Pleiotropic effect analysis and marker development for grain zinc and iron concentrations in spring wheat

Wheat (Triticum aestivum L.) is one of the main food crops in the world and a primary source of zinc (Zn) and iron (Fe) in the human body. The genetic mechanisms underlying related traits have been clarified, thereby providing a molecular theoretical foundation for the development of germplasm resources. In this study, a total of 23,536 high-quality DArT markers was used to map quantitative trait loci (QTL) of grain Zn (GZn) and grain Fe (GFe) concentrations in recombinant inbred lines crossed by Avocet/Chilero. A total of 17 QTLs was located on chromosomes 1BL, 2BL, 3BL, 4AL, 4BS, 5AL, 5DL, 6AS, 6BS, 6DS, and 7AS accounting for 0.38–16.62% of the phenotypic variance. QGZn.haust-4AL, QGZn.haust-7AS.1, and QGFe.haust-6BS were detected on chromosomes 4AL, 6BS, and 7AS, accounting for 10.63–16.62% of the phenotypic variance. Four stable QTLs, QGZn.haust-4AL, QGFe.haust-1BL, QGFe.haust-4AL, and QGFe.haust-5DL, were located on chromosomes 1BL, 4AL, and 5DL. Three pleiotropic effects loci for GZn and GFe concentrations were located on chromosomes 1BL, 4AL, and 5DL. Two high-throughput Kompetitive Allele Specific PCR markers were developed by closely linking single-nucleotide polymorphisms on chromosomes 4AL and 5DL, which were validated by a germplasm panel. Therefore, it is the most important that quantitative trait loci and KASP marker for grain zinc and iron concentrations were developed for utilizing in marker-assisted breeding and biofortification of wheat grain in breeding programs.


Introduction
Wheat (Triticum aestivum L.) is a vital crop in the food production industry. Wheat grains contain various mineral elements, among which, zinc (Zn) and iron (Fe) have important clinical significance for global public health (Maathuis 2009;Nikwan et al. 2021). Malnutrition caused by mineral deficiencies afflicts > 2 billion people worldwide (Zhang and Gladyshev 2011;Stevens et al. 2013;Shivangi et al. Abstract Wheat (Triticum aestivum L.) is one of the main food crops in the world and a primary source of zinc (Zn) and iron (Fe) in the human body. The genetic mechanisms underlying related traits have been clarified, thereby providing a molecular theoretical foundation for the development of germplasm resources. In this study, a total of 23,536 high-quality DArT markers was used to map quantitative trait loci (QTL) of grain Zn (GZn) and grain Fe (GFe) concentrations in recombinant inbred lines crossed by Avocet/Chilero. A total of 17 QTLs was located on chromosomes 1BL, 2BL, 3BL, 4AL, 4BS, 5AL, 5DL, 6AS, 6BS, 6DS, and 7AS accounting for 0.38-16.62% of the phenotypic variance. QGZn. haust-4AL, QGZn.haust-7AS.1, and QGFe.haust-6BS 2021). Statistically, Zn deficiency threatens 17.3% of the world's population and roughly 433,000 children under the age of 5 die every year due to Zn deficiency. Fe deficiency is the thirteenth greatest health risk factor in the world with 24.8% of the global population and > 65% of preschool children in Africa and Southeast Asia suffering from Fe deficiency anemia (Stevens et al. 2013;Liu et al. 2019). Many people in developing countries supplement Zn and Fe through the consumption of wheat foods. Genetic modification is the primary method by which grain Zn (GZn) and grain Fe (GFe) concentrations are increased and have become an increasingly popular area of research in crop science (Xue et al. 2012).
At the beginning of the twenty-first century, while GZn and GFe concentrations' harvestplus in wheat was being promoted, different researchers presented various findings (Ryan et al. 2010). In earlier years, researchers were focused on increasing agricultural production. Kutman et al. (2011) showed that nitrogen (N) supply improved the rhizosphere environment of wheat and changed the rhizosphere morphology, thereby facilitating Zn and Fe absorption by the root system, which is then transferred to the grains. Ryan et al. (2010) found that phosphorous (P) fertilizer application had different effects on Zn and Fe absorption when compared to N fertilizer. If the P concentration in the soil was too high, Zn transport was inhibited. Srivastava et al. (2015) showed that potassium (K) application significantly enhanced Zn absorption in wheat grains and straws. Furthermore, Jalal et al. (2020) found that GZn and GFe concentrations significantly increased after spraying Zn and Fe fertilizer on plant leaves. However, these traditional fertilization methods are costly and easily poison the soil and plants. Currently, the safest and most effective method for addressing Zn and Fe deficiencies in humans is breeding Zn-and Fe-rich wheat varieties.
Recently, after the publication of the Chinese Spring Reference Genome, several gene loci related to GZn and GFe concentrations have been mined through linkage analyses and genome-wide association studies (GWAS) (Rasheed et al. 2017;Velu et al. 2019;Tong et al. 2020). Specifically, Crespo-Herrera et al. (2016) detected three main quantitative trait loci (QTL) for GZn concentrations and three for GFe concentrations on chromosomes 4B and 7D using recombinant inbred line (RIL) populations of spring wheat, which accounted for 10.7-19.6% of the phenotypic variation. Velu et al. (2016) detected two major QTLs for GZn concentrations and three for GFe concentrations on chromosomes 1B, 3A, 5B, and 6B using a total of 105 tetraploid RILs, which accounted for 10.6-16.9% of the phenotypic variation. In a separate study, eight major QTLs for GZn concentrations and four for GFe concentrations were detected on chromosomes 1B, 1D, 2B, 3A, 3D, 6B, 7A, and 7B using a total of 127 hexaploid RILs, which accounted for 10.0-31.0% of the phenotypic variation. Wang et al. (2021) detected one major QTL for GZn concentrations and one for GFe concentrations on chromosome 4D, which was a pleiotropic QTL. Liu et al. (2019) detected 1 major QTL for GZn concentrations and 2 for GFe on chromosomes 2B, 3B, and 5A, which accounted for 10.8-14.6% of the phenotypic variation. Rathan et al. (2021) identified seven new pleiotropic QTL for GZn and GFe concentrations on chromosomes 1B, 1D, 2B, 6A, and 7D. Genc et al. (2009) detected four major QTLs for GZn concentrations on chromosomes 3D, 4B, 6B, and 7A using a doubled haploid population. Each QTL exerted little effect but the aggregation of four high Zn alleles increased GZn concentrations by 23%.
Several pieces of research have proven very effective in increasing Zn and Fe content by wheat breeding method (Shivangi et al. 2021). Based on advancements using different populations for gene mapping, several QTLs that control GZn and GFe concentrations have been identified; however, the number of cloned genes is relatively scarce. Therefore, in this study, a genetic analysis was carried out using new populations to serve as a reference for the mining of effective candidate genes in future studies. The high-density maps were constructed to mine important QTLs and develop molecular markers that can be used for breeding and are of great significance for the genetic improvement of GZn and GFe concentrations' nutritional quality. Thus, we aimed to identify QTLs for GZn and GFe concentrations using a total of 164 RILs derived from Avocet/Chilero. This study provides a theoretical basis and technical support for high Fe and Zn concentrations by utilizing wheat biotechnological breeding.

Plant materials
A total of 164 F6 RILs derived from crossing Avocet/Chilero were used for the QTL mapping of GZn and GFe concentrations. The population was developed by the International Maize and Wheat Improvement Center (CIMMYT) with a single seed descent approach described (Basnet et al. 2014), and was used previously to characterize leaf rust, stripe rust resistance, and fructan content (Ponce-Molina et al.2018;Zeng et al. 2020). A germplasm panel, including 54 domestic and 97 foreign germplasm resources, was used to validate QTLs for GZn and GFe concentrations. Materials were provided by the International Maize and Wheat Improvement Center (CIMMYT) and the wheat germplasm innovation and molecular breeding project of Henan University of Science and Technology, China.

Field trials
The RIL population was tested under a total of three environments in three sites. Field trials were conducted in the experimental field at the Agricultural College of Henan University of Science and Technology (E1; 34°62′N, 112°45′E; the concentrations of alkali-hydrolyzable nitrogen, available zinc, and available iron were 64.75 mg kg −1 , 1.73 mg kg −1 , and 4.21 mg kg −1 respectively) during the 2019-2020 cropping seasons, Luoning (E2; 34°38′N, 111°65′E; the concentrations of alkali-hydrolyzable nitrogen, available zinc, and available iron were 60.08 mg kg −1 , 1.48 mg kg −1 , and 6.23 mg kg −1 respectively) during the2019-2020 cropping seasons, and Mengjin (E3; 34°49′N, 112°26′E; the concentrations of alkalihydrolyzable nitrogen, available zinc, and available iron were 55.85 mg kg −1 , 1.36 mg kg −1 , and 5.83 mg kg −1 respectively) during the2020-2021 cropping seasons in Henan Province, China. Each plot consisted of 2 m rows that were spaced 25 cm. The experiment was arranged in a randomized block design with three replicates. Standard agronomic wheat management practices were applied without using fertilizers to avoid the effect of additional fertilizers on the actual Zn and Fe concentrations. There were no serious pests or diseases and lodging throughout the entire growth period.

Measuring of grain Zn and Fe concentrations
After wheat plants matured, we randomly selected normal grains weighing ~ 5 g, placed the samples in a paper bag, and oven-dried the grains at 80 °C to a constant weight. Dried samples were crushed with a tungsten carbide ball mill (MM 400; Retsch, Haan, Germany) to obtain whole wheat flour. Whole wheat flour weighing 0.2 g was placed in a digestion tube, to which, 5 mL HNO 3 (analytical grade) and 1 mL H 2 O 2 (analytical grade) were added. Samples were digested in a microwave digestion machine (Multiwave PRO; Anton Paar, Shanghai, China) until no visible tissue pieces were left in the tube. After cooling, the volume was fixed to 25 mL. An inductively coupled plasma emission spectrometer (5110 ICP-OES; Agilent, CA, USA) was used to determine the GZn and GFe concentrations (Lu et al. 2010;Vijay et al. 2009). Analysis of variance (ANOVA) and normal distribution tests were performed using Ici-Mapping v4.1 (http:// www. isbre eding. net) and SPSS v21.0 (IBM Inc., Chicago, USA) (Meng et al. 2015;Yan et al. 2013). Student's t-test was used to calculate whether there were significant differences among different genotypes in germplasm panels (Velu et al. 2014). Broad-sense heritability (h 2 ) was estimated using the following equation which was calculated following Holland et al. (2003): g represents the number of varieties, e represents the environments, and MS represents the mean square.

Single-nucleotide polymorphism (SNP) genotyping analysis
DArT markers covering the whole wheat genome with a total of 23,536 detection probes were used in this study, which were measured by Triticarte Co., Ltd. (Canberra, Australia). The molecular marker data was provided by Dr. Ravi P. Singh's team of CIMMYT.

QTL mapping
Statistical processing of the markers was performed using Excel 2010 (Microsoft, WA, USA) to screen polymorphic markers with deletion rates < 10% and minimum alleles > 30%. IciMapping v4.1 software was used to construct the linkage maps and conduct the QTL analysis Wang 2009). Join-Map v4.0 (https:// joinm ap. softw are. infor mer. com) was used to calibrate the genetic lengths and obtain the final genetic map (Zeng et al. 2020). QTL mapping was performed by inclusive composite interval mapping (ICIM). The linkage marker sequence was compared to the Chinese Spring Reference Genome to obtain the physical location of the marker. Physical maps for the positional comparisons of GZn and GFe concentrations QTLs with previous reports were visualized using MapChart v2.3 (https:// mapch art. net) ).

Conversion of SNPs to Kompetitive Allele Specific PCR (KASP) markers
SNPs closely linked to the target QTLs were transformed into KASP markers. PolyMarker (http:// www. polym arker. info/) was used to design KASP primers . A fluorescent quantitative realtime (qRT)-PCR instrument (C1000 Touch; Bio-Rad, CA, USA) was used for gel-free fluorescent signal scanning and allele separation (Chandra et al. 2017). KASP reaction procedures were conducted following previously described methods (Rasheed et al. 2016).

Phenotypic evaluation
Concentrations and variation ranges of GZn and GFe concentrations in the RIL population and the parental lines are provided in Table 1. There was a significant difference between the GFe concentrations of Avocet and Chilero, where the GFe concentrations of Avocet were higher than Chilero in three environments and there were significant differences between the GFe concentrations of the RILs. There were no significant differences between the GZn concentrations of Avocet and Chilero in three environments, but the GZn concentrations among the RILs differed considerably. Thus, it was inferred that the GZn concentration is a quantitative trait controlled by polygenes. GZn and GFe concentrations in the RIL population had a continuous normal distribution (Fig. 1). Results showed that genotype, environment, and their interactions had significant effects on the GZn and GFe concentrations in the RIL population. Both GZn and GFe concentrations had a heritability of 55% (Table 2).

Linkage map construction
Of the 23,536 DArT markers (genotypic data is available for the RILs), 3290 polymorphic molecular markers were used to construct the genetic linkage map. The map covered 15,281.85 cM, and the average genetic distance was 4.64 cM. Except for group 4 consisting of homoeologous chromosomes, the number of markers for other groups was roughly the same. Chromosome 4D had the least label coverage with 41 linkage markers and chromosome 6B had the most with 266 linkage markers. Chromosome 4D had the longest average genetic distance of 8.68 cM and chromosome 3B had the shortest at 3.71 cM (Table 3).

QTL mapping of GZn and GFe concentrations
A total of 10 QTLs of GZn concentrations were located on chromosomes 1B, 2B, 3B, 4A, 4B, 5A, 5D, 6A, and 7A, and identified in different environments ( Fig. 2; Table 4 haust-7AS.1 was 11.87% and 16.62%, respectively, indicating that they are major QTLs (Fig. 4). QGZn. haust-4AL was bound to the DArT marker interval SNP100512195-1,721,210 with a physical interval  haust-5DL, and QGFe.haust-6DS exhibited increasing GFe concentrations. The favorable alleles originated from Avocet and the phenotypic variation of each QTL was < 10%. The phenotypic variation of QGFe. haust-6BS was 10.63%, respectively, indicating that it is a major QTL. QGFe.haust-1BL was bound to the DArT marker interval 2,268,342-SNP100598436 with a physical interval ranging from 554.7 to 635.1 Mb and was detected in two of the three environments. QGFe. haust-4AL and QGFe.haust-5DL were bound to the DArT marker intervals SNP100512195-1,721,201 and 4,440,045-4,989,564 with physical intervals ranging from 720.1 to 731.0 and 432.8 to 554.8 Mb, respectively, and were detected in three environments. These findings indicated that the 3 loci were stable major QTLs for controlling GFe concentrations (Fig. 4).

QTL pyramids and validation
The chip markers located on the 1BL chromosome could not be converted into specific markers. The flanking SNPs of the pleiotropic loci (QGZn.haust-4AL and QGFe.haust-4AL; QGZn.haust-5DL and QGFe. haust-5DL) located on chromosomes 4AL and 5DL were converted into KASP markers, named Kasp_4A_ QGZnFe and Kasp_5D_QGZnFe (Table 6). The markers were used to detect genotypes in the germplasm panels. Research indicates that both Kasp_4A_QGZnFe and Kasp_5D_QGZnFe had obvious genotypes in the germplasm panels (Fig. 3). Among the 151 germplasm panels, there was no significant difference (P < 0.05) in the GG genotype with mean differences of 3.57 and 3.01 mg kg −1 , respectively (Table 7). The frequency of germplasm carrying favorable allelic variants of CC was lower than GG allele, indicating that the pleiotropic loci (QGZn.haust-5DL and QGFe.haust-5DL) have considerable application value in breeding.

Methods for increasing GZn and GFe concentrations
The GZn and GFe concentrations in wheat can be increased through several methods (Kopittke et al. 2018). Kutman et al. (2011) showed that the application of N fertilizer improved GZn and GFe concentrations in wheat. Srivastava et al. (Srivastava et al. 2015) showed that K fertilizer application improved GZn absorption in wheat. Furthermore, Ryan et al. (2010) found that P fertilizer inhibited the absorption of GZn and GFe  Fig. 3 Genotyping of KASP molecular marker in the germplasm concentrations in wheat. However, GZn and GFe concentrations are mainly absorbed from and accumulate in the soil, and the Zn and Fe contents in the soil that can be absorbed and transformed by plants are generally low. Therefore, only applying NPK fertilizer to improve GZn and GFe concentrations is inefficient and could easily cause soil toxicity. Jalal et al. (2020) found that the foliar application of ZnSO 4 ·7H 2 O and FeSO 4 increased GZn and GFe concentrations in wheat. An (2011) demonstrated that the combination of spraying Zn and bulk element fertilizers effectively improved crop yields and the GZn concentrations when compared to spraying Zn fertilizer alone. Although the application of Zn and Fe fertilizers increases the GZn and GFe concentrations,    their cost is too high, and excessive Zn and Fe lead to toxicity that affects plants and humans (Uauy et al. 2006;Peleg et al. 2009). Thus, excavating for QTLs or genes with high Zn and Fe efficiencies to improve GZn and GFe accumulation in wheat is a safe and efficient biotechnology method. The use of genetic engineering has uncovered a new method for increasing the accumulation of GFe and GZn in wheat (Connorton andBalk 2019;Kathleen 2019;Henry et al. 2018).

Comparisons with QTL mapping
The genetic basis for and QTL mapping of GZn and GFe accumulation will serve as important references for selecting suitable breeding strategies through molecular marker-assisted breeding to improve GZn and GFe concentrations in wheat. In previous studies, if the genetic loci of GZn and GFe concentrations were located on the same or similar chromosomal regions, the loci had stable genetic effects (Gupta et al. 2020;Cao et al. 2020). In this study, the variation distributions of GZn and GFe concentrations of the RILs were continuous and susceptible to environmental factors, indicating they were quantitative traits controlled by multiple genes (Fig. 1), which was consistent with the findings of Velu et al. (2016). A total of 17 QTLs were detected that controlled GZn and GFe concentrations (Fig. 2 (Tong et al. 2020). However, they were similar to the loci of GZn-Xwmc357 and GFe-Xwmc357 detected by GWAS, indicating that they may be the same QTL (Gorafi et al. 2016 (Tong et al. 2020).

Candidate gene analysis
There are nearly 200 Fe/Zn-related genes in Arabidopsis and rice, which a previous study used in a BLAST analysis against the wheat database in EnsemblPlants (Tong et al. 2020). Using the QTLs identified in this study, we compared the homologous genes controlling GZn and GFe concentrations in rice and Arabidopsis. TraesCS1B02G359000 is located in the physical interval of QGZn.haust-1BL. TraesCS2B02G530500, TraesCS2B02G543700, and TraesCS2B02G543800 are located in the physical interval of QGZn.haust-2BL. TraesCS2B02G443800 and TraesCS2B02G466500 are located in the physical interval of QGFe.haust-2BL. TraesCS3B02G562500 is located in the physical interval of QGZn.haust-3BL. TraesCS4B02G056600 and TraesCS4B02G081200 are located in the physical interval of QGZn.haust-4BS. TraesCS5A02G383400 and TraesCS5A02G391000 are located in the physical interval of QGZn.haust-5AL. TraesCS6B02G056600 and TraesCS4B02G081200 are located in the physical interval of QGZn.haust-5AL. TraesCS6B02G058700 and TraesCS6B02G391000 are located in the physical interval of QGZn.haust-5AL. By editing these genes, GZn and GFe concentrations in wheat can theoretically be effectively regulated. However, this biotechnology method is rarely used for the improvement of GZn and GFe concentrations in wheat. When OsNAS2 and PvFERRITIN are overexpressed in wheat, the GZn concentrations in wheat is thereby improved, thus laying a foundation for further research on Zn/Fe biofortification in wheat (Beasley et al. 2019;Singh et al. 2017).

Pleiotropic analysis of GZn and GFe concentrations
In this study, we identified QTLs on chromosomes 1B, 4A, and 5D that simultaneously control GZn and GFe concentrations in wheat (Table 5). These loci are pleiotropic QTLs. Moreover, Rathan et al. (2021) showed that nine RILs with the best combination of pleiotropic QTL for Zn and Fe have been identified to be used in future crossing programs and to be screened in elite yield trials before releasing as biofortified varieties. Gorafi et al. (2016) found a pleiotropic locus that controls GZn and GFe concentrations on chromosome 5D. Velu et al. (2016) found a pleiotropic locus on chromosome 2B that controls GZn and GFe concentrations, and demonstrated that a single locus could simultaneously increase GZn and GFe concentrations. Peleg et al. (2009) found pleiotropic QTLs that control GZn and GFe concentrations in tetraploid wheat. Therefore, it can be inferred that the accumulation of GZn and GFe concentrations has a similar genetic basis. Hao et al. (2021) reported that in the past 80 years, the GZn concentration increased but the GFe concentration decreased in Chinese wheat breeding. However, other studies showed that there was a significant positive correlation between GZn and GFe concentrations in wheat (Zhao et al. 2009;Morgounov et al. 2007). It was speculated that this result was due to the fact that Zn and Fe fertilizers were not applied in the experiment, which did not lead to coion antagonism effect. The effects of the pleiotropic loci on chromosome 5DL (QGZn.haust-5DL and QGFe. haust-5DL) in the germplasm panels corroborated these findings. The frequency of germplasm panels carrying favorable allelic variants of CC was lower than GG alleles, indicating that the pleiotropic loci (QGZn.haust-5DL and QGFe.haust-5DL) have considerable application value in breeding. The development of KASP markers can be effectively used for molecular marker-assisted breeding for the enhancement of GZn and GFe in wheat. Some transporters, chelators, and genes simultaneously regulate GZn and GFe concentrations in wheat (Shanmugam et al. 2013;Tong et al. 2020). These results indicated that GZn and GFe concentrations can be improved simultaneously in breeding programs aimed at the biofortification of trace elements. Therefore, it is most important that quantitative trait loci and KASP marker for grain zinc and iron concentrations were developed for utilizing in marker-assisted breeding and biofortification of wheat grain in breeding programs.

Conclusion
In this study, QTLs for GZn and GFe concentrations were identified using 164 RILs derived from Avocet/Chilero. Four stable QTLs, QGZn.haust-4AL, QGFe.haust-1BL, QGFe.haust-4AL, and QGFe. haust-5DL, were located on chromosomes 1BL, 4AL, and 5DL. Three pleiotropic effects loci for GZn and GFe concentrations were located on chromosomes 1BL, 4AL, and 5DL. Two high-throughput Kompetitive Allele Specific PCR markers were developed by closely linking single-nucleotide polymorphisms on chromosomes 4AL and 5DL, which were validated by a germplasm panel, further expanding upon the theoretical foundation for Zn and Fe biofortification and genetic improvement.