Edited by:
Reviewed by:
*Correspondence:
This article was submitted to Plant Breeding, a section of the journal Frontiers in Plant Science
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Mineral concentrations in cereals are important for human health, especially for people who depend mainly on consuming cereal diet. In this study, we carried out a genome-wide association study (GWAS) of calcium concentrations in wheat (
Hexaploid wheat (
Several studies on different plant species and crops identified putative quantitative trait loci (QTL) for calcium in grains of wheat, rice, sorghum, barley, maize, pearl millet, or beans (
European countries are among the top wheat producers and exporters in the world (FAO, 2017)
A European wheat panel consisting of 353 varieties mainly coming from Germany and France was used in this study. This panel included 339 WW and 14 varieties of SW. Field experiments were carried out at the Leibniz Institute of Plant Genetics and Crop Plant Research in Gatersleben, Germany (51°490N, 11°170E, 112 m), during two consecutive seasons (2014/2015 and 2015/2016). The individual plot size was 1 m × 1.5 m with four rows spaced 0.20 m apart. All varieties were sown in autumn and subjected to standard agronomic wheat management practices.
Phenotypic analysis was conducted for the whole set of wheat varieties in each season. For each sample, 50 kernels were counted using a digital seed analyzer/counter Marvin (GTA Sensorik GmbH, Neubrandenburg, Germany) and the thousand-grain weight (TGW) was estimated. The samples were milled using a Retsch mill (MM300, Germany) and the milled samples were dried overnight at 40°C. Calcium concentrations were measured by inductively coupled plasma optical emission spectrometry (ICP-OES, iCAP 6000, Thermo Fisher Scientific, Germany) combined with a CETAC ASXPRESSTM PLUS rapid sample introduction system and a CETAC autosampler (CETAC Technologies, Omaha, United States). Fifty micrograms of dried and ground samples from each variety were wet digested in 2 ml nitric acid (HNO3, 69%, Bernd Kraft GmbH, Germany) using a high-performance microwave reactor (UltraClave IV, MLS, Germany). Digested samples were filled up to 15 ml final volume with de-ionized distilled water (Milli-Q® Reference System, Merck, Germany). Element standards were prepared from Bernd Kraft multi-element standard solution (Germany). Calcium as an external standard and Y (ICP Standard Certipur®, Merck, Germany) were used as internal standards for matrix correction.
The resulting calcium values for wheat grains of each variety and environment were used to calculate the best linear unbiased estimates (BLUEs), by applying the residual maximum likelihood (REML) algorithm with mixed linear models (MLMs) function (
The broad sense heritability of Ca was calculated using the equation:
where
Analyses of variance (ANOVA) and Pearson’s correlation coefficient were calculated for the calcium trait across the two environments with SigmaPlot package 13.
All wheat varieties were genotyped by the company Trait Genetics GmbH, Gatersleben, Germany
GWAS analysis for the phenotypic and genotypic dataset was performed by using GenStat v16 software (VSN International, Hemel Hempstead, Hertfordshire, United Kingdom). Association analysis was performed using the “Single trait association analysis” function with Kinship matrix (K) as a relationship model to control for population structure by GenStat v16, though no obvious population structure was observed in the described population (
To declare the significant marker–traits associations (MTAs), we considered a threshold
BLUEs of the trait, each variety and across the seasons (2015 and 2016) were calculated by applying the “mixed models REML” module with the “linear mixed models” of GenStat v16.
Linkage disequilibrium (LD) which is the non-random association between pairs of loci was studied in the whole panel, observed by using squared allele frequency correlation and calculated within each chromosome. Loci in the LD region were determined according to the squared allele frequency correlations (
While in GWAS analysis the marker data were connected with the phenotypic data in order to identify significant MTAs, in this step we identified the flanking sequence of SNP markers defining significant associations with the calcium trait. Markers which were located in significant LD regions were obtained from the wheat 90k database (
Calcium measurements were performed for the whole set of European wheat varieties (WW = 339, SP = 14) grown in two seasons (2015 and 2016) (Supplementary Table
Calcium concentrations of the top 6 accessions [Ca > 600 (μg g-1 DW)] based on BLUE values ± SE. Numbers on the
GWAS analysis was performed using a MLM with 90k and 35k SNP markers for the calcium data from the two growing seasons 2015 and 2016. Additionally the MTAs for the BLUEs from both years were calculated. Our analysis detected 485 significant [-log10 (
Manhattan plots with –log
Summary of consistently significant markers detected in the two environments and BLUEs.
Marker∗ | Chromosome | Position (cM) | -log10 ( |
Effect BLUEs | % |
---|---|---|---|---|---|
BS00049644_51 | 2A | 66.6 | 8.46 | –20.89 | 4.60 |
RAC875_c24517_558 | 2A | 64.3 | 4.73 | –15.17 | 1.68 |
Kukri_c40035_258 | 2A | 64.3 | 4.98 | –15.60 | 1.85 |
AX-95169653 | 2A | 66.6 | 6.27 | 18.27 | 3.12 |
AX-94940052 | 2A | 66.6 | 5.97 | –17.96 | 1.61 |
AX-94536561 | 2A | 66.6 | 8.11 | –20.42 | 3.44 |
AX-94881950 | 2A | 64.3 | 5.17 | –16.02 | 2.25 |
AX-94850365 | 2A | 64.3 | 5.06 | –15.85 | 2.19 |
AX-94560505 | 2A | 64.3 | 4.42 | –14.61 | 1.70 |
AX-94544896 | 2A | 64.3 | 4.53 | –14.78 | 1.78 |
AX-94404038 | 2A | 64.3 | 4.96 | –15.55 | 2.13 |
RFL_Contig1175_354 | 3A | 109 | 4.47 | 19.60 | 0.81 |
wsnp_Ex_c20899_30011827 | 5A | 117.7 | 7.87 | 45.80 | 7.83 |
RAC875_c8642_231 | 5A | 114.5 | 12.31 | 43.28 | 11.27 |
AX-95077733 | 5A | 117.7 | 7.87 | 45.80 | 8.26 |
snp_CAP8_c1210_739429 | 5B | 149.8 | 4.61 | –33.57 | 1.86 |
CAP7_c5481_96 | 5B | 149.8 | 4.61 | –33.57 | 1.86 |
RAC875_c30011_426 | 5B | 78.7 | 6.93 | –18.44 | 3.44 |
BS00062731_51 | 5B | 78.7 | 6.41 | –17.70 | 1.03 |
GENE-0168_7 | 5B | 75.5 | 7.81 | –21.80 | 4.67 |
AX-94644169 | 5B | 103.2 | 8.08 | –21.88 | 4.43 |
AX-94541836 | 5B | 101.7 | 7.98 | –20.14 | 2.80 |
AX-94547820 | 5B | 100.9 | 7.32 | –18.95 | 3.61 |
AX-94452355 | 5B | 100.9 | 6.56 | –17.89 | 2.40 |
Jagger_c8037_96 | 5D | 167 | 3.87 | 13.73 | 1.64 |
wsnp_Ex_c17575_26300030 | 6A | 37.3 | 7.27 | 24.98 | 3.55 |
wsnp_Ex_c17575_26299925 | 6A | 37.3 | 7.33 | 25.12 | 3.72 |
Tdurum_contig62141_496 | 6A | 37.3 | 7.27 | 24.98 | 3.55 |
Kukri_rep_c104648_439 | 6A | 37.3 | 7.27 | 24.98 | 3.55 |
Kukri_c35661_63 | 6A | 37.3 | 7.27 | 24.98 | 3.55 |
AX-94415776 | 6A | 37.3 | 7.04 | 24.69 | 3.88 |
The additive effects of five representative significant markers based on BLUEs are depicted in
Box plot analysis for five significant markers based on BLUEs for calcium trait (MAF = AF). ∗∗∗denotes 0.001.
In order to identify potential candidate genes for calcium concentrations in wheat grains, the significant SNP markers [–log10 (
Blast analysis of these markers using POPSEQ showed that chromosomes 2A, 5A, 5B, 5D, and 6A harbored many calcium- transporting genes. The most significant SNP (RAC875_ c8642_231) in our analysis was located on chromosome 5A (114.5 cM). The gene underlying this marker encodes a cation/sugar symporter, while the second significant locus (wsnp_Ex_c20899_30011827) on the same chromosome (117.7 cM) carries a gene that encodes an AP2-type transcription factor. We further detected two genes, which may be related to calcium transport near this significant region (114.5–117.7 cM): one gene (Traes_5AL_898DAA873) is related to plasma membrane ATPases while another gene (Traes_5AL_637EB761F) encodes an H+-ATPase. Furthermore, in the same region, we found a gene (Traes_5AL_AE6B41A0A) related to divalent metal cation transport together with two further genes (Traes_5AL_6C9A5537F and Traes_5AL_6C8BD96CB) related to heavy metal transport/detoxification. Along this chromosome we were able to find loci associated with Ca-permeable ion channels, such as Traes_5AL_E1F7DD9EA and Traes_5AL_C89AC9640 coding for cyclic nucleotide-gated channels (CNGCs) or Traes_5AL_F1522B81F and Traes_5AL_98814295D encoding mechanosensitive ion channels. However, these genes were not closely located to the significant markers (Supplementary Table
In the LD region of chromosome 2A, we found a number of genes, which are related to calcium transport functions, such as Traes_2AL_72F83E7B0 and Traes_2AL_6069A8864 (mechanosensitive ion channel family protein), Traes_2AL_ D33454518 (cation/H+ antiporter), and Traes_2AL_CF9F964E6 underlying a calcium-transporting ATPase. The most significant association on chromosome 5B was the locus Traes_5BL_DF8D1B819, which is related to an ammonium transporter. Other closely linked genes are Traes_5BS_4AEE5C2AE encoding a mechanosensitive ion channel family protein and Traes_5BS_98C73F5CA, Traes_5BS_06F7D0060, Traes_5BS_272FDBF9D, Traes_5BL_4AACBDDAA, and Traes_5BL_E4BE45756, all related to cation/H+ antiporters. Near to the highly significant markers of chromosome 5B (depicted in yellow in Supplementary Table
Genetic fortification strategies are highly suitable for developing wheat varieties with high mineral element contents. Therefore, this study focused on investigating the natural genetic variation in European wheat varieties and on identifying candidate genes contributing to calcium accumulation in wheat grains. Phenotypic analysis for calcium concentrations showed a wide variation between the varieties based on BLUEs which ranged from 288.2 to 647.5 μg g-1 DW. The heritability was high (0.73) indicating that the major part of the variability was due to genotypic effects, which is in agreement with previous studies (
In the present study, genome mapping revealed that most of the significant MTAs for the consistently significant markers in 2015, 2016, and BLUEs (
In general, calcium transporters are involved in the cellular compartmentalization of calcium in different plant organs. Three major gene families of calcium transporter proteins have been described: (i) Ca2+-transporting P-type-ATPases [endoplasmic reticulum-type Ca2+-ATPase (ECA/IIA Type) and autoinhibited Ca2+-ATPase (ACA/IIB-type)], (ii) divalent cation–H+ antiporters/exchangers [cation/H+ antiporters (CAX), CCX and CHX], and (iii) Ca-permeable ion channels that include mechanosensitive calcium-permeable channels (MSCCs), glutamate receptors (GLRs), CNGCs and two-pore channels (TPC) (
Putative candidate genes for Ca in the vicinity of highly significant associated SNP-marker.
Significant marker | Chromosome, | ( |
–log10( |
Effect | % |
Gene name | Human_Readable_Description |
---|---|---|---|---|---|---|---|
position (cM) | bw_SNP | BLUES | (IWGSC_v1) | ||||
RAC875_c8642_231 | 5A, -114.5 | 0.46 | 12.31 | 43.72 | 11.27 | Traes_5AL_F49663738 | Sugar transporter/solute-cation symport |
wsnp_Ex_c20899_30011827 | 5A, -117.7 | 1.00 | 7.87 | 46.14 | 7.83 | Traes_5AL_19637DE03 | AP-2 complex subunit alpha-1 |
– | 5A | – | – | – | – | Traes_5AL_6C8BD96CB | Heavy metal transport/detoxification superfamily protein |
– | 5A | – | – | – | – | Traes_5AL_898DAA873 | Plasma membrane ATPase 1 |
– | 5A | – | – | – | – | Traes_5AL_637EB761F | H(+)-ATPase 11 |
– | 5A | – | – | – | – | Traes_5AL_AE6B41A0A | Divalent metal cation transporter MntH |
– | 5A | – | – | – | – | Traes_5AL_6C9A5537F | Heavy metal transport/detoxification |
– | 5A | – | – | – | – | Traes_5AL_6C8BD96CB | Heavy metal transport/detoxification |
AX-94940052 | 2A, -66.6 | 0.48 | 5.97 | –17.96 | 1.61 | Traes_2AL_156C770EE | Receptor-like protein kinase 2 |
– | 2A | – | – | – | – | Traes_2AL_542E74269 | Potassium channel AKT1 |
Kukri_c40035_258 | 2A, -64.0 | 0.10 | 4.98 | –15.60 | 1.85 | Traes_2AS_0C87833C6 | Phosphatidylinositol-4-phosphate 5-kinase family protein |
– | 2A | – | – | – | – | Traes_2AL_72F83E7B0 | Mechanosensitive ion channel family protein |
– | 2A | – | – | – | – | Traes_2AL_6069A8864 | Mechanosensitive ion channel family protein |
– | 2A | – | – | – | – | Traes_2AL_D33454518 | Cation/H+ antiporter |
– | 2A | – | – | – | – | Traes_2AS_B264257CD | Flavin-binding monooxygenase family protein |
– | 2A | – | – | – | – | Traes_2AL_CF9F964E6 | Calcium-transporting ATPase |
– | 2A | – | – | – | – | Traes_2AS_95611CAD2 | Heavy metal transport/detoxification superfamily |
– | 2A | – | – | – | – | Traes_2AL_6DD37E6BE | Heavy metal transport/detoxification superfamily |
– | 2A | – | – | – | – | Traes_2AL_9B175F3Da | Heavy metal transport/detoxification superfamily |
– | 2A | – | – | – | – | Traes_2AL_F360E3FE3 | Heavy metal transport/detoxification superfamily |
– | 2A | – | – | – | – | Traes_2AL_13CBA4FEA | Heavy metal transport/detoxification superfamily |
– | 2A | – | – | – | – | Traes_2AS_AA84E72D4 | Heavy metal transport/detoxification superfamily |
– | 2A | – | – | – | – | Traes_161086245 | Heavy metal transport/detoxification superfamily |
AX-94547820 | 5B, -100.9 | 0.03 | 7.20 | –18.80 | 3.61 | Traes_5BL_DF8D1B819 | Ammonium transporter 2 |
– | 5B | – | – | – | – | Traes_5BL_411EF97B9 | Cyclic nucleotide-gated channel |
Based on our investigations, we found 41 potential calcium-transporting genes distributed over six chromosomes (2A, 3A, 5A, 5B, 5D, and 6A). These include 8 Ca/proton exchangers, 4 Ca-ATPases, and 31 channels in addition to other genes that are putatively related to calcium transport, such as Traes_5AL_6C9A5537F which is annotated as heavy metal transporter (Supplementary Table
Apart from focusing on the concentrations of iron and zinc in wheat, which has taken much attention in previous studies, only few genetic studies are available on calcium concentrations in wheat grains are available. Improving levels of grain calcium concentration in hexaploid wheat remains one of the most important breeding objectives for the nutritional security of the whole population and especially for the poor from the nations where wheat is the main source of calories. Overall, through measurable phenotypic and genotypic variation for grain calcium concentrations as well as by considering the detected favorable QTLs distributed across various chromosomes and potentially responsible genes in the current research, we aimed to deepen the understanding of the genetic basis of calcium accumulation in wheat grains and to open the door to more efficient ways to increase calcium concentration in the grain and thereby overall wheat quality.
DA performed the data analysis including genome-wide association scan and related analyses. KE and NvW participated in calcium concentration measurements. KP and MR designed the experiment. MR conceived the idea and participated in the interpretation of results. DA and MR wrote the manuscript. All authors read and approved the final manuscript.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The authors would like to acknowledge Ahmad M. Alqudah for the helpful assistance in software learning and giving good advices. Authors also thank Quddoos Muqaddasi for his help in the software tools. We also thank Ellen Weiß and Yudelsy Antonia Tandron Moya for excellent technical assistance.
The Supplementary Material for this article can be found online at: