Background

Wheat quality is mainly based on wheat gluten protein, which has a wide range of effects on dough properties, and protein content and composition are key quality-determining parameters [1]. Gluten is a typical water-insoluble protein polymer composed of disulfide bonds and noncovalent hydrogen bonds between polymeric glutenin and monomeric gliadins [2]. Glutenin influences the strength and elasticity of wheat dough [3], whereas gliadins are responsible for its extensibility and viscosity [4]. Glutenin proteins include two major subunits, high-molecular-weight glutenin subunits (HMW-GSs) and low-molecular-weight glutenin subunits (LMW-GSs), which are the major proteins affecting the end-use quality of wheat [3, 5]. The genes encoding the HMW-GS, namely, Glu-A1, Glu-B1, and Glu-D1, are located on the long arms of chromosomes 1A, 1B, and 1D, respectively [4, 6]. Each Glu-1 locus consists of two tightly linked genes, designated as the x-type and y-type subunits that are highly conserved, contain repeated domains, and exhibit multiple alleles [7,8,9]. HMW-GSs account for only about 5%–10% of grain protein, but allelic variations in HMW-GSs have been reported to account for up to 50–70% of the variation in bread-making quality [10,11,12,13,14,15]. Three, 11, and six alleles at the Glu-A1, Glu-B1, and Glu-D1 loci, respectively, were systematically identified in 1983 by isolating HMW-GSs using sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) [16]. Extensive polymorphisms were detected at all three Glu-1 loci. The degree of polymorphism has continued to increase with analyses of different landraces, wild species, and wheat relatives [5, 17]. The Grain Genes 2.0 database identified 22 alleles for Glu-A1, 52 for Glu-B1, and 36 for Glu-D1 [18].

Various mass spectrometry techniques and SDS-PAGE analyses have been developed and used to identify HMW-GSs [19,20,21,22]. Some HMW-GSs with similar mobilities are indistinguishable on SDS-PAGE [23, 24], but alleles such as Glu-1Bx7 and Glu-1Bx7*, Glu-1By8 and Glu-1By8*, and Glu-1Dx2 and Glu-1Ax2* show differences in surface hydrophobicity on ultra-performance liquid chromatography (UPLC) [25]. However, despite small differences in electrophoretic mobility between the Glu-1Bx7 and Glu-1Bx7* subunits, these subunits cannot be differentiated on the basis of the elution time in reversed-phase high-performance liquid chromatography (RP-HPLC) [26].

Molecular markers such as allele-specific polymerase chain reaction (AS-PCR) based on single-nucleotide polymorphisms (SNPs) and insertions and deletions (indels) have been used for fast and reliable screening of HMW-GSs in wheat breeding. Several AS-PCR markers have been developed for different HMW-GS genes, including Glu-A1, Glu-B1, and Glu-D1 [23, 27,28,29,30,31,32,33,34,35,36,37,38,39]. Several indels and duplications have also been reported in the promoter regions of Glu-B1 alleles. The 43 bp indel, 185 bp indel, and 54 bp duplication of the promoter can distinguish Glu-1Bx6, Glu-1Bx7, Glu-1Bx13, Glu-1Bx14, and Glu-1Bx20 from other Glu-1Bx genes. Glu-1Bx7 OE and some types of Glu-1Bx6 and Glu-1Bx14 with a 43 bp insertion in the promoter region [40] can be selected using a marker reported by Ragupathy et al. [33]. The difference between Glu-1Bx7 and Glu-1Bx7 OE is a 43 bp insertion in the promoter region and gene duplication [23, 33]. Glu-1Bx7* shows an 18 bp deletion in the coding region of the gene. Agarose gel-based DNA markers have been used to accurately discriminate the Glu-B1 locus subunits Glu-1Bx7 and Glu-1Bx7*, Glu-1Bx7 and Glu-1Bx17, Glu-1By9 and Glu-1By8, Glu-1Bx6 and non-Glu-1Bx6, and Glu–1Bx23 and Glu-1Bx23* by size polymorphisms of 18, 108, 45, 15, and 18 bp, respectively [24, 31, 32, 35, 37, 38]. In addition, AS-PCR markers have been used to discriminate between Glu-1By8 and Glu-1By18 using SNPs [24], and two Glu-1By18-specific SNP-based markers were developed by Liang et al. [34].

Identification of HMW glutenin in wheat remains a high priority in wheat-breeding programs to determine which genotypes should be used in breeding programs because of the large variations in allelic composition among varieties. Although new genotypes continue to be reported, PCR markers cannot distinguish among all genotypes. In addition, the discriminant information of previously reported markers was limited to only a few alleles. SDS-PAGE and HPLC/UPLC analyses are not only time-consuming, but in some cases size-indistinguishable, necessitating the development of PCR markers that can distinguish the genotype of a breed. In this study, we developed PCR markers to distinguish nine Glu-1Bx alleles and seven Glu-1By alleles located at the Glu-B1 locus by using several specific SNPs and indels in 25 wheat varieties. These markers could be useful for marker-assisted selection of Glu-B1 alleles in wheat quality improvement programs.

Methods

Plant materials

A total of 25 wheats were provided by the Korean Agricultural Culture Collection (KACC) (htpp://genebank. rda. go. kr). Formal identification of wheat varieties used in this study were performed by KACC. Three IT numbers (registration numbers for plants at the NARO Institute of Agrobiological Sciences National) were assigned to the Avalone cultivar. Information of HMW-GS allele compositions and IT numbers is listed in Table 1. The respective countries of origin and published and corrected Glu-B1 alleles are also indicated (Table 1).

Table 1 List of wheat varieties used in this study. The respective countries of origin and published and corrected Glu-B1 alleles are indicated
Full size table

Genomic DNA extraction

A total of 27 hexaploid wheat (Triticum aestivum L.) plants were grown in a Petri dish with moisture paper. Genomic DNA was extracted from the leaf tissue using the Higene™ Genomic DNA Prep kit (solution type; BioFACT, Korea). The quality of the genomic DNA was assessed using a NanoDrop 1000 spectrophotometer (Thermo Scientific, MA, USA), and the integrity of the DNA was checked using 1% agarose gel electrophoresis.

Sequence alignment and phylogenetic tree analysis

Sequences of Glu-1Bx6 (KX454509.1), Glu-1Bx7 (BK006773.1), Glu-1Bx7* (X13927.3), Glu-1Bx13 (EF540764.1), Glu-1Bx14 (AY367771.1 and KF733216.1), Glu-1Bx17* (KF547469.1), Glu-1Bx17 (KC254854.1), Glu-1Bx20 (AJ437000.2), Glu-1Bx23 (AY553933.1), and Glu-1Bx23* (KF995273.1) from Glu-1Bx and sequences of Glu-1By8 (JN255519.1), Glu-1By9 (X61026.1), Glu- 1By15 (EU137874.1), Glu-1By15* (KJ579440.1), Glu-1By16 (EF540765.1), Glu-1By18 (KF430649.1), and Glu-1By20 (KU886033.1) from Glu-1By were collected from National Center for Biotechnology information (NCBI). The gene sequence of Glu-1By8* was not available in the public databases. In this study, Glu-1Bx14 alleles with accession numbers KF733216.1 and AY367771.1 were named Glu-1Bx14 (−) and Glu-1Bx14 (+), respectively. Multiple sequence alignments of the full-length nucleotide sequences of the Glu-1Bx and Glu-1By alleles were performed using Clustal Omega (www.ebi.ac.uk/Tools/msa/Clustalo). Phylogenetic tree was generated using Molecular Evolutionary Genetics Analysis version 11 (MEGA11) [41]. Bootstrap values were calculated on the basis of 1,000 replications.

PCR conditions

PCR was performed under the following conditions: 5 min at 95 °C, 30 cycles of 30 s at 94 °C, 30 s at 55–68 °C, and 30 s at 72℃ on a thermal cycler (Applied Biosystem, MA, USA) with 20 μL reaction volumes for each sample containing 50 ng of template DNA, 10 pmol of each primer with 1 × master mix solution (i-MAX II DNA polymerase, Intronbio, Korea) or 0.1 μL (5 units/μL) of Takara Ex Taq DNA polymerase, 1.6 μL of dNTP mixture (2.5 mM each), and 1 × Ex Taq Buffer (Mg2+ free) (Takara, Japan). PCR products were electrophoresed on 1%–3% agarose gels, and visualized using a Davinch-K Gel imaging system (Davinch-K, Korea). The primers and PCR conditions are listed in Table 2.

Table 2 Primers used in this study for detection of Glu-1Bx alleles
Full size table

Glutenin protein analysis using SDS-PAGE

Glutenin was extracted from single wheat grains using a previously reported HMW-GS extraction protocol [42, 43]. Protein (approximately 10 μg) extracted from grains of different varieties was separated using 10% SDS-PAGE and visualized using Coomassie Brilliant Blue R + 250 staining solution (Bio-Rad, CA, USA).

Glutenin analysis using ultra-performance liquid chromatography

Glutenin analysis was performed by crushing single grains of wheat, and the glutenin extraction method was performed as described previously [42, 43]. The extracted glutenin was analyzed using a UPLC system (Alliance e2695, Waters Corp., MA, USA) with an ACQUITY UPLC Peptide BEH C18 column (300A, 1.7 μm, 2.1 mm × 50 mm) and a photodiode array detector. The mobile phases were H2O containing 0.1% trifluoroacetic acid (A) and acetonitrile containing 0.1% trifluoroacetic acid (B). The injection volume of the dissolved samples was 3 μL and the flow rate was 0.55 μL/min. The solvent gradient was changed from 21 to 47% (B) from 0 to 30 min, and the column and sample temperatures were set to 55 °C and 10 °C, respectively.

Results

Sequence alignment and phylogenetic analysis of Glu-1Bx subunits

The nucleotide sequences of 11 Glu-1Bx subunits (Glu-1Bx6, Glu-1Bx7*, Glu-1Bx7, Glu-1Bx13, Glu-1Bx14*, two Glu-1Bx14, Glu-1Bx17, Glu-1Bx20, Glu-1Bx23*, and Glu-1Bx23) were aligned using Clustal Omega (Fig. S1). Among these, the base sequences of Glu-1Bx14 (AY367771.1) and Glu-1Bx20 showed a high degree of identity (99%). The base sequences of Glu-1Bx14* and Glu-1Bx23* also showed high identity (99%). In contrast, the base sequences of the two accessions of Glu-1Bx14 showed a relatively low identity of 96% (Fig. S1). The complete coding sequences of the 11 HMW-GS genes were used to construct a neighbor-joining tree to investigate the phylogenetic relationships among the HMW-GS Glu-1Bx genes (Fig. 1). The genes encoding Glu-1Bx14 (+) and Glu-1Bx20, Glu-1Bx14* and Glu-1Bx23*, and Glu-1Bx7* and Glu-1Bx17 were closely related (Fig. 1), indicating similar structural features and close phylogenetic evolutionary relationships.

Fig. 1
figure 1

Phylogenetic tree of 11 HMW-GS Glu-1Bx alleles constructed using complete coding DNA sequences. HMW-GS Glu-1Bx alleles included Glu-1Bx6 (KX454509.1), Glu-1Bx7 (BK006773.1), Glu-1Bx7* (X13927.3), Glu-1Bx13 (EF540764.1), Glu-1Bx14 (−) (KF7333216.1), Glu-1Bx14.(+) (AY367771.1), Glu-1Bx14* (KJ579439.1), Glu-1Bx17 (KC254854.1), Glu-1Bx20 (AJ437000.2), Glu-1Bx23 (AY553933.1), and Glu-1Bx23* (KF995273.1)

Full size image

Detection of Glu-1Bx subunits

Eight Glu-1Bx subunits were primarily investigated using PCR-based markers in the 24 cultivars (Table 1). Previously reported molecular markers identifying the Glu-1Bx7 allele were > 2,000 bp (2,373/2,500 bp) in size [36, 44]. Therefore, in this study, we designed a PCR primer set (PS1) to detect Glu-1Bx7 homologs (Glu-1Bx7, Glu-1Bx7*, and Glu-1Bx7 OE). However, Glu-1Bx17 (Joongmo2008, Suwon105, and Suwon92) and Glu-1Bx14 (−) (Troll, Hanno, and Imbros) were also detected in PS1 (Fig. 2a). Next, to examine Glu-1Bx7 OE, we detected a 43 bp insertion in the promoter region by using the marker reported by Butow et al. [23] (PS2, Fig. 2b). In a previous report, a 43 bp insertion in the promoter was observed in some types of Glu-1Bx6 and Glu-1Bx14. Additionally, some types of Glu-1Bx14 and Glu-1Bx20 contain a 185 bp insertion in the promoter region [40, 45]. Therefore, the PCR products were expected to be of four sizes: 520 bp (non-43 bp insertion), 563 bp (43 bp insertion), 705 bp (185 bp insertion), and 748 bp (43 bp and 185 bp insertions). However, a 43 bp insertion was detected in the promoter region of Glu-1Bx7 OE (Chisolm, MT8191, and KS85WGRC01), but not in Glu-1Bx6 (Avalon) or Glu-1Bx14 (−) (Troll, Hanno, and Imbros). In addition, in Glu-1Bx20 (Suwon15, Suwon28, and Suwon42), only 185 bp was detected without the 43 bp insertion (Fig. 2b). Next, two Glu-1Bx7 gene duplication markers (left- and right-junction markers) were tested to confirm the cultivars with the 43 bp insertion contained Glu-1Bx7 OE [33]. Gene duplication was detected in three cultivars (Chisolm, MT8191, and KS85WGRC01), suggesting that they contained Glu-1Bx7 OE (Fig. 2c, d). Therefore, we developed a PCR-based marker that detects a 185 bp insertion without a 43 bp insertion in the promoter (Fig. 2e, Fig. S2). Similar to the marker reported by Xu et al. [31] (PS6, Fig. 2f), this marker showed good detection ability for the 185 bp insertion in Glu-1Bx20 (Suwon15, Suwon28, and Suwon42).

Fig. 2
figure 2

PCR analysis of Glu-1Bx alleles in wheat varieties. The numbers above the figure are the same as the variety numbers in Table 1. The primer set numbers in the figure are the same as in Table 2. DM, DNA size marker; PS, primer set. Arrows indicate indels

Full size image

One of the Glu-1Bx7 variants, Glu-1Bx7*, is characterized by an 18 bp deletion in the repeat domain corresponding to an additional hexapeptide motif [46], and a Glu-1Bx7* detection marker using this 18 bp deletion has been reported by Espi et al. [35]. The reverse primer for Glu-1Bx7* reported by Espi et al. [35] was modified in this study. However, Glu-1Bx17 also lacks an 18 bp repeat domain corresponding to an extra hexapeptide motif and was detected by PS7 (Fig. 2g). In addition to the 18 bp deletion, Glu-1Bx17 is characterized by a 108 bp deletion in the coding sequence; therefore, a marker capable of distinguishing Glu-1Bx17 with a 108 bp deletion from other Glu-1Bx genes has been reported [28, 31] (PS8, Fig. 2h). With this marker, Glu-1Bx6 was detected to be 45 bp larger than other Glu-1Bx genes except for Glu-1Bx17. Additionally, we found that Gobun, Eunpa, and Anbaek contained Glu-1Bx7* instead of Glu-1Bx7 [47, 48] (Fig. 2g).

Detection of the Glu-1Bx6 subunit

Glu-1Bx6 was detected as a larger PCR fragment than the other Glu-1Bx alleles in PS8. Additionally, Glu-1Bx6 was distinguished from other Glu-1Bx, except for Glu-1Bx20, by a 15 bp insertion in the coding regions [32] (PS9, Fig. 2i). However, Glu-1Bx20 also contained a 15 bp insertion in the coding region (Fig. 2i). Therefore, a dominant marker for Glu-1Bx6 was developed to detect a size of 457 bp using SNP (PS10). As shown in Fig. 2j, the PCR products from three Avalon cultivars were detected using this marker.

Detection of Glu-1Bx13 alleles

The tandem 54 bp duplication at position –400 of the promoter region contains a “cereal box”, which has been implicated in seed-specific expression [49]. However, the promoter region of Glu-1Bx13 does not contain a 54 bp replication [40]. To discriminate of Glu-1Bx13 from other Glu-1Bx alleles in the promoter, we developed a co-dominant primer set to detect Glu-1Bx13 and non-Glu-1Bx13 with sizes of 365 and 419 bp, respectively (PS11, Fig. 2k). Three cultivars (Jeokjoong, Baekjoong, and Joeun) showed a 54 bp deletion in comparison with the other cultivars. In addition, we developed a PCR primer set to detect the Glu-1Bx13 coding region. This primer set dominantly detected Glu-1Bx13 with a size of 254 bp (PS12, Fig. 2l).

Detection of the Glu-1Bx20 subunit

Glu-1Bx20 contains a 185 bp insertion in the promoter region and was detected by PS2, PS5, and PS6. However, some types of Glu-1Bx14 have been reported to contain 185 bp insertion in the promoter region [40, 45]. Therefore, we developed a set of primers (PS13) for the detection of the Glu-1Bx20 coding region with a size of 501 bp using an SNP (Fig. 2m). This primer set specifically detected Glu-1Bx20 in three cultivars (Suwon15, Suwon28, and Suwon42).

Detection of Glu-1Bx14 subunits

Two German bread wheat cultivars, Hanno and Imbros, carry Glu-1Bx14 (−) along with the Glu-1By15 subunits [21]. The Glu-1Bx14 (+) sequence was highly similar with to that of Glu-1Bx20 [31, 39]. However, the three variants (Troll, Hanno, and Imbros) not only did not contain insertions of 43 bp and 185 bp in the promoter region but were also specifically detected by PS1 in the detection of Glu-1Bx7 homologs (Fig. 2). Since two accession numbers were registered in NCBI, we compared the sequences corresponding to the accession numbers AY367771 and KF733216 (Fig. S3). These two accession numbers showed 94% similarity in the nucleotide sequence, and the KF733216 sequence showed an 18 bp deletion in the coding sequence in comparison with other Glu-1Bx alleles. Next, we developed a PCR marker for detection of Glu-1Bx14 (−) using the 18 bp indel (PS14, Fig. 2n). This marker differentiated Glu-1Bx14 (−) from Glu-1Bx7 homologs and Glu-1Bx17. Three cultivars (Troll, Hanno, and Imbros) showed a 18 bp deletion. We also tested RANEE (Glu-1Bx14 + Glu-1By15), a cultivar containing Glu-1Bx14. A 185 bp insertion was detected in the promoter region of RANEE, similar to Suwon15 (PS2), but an 18 bp deletion was not detected in the coding region (PS7, PS14), indicating Glu-1Bx14 (+) (Fig. 3a). However, both Glu-1Bx20 and Glu-1Bx14 (+) were detected using the Glu-1Bx20 detection marker PS13. Glu-1Bx14 (+) and Glu-1Bx20 contain several SNPs in the coding sequence but have 99% nucleotide sequence identity; therefore, in this study, several AS-PCR markers were tested, but Glu-1Bx14 (+) and Glu-1Bx20 could not be distinguished.

Fig. 3
figure 3

Analysis of two Glu-1Bx14 alleles. (A), PCR amplification of six cultivars. DM, DNA size marker. Allows indicate indels. (B), HMW-GS from four cultivars identified by SDS-PAGE. PM, protein size marker; KK, Keumkang; Ha, Hanno; RA, RANEE. Arrows indicate Glu-1Bx and Glu-1By. (C), HMW-GS from three cultivars identified by UPLC. AU, arbitrary units; RT, retention time

Full size image

In SDS-PAGE, the size of Glu-Bx14(−) was different from that of Glu-1Bx14(+) (Fig. 3b), and the extraction time of Glu-Bx14(−) was faster than of Glu-Bx14(+) (Fig. 3c).

Sequence alignment and phylogenetic analysis of Glu-1By subunits

The nucleotide sequences of seven Glu-1By subunits (Glu-1By8, Glu-1By9, Glu-1By15, Glu-1By15*, Glu-1By16, Glu-1By18, and Glu-1By20) were aligned (Fig. S4). Of these, Glu-1By8 and Glu-1By18, and Glu-1By15 and Glu-1By20 were highly conserved with 99% sequence identity. To investigate the phylogenetic relationships among the HMW-GS Glu-1By genes, the complete coding sequences of the seven HMW-GS genes were used to construct a neighbor-joining tree (Fig. 4). The genes encoding Glu-1By15 and Glu-1By20 were more closely related. Additionally, the genes encoding Glu-1By8 and Glu-1By18 were also more closely related, indicating similar structural features and close phylogenetic evolutionary relationships (Fig. 4).

Fig. 4
figure 4

Phylogenetic tree of seven HMW-GS Glu-1By alleles constructed using complete coding DNA sequences. HMW-GS alleles included Glu-1By8 (JN255519.1), Glu-1By18 (KF430649.1), Glu-1By9 (X61026.1), Glu-1By15 (EU137874.1), Glu-1By15* (KJ579440.1), Glu-1By16 (EF540765.1), and Glu-1By20 (KU886033.1)

Full size image

Detection of Glu-1By8 and Glu-1By18 subunits

Lei et al. [28] reported two markers (PS15 and PS16) that could differentiate Glu-1By8 from Glu-1Bx8* and Glu-1Bx18 alleles, which are generally difficult to distinguish using SDS-PAGE. A pair of AS-PCR primers (PS15) discriminated Glu-1By8, which produced a 527 bp fragment, while non-Glu-1By8 alleles showed negative results for PS15 (Fig. 5a). Previously, a pair of AS-PCR primers (PS16) was used to discriminate between Glu-1By18 and Glu-1By8 genes. In this study, no PCR product was detected from Glu-1By8 using this primer set; however, a PCR product of 527 bp was amplified from the other Glu-1By alleles, but the amplified Glu-1By15 and Glu-1By20 PCR products were weak (Fig. 5b). These markers could distinguish Glu-1By8 from other Glu-1By alleles, but could not distinguish Glu-1By8* and Glu-1By18 from other Glu-1By alleles (Fig. 5a, b). Therefore, in this study, we developed a set of specific primers to detect a 543 bp PCR product to distinguish Glu-1By18 from other Glu-1By alleles (PS17, Fig. 5c). With PS17, a single band of 543 bp was specifically detected in three cultivars (Joongmo2008, Suwon92, and Suwon105) containing Glu-1By18.

Fig. 5
figure 5

PCR analysis of Glu-1By alleles in wheat varieties. The numbers above the figure are the same as the variety numbers in Table 1. The primer set numbers in the figure are the same as in the Table 3. DM, DNA size marker; PS, primer set. Arrows indicate indels

Full size image
Table 3 Primers used in this study for detection of Glu-1By alleles
Full size table

Detection of Glu-1By9 and Glu-1By15/20 subunits

Two specific primer sets (PS18 and PS19) for Glu-1By9 allele detection have been reported to assay a 45 bp deletion [28, 37]. In this study, PS18 identified a PCR product that was 45 bp smaller in Glu-1By9 than in non-Glu-1By9, and two weak bands were detected for the Glu-1By15 and Glu-1By20 alleles (Fig. 5d). PS19 also identified a PCR product 45 bp smaller in Glu-1By9 than in non-Glu-1By9. One and two larger bands were detected for Glu-1By15 and Glu-1By20 alleles, respectively (Fig. 5e). These two primer sets detected a size of approximately 900 bp, and prolonged electrophoresis was required to distinguish the indels. In addition, the nucleotide sequences of Glu-1By15 and Glu-1By20 were highly similar and could not be distinguished. Therefore, in this study, we designed two additional primer sets to distinguish between Glu-1By15 and Glu-1By20, which produced a smaller PCR product with a 45 bp deletion (PS20–PS21, Fig. 5f-g). These markers also showed a 45 bp indel in Glu-1By9, but one large-sized band (PS20–PS21) were detected for Glu-1Bx15 and Glu-1Bx20, respectively. Therefore, these primer sets (PS18–21) could distinguish Glu-1By15 and Glu-1By20 from other alleles; however, Glu-1By15 and Glu-1By20 were detected in the same pattern and could not be distinguished from each other.

Detection of the Glu-1By16 subunit

Lei et al. [24] previously reported a PCR marker (PS22) for the detection of the Glu-1By16 allele. However, this marker showed multiple bands (three bands for Glu-1Bx16, zero or one band for Glu-1Bx15 and Glu-1Bx20, and two bands for the other Glu-1By alleles) (Fig. 5h). Therefore, we developed a Glu-1By16-specific PCR-based marker with a product size of 558 bp (PS23, Fig. 5i). This primer set specifically detected the Glu-1By16 allele in three cultivars: Jeokjoong, Baekjoong, and Joeun.

Analysis of HMW-GS protein using UPLC

HMW-GS glutenin subunits in wheat grains were confirmed by UPLC analysis (Fig. 6). The retention times of Glu-1Bx7, Glu-1Bx7*, and Glu-1Bx17 were 11.448 ± 0.027, 11.618 ± 0.067, and 11.670 ± 0.074 min, respectively, making it difficult to distinguish them. The retention times of Glu-1By8, Glu-1By8*, and Glu-1By18 were 10. 649 ± 0.022, 9.714 ± 0.048, and 9.676 ± 0.044 min, respectively. Glu-1By8* and Glu-1By18 were extracted earlier than Glu-1Dx, while Glu-1By8 was extracted later than Glu-1Dx. Thus, distinguishing Glu-1Bx8 from Glu-1Bx8* and Glu-1Bx18 may be possible using UPLC; however, it is difficult to distinguish Glu-1By8* and Glu-1By18 from each other. In addition, the extraction times of Glu-1Bx14(−), Glu-1Bx14(+), and Glu-1Bx20 were 11.549 ± 0.023, 12.319 ± 0.105, and 12.173 ± 0.029 min, respectively. The extraction time of Glu-1Bx14(−) was shorter than that of Glu-1Bx14(+) and Glu-1By20, but the extraction times of Glu-1Bx14(+) and Glu-1By20 were very similar, making it difficult to distinguish them. Likewise, the extraction times of Glu-1By15 and Glu-1By20 were very similar (9.687 ± 0.027 min and 9.645 ± 0.019 min, respectively). Therefore, the distinction between the allelic combinations of Glu-1Bx14(+) + Glu-1By15 and Glu-1Bx20 + Glu-1By20 is difficult, even using UPLC analysis.

Fig. 6
figure 6

Identification of HMW-GS Glu-1Bx and Glu-1By in 24 wheat cultivars by ultra- performance liquid chromatography. Glu-1Bx and Glu-1By alleles are shown in blue. AU, arbitrary units; RT, retention time

Full size image

Analysis of HMW-GS proteins using SDS-PAGE

HMW-GS glutenin subunits were compared using SDS-PAGE to confirm the presence of PCR markers in the wheat grains (Fig. 7). Glu-1Bx7 and Glu-1Bx7* differed by five amino acids and showed calculated molecular weights of 85.31 kDa and 84.71 kDa (www.bioinformatics.org), respectively, and the two proteins could not be distinguished by SDS-PAGE. Similarly, the molecular weights of Glu-1Bx14(−), Glu-1Bx14(+), and Glu-1Bx20 were calculated to be 84.53 kDa, 86.23 kDa, and 86.11 kDa, respectively, which were distinct from those of Glu-1Bx14(−) and Glu-1Bx20. However, Glu-1Bx14(+) and Glu-1Bx20 could not be distinguished by SDS-PAGE. Additionally, the molecular weights of Glu-1By8 and Glu-1By18 were calculated to be 77.38 kDa and 77.44 kDa, respectively. Glu-1By8 and Glu-1By8* proteins were highly identical in size and could not be distinguished by SDS-PAGE. The molecular weights of Glu-1By15 and Glu-1By20 were calculated to be 77.40 kDa and 77.43 kDa, respectively, and were also not distinguishable from each other.

Fig. 7
figure 7

Identification of HMW-GS Glu-1Bx and Glu-1By by SDS-PAGE in 25 wheat cultivars. The numbers above the figure are the same as the variety numbers in Table 1. x + y, Glu-1Bx + Glu-1By; HMW-GS, high-molecular-weight glutenin subunits; LMW-GS, low-molecular-weight glutenin subunits; PM, protein size marker. Arrows indicate Glu-1Bx and Glu-1By alleles

Full size image

Discussion

Allele variations in HMW-GSs are highly related to wheat baking quality, and among the three Glu-1 loci, Glu-B1 shows the greatest allele variations in both tetraploid and hexaploid wheat [50]. Therefore, various mass spectrometry techniques, SDS-PAGE analyses, and molecular markers have been developed to identify HMW-GSs. However, new genotypes continue to be reported, and there are few genotypes that can be distinguished using the developed markers. In addition, many previously reported PCR-based markers often performed simple relative comparison analyses without comparing various alleles. Moreover, the primer sequences and accession numbers were incorrect in some reports [35, 39]. In addition, in some cases, in genotypes of alleles that could not be distinguished by SDS-PAGE and LC analyses were re-evaluated [43, 47].

In this study, 11 novel Glu-B1 allele identification markers were developed, and together with previously reported markers, they could be used to distinguish nine Glu-1Bx alleles (Glu-1Bx6, Glu-1Bx7, Glu-1Bx7*, Glu-1Bx7 OE, Glu-1Bx13, Glu-1Bx14 (−), Glu-1Bx17, and Glu-1Bx14 (+)/20) and seven Glu-1By alleles (Glu-1By8, Glu-1By8*, Glu-1By9, Glu-1Bx16, Glu-1By18, and Glu-1By15/20). These findings confirmed that the Glu-1By allele of Avalone was Glu-1By8*, not Glu-1By8, and the Glu-1Bx allele of the three cultivars, Gobun, Eunpa, and Anbaek, was confirmed to be Glu-1Bx7*, not Glu-1By7. Geng et al. (2014) reported three Chinese and 11 European cultivars among 505 Chinese and 160 European cultivars with a 43 bp insertion. Cultivars containing 43 bp insertion were rare and had a high proportion of the Glu-1Bx7 gene. Among them, Glu-1Bx6 containing 43 bp insertion was identified in European cultivars ‘GK Bence’ and ‘Komorowska-pol’, and Glu-1Bx14 containing 43 bp insertion was found in European cultivars ‘Funo’ and ‘Amarelo de barba branca’ [40]. The present study showed that the Glu-1Bx6 allele from Avalon and the Glu-1Bx14 allele from Troll, Hanno, and Imbros are cultivars that do not contain 43 bp insertion in the promoter. Therefore, two types of promoters possibly exist in the Glu-1Bx6 and Glu-1Bx14 alleles. Cases with and without 43 bp insertions were identified; however, Glu-1Bx6 and Glu-1Bx14 with a 43 bp insertion are rare. Additionally, two Glu-1Bx14 allele accession numbers have been registered with NCBI and were distinguishable by 18 bp indels. Three cultivars—Hanno, Imbros, and Troll, showed a 18 bp deletion in Glu-1Bx14 in comparison with the other Glu-1Bx alleles, but RNAEE did not, allowing the distinction of these cultivars using the markers we developed.

In this study, two DNA polymerases were used for amplification. Unlike the conditions described in the previous studies, the annealing temperature for the optimal conditions differed depending on the DNA polymerase. Therefore, when analyzing markers, the annealing temperature must be set according to the DNA polymerase and equipment.

Glu-1Bx14(+) and Glu-1By20, and Glu-1By15 and Glu-1By20 could not be distinguished using UPLC analysis. Additionally, it was not easy to distinguish the sizes of the two proteins in SDS-PAGE analysis. These findings highlight the need to develop PCR-based markers that can easily distinguish between these two allele combinations. However, we were could not distinguish Glu-1Bx14 (+) from Glu-1Bx20, nor could it distinguish Glu-1By15 from Glu-1By20 with the PCR-based marker. In addition, since many alleles were not tested in this study and more alleles may occur, the primers developed here may not be fully applicable. However, the most commonly used allele combinations can be distinguished by PCR-based markers developed in this study. Additionally, these results suggesting that the Glu-A1 and Glu-D1 alleles also need to be reassessed through PCR-based markers.

Conclusions

HMW-GS allele composition is a crucial factor in determining end-use quality, and allele identification is an essential task in wheat breeding programs. Seven Glu-1Bx and four Glu-1By allele detection markers were developed to detect nine Glu-1Bx and seven Glu-1By locus alleles, respectively. The discrimination of Glu-B1 locus alleles can be improved and the most commonly used allele combinations can be identified by integrating previously reported markers and 11 newly developed PCR markers. However, these PCR markers cannot distinguish the Glu-1Bx14 (+) + Glu-1By15 and Glu-1Bx20 + Glu-1By20 combination; therefore, further research is needed. The developed markers can facilitate more effective analysis of molecular variations in the Glu-B1 allele, thereby improving the end-use quality of common wheat.