Skip to main content
NCBI home page
Springer logoLink to Springer
. 2025 Jan 23;138(1):32. doi: 10.1007/s00122-025-04817-y

Unveiling the genetic architecture of barley embryo: QTL mapping, candidate genes identification and its relationship with kernel size and early vigour

Xiaoxia Chen 1,2,#, Zhouyang Su 1,#, Yunpu Zheng 1,3, Cong Li 4, Jun Ma 5, Jian Ma 4, Fusun Shi 2, Haiyan Hu 6, Chunji Liu 1,6,✉, Zhi Zheng 1,✉
PMCID: PMC11754356  PMID: 39843841

Keymessage

In this first QTL mapping study of embryo size in barley, novel and stable QTL were identified and candidate genes underlying a significant locus independent of kernel size were identified based on orthologous analysis and comparison of the whole-genome assemblies for both parental genotypes of the mapping population.

Embryo, also known as germ, in cereal grains plays a crucial role in plant development. The embryo accounts for only a small portion of grain weight but it is rich in nutrients. Larger embryo translates to a more nutritious grain and larger store of energy reserves, which can benefit seed germination and seedling establishment. However, reports on quantitative trait loci (QTL) affecting embryo size in barley is rare. To understand the genetic basis of embryo size in barley, a population consisting of 201 F9 recombination inbred lines (RILs) was assessed in four environments. Three regions affecting various characteristics of embryo size including embryo length (EL), embryo width (EW) and embryo area (EA) were consistently identified. They located on chromosomes 2H, 4H and 7H, respectively. Among them, the QTL on 7H was not significantly affected by kernel size. Phenotypic variances explained by this QTL for EL, EW and EA were 11.8%, 9.3% and 12.7%, respectively. Taken advantage of the available genomic assemblies of the two parental genotypes, candidate genes for this locus on 7H were identified. In addition, significant correlations between embryo size and early vigour and kernel traits were detected. To our knowledge, the present study is for the first time reporting QTL conferring embryo size by directly measuring the characteristics as quantitative trait in barley, which would broaden our understanding of the genetic basis of barley embryo size and offer valuable targets for future breeding programmes.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00122-025-04817-y.

Introduction

Barley, ranked as the fourth most significant cereal crop in terms of both production and economic impact, serves as a crucial component in stockfeed and forms the foundation for malting and brewing (Walker et al. 2013). Cereal grains consist of four main components: the hull or husk, the bran layer, the endosperm, and the embryo or germ. The embryo constitutes a relatively small portion (2–3%) of the total kernel weight but it is packed with essential nutrients that are vital for plant growth and development including protein, vitamins, oil and non-starch carbohydrates (Raj et al. 2023). Increasing relative embryo size is known to enhance grain quality and showed nutritional benefits for human food and animal feed (Lee et al. 2016; Jung et al. 2017). As the demand for high-quality grains surges, cereal varieties with larger embryos present an exciting avenue for enhancing the nutritional value of our food sources (Orman-Ligeza et al. 2020). While embryo size is not the sole determinant of yield, it can significantly impact grain weight and overall crop productivity.

Embryo size has been a key indicator for agronomic improvement on early vigour (Zhao et al. 2019). Results from previous studies showed that larger embryos in cereals are associated with faster seed germination, greater crop establishment, increased leaf width, enhanced seedling shoot growth and improved root development (Forbis et al. 2002; Richards and Lukacs 2002; Rebetzke et al. 2008; Maydup et al. 2012; Finch-Savage and Bassel 2016; Hendriks et al. 2022). Rebetzke et al. (2022) suggested that these traits increase resource capture and weed competitiveness, further improving crop performance and grain yield under water-limited conditions (Li et al. 2019; Rebetzke et al. 2022). While embryo size is associated with kernel size (López‐Castañeda et al. 1996; Richards and Lukacs 2002; Li et al. 2022), it is suggested that embryo size independent of kernel size may be an important determinant of final number and size of seminal roots (Rebetzke et al. 2022). In barley, embryo size is an important determinant of early viability (López‐Castañeda et al. 1996; Moore and Rebetzke 2015). Therefore, understanding of genetic architecture for barley embryo size and its genetic association to kernel size and early vigour would accelerate the improvement of barley quality and seedling establishment.

Linkage mapping is one of the main approaches for identifying loci for a targeted trait. Compared to extensive investigations in other cereal crops such as rice, maize and wheat (Moore and Rebetzke 2015; Lee et al. 2019; Yuan et al. 2019; Li et al. 2022; Rebetzke et al. 2022; Chen et al. 2023; Katral et al. 2023; Wang et al. 2023), the genetic studies on barley embryo size is rare. The only genetic study on this trait in barley was based on a mutagenized germplasm described by Orman-Ligeza et al. (2020). In that study, the embryo size was assessed as a qualitative trait using an F2 population. A high-lysine locus (LYS3) encoding a prolamin-box-binding transcription factor was identified as controlling embryo size on chromosome arm 5HL. However, there have been few attempts to elucidate the genetic basis of embryo size based on natural variation. Despite of the small size of the embryo, its enclosure within the hull poses significant challenges for direct measurement, hindering genetic studies on embryo in hulled barley.

In this study, we for the first time attempted to identify novel QTL conferring embryo size including embryo length (EL), embryo width (EW) and embryo area (EA) in barley. To achieve this, seeds from a population consisting of 201 F9 recombinant inbred lines (RILs) were harvested and assessed across four field sites during the 2022 and 2023 seasons. In addition, we investigated its correlations with kernel size and early vigour traits. Moreover, we employed comparative analysis of genes with known effects in other crop species like rice between the two parental genotypes. This approach has proven to be powerful for identifying candidate genes within targeted regions (Gao et al. 2024; Zheng et al. 2022; Zhou et al. 2021). Leveraging the high-quality genome assemblies available for the two parental genotypes of this population, we successfully predicted candidate genes underlying a locus independent of kernel size. The findings of this research are detailed in this publication.

Materials and methods

Plant materials

A RIL population consisting of 201 F9 lines was assessed in this study. This RIL population was generated from a cross between Morex, a six-row malting barley, and AWCS276, a two-row wild barley (Zhou et al. 2021). Single seed descendent method and fast generation procedure were used to advance this population (Zheng et al. 2013).

Phenotypic assessment

Data on embryo size were collected from four field trials. The mapping population, along with the two parents, were assessed in each of the field trials. The field trials were carried out at CSIRO Forest Hill Research Station (27°33′S, 152°16′E) and Boorowa Research Station (34°47′S, 148°41′E) in 2022 and 2023. These trials were designated as FH22, FH23, BO22 and BO23, respectively. Each trial contained two replicates, with each replicate consisting of ten seedlings planted at a spacing of 20 cm within a single row. Row was spaced 25 cm apart. Field management was according to local agricultural practices.

At Zadoks 92, five well-pollinated spikes from each line were harvested from the main tillers of each plant and put into a 37 °C oven for four days. Eight well-developed kernels from the middle sections of central spikelets from each of the five spikes were selected and then carefully removed the hull without damaging the embryos. Then, the hump of barley kernels was adhered to double-sided tape in Petri dishes to facilitate the measurement. Embryo characteristics including EL and EW were measured using a stereomicroscope (OLYMPUS-SZX7; Olympus, Tokyo, Japan) as described by Wang et al. (2023). The embryo area (EA) is calculated as EA = EL*EW*0.72 as described by Moore and Rebetzke (2015). Kernel characteristics of the same kernels, including kernel length (KL), kernel width (KW) and kernel area (KA), were measured using an SC6000R digital image analyser (Next Instruments, Condell Park, Australia). Maximum root length of seedlings (MRL) was determined by measuring the length of the longest root from crown to root tip at 14 days by growing the seedlings in water-moisture paper rolls. Thousand kernel weight (TKW) and early vigour traits including coleoptile length (CL), third leaf thickness of seedling (S3LT), third leaf length of seedlings (S3LL), third leaf width of seedling (S3LW), third leaf area (S3LA), and ratios of third leaf length and width of seedlings (S3LWR) were retrieved from previous studies (Zheng et al. 2022; Gao et al. 2024).

Statistical analysis

The average values of each line in a single environment and the best linear unbiased prediction (BLUP) value estimated from average values of different environments were used for QTL identification and further analysis. BLUP of target traits and the broad-sense heritability (H2) were calculated using the PROC MIXED and VARCOMP procedures of SAS V8.0 (SAS Institute, Cary, NC, USA; https://www.sas.com). SPSS18.0 software (SPSS, Chicago, IL, USA) was used to perform normal distribution test, Student’s t test (P < 0.05) and correlation analysis of phenotype values in different trials.

QTL analysis

A high-density genetic map of the mapping population based on genotyping by sequencing (GBS) data was constructed according to the previous study (Zhou et al. 2021). A total of 1,140 polymorphic markers were used in this study, resulting in the linkage map spanning approximately 1022.4 cM with an average marker distance of 0.7 cM. MapQTL 6.0 (Van Ooijen and Kyazma 2009) was employed to identify putative QTL conferring embryo size. A test of 1000 permutations were performed to identify the LOD threshold corresponding to a genome-wide false discovery rate of 5% (P < 0.05) for each trial. Interval mapping was then used to identify QTL and MapChart was used to draw linkage maps with QTL positions (Voorrips 2002). Confidence interval is defined as the region on either side of the QTL peak where the LOD score drops by 1.5 compared to the peak LOD score (Visscher et al. 1996). The flanking markers were identified as the nearest markers bordering the support interval.

Conditional QTL analysis

To understand the complex relationships between embryo size and other agronomic traits, conditional QTL analysis was employed using QGAStation 2.0 software (http://ibi.zju.edu.cn/software/) (Zhu 1995) to identify QTL while considering the influence of other traits. For example, A|B indicates that A is conditional on B. BLUP datasets were used to generate the conditional phenotypic values. The interpretation of conditional QTL results can be categorized into four scenarios: (1) QTL for A detected only in traditional analysis, suggesting that B influences A through this locus (2) QTL for A detected in both non-conditional and conditional analyses with similar effect values. This indicates that the locus affects trait A independently; (3) QTL for A detected in both non-conditional and conditional analyses, with significant changes in effect values, which implies that the locus impacts both A and B simultaneously; and (4) QTL for A detected only in conditional analysis, indicating that the effect of this locus on A is masked by B.

Identification of candidate genes underlying QTL for embryo size

Candidate genes were identified in the targeted QTL independent of kernel size as described by Zheng et al. (2022). Briefly, sequences of two markers flanking the confidence interval were used to blast against genome assemblies of barley pseudomolecules Morex (Mascher et al. 2021) to define the physical interval. Coding sequence and protein sequences of predicted genes in the identified QTL regions were retrieved from ftp://ftp.ensemblgenomes.org/pub/release-44/plants/gff3/hordeum_vulgare for Morex and NCGR wild barley database http://db.ncgr.ac.cn/wild_barley/ for AWCS276 (Liu et al. 2020). Gene sequences known to related to embryo and kernel size in rice were used to identify potentially homologous genes with the Morex genome assembly. Candidate genes detected in Morex were then blasted against the genome assembly of AWCS276 for differences. Reciprocal best hits analysis using DIAMOND v2.0 (Buchfink et al. 2021) was performed to understand the potential functional roles of the identified protein sequences.

Results

Phenotypic data analysis in the mapping population

Values of the three embryo size traits (EL, EW and EA) showed highly significant and positive correlation across four different trials and the BLUP dataset (P < 0.01), with correlation coefficients varying from 0.76 to 0.98 (Table S1). In all four trials conducted, AWCS276 had significantly greater EL but smaller EW and EA compared to Morex (Fig. 1; Table 1). The observed variation within the mapping population for EL ranged from 1.18 to 2.05 mm and the standard deviation (SD) of 0.16 mm in the BLUP dataset, while the EW varied from 0.91 to 1.65 mm with the SD of 0.12 mm in the BLUP dataset (Table 1). The H2 estimates for EL, EW and EA were 0.82, 0.90 and 0.89, respectively. These high heritability values suggest that genetic factors played major roles in determining these embryo characteristics (Table 1). Moreover, the frequency distribution of all these three traits displayed continuous variation with transgressive segregation (Fig. 2).

Fig. 1.

Fig. 1

Open in a new tab

Morphologies of the two parental genotypes, AWCS276 and Morex, showing the differences in embryo length (L) and width (W)

Table 1.

Phenotypic variation and heritability of embryo size for the parents and population assessed in different environments

#Trait Trial Parents Population
Morex AWCS276 Min Max Mean SD CV(%) H2
EL(mm) FH22 1.23 1.51 1.17 2.04 1.62 0.17 10.20
BO22 1.28 1.53 1.18 2.06 1.63 0.17 10.34
FH23 1.19 1.58 1.17 2.04 1.62 0.17 10.23
BO23 1.22 1.52 1.17 2.08 1.63 0.17 10.51
BLUP 1.24 1.55 1.18 2.05 1.63 0.16 10.03 0.82
EW(mm) FH22 1.18 0.92 0.91 1.63 1.21 0.12 10.26
BO22 1.16 0.88 0.91 1.64 1.20 0.13 10.41
FH23 1.19 0.91 0.92 1.63 1.21 0.12 10.28
BO23 1.21 0.89 0.89 1.65 1.21 0.13 10.47
BLUP 1.20 0.90 0.91 1.64 1.21 0.12 10.07 0.90
EA(mm2) FH22 1.18 0.93 0.80 2.01 1.42 0.24 16.72
BO22 1.21 0.96 0.79 2.01 1.41 0.24 16.86
FH23 1.15 1.00 0.79 2.02 1.42 0.24 16.74
BO23 1.08 0.88 0.79 2.01 1.42 0.24 17.04
BLUP 1.13 0.96 0.80 2.01 1.42 0.23 16.41 0.89
Open in a new tab

#EL embryo length, EW embryo width, EA embryo area, FH22, BO22, FH23 and BO23 four independent trials conducted for embryo size, BLUP best linear unbiased prediction, SD standard deviation, CV coefficient of variation, H2 the broad-sense heritability

Fig. 2.

Fig. 2

Open in a new tab

Frequency distributions for embryo length (EL), embryo width (EW), and embryo area obtained from the population of Morex/AWCS276 using BLUP datasets

Correlations between embryo size and kernel size and early vigour traits

Significant and positive relationships were detected among all three embryo characteristics (P < 0.001). Potential correlations between embryo size and 11 other traits of agronomic importance were also analysed using the BLUP datasets (Table 2). These agronomic traits included early vigour traits (CL, MRL, S3LT, S3LL, S3LW, S3LA, and S3LWR) and kernel traits (KL, KW, KA and TKW), which were collected from independent experiments using the same population. All three embryo characteristics showed positive and significant correlations with all four kernel traits, except for EL which showed no correlation with KW and TKW (Table 2). In addition, both EW and EA were significantly and positively correlated with S3LL, S3LA, MRL and TKW, while EW was also significantly associated with S3LW, EL and EA were positively correlated with S3LWR. Furthermore, none of the three embryo traits showed significant correlations with S3LT and CL (Table 2).

Table 2.

Coefficients of pairwise Pearson correlations between embryo size and other traits of agronomic importance

Group Traits EL EW EA
Embryo traits EL 1.00
EW 0.35*** 1.00
EA 0.82*** 0.83*** 1.00
Kernel traits KL 0.41*** 0.18* 0.39***
KW 0.02 ns 0.54*** 0.34***
KA 0.29*** 0.47*** 0.46***
TKW 0.03 ns 0.42*** 0.27***
Early vigour traits S3LT –0.04 ns -0.03 ns -0.04 ns
S3LL 0.12 ns 0.14* 0.16*
S3LW –0.07 ns 0.12* 0.03 ns
S3LA 0.06 ns 0.16* 0.13*
S3LRW 0.20** 0.06 ns 0.17**
MRL 0.12 ns 0.21** 0.20**
CL 0.10 ns 0.01 ns 0.06 ns
Open in a new tab

#Correlations were calculated using BLUP datasets. EL embryo length, EW embryo width, EA embryo area, KL kernel length, KW kernel width, KA kernel area, TKW thousand kernel weight S3LT 3rd leaf thickness from seedling, S3LL 3rd leaf length from seedling, S3LW 3rd leaf width from seedling, S3LA 3rd leaf area from seedling, S3LWR 3rd leaf length and width ratio from seedling, MRL maximum root length, CL coleoptile length,

Note: ‘ns’, ‘*’, ‘**’, and ‘***’ refer to significance of correlations (ns not significant, P < 0.05, P < 0.01, P < 0.001)

QTL identification for embryo size

Assessment of the whole mapping population detected three putative QTL regions associated with all three embryo characteristics on chromosomes 2H, 4H and 7H. The allele for large embryo size of the 7H locus was derived from the parent Morex, and the alleles for large embryo size of the 2H and 4H loci were derived from AWCS276. These QTL were consistently detected in all four trials as well as the BLUP dataset (Table 3). The most significant QTL for EL was located on chromosome 7H (designated as Qes.caf-7H) explaining up to 11.8% of the phenotypic variance with a maximum LOD value of 5.2, while the largest effect QTL targeting EW was identified on 2H (designated as Qes.caf-2H) accounting for 15.6% of the phenotypic variance with a LOD value of 7.1. Regarding EA, the largest effect QTL was detected on the short arm of chromosome 4H (designated as Qes.caf-4H), which explained the phenotypic variance ranging from 13.7 to 14.8% with the LOD values varying from 6.2 to 6.7 (Fig. S1, Table 3).

Table 3.

QTL for embryo size identified in the population of Morex/AWCS276#

Traits Trials QTL Chr Linkage map interval (cM) Physical map interval (Mbp) LeftMarker RightMarker LOD PVE (%) Add
EL FH22 Qes.caf-2H 2H 52.1–59.3 529.1–542.6 GBS_MST812 GBS_MST1258 4.0 9.2 –0.050
Qes.caf-4H 4H 15.8–20.9 13.2–23.0 GBS_MST2470 GBS_MST2490 3.4 7.8 –0.046
Qes.caf-7H 7H 89.8–96.3 597.9–615.1 GBS_MST4580 GBS_MST4562 5.2 11.8 0.057
BO22 Qes.caf-2H 2H 52.9–59.5 529.1–542.6 GBS_MST828 GBS_MST1262 3.3 7.7 –0.047
Qes.caf-4H 4H 15.8–20.9 13.2–23.0 GBS_MST2470 GBS_MST2490 3.0 6.8 –0.044
Qes.caf-7H 7H 89.8–96.6 597.9–615.1 GBS_MST4580 GBS_MST4560 4.6 10.4 0.055
FH23 Qes.caf-2H 2H 53.2–59.3 529.1–542.6 GBS_MST833 GBS_MST1258 3.7 8.7 –0.049
Qes.caf-4H 4H 15.8–24.9 13.2–24.3 GBS_MST2470 GBS_MST2491 3.4 7.7 –0.046
Qes.caf-7H 7H 89.5–96.3 597.9–615.1 GBS_MST4580 GBS_MST4562 5.1 11.4 0.057
BO23 Qes.caf-2H 2H 52.9–59.5 529.1–542.6 GBS_MST821 GBS_MST1262 3.0 6.9 –0.045
Qes.caf-4H 4H 12.9–23.3 11.2–24.3 GBS_MST2469 GBS_MST2490 2.8 6.6 –0.044
Qes.caf-7H 7H 91.5–98.8 597.9–615.1 GBS_MST4573 GBS_MST4554 4.1 9.3 0.053
BLUP Qes.caf-2H 2H 52.9–59.3 529.1–542.6 GBS_MST828 GBS_MST1258 3.5 8.2 –0.047
Qes.caf-4H 4H 12.9–24.9 11.5–24.3 GBS_MST2469 GBS_MST2491 3.1 7.2 –0.044
Qes.caf-7H 7H 89.5–96.3 597.9–615.1 GBS_MST4580 GBS_MST4562 4.8 10.7 0.055
EW FH22 Qes.caf-2H 2H 67.4–72.0 568.8–586.1 GBS_MST1320 GBS_MST1355 7.1 15.6 –0.049
Qes.caf-4H 4H 15.8–19.8 13.2–22.7 GBS_MST2470 GBS_MST2477 6.4 14.2 –0.047
Qes.caf-7H 7H 72.7–91.2 571.7–603.8 GBS_MST4649 GBS_MST4574 4.0 9.1 0.037
BO22 Qes.caf-2H 2H 67.4–71.7 568.8–586.1 GBS_MST1320 GBS_MST1352 6.9 15.3 –0.049
Qes.caf-4H 4H 15.8–19.8 13.2–22.7 GBS_MST2470 GBS_MST2477 6.5 14.5 –0.048
Qes.caf-7H 7H 76.9–85.9 581.1–593.4 GBS_MST4621 GBS_MST4586 4.1 9.3 0.038
FH23 Qes.caf-2H 2H 67.4–72.0 568.8–586.1 GBS_MST1320 GBS_MST1355 7.0 15.5 –0.049
Qes.caf-4H 4H 15.8–19.8 13.2–22.7 GBS_MST2470 GBS_MST2477 6.4 14.2 –0.047
Qes.caf-7H 7H 74.6–85.9 578.2–593.4 GBS_MST4633 GBS_MST4586 4.0 9.1 0.037
BO23 Qes.caf-2H 2H 67.4–71.7 568.8–584.1 GBS_MST1320 GBS_MST1352 7.2 15.8 –0.050
Qes.caf-4H 4H 15.8–19.8 13.2–22.7 GBS_MST2470 GBS_MST2477 6.3 14.0 –0.047
Qes.caf-7H 7H 76.9–85.9 581.1–593.4 GBS_MST4621 GBS_MST4586 4.1 9.2 0.039
BLUPUP Qes.caf-2H 2H 67.4–72.0 568.8–586.1 GBS_MST1320 GBS_MST1355 7.1 15.5 –0.049
Qes.caf-4H 4H 15.8–19.8 13.2–22.7 GBS_MST2470 GBS_MST2477 6.4 14.2 –0.047
Qes.caf-7H 7H 76.9–85.9 581.1–593.4 GBS_MST4621 GBS_MST4586 4.1 9.2 0.038
EA FH22 Qes.caf-2H 2H 54.9–60.3 529.1–542.6 GBS_MST1160 GBS_MST1264 5.4 12.2 –0.083
Qes.caf-4H 4H 15.8–19.8 13.2–22.7 GBS_MST2470 GBS_MST2477 6.6 14.7 –0.091
Qes.caf-7H 7H 91.5–94.5 602.9–613.1 GBS_MST4573 GBS_MST4562 5.6 12.7 0.086
BO22 Qes.caf-2H 2H 54.2–61.4 529.1–542.6 GBS_MST1049 GBS_MST1274 4.7 10.7 –0.078
Qes.caf-4H 4H 15.8–19.8 13.2–22.7 GBS_MST2470 GBS_MST2477 6.5 14.3 –0.090
Qes.caf-7H 7H 91.5–96.6 602.9–613.1 GBS_MST4573 GBS_MST4560 5.0 11.4 0.082
FH23 Qes.caf-2H 2H 54.9–60.3 529.1–542.6 GBS_MST1160 GBS_MST1264 5.2 11.6 –0.080
Qes.caf-4H 4H 15.8–19.8 13.2–22.7 GBS_MST2470 GBS_MST2477 6.7 14.8 –0.091
Qes.caf-7H 7H 91.5–96.3 602.9–613.1 GBS_MST4573 GBS_MST4562 5.4 12.1 0.084
BO23 Qes.caf-2H 2H 54.9–62.0 529.1–542.6 GBS_MST1160 GBS_MST1276 4.5 10.3 –0.077
Qes.caf-4H 4H 15.8–19.8 13.2–22.7 GBS_MST2470 GBS_MST2477 6.2 13.7 –0.089
Qes.caf-7H 7H 91.5–96.3 602.9–613.1 GBS_MST4573 GBS_MST4562 4.5 10.3 0.079
BLUP Qes.caf-2H 2H 54.9–61.4 529.1–542.6 GBS_MST1160 GBS_MST1274 5.0 11.2 –0.079
Qes.caf-4H 4H 15.8–19.8 13.2–22.7 GBS_MST2470 GBS_MST2477 6.5 14.4 –0.090
Qes.caf-7H 7H 91.5–96.3 602.9–613.1 GBS_MST4573 GBS_MST4562 5.2 11.6 0.082
Open in a new tab

#EL embryo length, EW embryo width, EA embryo area, Chr chromosome, FH22, BO22, FH23 and BO23 four independent trials conducted for embryo size, BLUP best linear unbiased prediction, cM centimorgan, PVE phenotype variance explained, Add Additive effect (positive values: alleles for long coleoptile from Morex, negative values: alleles for long coleoptile from AWCS276)

Conditional QTL detection for embryo size

Conditional QTL analysis was conducted using the BLUP datasets and identified five QTL regions associated with embryo size, including three regions previously detected in the traditional QTL analysis and two newly discovered regions. For all three embryo characteristics, Qes.caf-7H remained significant with slight changes in LOD values and phenotypic variation explained (PVE) when conditioned on all kernel traits (Fig. 3). This result suggested that this QTL was potentially independent of kernel traits. Qes.caf-2H and Qes.caf-4H showed no significant effects on EL when conditioned on KL and KA but remained significant when conditioned on KW and TKW, suggesting that these two QTL were probably dependent on KL and KA and independent of KW and TKW. Regarding EW and EA, while the LOD values of Qes.caf-4H did not change significantly (when they were conditional on all kernel traits), those of Qes.caf-2H decreased dramatically (LOD < 2.7) and only remain significant on EW when conditional on KL (Table S2). Two QTL conferring EW, one conferring EL and one QTL conferring EA were newly identified on chromosomes 1H and 6H when removing the effects of KW, KA and TKW. These results indicated that the effects of the newly detected QTL were masked by these kernel traits (Fig. S1, Table S2).

Fig. 3.

Fig. 3

Open in a new tab

Conditional QTL analysis of Qes.caf-7H with kernel-related traits and 1000-kernel weight, showing the changes of LOD value when a embryo length (EL), b embryo width (EW) and c embryo area (EA) was conditional on kernel length (KL), kernel width (KW), kernel area (KA) and 1000-kernel weight (TKW), respectively

Candidate genes underlying the major locus on chromosome 7H

Sequences for 488 genes associated with embryo and kernel size from rice were obtained and blasted against the assemblies of Morex and AWCS276 (Table S3). This orthologous analysis identified four candidate genes in the 7H region, including HORVU.MOREX.r3.7HG0736160, HORVU.MOREX.r3.7HG0742750, HORVU.MOREX.r3.7HG0744200, and HORVU.MOREX.r3.7HG0746000. Genomic comparation of the two parental genotypes detected insertion-deletion (Indel) in all the four candidate genes (Fig. 4). HORVU.MOREX.r3.7HG0736160 contained two deletions (39 bp and 48 bp) and one SNP variation in the first exon and one SNP substitution in each of the exon 2, 14 and 20, respectively (Fig. 4). Two deletions (one 27 bp and the other one 51 bp) were identified in exon 1 of HORVU.MOREX.r3.7HG0744200 between the two parental genotypes, resulting in a 9-amino acid deletion and an 18-amino acid deletion. When comparing the two parents, four Indels (3 bp, 6 bp, 3 bp and 39 bp, respectively) and four SNP variations in the first exon as well as one SNP in the eighth exon and one Indel and three SNPs in the tenth exon were observed from HORVU.MOREX.r3.7HG0746000. The fourth gene HORVU.MOREX.r3.7HG0742750 in AWCS276 contained a 3 bp insertion and three SNP variations in the first exon as well as a 6 bp deletion and two SNP variations in the third exon when compared to Morex. Importantly, all these Indels between the parents resulted in alterations to the protein domain structure and integrity.

Fig. 4.

Fig. 4

Open in a new tab

Gene structures and genome sequence alignment of candidate genes for Qes.caf-7H between Morex and AWCS276

Discussion

Embryo size is a critical trait influencing early seedling vigour, germination rate, overall crop establishment and quality in cereal crops. Despite its importance, the genetic control of embryo size remains poorly understood in barley. To date, only one gene associated embryo size has been reported as a qualitative trait on chromosome 5H derived from a mutagenized barley germplasm (Orman-Ligeza et al. 2020). Thus, all the QTL identified in this study are novel and have not been previously reported. Unlike that study, we for the first time identified three novel QTL regions associated with embryo size from natural variation using direct measurements in barley. Among these QTL, the one located on 7HL chromosome arm remained significant after removing the effects of kernel size. Further comparative genomic analysis identified four candidate genes in the 7H interval. These genes all contain large Indels leading to amino acid changes, making them primary targets for future cloning efforts to identify the gene(s) underlying this locus. Our study provides valuable insights into the genetic architecture of barley embryo size and offers promising targets for future research and breeding efforts.

Unlike the oblique orientation of wheat embryo, barley embryo lies relatively flat within the seed. However, most barley is ‘covered barley’, which has a tough, inedible outer hull surrounding the barley kernel. This hull is beneficial for yield and is in high demand for malting due to its role in the malting process (Hebelstrup et al. 2017; Raj et al. 2023), nevertheless the presence of this hull poses a significant challenge for understanding the complex genetics underlying embryo size. To overcome this challenge, we carefully removed the hull from eight well-developed kernels, primarily located in the centre of the spike to enable accurate measurement of embryo size. This method, previously used to identify major and stable QTL conferring embryo size in wheat (Wang et al. 2023), was slightly modified for barley. By adhering the hump of barley kernels to double-sided tape in Petri dishes, we were able to effectively measure embryo size using stereomicroscope and identified consistent QTL conferring this trait.

Given that the embryo is part of seed, it is understandable that significantly positive correlations were detected between them. This suggests that selecting for larger seeds can lead to larger embryos (Wang et al. 2023). However, conditional QTL analysis revealed that Qes.caf-7H remained significant with only slight changes in LOD values and PVE when removing the effects of kernel size, suggesting its potential independence from kernel size. Therefore, Qes.caf-7H could be a valuable target for further improving embryo size in the breeding programs. Moreover, the present study also detected significantly positive relationships between embryo size and early vigour traits. For example, EW was significantly correlated with S3LL, S3LW, S3LA and MRL (Table 2). Previous studies have reported that larger embryos are associated with increased leaf width and seedling root growth (Rebetzke et al. 2008; Maydup et al. 2012; Richards and Lukacs 2002; Hendriks et al. 2022). As reported by Rebetzke et al. (2022), embryo size independent of kernel size may be a crucial determinant of the final number and size of seminal roots in wheat. By incorporating the novel QTL identified in this study, breeding programs can potentially improve resource capture, weed competitiveness, and ultimately improved crop performance and grain yield under water-limited conditions (Li et al. 2019; Rebetzke et al. 2022).

Orthologous and comparative analysis identified four candidate genes within the Qes.caf-7H region: HORVU.MOREX.r3.7HG0736160, HORVU.MOREX.r3.7HG0744200, HORVU.MOREX.r3.7HG0746000, HORVU.MOREX.r3.7HG0742750. These genes are orthologous to TWINKLE/RECA3, HOS59, AP2-4 and GW6a/OsglHAT1 in rice, respectively. It has been reported that TWINKLE/RECA3 interacting with OsmtSSB1 impacted kernel size and weight (Li et al. 2021), which may in turn impact embryo size due to their correlations. HOS59, a rice plant knotted1-like homeobox (KNOX) Class II subfamily gene, is known to regulate grain size and embryonic/postembryonic development in various plants (Tsuda and Hake 2015; Sheng et al. 2022). Previous studies have also reported AP2 gene family mediates kernel size by affecting embryo, endosperm and seed coat development in Arabidopsis (Jofuku et al. 2005; Ohto et al. 2005, 2009). In addition, GW6a/OsglHAT1 in rice is associated with kernel length and weight, encoding a novel GNAT-like protein with intrinsic histone acetyltransferase activity (Song et al. 2015). It was shown that GW6a/OsglHAT1 interacts with HDR3 (a ubiquitin-containing receptor) regulating kernel size by positively regulating cell number and grain filling rate (Gao et al. 2021). Further investigation is warranted to elucidate the precise functions of these candidate genes and their interactions in regulating embryo development.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file1 (XLSX 30 kb) (30.7KB, xlsx)
Supplementary file2 (DOCX 203 kb) (203.6KB, docx)

Acknowledgements

This study was supported by CSIRO Agriculture and Food (Grant No. R-10191-01). XC and YZ are grateful to the China Scholarship Council for financial support of their visit to CSIRO.

Author contributions

CL and ZZ conceived the study and designed the experiments. XC, YZ, and ZZ conducted the experiments, collected data. JM, JM, ZL, ZS, HH and ZZ conducted the data analyse. XC, YZ and ZZ wrote the article with contributions from all the authors. HH, CL and ZZ supervised and complemented the writing. All authors approved the final manuscript.

Funding

Open access funding provided by CSIRO Library Services. This study was supported by CSIRO Agriculture and Food.

Data availability

The datasets generated during and/or analysed during the current study are available from the corresponding author upon reasonable request.

Declarations

Conflict of interest

The authors declare that there are no conflicts of interest.

Footnotes

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Xiaoxia Chen and Zhouyang Su have contributed equally to this work.

Contributor Information

Chunji Liu, Email: chunji.liu@csiro.au.

Zhi Zheng, Email: zhi.zheng@csiro.au.

References

  1. Buchfink B, Reuter K, Drost HG (2021) Sensitive protein alignments at tree-of-life scale using DIAMOND. Nat Methods 18:366–368. 10.1038/s41592-021-01101-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Chen JA, Zhou JG, Qu XR, Wang SR, Tang HP, Lan XJ, Wei YM, Ma J, Jiang Y, Tang LW (2023) Mapping and analysis of QTL for embryo size-related traits in tetraploid wheat. Sci Agric Sin 56(2):203–216. 10.3864/j.issn.0578-1752.2023.02.001 [Google Scholar]
  3. Finch-Savage WE, Bassel GW (2016) Seed vigour and crop establishment: extending performance beyond adaptation. J Exp Bot 67:567–591. 10.1093/jxb/erv490 [DOI] [PubMed] [Google Scholar]
  4. Forbis TA, Floyd SK, de Queiroz A (2002) The evolution of embryo size in angiosperms and other seed plants: implications for the evolution of seed dormancy. Evolution 56:2112–2125. 10.1111/j.0014-3820.2002.tb00137.x [DOI] [PubMed] [Google Scholar]
  5. Gao Q, Zhang N, Wang WQ, Shen SY, Bai C, Song XJ (2021) The ubiquitin-interacting motif-type ubiquitin receptor HDR3 interacts with and stabilizes the histone acetyltransferase GW6a to control the grain size in rice. Plant Cell 33:3331–3347. 10.1093/plcell/koab194 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Gao S, Su Z, Ma J, Ma J, Liu C, Li H, Zheng Z (2024) Identification of a novel and plant height-independent QTL for coleoptile length in barley and validation of its effect using near isogenic lines. Theor Appl Genet 137:1–11. 10.1007/s00122-024-04561-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Hebelstrup KH, Nielsen MM, Carciofi M, Andrzejczak O, Shaik SS, Blennow A, Palcic MM (2017) Waxy and non-waxy barley cultivars exhibit differences in the targeting and catalytic activity of GBSS1a. J Exp Bot 68:931–941. 10.1093/jxb/erw503 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Hendriks P-W, Gurusinghe S, Ryan PR, Rebetzke GJ, Weston LA (2022) Competitiveness of early vigour wheat (Triticum aestivum L) genotypes is established at early growth stages. Agronomy 12:377. 10.3390/agronomy12020377 [Google Scholar]
  9. Jofuku KD, Omidyar PK, Gee Z, Okamuro JK (2005) Control of seed mass and seed yield by the floral homeotic gene APETALA2. Proc Natl Acad Sci USA 102:3117–3122. 10.1073/pnas.040989310 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Jung WY, Kim SG, Lee JS, Kim HK, Son BG, Kim JW, Suh JW (2017) Effect of feeding high gamma-aminobutyric acid-containing giant embryo black sticky rice (Oryza sativa L.) on anxiety-related behavior of C57BL/6 mice. J Med Food 20:777–781. 10.1089/jmf.2017.3928 [DOI] [PubMed] [Google Scholar]
  11. Katral A, Hossain F, Gopinath I, Chand G, Mehta BK, Kamboj MC, Zunjare RU, Yadava DK, Muthusamy V (2023) Genetic dissection of embryo size and weight related traits for enhancement of kernel oil in maize. Plant Physiol Biochem 197:107668. 10.1016/j.plaphy.2023.107668 [DOI] [PubMed] [Google Scholar]
  12. Lee YM, Han SI, Won YJ, Lee E, Park E, Hwang SY, Yeum KJ (2016) Black rice with giant embryo attenuates obesity-associated metabolic disorders in ob/ob mice. J Agric Food Chem 64:2492–2497. 10.1021/acs.jafc.5b05361 [DOI] [PubMed] [Google Scholar]
  13. Lee G, Piao R, Lee Y, Kim B, Seo J, Lee D, Jang S, Jin Z, Lee C, Chin JH (2019) Identification and characterization of LARGE EMBRYO, a new gene controlling embryo size in rice (Oryza sativa L.). Rice 12:1–12. 10.1186/s12284-019-0277-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Li X, Ingvordsen C, Weiss M, Rebetzke GJ, Condon AG, James RA, Richards RA (2019) Deeper roots associated with cooler canopies, higher NDVI and greater yield in three wheat populations grown on stored soil water. J Exp Bot 24:4963–4974. 10.1093/jxb/erz232 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Li DQ, Wu XB, Wang HF, Feng X, Yan SJ, Wu SY, Liu JX, Yao XF, Bai AN, Zhao H (2021) Defective mitochondrial function by mutation in THICK ALEURONE 1 encoding a mitochondrion-targeted single-stranded DNA-binding protein leads to increased aleurone cell layers and improved nutrition in rice. Mol Plant 14:1343–1361. 10.1016/j.molp.2021.05.016 [DOI] [PubMed] [Google Scholar]
  16. Li X, Wang M, Zhang R, Fang H, Fu X, Yang X, Li J (2022) Genetic architecture of embryo size and related traits in maize. Crop J 10:204–215. 10.1016/j.cj.2021.03.007 [Google Scholar]
  17. Liu M, Li Y, Ma Y, Zhao Q, Stiller J, Feng Q, Tian Q, Liu D, Han B, Liu C (2020) The draft genome of a wild barley genotype reveals its enrichment in genes related to biotic and abiotic stresses compared to cultivated barley. Plant Biotechnol J 18:443–456. 10.1111/pbi.13210 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. López-Castañeda C, Richards R, Farquhar G, Williamson R (1996) Seed and seedling characteristics contributing to variation in early vigor among temperate cereals. Crop Sci 36:1257–1266. 10.2135/cropsci1996.0011183X003600050031x [Google Scholar]
  19. Mascher M, Wicker T, Jenkins J, Plott C, Lux T, Koh CS, Ens J, Gundlach H, Boston LB, Tulpová Z (2021) Long-read sequence assembly: a technical evaluation in barley. Plant Cell 33:1888–1906. 10.1093/plcell/koab077 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Maydup ML, Graciano C, Guiamet JJ, Tambussi EA (2012) Analysis of early vigour in twenty modern cultivars of bread wheat (Triticum aestivum L.). Crop Pasture Sci 63:987–996. 10.1071/CP12169 [Google Scholar]
  21. Moore C, Rebetzke G (2015) Genomic regions for embryo size and early vigour in multiple wheat (Triticum aestivum L.) populations. Agronomy 5:152–179. 10.3390/agronomy5020152 [Google Scholar]
  22. Ohto M-a, Fischer RL, Goldberg RB, Nakamura K, Harada JJ (2005) Control of seed mass by APETALA2. Proc Natl Acad Sci USA 102:3123–3128. 10.1073/pnas.0409858102 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Ohto M-a, Floyd SK, Fischer RL, Goldberg RB, Harada JJ (2009) Effects of APETALA2 on embryo, endosperm, and seed coat development determine seed size in Arabidopsis. Sex Plant Reprod 22:277–289. 10.1007/s00497-009-0116-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Orman-Ligeza B, Borrill P, Chia T, Chirico M, Doležel J, Drea S, Karafiátová M, Schatlowski N, Solomon CU, Steuernagel B (2020) LYS3 encodes a prolamin-box-binding transcription factor that controls embryo growth in barley and wheat. J Cereal Sci 93:102965. 10.1016/j.jcs.2020.102965 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Raj R, Shams R, Pandey VK, Dash KK, Singh P, Bashir O (2023) Barley phytochemicals and health promoting benefits: a comprehensive review. J Agric Food Res 14:100677. 10.1016/j.jafr.2023.100677 [Google Scholar]
  26. Rebetzke G, López-Castañeda C, Acuna TB, Condon A, Richards R (2008) Inheritance of coleoptile tiller appearance and size in wheat. Aust J Agric Res 59:863–873. 10.1071/AR07397 [Google Scholar]
  27. Rebetzke G, Zhang H, Ingvordsen C, Condon A, Rich S, Ellis M (2022) Genotypic variation and covariation in wheat seedling seminal root architecture and grain yield under field conditions. Theor Appl Genet 135:3247–3264. 10.1007/s00122-022-04183-z [DOI] [PubMed] [Google Scholar]
  28. Richards R, Lukacs Z (2002) Seedling vigour in wheat-sources of variation for genetic and agronomic improvement. Aust J Agric Res 53:41–50. 10.1071/AR00147 [Google Scholar]
  29. Sheng M, Ma X, Wang J, Xue T, Li Z, Cao Y, Yu X, Zhang X, Wang Y, Xu W (2022) KNOX II transcription factor HOS59 functions in regulating rice grain size. Plant J 110:863–880. 10.1111/tpj.15709 [DOI] [PubMed] [Google Scholar]
  30. Song XJ, Kuroha T, Ayano M, Furuta T, Nagai K, Komeda N, Segami S, Miura K, Ogawa D, Kamura T (2015) Rare allele of a previously unidentified histone H4 acetyltransferase enhances grain weight, yield, and plant biomass in rice. Proc Natl Acad Sci USA 112:76–81. 10.1073/pnas.142112711 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Tsuda K, Hake S (2015) Diverse functions of KNOX transcription factors in the diploid body plan of plants. Curr Opin in Plant Biol 27:91–96. 10.1016/j.pbi.2015.06.015 [DOI] [PubMed] [Google Scholar]
  32. Van Ooijen J, Kyazma B (2009) MapQTL 6 Software for the mapping of quantitative trait loci in experimental populations of diploid species Kyazma BV. Wageningen, Netherlands [Google Scholar]
  33. Visscher PM, Thompson R, Haley CS (1996) Confidence intervals in QTL mapping by bootstrapping. Genetics 143:1013–1020. 10.1093/genetics/143.2.1013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Voorrips RE (2002) MapChart: software for the graphical presentation of linkage maps and QTLs. J Hered 93:77–78. 10.1093/jhered/93.1.77 [DOI] [PubMed] [Google Scholar]
  35. Walker CK, Ford R et al (2013) The detection of QTLs in barley associated with endosperm hardness grain density grain size and malting quality using rapid phenotyping tools. Theor Appl Genet 126:2533–2551. 10.1007/s00122-013-2153-2 [DOI] [PubMed] [Google Scholar]
  36. Wang S, Wang T, Xuan Q, Qu X, Xu Q, Jiang Q, Pu Z, Li Y, Jiang Y, Chen G (2023) Major and stably expressed QTL for traits related to the mature wheat embryo independent of kernel size. Theor Appl Genet 136:90. 10.1007/s00122-023-04346-6 [DOI] [PubMed] [Google Scholar]
  37. Yuan N, Wang J, Zhou Y, An D, Xiao Q, Wang W, Wu Y (2019) EMB-7L is required for embryogenesis and plant development in maize involved in RNA splicing of multiple chloroplast genes. Plant Sci 287:110203. 10.1016/j.plantsci.2019.110203 [DOI] [PubMed] [Google Scholar]
  38. Zhao Z, Rebetzke GJ, Zheng B, Chapman SC, Wang E (2019) Modelling impact of early vigour on wheat yield in dryland regions. J Exp Bot 70:2535–2548. 10.1093/jxb/erz069 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Zheng Z, Wang HB, Chen GD, Yan GJ, Liu CJ (2013) A procedure allowing up to eight generations of wheat and nine generations of barley per annum. Euphytica 191:311–316. 10.1007/s10681-013-0909-z [Google Scholar]
  40. Zheng Z, Hu H, Gao S, Zhou H, Luo W, Kage U, Liu C, Jia J (2022) Leaf thickness of barley: genetic dissection, candidate genes prediction and its relationship with yield-related traits. Theor Appl Genet 135:1843–1854. 10.1007/s00122-022-04076-1 [DOI] [PubMed] [Google Scholar]
  41. Zhou H, Luo W, Gao S, Ma J, Zhou M, Wei Y, Zheng Y, Liu Y, Liu C (2021) Identification of loci and candidate genes controlling kernel weight in barley based on a population for which whole genome assemblies are available for both parents. Crop J 9:854–861. 10.1016/j.cj.2020.07.010 [Google Scholar]
  42. Zhu J (1995) Analysis of conditional genetic effects and variance components in developmental genetics. Genetics 141:1633–1639. 10.1093/genetics/141.4.1633 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary file1 (XLSX 30 kb) (30.7KB, xlsx)
Supplementary file2 (DOCX 203 kb) (203.6KB, docx)

Data Availability Statement

The datasets generated during and/or analysed during the current study are available from the corresponding author upon reasonable request.

Articles from TAG. Theoretical and Applied Genetics. Theoretische Und Angewandte Genetik are provided here courtesy of Springer

RESOURCES