2018 年 93 巻 1 号 p. 9-20
The wheat seed storage proteins gliadin and glutenin are encoded by multigenes. Gliadins are further classified into α-, γ-, δ- and ω-gliadins. Genes encoding α-gliadins belong to a large multigene family, whose members are located on the homoeologous group 6 chromosomes at the Gli-2 loci. Genes encoding other gliadins are located on the homoeologous group 1 chromosomes at the Gli-1 loci. Two-dimensional polyacrylamide gel electrophoresis (2-DE) was used to characterize and profile the gliadins. The gliadins in aneuploid Chinese Spring wheat lines were then compared in this study. Gliadin proteins separated into 70 spots after 2-DE and a total of 10, 10 and 16 spots were encoded on chromosomes 6A, 6B and 6D, respectively, which suggested that they were α-gliadins. Similarly, six, three and seven spots were encoded on chromosomes 1A, 1B and 1D, respectively, which indicated that they were γ-gliadins. Spots that could not be assigned to chromosomes were N-terminally sequenced and were all determined to be α-gliadins or γ-gliadins. The 2-DE profiles showed that specific α-gliadin spots assigned to chromosome 6D were lost in tetrasomic chromosome 2A lines. Furthermore, western blotting against the Glia-α9 peptide, an epitope for celiac disease (CD), suggested that α-gliadins harboring the CD epitope on chromosome 6D were absent in the tetrasomic chromosome 2A lines. Systematic analysis of α-gliadins using 2-DE, quantitative RT-PCR and genomic PCR revealed that tetrasomic 2A lines carry deletion of a chromosome segment at the Gli-D2 locus. This structural alteration at the Gli-D2 locus may provide a genetic resource in breeding programs for the reduction of CD immunotoxicity.
Wheat flour is processed into various foods and eaten all over the world. Gluten plays an important role in the unique properties of flour. It is a composite of glutenins and gliadins, which are insoluble in water. This means that they are classified as prolamin proteins and act as seed storage proteins in cereals (Shewry et al., 1995). Glutenins are composed of high molecular weight glutenin subunits (HMW-GS) and low molecular weight glutenin subunits (LMW-GS) (Jackson et al., 1983). Gliadins are generally separated into three classes, α-, γ- and ω-gliadin, based on their molecular weight and amino acid composition. Recently, one more class of gliadin, δ-gliadin, was found after DNA sequence analysis (Anderson et al., 2012). The genes encoding δ-, γ- and ω-gliadins, and LMW-GS, are tightly linked and are part of the Gli-1 and Glu-3 loci, which are found on the short arm of the homoeologous group 1 chromosomes (Payne et al., 1982; Anderson et al., 2012). The genes encoding HMW-GS are part of the Glu-1 loci and are located on the long arm of the homoeologous group 1 chromosomes (Payne et al., 1982). The α-gliadin genes are part of the Gli-2 loci, which are on the short arm of the homoeologous group 6 chromosomes (Harberd et al., 1985).
The genes encoding these gluten proteins belong to multigene families. Gliadins are encoded by a particularly large number of multigenes. The copy number differs among cultivars or varieties, and the number in each genome has been estimated by their restriction fragment length polymorphism patterns and by genomic PCR analyses. The α-gliadin gene copy number was estimated to be about 150 in the bread wheat cultivar Cheyenne (Anderson et al., 1997). The numbers of γ- and ω-gliadin genes were estimated to be 39 and seven, respectively, in the bread wheat cultivar Chinese Spring (Sabelli and Shewry, 1991). Large DNA sequence datasets, such as expressed sequence tags (ESTs) and next-generation sequencing (NGS), allow us to estimate almost complete gene sets. A comprehensive EST analysis for Chinese Spring wheat showed that 36 distinct α-gliadin genes were expressed during seed development (Kawaura et al., 2005). A total of 13 unique γ-gliadin-encoding genes have been reported to be present in the Chinese Spring wheat genome, and 11 of them are active according to EST and NGS analysis (Anderson et al., 2013). In the cultivar Butte 86, nine γ-gliadin genes encoding full-length proteins were identified by EST analysis (Altenbach et al., 2010). RNA sequencing using a third-generation sequencer determined that 25, 11, one and five genes coded for α-, γ-, δ- and ω-gliadins, respectively, in the cultivar Xiaoyan 81 (Wang et al., 2017).
Gluten proteins are responsible for wheat flour characteristics, and some of them are celiac disease (CD) elicitors (Vader et al., 2003). A synthetic peptide corresponding to a deamidated α-gliadin peptide, produced by transglutaminase, enhances CD T-cell responses (Anderson et al., 2000). Some α-gliadins contain a peptide that is a CD epitope, and genes on chromosome 6D preferentially code for immunodominant peptide fragments (Molberg et al., 2005; van Herpen et al., 2006). Large-scale amplicon sequencing of α-gliadin genes from Triticum and Aegilops showed that α-gliadin genes were classified into six types (Ozuna et al., 2015). One type encoded a major immunogenic peptide for CD and is only found in T. aestivum and Ae. tauschii, which is the D genome ancestor of T. aestivum (Ozuna et al., 2015).
To elucidate the expressed gliadin proteins encoded by these extremely large multigene families, we conducted two-dimensional polyacrylamide gel electrophoresis (2-DE) analysis of gliadins in aneuploid lines of the cultivar Chinese Spring. Systematic analyses of each gene and gene product for α-gliadin using 2-DE, quantitative RT-PCR and genomic PCR showed that almost all spots can be assigned to the respective chromosomes, and that tetrasomic 2A lines carry a deletion of the chromosome segment at the Gli-D2 locus harboring an epitope for CD.
Triticum aestivum L. cv. Chinese Spring and its aneuploid lines were provided by the National BioResource Project-Wheat, Japan. The plants were grown in a field at the Kihara Institute for Biological Research, Yokohama, Japan, during the 2015 growing season. Immature seeds were sampled and immediately stored at –80 ℃ for gene expression analysis.
Gliadin proteins were extracted according to DuPont et al. (DuPont et al., 2005), with modification. Powdered endosperm (15 mg) from mature seeds was incubated for 30 min at room temperature in 300 μl of 0.3 M sodium iodide buffer containing 7.5% 1-propanol. The sample was centrifuged at 20,600 g for 10 min at room temperature and 200 μl of the supernatant was collected. The pellet was dispersed again in the 0.3 M sodium iodide buffer and centrifuged as above. The supernatant was mixed with the earlier 200-μl sample and precipitated with four volumes of acetone for 4 h at –30 ℃. Gliadin proteins were collected by centrifuging the sample at 20,600 g for 10 mi at 4 ℃.
The gliadins were dissolved in 250 μl of isoelectric focusing (IEF) sample buffer containing 16% isopropanol, 8.5 M urea, 4% (w/v) CHAPS, 25 mM DTT and 0.5% (v/v) IEF buffer (GE Healthcare, Buckinghamshire, UK) (Ikeda et al., 2006). After shaking for 30 min and centrifuging at 20,600 g for 1 min at room temperature, 200 μl of the supernatant was applied to an Immobiline Dry-Strip pH 6–11 (GE Healthcare). The 18-cm strip was cut on the acidic side to 11.2 cm. In-gel rehydration was conducted for 14 h at 20 ℃ using a PROTEAN IEF Cell (Bio-Rad Laboratories, Hercules, CA, USA). IEF was carried out at 250 V for 15 min, at 6000 V for 12 kVh, and finally at 8000 V for 35 kVh. The gel was equilibrated with 6 M urea, 2% (v/w) SDS, 20% glycerol and 50 mM Tris-HCl (pH 6.8), and then stored at –30 ℃. Second-dimension SDS-PAGE was undertaken using a 15% acrylamide gel (14 cm × 14 cm × 1 mm) for 5 h at 200 V. The gel was incubated twice with 50% methanol and 7% acetic acid for 30 min each time and stained with SYPRO RUBY Protein Gel Stain (Thermo Fisher Scientific, Waltham, MA, USA) overnight. Fluorescent signals were scanned using a Typhoon9400 (GE Healthcare). N-terminal amino acid sequence analyses were performed as described previously (Ikeda et al., 2006).
Proteins were extracted from mature seeds with 62.5 mM Tris-HCl (pH 6.8), 10% glycerol, 1% SDS and 2% mercaptoethanol for 2 h. The samples were incubated at 96 ℃ for 2 min and centrifuged at 20,600 g for 10 min. The supernatant was then used for SDS-PAGE, which was performed using a 12.5% acrylamide gel at 200 V for 2 h. The proteins were transferred onto a PVDF membrane (Immobilon-P, Merck Millipore, Darmstadt, Germany) by a Trans-Blot SD Semi-Dry Electrophoretic Transfer Cell (Bio-Rad) according to the manufacturer’s instructions. The blot was blocked in Tris-buffered saline (TBS) solution containing 0.1% (v/v) Tween-20 and 5% (w/v) skimmed milk for 1 h and then rinsed in TBS for 5 min. The membrane was incubated with primary antibody (Eurofins Genomics, Tokyo, Japan; diluted 1:10000) overnight. The primary antibody was a rabbit polyclonal antibody against an oligopeptide sequence of Glia-α9 (PFPQPQLPY) (Arentz-Hansen et al., 2000), which is an epitope for CD in α-gliadin. After the membrane had been washed three times with TBS for 5 min each time, it was incubated with donkey anti-rabbit IgG (Promega, Madison, WI, USA; diluted 1:10000) for 30 min. Chemiluminescence was developed by Pierce Western Blotting Substrate Plus (Thermo Fisher Scientific) according to the manufacturer’s instructions. Chemiluminescence was scanned using a Typhoon9400.
Total RNA was isolated from immature seeds using PureLink Plant RNA Reagent (Thermo Fisher Scientific) following the manufacturer’s instructions. The RNA was treated with DNase I (Thermo Fisher Scientific) at 37 ℃ for 2 h and incubated at 80 ℃ for 10 min to inactivate the enzyme. cDNA was synthesized using an oligo(dT)20 primer and Superscript III (Thermo Fisher Scientific) at 50 ℃ for 3 h. Gene expression level was measured using SYBR Premix Ex Taq II (Takara Bio, Kusatsu, Japan) and a Thermal Cycler Dice Real Time System TP800 (Takara Bio). The primers for each gene group were as previously reported (Noma et al., 2016). Gene expression was normalized to that of the actin gene (Actin-F: AAGTACAGTGTCTGGATTGGAGGG, Actin-R: TCGCAACTTAGAAGCACTTCCG).
Total DNAs were extracted from fresh leaves with a DNeasy Plant Mini Kit (Qiagen, Hilden, Germany). The primers and PCR conditions for each gene group were as previously reported (Noma et al., 2016).
Gliadin proteins were extracted from mature seeds using the propanol and sodium iodide buffer method (DuPont et al., 2005). Proteins extracted from Chinese Spring wheat seeds were separated into 70 spots by 2-DE (Fig. 1). These major spots were from 30 to 50 kDa in size at pH 6.0 to pH 8.5, which suggested that they were α-gliadins and γ-gliadins. This 2-DE profile of euploid Chinese Spring wheat was compared to the profile for each chromosome in the aneuploid lines.
2-DE profile of gliadin proteins in Chinese Spring wheat. (A) Gliadin spots were separated by 2-DE. Spots were numbered from 1 to 70. (B) Spots were assigned to each chromosome.
The 2-DE profiles of the aneuploid lines of group 1 chromosomes revealed each γ-gliadin locus because genes encoding γ-gliadin are located at the Gli-A1, Gli-B1 and Gli-D1 homoeoloci of group 1 chromosomes. Six of the 70 spots were absent in the profile of the nullisomic-1A tetrasomic-1B line (N1AT1B) (Fig. 2A, Table 1), which indicated that these proteins are encoded by Gli-A1. Likewise, three spots were absent in the profiles of N1BT1A and N1BT1D, and seven spots were absent in N1DT1A and N1DT1B, which suggested that the three and seven proteins are encoded by Gli-B1 and Gli-D1, respectively (Fig. 2B–E, Table 1).
Gliadin 2-DE profiles of Chinese Spring and group 1 chromosome aneuploid lines. A number indicates an absent spot. (A) N1AT1B; (B) N1BT1A; (C) N1BT1D; (D) N1DT1A; and (E) N1DT1B.
+: present, –: absent.
Genes for α-gliadins are located at the Gli-A2, Gli-B2 and Gli-D2 homoeoloci of group 6 chromosomes. The 2-DE profiles showed that 10, 10 and 16 spots were absent in nullisomic 6A, nullisomic 6B and nullisomic 6D lines, respectively (Fig. 3, Table 1), and that these could be assigned to α-gliadins derived from genes at Gli-A2, Gli-B2 and Gli-D2, respectively. Interestingly, 16 spots, which may be derived from chromosome 6D, were depressed in the N2BT2A and N2DT2A tetrasomic 2A lines (Fig. 3, Table 1). In groups 3, 4, 5 and 7 chromosomes of the aneuploid lines, the 2-DE profiles did not show any spots that were different to Chinese Spring in the α-gliadin and γ-gliadin areas. However, 18 spots were present in all aneuploid lines according to this study.
Gliadin 2-DE profile of groups 6 and 2 chromosome aneuploid lines. A number indicates an absent spot. (A) N6AT6B; (B) N6AT6D; (C) N6BT6A; (D) N6DT6A; (E) N6DT6B; (F) N2BT2A; (G) N2BT2D; (H) N2DT2A; and (I) N2DT2B.
N-terminal amino acid sequencing was conducted to characterize the spots separated by 2-DE. A total of 18 spots were separated into 13 samples because spots ID 12/13/14, 32/33, 35/36 and 37/38 were too close to separate from each other (Fig. 1). Additionally, six spots that had already been assigned to their derived chromosomes were sequenced. A total of 19 samples were sequenced and five to 20 amino acid residues were determined (Table 2), except for spot ID 49, which was blocked at the N-terminus.
Among these 19 samples, 10 corresponded to the common N-terminal domain of α-gliadin and at least nine amino acid residues were completely conserved (VRVPVPQLQ) (Table 2). The samples containing two or three spots (excluding spots ID 12/13/14) had the same sequences as other samples containing just one spot, which indicated that they were a mixture of α-gliadins that have identical N-terminal sequences. Four spots (spots ID 3, 27, 45 and 46), which have been assigned to derived chromosomes 6B, 6D, 6A and 6A, respectively, were confirmed as α-gliadins. The other eight samples showed homology to the N-terminal sequences of γ-gliadin. They had almost the same sequences, but spot ID 69 had proline (P) at its third residue instead of glutamine (Q). Spots ID 63 and 70 were assigned to the 1B and 1A chromosomes, respectively, by the aneuploid 2-DE analysis and by their possession of γ-gliadin N-terminal sequences.
The N-terminal amino acid sequences for all samples were either α-gliadin or γ-gliadin sequences (Table 3). Finally, the numbers of α-gliadin and γ-gliadin protein spots were 48 and 22, respectively. The chromosome locations for 12 α-gliadin spots and six γ-gliadin spots could not be determined, which suggested that the spots overlapped with those from a different chromosome. However, the other α- and γ-gliadin spots did not overlap with each other. Therefore, the number of distinct α- and γ-gliadins may be more than the number of spots separated by 2-DE.
Sets of genes encoding α-gliadin and γ-gliadin have been reported previously by transcriptome analyses (Kawaura et al., 2005; Anderson et al., 2013). These studies estimated the numbers of expressed genes at each locus (Table 3). The total numbers of genes were more than the number of spots characterized by 2-DE for both α- and γ-gliadins. Almost an equal number of genes were estimated by genomic PCR (Noma et al., 2016) (Table 3). However, the copy number for the gliadin genes could not be predicted accurately from the reference genome data (IWGSC 2014) because the multigenes are very complex.
Some α-gliadins have epitopes that are toxic to CD patients. One of the immunodominant 33-mer sequences is encoded by α-gliadin genes on chromosome 6D (Molberg et al., 2005). Therefore, we conducted a western blot analysis using the Glia-α9 antibody, which binds to an epitope for CD in the 33-mer fragment. Total proteins were extracted from seeds of Chinese Spring wheat and aneuploid lines of groups 2 and 6 chromosomes. These were then used for western blot analysis (Fig. 4). Signal intensities were considerably weaker in nullisomic 6D lines (N6DT6A, N6DT6B) and tetrasomic 2A lines (N2BT2A, N2DT2A), indicating that nullisomic 6D and tetrasomic 2A lines reduced the α-gliadins encoded on chromosome 6D, which harbors the CD epitope.
Western blot analysis in aneuploid lines of groups 2 and 6 chromosomes using Glia-α9 antibody against a CD epitope peptide. (A) SDS-PAGE analysis. (B) Western blot analysis using the Glia-α9 antibody.
The 2-DE profiles and western blot analysis showed that specific α-gliadins were reduced in tetrasomic chromosome 2A lines (Figs. 3 and 4). Gene expression was measured by quantitative RT-PCR using allele-specific α-gliadin primers (Noma et al., 2016). The total expression levels of the α-gliadin genes did not differ greatly between the Chinese Spring and aneuploid lines (Fig. 5). The expression levels of genes transcribed from chromosomes 6A and 6B were shown as those in the Chinese Spring euploid line (Fig. 5). In contrast, the α-gliadin genes from chromosome 6D were not expressed in N6DT6A, N2BT2A or N2DT2A (Fig. 5). The same primer sets used for quantitative RT-PCR were applied for genomic PCR. The PCR products of α-gliadin genes could be assigned to each chromosome (Fig. 6, Noma et al., 2016), while the disappeared PCR products in tetrasomic 2A lines corresponded to those in nullisomic 6D lines, showing deletion of chromosome segment(s) at the Gli-D2 locus.
Gene expression patterns for α-gliadin genes in developing seeds. Allele-specific primers (AS1–AS11) have been reported previously (Noma et al., 2016). The α-gliadin primer was designed using a site that was common to the α-gliadin genes. The 6A, 6B and 6D chromosomes shown in parentheses indicate primers specific for the α-gliadin genes encoded on each chromosome. DPA, days post anthesis. Data are mean values ± SE from three independent samples.
Genomic PCR of α-gliadin genes in Chinese Spring and its aneuploid lines. The same allele-specific primers (AS1-AS11) for quantitative RT-PCR were used.
Gliadins in Chinese Spring wheat, which is the experimental standard cultivar, were characterized by 2-DE in this study. The genes encoding gliadins are mainly located on the short arms of group 1 chromosomes, namely 1AS, 1BS and 1DS, and on the short arms of group 6 chromosomes, namely 6AS, 6BS and 6DS (Payne, 1987). They form clusters at each locus and many genes have a high similarity to each other (Kawaura et al., 2012). Therefore, to clarify how many gliadins are expressed at each locus, we first conducted a 2-DE analysis of specifically extracted gliadin fractions from groups 1 and 6 chromosomes in the aneuploid lines (Figs. 1, 2 and 3). The gliadin proteins were selectively extracted using sodium iodide buffer from euploid Chinese Spring and 2-DE was used to separate the compounds into 70 spots that were of 30–50 kDa in size. Among them, 10, 10 and 16 spots were encoded on chromosomes 6A, 6B and 6D, respectively, which suggested that they were α-gliadins expressed at the Gli-2 loci (Table 1). Similarly, six, three and seven spots were encoded on chromosomes 1A, 1B and 1D, respectively, which indicated that they were γ-gliadins expressed at the Gli-1 loci (Table 1). Eighteen spots were expressed in all the relevant chromosomes from the aneuploid lines (Figs. 1, 2 and 3). After sequencing the N-terminal amino acids, we identified 48 spots as α-gliadins and 22 spots as γ-gliadins (Table 1). These results suggested that the 18 spots not assigned to any loci had overlapped in the 2-DE separation. Therefore, the actual number of proteins should be more than 48 and 22 spots for α- and γ-gliadins, respectively. Proteome analysis of wheat flour using 2-DE showed that 22 and 16 spots were assigned as α- and γ-gliadins, respectively, in cultivar Butte 86 (DuPont et al., 2011). Recently, gliadin spots were characterized by 2-DE and MS/MS in the cultivar Xiaoyan 81 using deletion lines (Wang et al., 2017). In Xiaoyan 81, 35 and 37 spots were assigned as α- and γ-gliadins, respectively (Wang et al., 2017). The number of gliadin spots in Chinese Spring was more than that estimated by EST analysis (Table 3) (Kawaura et al., 2005; Anderson et al., 2013) and almost equal to the number identified by genomic amplicon sequencing (Noma et al., 2016). The number of gliadins in Xiaoyan 81 was confirmed by second- and third-generation RNA sequencing (Wang et al., 2017). This suggested that the number of gliadin spots produced by Chinese Spring, Butte 86 and Xiaoyan 81 wheat were different: Chinese Spring had more α-gliadins than Butte 86 and Xiaoyan 81, whereas it had more γ-gliadins than Butte 86, but fewer than Xiaoyan 81.
Gliadins of the aneuploid lines for the group 1 and group 6 chromosomes, and all other chromosome aneuploid lines, were compared. The results revealed that the accumulation of α-gliadins assigned to chromosome 6D was also absent in tetrasomic chromosome 2A lines (Fig. 3). Expression of α-gliadin genes located on chromosome 6D was not detected (Fig. 5). Furthermore, the α-gliadin genes located on chromosome 6D were not amplified by genomic PCR (Fig. 6). These lines of evidence indicate that disappearance of α-gliadin spots corresponding to the genes on chromosome 6D in tetrasomic 2A lines must be the result of the structural alteration of chromosome 6D, and not of subgenomic regulation of α-gliadin gene expression. Since a previous 2-DE analysis showed that group 2 chromosomes affected seed proteins produced by chromosome 6 (Brown and Flavell, 1981), tetrasomic 2A lines (N2BT2A and N2DT2A) harbor the deletion of a chromosome segment at the Gli-D2 locus.
Transcription factors that spatially and temporally regulate the transcription of seed storage protein genes have been reported. In wheat, seed protein activator (SPA), wheat prolamin box binding factor (WPBF), GAMYB and TaFUSCA3 are known to bind to a specific cis-element in prolamin gene promoters (Albani et al., 1997; Ravel et al., 2006; Haseneyer et al., 2008; Sun et al., 2017). SPA is a homolog of OPAQUE2 (O2), a bZIP protein that regulates zein synthesis in maize (Albani et al., 1997). In barley, two O2-like bZIP proteins, BLZ1 and BLZ2, have been identified and their genes are located on chromosomes 5H and 1H, respectively (Vicente-Carbajosa et al., 1998; Oñate et al., 1999). Genes encoding SPA are located on the group 1 chromosomes (Guillaumie et al., 2004). Other O2 homologs were found on the group 5 chromosomes by a BLAST search of the reference genome database of the International Wheat Genome Sequencing Consortium (K. K., unpublished). WPBF is a homolog of maize PBF (Vicente-Carbajosa et al., 1997), which encodes a plant-specific DOF domain-containing protein (Yanagisawa and Schmidt, 1999). Furthermore, WPBF homoeologous genes are located in the centromeric region of the group 5 chromosomes (Ravel et al., 2006). GAMYB is involved in gibberellin signaling in seed aleurone cells and interacts with barley PBF in barley (Diaz et al., 2002). GAMYB is conserved between barley and wheat and encoded by genes at a chromosome 3 locus (Haseneyer et al., 2008). TaFUSCA3 is a B3-superfamily transcription factor located on chromosomes 3AL, 3B and 3DL (Sun et al., 2017). It can activate HMW-GS gene expression and interact with SPA (Sun et al., 2017). The 2-DE profiles of gliadins were not affected in the aneuploid lines of group 3 and 5 chromosomes, on which the genes encoding these transcription factors are located.
Western blotting using the Glia-α9 antibody showed that the gliadins from chromosome 6D had CD epitopes (Fig. 4). Since CD epitopes were preferentially found in α-gliadins encoded on chromosome 6D (Molberg et al., 2005; van Herpen et al., 2006), wheat lines containing a tiny chromosome deletion at the Gli-D2 locus could provide a valuable genetic resource for future breeding programs to reduce CD immunotoxicity caused by flour.
The aneuploid lines were provided by the National BioResource Project-Wheat with support from the National BioResource Project of MEXT, Japan. This research was supported by JSPS KAKENHI grant numbers 24580011 and 15K07261.