus symptoms were not seen in the plots. Plots were machine harvested at maturity

PhenologySenescence was visually evaluated and recorded when 50% of the flag leaves or spikes in a plot had lost their green color.

Soil Sampling and Nutrient Plot EstimatesSoil macro- and micronutrients (N, P, K, S, Zn, Fe, Mn, Cu, Ca, Mg, Na, and B) and pH characteristics were assessed at ?50 sampling points per location and were recorded with a Trimble Geo GPS. Soil samples were collected at two depths (0–15 and 15–30 cm) per sampling point with a 2.54-cm-wide soil probe. Soil chemical analyses were obtained from Ward Labo-ratories, Kearney, NE. Ten grams of soil sample was mixed for 2 h with 20 mL of diethylenetriaminepentaacetic acid (DTPA) extracting solution. The soil DTPA solution was then filtered and centrifuged. The extracted solution was analyzed using an inductively coupled Ar cooled plasma spectrometer (ICAP) (Lindsay and Norvell, 1978). GPS coordinates and soil chemical results were used to estimate plot soil nutrient availability using the geostatistics software GS+ (Robertson, 2008).

Grain Evaluation and Grain Mineral AnalysisAfter harvest, grain samples were air cleaned and grain volume weight was measured using a USDA-approved test

crop science, vol. 58, september–october 2018 www.crops.org 1945

classifications (LPA-GPC, LPA-WT, WT-GPC, and WT-WT), nine controls, 39 RILs, four locations, and 2 yr. The numbers of RILs per genotypic group were eight for LPA-GPC, seven for LPA-WT, 12 for WT-GPC, and 12 for WT-WT. Blocks and RILs and were considered as random effects. Data from two locations (Lincoln and Sidney) in 2017 were not analyzed for grain total Fe concentration due to the inconsistency of data and presence of many outliers. Therefore, six locations were considered as environments for the grain total Fe variable, and RILs, blocks, and environments were considered as random effects. A gene main effect and gene ´ gene interaction analysis was performed using contrasts between the lpa1-1 and Gpc-B1 genes. Soil plot estimates were used as covariates in an analysis of covariance (ANCOVA) to evaluate any confounding effects of soil nutrients on the total grain nutrients. Total grain nutri-ents were also used as covariates for dialyzed nutrient data analysis. Genetic correlations and correlation heat maps among traits were analyzed using the combined least squares means of RILs for all the locations and years with the cor, reshape2 and ggplot2 functions in R package 3.3.0. (R Core Team, 2014).

Broad-sense heritability for all the variables was calculated on an entry-mean basis, adjusted using covariates, as follows (Guttieri et al., 2015):

´

s=

s ss + +

2g2

2 2(g env)2 error

g env (env)

H

r

where s2g is the genetic variance, s2

(g´env) is the G ´ E variance, env is the number of environments, s2

error is the error variance, and r(env) is the number of replications per environment. Genetic, G ´ E, and error variances were obtained in a combined analysis of the four locations and 2 yr by using ASREML package in R

(Butler et al., 2009). Control genotypes were evaluated as fixed effects, whereas eight environments (four locations ´ 2 yr) and 39 RILs were evaluated as random effects.

RESULTSGenotypic Group DifferencesNo significant differences among genotypic groups were detected in many of the minerals noted in the Materials and Methods. Results, therefore, will be limited to those minerals and traits demonstrating significant differ-ences or those most important to human nutrition or agronomy. There were no significant differences among the four genotypic groups for grain yield, grain total Fe, Zn, Ca, and K, and dialyzed Fe and P concentrations (Table 1, Fig. 1–3). There were significant differences among the genotypic groups, and between geno-typic groups and the controls for senescence (Table 1, Fig.  1b), grain volume weight, grain protein, grain total P, K, S, Mg, and Mn, and dialyzed Zn, Ca, Mn, and Cu concentrations (Table 1, Fig. 1–3). The lpa1-1 and Gpc-B1 allele combination significantly reduced grain volume weight and grain total Mg concentration (Fig. 1d and 2g). Therefore, the expression of these two alleles could affect grading, milling, or baking proper-ties when they are combined in at least some genetic backgrounds. The lpa1-1 allele alone (LPA-WT group) significantly reduced grain protein concentration and S concentration (Fig. 1c and 2f ). The lpa1-1 allele alone and in combination with the Gpc-B1 allele (LPA-GPC group) significantly increased dialyzed Zn, Ca, and Mn

Table 1. Analysis of variance components for yield, senescence, grain protein concentration, grain volume weight, and grain total and dialyzed mineral concentrations across genotypic groups (lpa1-1 [LPA]- Gpc-B1 [GPC], LPA-wild type [WT], WT-GPC, and WT-WT). Genotype ´ environment (G ´ E) interactions and heritabilities of all variables are reported.

F valueVariable Genotypic group G ´ E Heritability (H2)Yield (kg ha−1) 1.20ns† 2.41** 0.11Senescence (d) 6.05*** 1.03ns 0.84Grain protein conc. (g kg−1) 4.00** 3.32*** 0.76Grain volume weight (kg hL−1) 5.62*** 1.74ns 0.91Grain total mineral concentration (mg kg−1): Fe 3.66ns 0.68ns 0.03 Zn 1.47ns 2.36** 0.14 P 2.44* 2.13** 0.53 Mn 6.04*** 0.99ns 0.41 Mg 3.44** 1.86* 0.66 S 3.12** 2.76*** 0.76Dialyzed mineral concentration (mg kg−1): Fe 0.83ns 0.94ns 0.05 Zn 3.42* 1.17ns 0.63 Ca 6.43** 0.68ns 0.60 Mn 2.60* 0.96ns 0.69 Cu 2.82* 1.89* 0.17

* Significant at the 0.05 probability level.

** Significant at the 0.01 probability level.

*** Significant at the 0.001 probability level.

† ns, not significant.

1946 www.crops.org crop science, vol. 58, september–october 2018

Fig. 1. Effects of genotypic groups (lpa1-1 [LPA], Gpc-B1 [GPC], and wild type [WT]) on (a) yield, (b) yield plus soil nutrient estimates used as a covariate, (c) grain protein concentration, and (d) grain volume weight. Means followed by the same letter are not significantly different according to protected t test (a = 0.05).

Fig. 2. Effects of genotypic groups (lpa1-1 [LPA], Gpc-B1 [GPC], and wild type [WT]) on grain total (a) Fe, (b) Zn, (c) Ca, (d) P, (e) K, (f) S, (g) Mg, and (h) Mn. Means followed by the same letter are not significantly different according to protected t test (a = 0.05).

crop science, vol. 58, september–october 2018 www.crops.org 1947

(Fig. 3b, 3c, and 3e) and decreased grain total P, Mg, and Mn (Fig. 2d, 2g, and 2h).

Grain total Zn, Fe, Ca, and Mn were used as covari-ates for the dialyzed fraction of Zn, Fe, Ca, and Mn, and an ANCOVA analysis was performed. The lpa1-1 and Gpc-B1 combination and the lpa1-1 allele alone main-tained their positive effect on dialyzable Zn after the ANCOVA analysis. In general, ANOVA and ANCOVA statistical analyses provided identical results; however, significant differences between genotypes for grain total P were found when soil P estimates were used as a covariate (Fig. 2d). Overall, these results demonstrate that Gpc-B1 and lpa1-1 alleles alone or in combination have no effect on grain yield, grain total Fe, Zn, Ca, and K, and dialyzed Fe and P concentrations (Fig. 1a, 2a, 2b, 2c, 2e, 3a, and 3d). Mean grain yield of each RIL geno-typic group were significantly lower than the mean grain yield of the controls.

According to the gene main effect analysis, lpa1-1 had significant negative effects on grain protein and total P, S, Mg, and Mn and had significant positive effects on total Ca, Pi, and dialyzed Zn, Ca, and Mn. (Table 2). The Gpc-B1 gene had a significant negative effect on grain volume weight and a significant positive effect on grain protein concentration. The gene ´ gene inter-action analysis found a significant epistatic interaction between lpa1-1 and Gpc-B1 for Pi concentration in 2016 (Table 2).

Heritabilities and Genotype Environment Interaction for Agronomic and Quality TraitsGrain volume weight and senescence had the highest heritabilities among all traits, followed by grain protein concentration, grain total S concentration, and dialyzed Mn concentration (Table 1). Grain total P, Mn, and Mg concentra-tions and grain dialyzed Zn and Ca had medium heritability values. Grain yield, grain total Fe and Zn, and dialyzed Fe and Cu had low heritability values. There was a significant G ´ E effect for grain yield, grain protein concentration, grain Zn, P, Mg, and S, and dialyzed Cu concentration among the genotypic groups. Grain volume weight, senescence, grain total Fe and Mn, and dialyzed Fe, Zn, Ca, and Mn concentrations were not significantly influenced by G ´ E interaction effects (Table 1). These results indicate potential for progress through selection with LPA-GPC germplasm for most of the traits except grain yield, grain total Fe and Zn, and dialyzed Fe and Cu. Additional rounds of mating and intro-gression of LPA into diverse genetic backgrounds are needed to improve grain yield, grain total Fe and Zn concentrations, and dialyzed Fe and Cu concentrations. Alternatively, one can anticipate using LPA-GPC to positively advance some traits, without consequent negative effects on grain yield and grain total Fe and Zn concentrations.

Correlations between VariablesGenetic correlations were calculated using LSmeans of RILs to evaluate the association between variables

Fig. 3. Effects of genotypic groups (lpa1-1 [LPA], Gpc-B1 [GPC], and wild type [WT]) on dialyzed (a) Fe, (b) Zn, (c) Ca, (d) P, (e) Mn, and (f) Cu. Means followed by the same letter are not significantly different according to protected t test (a = 0.05).

1948 www.crops.org crop science, vol. 58, september–october 2018

(Fig. 4). Grain yield was negatively correlated with protein concentration, senescence, and grain total Zn, P, K, and S concentrations and was positively correlated with grain volume weight. Grain volume weight was positively corre-lated with grain total Mg and dialyzed Zn, Ca, and P but was strongly negatively correlated with senescence, grain protein concentration, and grain total K. Grain protein concentration was positively correlated with grain total P, K, S, and Mn. Grain total Fe did not correlate with any of the tested variables. Grain total Zn concentration was

positively correlated with grain total P, S, and Mg and was negatively correlated with dialyzed Fe.

Grain total P was positively correlated with all the grain total minerals except grain total Fe. Dialyzed Zn was positively correlated with dialyzed Fe, Ca, and P but was negatively correlated with grain total K and S. Dialyzed Fe was negatively correlated with grain total Zn. The negative correlation between grain yield and mineral and protein concentrations indicates that genetic improvement for quality traits could reduce grain yield.

Table 2. Gene main effects and gene ́ gene interaction effects on grain volume weight, grain protein concentration, grain total P, S, Ca, Mg and Mn concentrations, dialyzed (dial.) Zn, Ca and Mn concentrations, and organic phosphate (Pi) concentration.

Parameter GVW† Protein Total P Total S Total Ca Total Mg Total Mn Dial. Zn Dial. Ca Dial. Mn Pilpa1-1 effect ns‡ *** *** *** * *** ** *** *** *** ***

Gpc-B1 effect *** * ns ns ns ns ns ns ns ns ns

Epistasis ns ns ns ns ns ns ns ns ns ns *

* Significant at the 0.05 probability level.

** Significant at the 0.01 probability level.

*** Significant at the 0.001 probability level.

† GVW, grain volume weight.

‡ ns, not significant.

Fig. 4. Genetic correlations between senescence, yield, grain volume weight, protein concentration, and grain total and dialyzed mineral concentrations of 39 recombinant inbred lines grown in Mead, Havelock, North Platte, and Sidney in Nebraska in 2016 and 2017. Positive correlations are shaded red, and negative correlations are shaded purple. * Significant at a = 0.05; ** Significant at a = 0.01; *** Significant at a = 0.001.

crop science, vol. 58, september–october 2018 www.crops.org 1949

DISCUSSIONGenotypic Group DifferencesThe nonsignificant differences in grain yield between the genotypic groups confirm those found in other studies when Gpc-B1 was introgressed into adapted material (Uauy et al., 2006a, 2006b; Brevis and Dubcovsky, 2010; Asplund et al., 2013). In these previous experiments, no significant reductions in grain yield in Gpc-B1 hexaploid experimental lines were reported. However, our study is the first to report that the combination of Gpc-B1 and lpa1-1 alleles in hard winter wheat had no significant effect on grain yield while maintaining protein concentration and some grain mineral concentrations (Fig. 1 and 2). Dialyzed Fe, P, and Cu concentrations were also main-tained when Gpc-B1 and lpa1-1 alleles were present in the same breeding line (Fig. 3a, 3d, and 3f ). Mean grain yields of the RIL genotypic classes were significantly lower than the mean grain yield of the control cultivars. Subsequent breeding work with these materials (Venegas et al., 2018) has demonstrated that this deficit, especially with LPA types, can be overcome with repeated rounds of mating to adapted parents, and selection for agronomic adaptation.

Unfortunately, the Gpc-B1 and lpa1-1 combination significantly reduced grain volume weight (Fig. 1d). A similar reduction in grain volume weight was found in hexaploid and tetraploid wheats when Gpc-B1 near-isogenic lines were tested in California (Brevis and Dubcovsky, 2010; Brevis et al., 2010). The reduction in grain volume weight was associ-ated with lower flour yield and increased ash concentration of flour and semolina. However, baking and pasta quality traits were improved by the Gpc-B1 introgression in those studies (Brevis et al., 2010). Nevertheless, the relationship between grain volume weight and flour yield is complex because it has also been reported that a reduction in grain volume weight was not associated with changes in flour yield in a study using Gpc-B1 RILs (Mesfin et al., 2000). These contrasting results showed that the negative effect on flour yield by the Gpc-B1 allele depends on the background used in the intro-gression and the environment where these genetic materials are grown. Grain volume weight, however, is a factor used in the grading of wheat for market (USDA, 2014), and released cultivars will need to overcome the grain volume weight deficit associated with Gpc-B1.

This study is the first to show that the mutant lpa1-1 allele alone causes a significant reduction in grain protein concentration in winter wheat germplasm. These results are consistent with those reported in nonmutant winter wheat populations where natural reduction in phytic acid led to undesirable reductions in grain protein concentra-tions (Raboy et al., 1991). Even though the lpa1-1 allele caused a negative effect on grain protein concentration in our study, the reduction of phytic acid in the kernel would positively affect gut absorption of minerals, including Fe and protein in human nutrition (Raboy, 2002).

The high-throughput dialysis protocol developed in this study was easy to use and was efficient for analyzing large numbers of samples. Previous investigators have applied dialysis to characterize interactions between tannins, phytate, and mineral accessibility in sorghum [Sorghum bicolor (L.) Moench] (Wu et al., 2016), Zn and Fe bioaccessibility in dry bean (Phaseolus vulgaris L.) (Martinez Meyer et al., 2013), and to characterize both cereals and legumes consumed in India (Hemalatha et al., 2007), among many other studies. Alternative methods to dialysis include the use of Caco-2 cells, or animal- or human-based feeding trials (Glahn et al., 1996; Moretti and Zimmermann, 2016). However, only dialysis is effi-cient as a high-throughput method (Etcheverry et al., 2012; Moretti and Zimmermann, 2016). This is espe-cially true in winter wheat breeding programs, where perhaps many hundreds of samples need be evaluated in a relatively narrow (2 mo in Nebraska, even less in more northern states and Canada) window between harvest and planting. Whereas dialysis methods may not predict Fe bioavailability as accurately as Caco-2 cells, the opposite is true for Zn (Moretti and Zimmermann, 2016). Dialysis has been found to be significantly correlated with human absorption for Zn (Moretti and Zimmermann, 2016) and actually is the current method of choice for such predic-tions (Etcheverry et al., 2012). The dialysis method was useful in that it allowed simultaneous evaluation of several minerals from a wide range of environmental conditions, an important consideration for any study investigating plant mineral composition. As the constant treatment imposed on the samples did identify statistically significant genotype-dependent differences in several key dialyzed minerals, this study could be viewed as a preliminary assessment identifying important genetic materials for subsequent studies using fewer samples, but more detailed procedures such as Caco-2 or animal feeding studies.

The significant increase in dialyzable Zn, Ca, and Mn caused by the Gpc-B1 and lpa1-1 allele combination and the lpa1-1 allele alone demonstrates the lpa1-1 allele’s ability to reduce phytic acid efficiently (Fig. 3b, 3c, and 3e). Reducing phytic acid potentially caused the free dialysis of Zn, Ca, and Mn across the dialysis membrane. In our study, we found a 17% increase in dialyzed Zn and a 16% increase for dialyzed Ca using least square means of LPA-WT and WT-WT genotypes. These results are similar to those found in previous bioavailability studies using in vivo trials. An increase of 11 to 13% in bone Ca and a 29 to 36% increase in blood Ca were reported when chickens were fed LPA and WT maize (Zea mays L.) (Ertl et al., 1998). In another study, Zn retention increased by 78% when LPA maize “polentas” were used in compar-ison with WT ones in a human trial (Adams et al., 2002). There are several reports showing that LPA maize increases Fe and P absorption (Ertl et al., 1998; Mendoza

1950 www.crops.org crop science, vol. 58, september–october 2018

et al., 1998); unfortunately, our protocol did not differ-entiate genotypes for dialyzed Fe and P. However, there was a clear positive effect of the lpa1-1 allele in the dialyz-ability of P, S, Mg, and Mn. Grain total P, S, Mg, and Mn concentrations were decreased by lpa1-1 (Fig. 2d, 2f, 2g, and 2h); however, this effect was reversed in the dialyzed fraction of these minerals, since no significant differ-ences were found between genotypic groups for dialyzed P, S, and Mg (Fig. 3d). The lpa1-1 allele caused a strong positive effect on dialyzed Mn (Fig. 3e). A similar reduc-tion of grain total P by the lpa1-1 allele was also reported in a study using a LPA soft white spring wheat in Idaho (Guttieri et al., 2004).

The gene main effects and gene ´ gene interac-tion effects explained the individual impact of lpa1-1 and Gpc-B1 on some of the tested variables (Table 2). Given the results of the ANOVA and gene main effect analysis, lpa1-1 caused a significant negative effect on grain protein concentration and grain total P, S, Mg, and Mn concen-trations, and a significant positive effect on grain total Ca and dialyzed Zn, Ca, Mn, and Pi concentrations. These results validated the findings of previous studies regarding the capacity of the lpa1-1-type mutants to increase the bioaccessible (dialyzable) fraction of the grain mineral concentration (Ertl et al., 1998; Adams et al., 2002). Gpc-B1 significantly reduced grain volume weight and significantly increased grain protein concentration. These results also validated previous studies that found that Gpc-B1 increased protein concentration (Uauy et al., 2006a, 2006b; Brevis and Dubcovsky, 2010; Asplund et al., 2013). The epistatic interaction found between lpa1-1 and Gpc-B1 on the Pi concentration could be due to the individual effect of lpa1-1 reducing phytate (Guttieri et al., 2004) and Gpc-B1 increasing P remobilization from the leaves to the kernel (Waters et al. 2009). Therefore, the interaction of both genes could change the phytate/Pi ratio in the wheat kernel.

Heritabilities and Genotype Environment Interaction for Agronomic and Quality TraitsWe expected to see a significant G ´ E interaction for yield and protein because the Gpc-B1 effect on protein and the significant G ´ E interaction of this trait have been reported in other studies (Brevis and Dubcovsky, 2010; Carter et al., 2012; Tabbita et al., 2013). Our results (Table 1, Fig. 1) also indicate that genotype (background) strongly affects the modulation of Gpc-B1 locus effect on grain yield. This G ´ E interaction could also be explained by the ability of Gpc-B1 to reduce the length of the grain-filling period, which results in reduced kernel size and grain yield even under favorable environmental conditions. However, reduced time to senescence (Fig. 1b) can also help the plant escape late-season moisture stress common in Nebraska, thereby potentially increasing

kernel size and yield in drought- and heat-stressed envi-ronments. The nonsignificant G ́ E interactions for grain volume weight concentration described in this study are similar to those reported using Gpc-B1 recombinant chro-mosome substitution lines in Israel (Distelfeld et al., 2007).

Medium and high heritabilities for most of the vari-ables within all four locations and over 2 yr show that progress through selection within LPA-GPC germplasm is possible in eastern and western Nebraska. However, because these two traits have significant G ´ E interac-tions (Guttieri et al., 2006; Brevis and Dubcovsky, 2010; Carter et al., 2012; Tabbita et al., 2013), the heritabili-ties are dependent on the genetic background where the Gpc-B1 and LPA traits are integrated. The low heritability (H2 = 0.03) found for grain total Fe and dialyzed Fe was mainly due to a low genetic variance observed in these traits in wheat, as reported elsewhere (Oury et al., 2006). Also, low heritability (H2 = 0.054) for grain Fe, caused by drought, was found in a winter wheat diversity panel planted in Oklahoma in 2012 and 2013 (Guttieri et al., 2015). The typical low heritabilities for Fe in hexaploid wheats have induced researchers to search for new sources of variation, either from relatives of wheat (Cakmak et al., 2004) or via genetic transformation (Borrill et al., 2014). Distelfeld et al. (2007) did find significant increases in Fe and Zn due to introgression of GPC-B1 in durum (Triticum turgidum var. durum) wheat backgrounds. The present investigation actually was undertaken with these observations in mind, with the intent being to determine whether lpa1-1, Gpc-B1, or their combination could induce this lacking variation in hexaploid wheats grown in Great Plains environments. Although significant effects on grain protein concentration were observed due to Gpc-B1, and significant effects on several mineral concentrations were observed due to lpa-1 (Table 2), no significant Fe variation was associated with either gene. This lack of variation could arise from masking effects of the hexaploid backgrounds or might simply be due to environmental conditions differing from those used by Distelfeld et al. (2007). Nonetheless, the number of beneficial effects observed for lpa1-1 and Gpc-B1 herein should encourage additional breeding programs to exploit these genes.

Correlation between VariablesGrain Zn and protein concentrations were negatively correlated with grain yield (r = −0.34 and −0.51, respec-tively). Negative correlations of grain yield with grain protein have been reported extensively in other studies (Oury et al., 2006; Garvin et al., 2006; Oury and Godin, 2007; Fan et al., 2008; McDonald et al., 2008; Zhao et al., 2009; Hussain et al., 2010). Despite the overall negative correlation of grain yield with grain total Zn and grain protein concentration found in our study, there were no significant differences between LPA-GPC RILs and

crop science, vol. 58, september–october 2018 www.crops.org 1951

WT-WT RILs for mean values of grain yield, grain protein concentration, and grain total Zn (Fig. 1a, 1c, and 2b). However, the LPA-GPC genotypic group, as well as all the experimental materials, had significantly lower grain yield than the control materials (Fig. 1a). Lower yield in LPA-GPC genotypes, when compared with controls, likely was caused by the less-than-optimal adaptation to the Nebraska Great Plains conditions, as the two parents used to generate this study’s biparental popu-lation were developed in Oregon and Idaho. In another study performed by us (Venegas et al., 2018), double-cross populations generated using GPC and LPA donor parents and Great Plains-adapted parents were tested in Nebraska using an augmented design with three environments and nine adapted cultivars as controls. Results of this study found no significant differences between mean grain yields of GPC-LPA genotypes and the mean of control genotypes. Therefore, these results indicate that breeding for improved adaptation plays an important role in the performance of unique alleles such as GPC and LPA.

There was no significant correlation between grain total Fe and Zn. These results are different than those found in other studies where Fe and Zn correlations were significant and strong positively correlated (Morgounov et al., 2007; Guttieri et al., 2015). In our study, the nonsig-nificant correlation between grain total Fe and Zn could be the result of the low correlation between soil Fe and Zn found among the four locations (r = 0.1). We also found variable correlations between senescence and grain total Fe, Zn, and protein. Senescence was negatively correlated with grain total Fe and Zn (r = −0.19 and −0.25, respec-tively); however, the correlation was not significant (P = 0.17 and 0.07, respectively). Additionally, senescence and grain protein concentration had a nonsignificant positive correlation (r = 0.16, P = 0.24). Other studies demon-strated that early senescence resulted in an increase in grain total Fe, Zn, and protein concentrations (Uauy et al., 2006b; Waters et al., 2009).

CONCLUSIONSResults of the present study suggest the introgression of the Gpc-B1 and lpa1-1 alleles into adapted Great Plains winter wheat materials can successfully maintain grain total Fe and Zn concentration and increase protein and dialyzed Zn, Ca, and Mn concentrations without reducing grain yield. Additional breeding work will be necessary to improve grain volume weight, which should consequently improve milling properties. Additionally, germplasm combining GPC-LPA, high yield, and high nutrient traits could be used as breeding materials in wheat breeding programs in developing countries where malnutrition is more severe than in developed areas of the world. Medium and high heritabilities across locations and variables found in our study demonstrate that progress

through selection within LPA-GPC germplasm is feasible in the Great Plains agroecosystems. Finally, the high-throughput dialysis protocol developed in this study has the potential to be used as a tool for breeding programs with biofortification endeavors.

Conflict of InterestThe authors declare that there is no conflict of interest.

AcknowledgmentsThe technical assistance of Mitch Montgomery, Greg Dorn, Lori Divis, and a host of graduate students and undergraduate workers of the University of Nebraska Wheat Breeding Pro-gram and the USDA-ARS Wheat Genetics Program at Lin-coln, NE, is gratefully acknowledged.

ReferencesAdams, C.L., M. Hambidge, V. Raboy, J.A. Dorsch, L. Sian,

J.L. Westcott, and N.F. Krebs. 2002. Zinc absorption from a low-phytic acid maize. Am. J. Clin. Nutr. 76:556–559. doi:10.1093/ajcn/76.3.556

Asplund, L., G. Bergkvist, M.W. Leino, A. Westerbergh, and M. Weih. 2013. Swedish spring wheat varieties with the rare high grain protein allele of NAM-B1 differ in leaf senescence and grain mineral content. PLoS One 8:e59704. doi:10.1371/jour-nal.pone.0059704

Borrill, P., J.M. Connorton, J. Balk, A.J. Miller, D. Sanders, and C. Uauy. 2014. Biofortification of wheat grain with iron and zinc: Integrating novel genomic resources and knowledge from model crops. Front. Plant Sci. 5:53. doi:10.3389/fpls.2014.00053

Brevis, J.C., and J. Dubcovsky. 2010. Effects of the chromosome region including the Gpc-B1 locus on wheat grain and protein yield. Crop Sci. 50:93–104. doi:10.2135/cropsci2009.02.0057

Brevis, J.C., C.F. Morris, F. Manthey, and J. Dubcovsky. 2010. Effect of the grain protein content locus Gpc-B1 on bread and pasta quality. J. Cereal Sci. 51:357–365. doi:10.1016/j.jcs.2010.02.004

Butler, D.G., B.R. Cullis, A.R. Gilmour, and B.J. Gogel. 2009. ASReml-R reference manual. State Queensland, Dep. Prim. Ind. Fish. Brisbane.

Cakmak, I., A. Torun, E. Millet, M. Feldman, T. Fahima, A. Korol, et al. 2004. Triticum dicoccoides: An important genetic resource for increasing zinc and iron concentration in modern cultivated wheat. Soil Sci. Plant Nutr. 50:1047–1054. doi:10.1080/00380768.2004.10408573

Carter, A.H., D.K. Santra, and K.K. Kidwell. 2012. Assessment of the effects of the Gpc-B1 allele on senescence rate, grain protein concentration and mineral content in hard red spring wheat (Triticum aestivum L.) from the Pacific Northwest region of the USA. Plant Breed. 131:62–68. doi:10.1111/j.1439-0523.2011.01900.x

Chatzav, M., Z. Peleg, L. Ozturk, A. Yazici, T. Fahima, I. Cak-mak, and Y. Saranga. 2010. Genetic diversity for grain nutri-ents in wild emmer wheat: Potential for wheat improvement. Ann. Bot. (Lond.) 105:1211–1220. doi:10.1093/aob/mcq024

Distelfeld, A., I. Cakmak, Z. Peleg, L. Ozturk, A.M. Yazici, H. Budak, et al. 2007. Multiple QTL-effects of wheat Gpc-B1 locus on grain protein and micronutrient concentrations. Physiol. Plant. 129:635–643. doi:10.1111/j.1399-3054.2006.00841.x

1952 www.crops.org crop science, vol. 58, september–october 2018

Distelfeld, A., C. Uauy, T. Fahima, and J. Dubcovsky. 2006. Physical map of the wheat high-grain protein content gene Gpc-B1 and development of a high-throughput molecu-lar marker. New Phytol. 169:753–763. doi:10.1111/j.1469-8137.2005.01627.x

Ertl, D.S., K.A. Young, and V. Raboy. 1998. Plant genetic approaches to phosphorus management in agricultural production. J. Environ. Qual. 27:299–304. doi:10.2134/jeq1998.00472425002700020008x

Etcheverry, P., M.A. Grusak, and L.E. Fleige. 2012. Application of in vitro bioaccessibility and bioavailability methods for calcium, carotenoids, folate, iron, magnesium, polyphenols, zinc, and vitamins B6, B12, D, and E. Front. Physiol. 3:317. doi:10.3389/fphys.2012.00317

Fan, M.-S., F.-J. Zhao, S.J. Fairweather-Tait, P.R. Poulton, S.J. Dunham, and S.P. McGrath. 2008. Evidence of decreas-ing mineral density in wheat grain over the last 160 years. J. Trace Elem. Med. Biol. 22:315–324. doi:10.1016/j.jtemb.2008.07.002

FAO. 2016. Food outlook: Global market analysis. FAO, Rome. http://www.fao.org/3/a-i6198e.pdf (accessed 23 Feb. 2017).

Garvin, D.F., R.M. Welch, and J.W. Finley. 2006. Historical shifts in the seed mineral micronutrient concentration of US hard red winter wheat germplasm. J. Sci. Food Agric. 86:2213–2220. doi:10.1002/jsfa.2601

GCARD. 2012. The Wheat Initiative: Second global confer-ence on agriculture research for development, Punta del Este, Uruguay. 29 Oct.–1 Nov. 2012. FAO, Rome. http://www.fao.org/docs/eims/upload/306175/Brief ing%20Paper%20(3)-Wheat%20Initative%20-%20H%C3%A9l%C3%A8ne%20Lucas.pdf (accessed 23 Feb. 2017).

Glahn, R.P., E.M. Wien, D.R. van Campen, and D.D. Miller. 1996. Caco-2 cell iron uptake from meat and casein digests parallels in vivo studies: Use of a novel in vitro method for rapid estimation of iron bioavailability. J. Nutr. 126:332–339. doi:10.1093/jn/126.1.332

Guttieri, M., D. Bowen, J.A. Dorsch, V. Raboy, and E. Souza. 2004. Identification and characterization of a low phytic acid wheat. Crop Sci. 44:418–424. doi:10.2135/cropsci2004.4180

Guttieri, M.J., P.S. Baenziger, K. Frels, B. Carver, B. Arnall, and B.M. Waters. 2015. Variation for grain mineral concentra-tion in a diversity panel of current and historical Great Plains hard winter wheat germplasm. Crop Sci. 55:1035–1052. doi:10.2135/cropsci2014.07.0506

Guttieri, M.J., K.M. Peterson, and E.J. Souza. 2006. Agronomic performance of low phytic acid wheat. Crop Sci. 46:2623–2629. doi:10.2135/cropsci2006.01.0008

Hallberg, L. 1987. Wheat fiber, phytates and iron absorp-tion. Scand. J. Gastroenterol. 22(sup129): 73–79. doi:10.3109/00365528709095855

Hemalatha, S., K. Platel, and K. Srinivasan. 2007. Zinc and iron content and their bioaccessibility in cereals and pulses con-sumed in India. Food Chem. 102:1328–1336. doi:10.1016/j.foodchem.2006.07.015

Hussain, A., H. Larsson, R. Kuktaite, and E. Johansson. 2010. Mineral composition of organically grown wheat genotypes: Contribution to daily minerals intake. Int. J. Environ. Res. Public Health 7:3442–3456. doi:10.3390/ijerph7093442

Lindsay, W.L., and W.A. Norvell. 1978. Development of a DTPA soil test for zinc, iron, manganese, and copper. Soil Sci. Soc. Am. J. 42:421–428. doi:10.2136/sssaj1978.03615995004200030009x

Luten, J., H. Crews, A. Flynn, P. Van Dael, P. Kastenmayer, R. Hurrell, et al. 1996. Interlaboratory trial on the deter-mination of the in vitro iron dialysability from food. J. Sci. Food Agric. 72:415–424. doi:10.1002/(SICI)1097-0010(199612)72:4<415::AID-JSFA675>3.0.CO;2-X

Martinez Meyer, M.R.M., A. Rojas, A. Santanen, and F.L. Stoddard. 2013. Content of zinc, iron and their absorp-tion inhibitors in Nicaraguan common beans (Phaseolus vulgaris L.). Food Chem. 136:87–93. doi:10.1016/j.food-chem.2012.07.105

McDonald, G.K., Y. Genc, and R.D. Graham. 2008. A simple method to evaluate genetic variation in grain zinc concentra-tion by correcting for differences in grain yield. Plant Soil 306:49–55. doi:10.1007/s11104-008-9555-y

Mendoza, C., F.E. Viteri, B. Lonnerdal, K.A. Young, V. Raboy, and K.H. Brown. 1998. Effect of genetically modified, low phytic acid maize on absorption of iron from tortillas. Am. J. Clin. Nutr. 68:1123–1127. doi:10.1093/ajcn/68.5.1123

Mesfin, A., R.C. Frohberg, K. Khan, and T.C. Olson. 2000. Increased grain protein content and its association with agro-nomic and end-use quality in two hard red spring wheat populations derived from Triticum turgidum L. var. dicoccoides. Euphytica 116:237–242. doi:10.1023/A:1004004331208

Mildon, A., N. Klaas, M. O’Leary, and M. Yiannakis. 2015. Can fortification be implemented in rural African communities where micronutrient deficiencies are greatest? Lessons from projects in Malawi, Tanzania, and Senegal. Food Nutr. Bull. 36:3–13. doi:10.1177/156482651503600101

Moretti, D., and M. Zimmermann. 2016. Assessing bioavailability and nutritional value of microencapsulated minerals. In: J.M. Lakkis, editor, Encapsulation and controlled release tech-nologies in food systems. 2nd ed. Wiley, West Sussex, UK. doi:10.1002/9781118946893.ch10.

Morgounov, A., H.F. Gómez-Becerra, A. Abugalieva, M. Dzhu-nusova, M. Yessimbekova, H. Muminjanov, et al. 2007. Iron and zinc grain density in common wheat grown in Cen-tral Asia. Euphytica 155:193–203. doi:10.1007/s10681-006-9321-2

Morris, C.F., S. Li, G.E. King, D.A. Engle, J.W. Burns, and A.S. Ross. 2009. A comprehensive genotype and environment assessment of wheat grain ash content in Oregon and Wash-ington: Analysis of variation. Cereal Chem. 86:307–312. doi:10.1094/CCHEM-86-3-0307

Oury, F.-X., and C. Godin. 2007. Yield and grain protein concen-tration in bread wheat: How to use the negative relationship between the two characters to identify favourable genotypes? Euphytica 157:45–57. doi:10.1007/s10681-007-9395-5

Oury, F.-X., F. Leenhardt, C. Remesy, E. Chanliaud, B. Duper-rier, F. Balfourier, and G. Charmet. 2006. Genetic variability and stability of grain magnesium, zinc and iron concentra-tions in bread wheat. Eur. J. Agron. 25:177–185. doi:10.1016/j.eja.2006.04.011

R Core Team. 2014. R: A language and environment for statistical computing. R Found. Stat. Comput., Vienna.

Raboy, V. 2002. Progress in breeding low phytate crops. J. Nutr. 132:503S–505S. doi:10.1093/jn/132.3.503S

Raboy, V. 2007. The ABCs of low-phytate crops. Nat. Biotechnol. 25:874–875. doi:10.1038/nbt0807-874

Raboy, V. 2009. Approaches and challenges to engineering seed phytate and total phosphorus. Plant Sci. 177:281–296. doi:10.1016/j.plantsci.2009.06.012

crop science, vol. 58, september–october 2018 www.crops.org 1953

Raboy, V., M.M. Noaman, G.A. Taylor, and S.G. Pickett. 1991. Grain phytic acid and protein are highly correlated in winter wheat. Crop Sci. 31:631–635. doi:10.2135/cropsci1991.0011183X003100030017x

Robertson, G.P. 2008. GS+: Geostatistics for the environmental sciences. Gamma Design Software, Plainwell, MI.

SAS Institute. 2011. SAS/STAT 9.3: User’s guide. SAS Inst., Cary, NC.

Tabbita, F., S. Lewis, J.P. Vouilloz, M.A. Ortega, M. Kade, P.E. Abbate, and A.J. Barneix. 2013. Effects of the Gpc-B1 locus on high grain protein content introgressed into Argentinean wheat germplasm. Plant Breed. 132:48–52. doi:10.1111/pbr.12011

Uauy, C., J.C. Brevis, and J. Dubcovsky. 2006a. The high grain protein content gene Gpc-B1 accelerates senescence and has pleiotropic effects on protein content in wheat. J. Exp. Bot. 57:2785–2794. doi:10.1093/jxb/erl047

Uauy, C., A. Distelfeld, T. Fahima, A. Blechl, and J. Dubcovsky. 2006b. A NAC gene regulating senescence improves grain pro-tein, zinc, and iron content in wheat. Science. 314:1298–1301.

USDA. 2014. Grain inspection handbook: Book II wheat. Grain Insp. Packers Stockyards Admin. https://www.gipsa.usda.gov/fgis/handbook/grain-insp/grbook2/wheat.pdf (accessed 26 Oct. 2017).

Venegas, J., R.A. Graybosch, P.S. Baenziger, G. Bai, and P. St. Amand. 2018. Release of Great Plains adapted reduced phy-tate winter wheat germplasm. J. Plant Reg. doi:10.3198/jpr2018.01.0004CRG (in press).

Waters, B.M., C. Uauy, J. Dubcovsky, and M.A. Grusak. 2009. Wheat (Triticum aestivum) NAM proteins regulate the translo-cation of iron, zinc, and nitrogen compounds from vegetative tissues to grain. J. Exp. Bot. 60:4263–4274. doi:10.1093/jxb/erp257

White, P.J., and M.R. Broadley. 2009. Biofortification of crops with seven mineral elements often lacking in human diets: Iron, zinc, copper, calcium, magnesium, selenium and iodine. New Phytol. 182:49–84. doi:10.1111/j.1469-8137.2008.02738.x

Wu, G., S.K. Johnson, J.F. Bornman, S.J. Bennett, V. Singh, A. Simic, and Z. Fang. 2016. Effects of genotype and growth temperature on the contents of tannin, phytate and in vitro iron availability of sorghum grains. PLoS ONE, 11:e0148712. doi:10.1371/journal.pone.0148712

Zhao, F.J., Y.H. Su, S.J. Dunham, M. Rakszegi, Z. Bedo, S.P. McGrath, and P.R. Shewry. 2009. Variation in min-eral micronutrient concentrations in grain of wheat lines of diverse origin. J. Cereal Sci. 49:290–295. doi:10.1016/j.jcs.2008.11.007