Introduction
Low molecular weight glutenin subunits (LWM-GSs) account for approximately 40% of total proteins in wheat grain and primarily determine dough extensibility and strength, thus playing a pivotal role in wheat flour processing quality (Lee et al., Reference Lee, Beom, Altenbach, Lim, Kim, Kang, Yoon, Gupta, Kim, Ahn and Kim2016; Li et al., Reference Li, Liu, Song, Zhang, Li and Gao2016; Xiang et al., Reference Xiang, Huang, Gong, Liu, Wang, Jin, He, He, Jiang, Zheng, Liu and Wu2019). The LMW-GSs encoding genes are located at loci of Glu-A3, Glu-B3 and Glu-D3 on the short arms of wheat chromosomes 1A, 1B and 1D, respectively (D'Ovidio and Masci, Reference D'Ovidio and Masci2004). The copy numbers of LMW-GS genes in bread wheat are varied from 10–20 to 30–40 (Harberd et al., Reference Harberd, Bartels and Thompson1985; Lee et al., Reference Lee, Beom, Altenbach, Lim, Kim, Kang, Yoon, Gupta, Kim, Ahn and Kim2016). The copy numbers of active LMW-GS genes in single accession were estimated to be 9 to 13 (Zhang et al., Reference Zhang, Liu, Zhang, Jiang, Luo, Yang, Sun, Tong, Cui and Zhang2013). Based on the first amino acid residue at the N-terminal domain of the mature protein, the LMW-GSs genes are traditionally categorized into LMW-methionine (m), LMW-serine (s) and LMW-isoleucine (i) types (D'Ovidio and Masci, Reference D'Ovidio and Masci2004). A fourth type of LMW-GS gene, LMW-leucine (l), was characterized from Aegilops comosa (Wang et al., Reference Wang, Gao, Wang, Zhang, Li, Zhang, Xie, Yan, Belgard and Ma2011b; Huang et al., Reference Huang, He, Jin, Wang, He, Feng, Liu and Wu2018).
Aegilops umbellulata Zhuk. (2n = 2x = 14, UU) is a wild diploid relative of cultivated wheat that harbours great variability in such valuable traits as disease resistance (Gill et al., Reference Gill, Sharma, Raupp, Browder, Heachett, Harvey, Moseman and Waines1985; Chhuneja et al., Reference Chhuneja, Kaur, Goel, Aghaee-Sarbaezeh, Parashar and Dhaliwal2008; Edae et al., Reference Edae, Olivera, Jin, Poland and Rouse2016) and grain quality-related traits (Law and Payne, Reference Law and Payne1983; Dai et al., Reference Dai, Zhao, Xue, Jia, Liu, Pu, Zheng and Yan2015; Wang et al., Reference Wang, Wang, Zhen, Li and Yan2018). Ae. umbellulata has crossability with tetraploid wheat, and therefore agronomic desirable traits from Ae. umbellulata can be introgressed into common wheat using synthetic wheat hexaploids with the AABBUU genome as bridges (Bansal et al., Reference Bansal, Kaur, Dhaliwal, Bains, Bariana, Chhuneja and Bansal2017; Song et al., Reference Song, Dai, Jia, Zhao, Kang, Liu, Wei, Zheng and Yan2019). PmY39 for powdery mildew resistance (Zhu et al., Reference Zhu, Zhou, Kong, Dong and Jia2006) and Lr9 for leaf rust resistance (Schachermayr et al., Reference Schachermayr, Siedler, Gale, Winzeller, Winzeller and Keller1994) were successfully transferred from Ae. umbellulata to wheat cultivars. The introgression of 1U genome of Ae. umbellulata into hexaploid wheat showed significantly improved dough rheological properties and bread-making quality (Wang et al., Reference Wang, Wang, Zhen, Li and Yan2018).
A pair of HMW-GS and their coding genes at Glu-U1 locus was characterized and cloned (Liu et al., Reference Liu, Zhang, Wan, Liu and Wang2002). Some Glu-U1 genes were transferred from Ae. umbellulata to the Chinese Spring wheat variety (Islam-Faridi, Reference Islam-Faridi1988). Characterization of LMW-GS genes from Ae. umbellulata was reported by limited studies (Li et al., Reference Li, Wang, Wang, Gao, Xie, Hsam, Zeller and Yan2010; Wang et al., Reference Wang, Li, Wang, Wang, Li, Zhang, Guo, Zeller, Hsam and Yan2011a). The numbers of LMW-GS genes reported in these studies varied from 1 to 5. In the present study, we isolated 35 novel LMW-GS genes (11–14 copies per accession) from three Ae. umbellulata accessions and their molecular characterization was investigated.
Materials and methods
Plant materials
Ae. umbellulata (2n = 2x = 14, UU) accessions (Fig. 1) numbered PI 554,396, AS3 and AS4 were used in this study. PI 554,396 was kindly provided by the National Plant Germplasm System of the USDA-ARS, USA. AS3 and AS4 were originally provided by Dr Sadao Sakamoto, Plant Germ-plasm Institute, Kyoto University, Kyoto, Japan. All these materials were kept at the Triticeae Research Institute, Sichuan Agricultural University, Chengdu, China.
Fig. 1. Morphology of Aegilops umbellulata spike. (a), spike; (b), spikelets; (c), seeds.
DNA extraction, PCR amplification and sequencing
Genomic DNA was extracted from seedling leaves of Ae. umbellulata accessions using the cetyltrimethylammonium bromide (CTAB) method (Rogers and Bendich, Reference Rogers and Bendich1985) with slight modification. One pair of allele-specific PCR (AS-PCR) primers (LMW-1: ATCATCACAAGCACAAGCATC and LMW-2: TTCTTATCAGTAGGCACCAAC) (Wang et al., Reference Wang, Gao, Wang, Zhang, Li, Zhang, Xie, Yan, Belgard and Ma2011b) was used to amplify LMW-GS genes. The PCR amplification was performed using the Veriti™ 96-Well Fast Thermal Cycler (Applied Biosystems, Foster City, CA, USA) in a 40 μl reaction volume as described by Huang et al. (Reference Huang, He, Jin, Wang, He, Feng, Liu and Wu2018). PCR products were separated using 1.5% agarose gel electrophoresis, stained with GelRed Nucleic Acid Gel Stain (Biotium, Hayward, CA, USA). These fragments were purified by using Sureclean (Bioline Reagents, London), ligated to pMD19-T vector (Takara, Dalian, China) and used to transform Escherichia coli DH5α cells. The positive clones were sequenced by Sangon Biotechnology Company (Shanghai, China). To avoid possible error, the final sequence of each LMW-GS gene was determined from the sequencing results of at least three independent clones.
Sequence analysis of LMW-GS genes
The assembly of LMW-GS sequences was completed using Lasergene software (DNASTAR; http://www.dnastar.com/). The sequence alignment of LMW-GS genes was performed with BioEdit (Hall, Reference Hall2007). The phylogenetic tree was constructed using the neighbour-joining method in the MEGA 6.0 software (Tamura et al., Reference Tamura, Stecher, Peterson, Filipski and Kumar2013). Bootstrap values were calculated from 1000 replications.
Results
Cloning and characterization of LMW-GS genes
PCR products from three Ae. umbellulata accessions are shown in Fig. 2. A single candidate amplification product with approximately 900 bp was obtained using AS-PCR primers LMW-1/LMW2. After sequencing, a total of 35 LMW-GS sequences were obtained (online Supplementary Table S1). GenBank database comparison showed that these genes were different from reported LMW-GS genes in Ae. umbellulata (Li et al., Reference Li, Wang, Wang, Gao, Xie, Hsam, Zeller and Yan2010; Wang et al., Reference Wang, Li, Wang, Wang, Li, Zhang, Guo, Zeller, Hsam and Yan2011a) and other Triticum species; thus, all these 35 genes were novel and were deposited in GenBank with the accession numbers (MZ424047-MZ424081).
Fig. 2. PCR amplification of LMW-GS genes from Aegilops umbellulata accessions by the AS-PCR primers LMW-1/LMW2. Lanes 1, 2 and 3 PCR products were amplified from PI 554,396, AS3 and AS4, respectively. M, DL2000 DNA marker.
In total, 14, 12 and 11 LMW-GS genes were revealed by sequencing of 42, 24 and 22 clones from PI554396, AS3 and AS4, respectively. MZ424067, which present in three Ae. umbellulata accessions, was found to be the most common LMW-GS gene (>54%). MZ424080 and MZ424081 in PI554396 and MZ424079 in AS3 are pseudogenes, the remaining 32 genes are complete active genes. The ORFs of the 32 active LMW-GS genes varied from 885 to 888 bp, which encoding 31 different LMW-GSs with 294 to 295 amino acid residues. The coding regions of these LMW-GS genes were all terminated by double stop codons.
Amino acid sequence alignment indicated that all subunits shared four main structural domains, including a signal peptide, an N-terminal region, a repetitive domain and a C-terminal domain with three sub-regions (C-terminal I, II and III). Since the first amino acid residue of the mature protein was methionine, the 31 LMW-GS were regarded as typical LMW-m-type subunits (Fig. 3). All LMW-m-type proteins begin with METSCIPGL except MZ424060 begin with METSCILGL. As typical for LMW-GS genes, 30 subunits had eight highly conserved cysteine residues, while MZ424050 containing nine cysteine residues with an extra cysteine residue located at the last amino acid residue of the conserved C-terminal III domain in LMW-m subunit (online Supplementary Table S1).
Fig. 3. Multiple alignments of the deduced amino acid sequences of LMW glutenin genes. Signal, signal peptide. The mature protein sequences were divided into: N-terminal domain; repetitive domain; C-terminal domains (I–III). Magenta shows the first amino acid residue of the mature proteins. Grey shading indicates the cysteine residues. Identical sequences and deletions were presented by dots and dashes, respectively.
Sequence variations of the LMW-GS genes
Multiple sequence alignment of the LMW-GS sequences was performed to identify the presence of single-nucleotide polymorphisms (SNPs) and insertions and deletions (InDels). As compared to the predominant MZ424067 in three Ae. umbellulata accessions, a total of 55 SNPs were detected at different positions (Table 1). One-base deletion (C) was found at the position 226 in MZ424081 and a three-base deletion (ACA) at 306–308 were detected in MZ424048 and MZ424074.
Table 1. SNPs and InDels identified among 35 LMW-GS genes from Ae. umbellulata
Phylogenetic analysis of LMW-GS genes
To further investigate the phylogenetic relationships among the LMW-GS genes at the Glu-3 loci, a phylogenetic tree was constructed based on 35 LMW-GS genes in this study and other 17 LMW-GS genes retrieved from GenBank. The results are shown in Fig. 4.
Fig. 4. Phylogenetic relationships of the 35 newly isolated and LMW-GS sequences retrieved from GenBank. The nucleotide sequences were specified by their corresponding GenBank accession numbers. The LMW-GS gene types and species are indicated. Bold, the 35 LMW-GSs identified in this study.
The phylogenetic tree was divided into three clear branches, with alleles of LMW-m-type genes at the top, LMW-s-type genes at the middle and LMW-i-type genes at the bottom. The 35 LMW-GS genes obtained in the current study were located in the LMW-m-type branch, which clustered with two published LMW-m-type genes EU571725 and EF649991 from Ae. umbellulata, further confirming that they are LMW-m-type genes.
Discussion
The LMW-GS genes are widely presented in genomes of wheat and related cereals (Ikeda et al., Reference Ikeda, Nagamine, Fukuoka and Yano2002; Henkrar et al., Reference Henkrar, El-Haddoury, Iraqi, Bendaou and Udupa2017; Huang et al., Reference Huang, He, Jin, Wang, He, Feng, Liu and Wu2018; Xiang et al., Reference Xiang, Huang, Gong, Liu, Wang, Jin, He, He, Jiang, Zheng, Liu and Wu2019). In the current study, we have isolated 32 novel active LMW-GS genes and three pseudogenes from the Ae. umbellulata genome using AS-PCR primers. We have found extensive allelic variations at Glu-3U that were mainly resulted from SNPs and InDels presented in the LMW-GS genes. A phylogenetic tree was constructed to understand the phylogenetic relationships within these LMW-GS genes.
To date, a few LMW-GS genes have been cloned from Ae. umbellulata accessions. For example, one LMW-GS gene EU571725 from Ae. umbellulata accession PI222762 (Li et al., Reference Li, Wang, Wang, Gao, Xie, Hsam, Zeller and Yan2010); five genes GQ980034, GQ980035, GQ870240-GQ870242 from PI573516 (Wang et al., Reference Wang, Li, Wang, Wang, Li, Zhang, Guo, Zeller, Hsam and Yan2011a). Previous studies showed that the copies of Glu-3 genes in common wheat are varied from 10–20 to 30–40 (Harberd et al., Reference Harberd, Bartels and Thompson1985; Lee et al., Reference Lee, Beom, Altenbach, Lim, Kim, Kang, Yoon, Gupta, Kim, Ahn and Kim2016) and active LMW-GS genes in single accession were estimated to be 9 to 13 (Zhang et al., Reference Zhang, Liu, Zhang, Jiang, Luo, Yang, Sun, Tong, Cui and Zhang2013). Wang et al. (Reference Wang, Wang, Zhen, Li and Yan2018) identified 10 abundant 1U-encoded LMW-GS subunits in CNU609 derived from crosses between Chinese Spring and the wheat-Ae. umbellulata 1U(1B) substitution line and cloned two Glu-U3-encoded LMW-m subunit genes. In the present study, we have isolated 14 (12 active), 12 (11 active) and 11 (11 active) LMW-GS genes from three Ae. umbellulata accessions, respectively. These results suggest that almost all active genes at Glu-3U loci were cloned and the Ae. umbellulata genome may have the similar Glu-3 gene copies to those of wheat.
Previous studies have shown that the cysteine residues were involved in the formation of inter- and intra-molecular disulphide bonds, and any modification in the number or location of cysteine residues could lead to functional variations (D'Ovidio and Masci, Reference D'Ovidio and Masci2004). The most mature Glu-3 proteins contain eight highly conserved cysteine residues. The first and seventh cysteine residues are involved in inter-molecular disulphide bonds, while the remaining six cysteine residues are participated in forming intra-molecular disulphide bonds. In this study, 30 subunits had eight cysteine residues, and the position of these cysteine residues was similar to that of the typical LMW-m-type subunits, whereas MZ424050 subunit had nine cysteine residues with an extra cysteine residue located in the conserved C-terminal III domain. The GQ870241 subunit from U genome with an extra cysteine residue at the C-terminal III was predicted to have positive effects on dough properties. The same number of cysteine residues was found in a LMW-m-type subunit AY263369, which was likely associated with good bread-making quality (Zhao et al., Reference Zhao, Wang, Guo, Hu and Sun2004; Xu et al., Reference Xu, Wang, Shen, Zhao, Sun, Zhao and Guo2006). It has been suggested that the extra cysteine residue might benefit the formation of larger glutenin polymers and contributed to superior gluten quality (Lan et al., Reference Lan, Feng, Xu, Zhao and Wang2013). Therefore, the MZ424050 subunit could contribute to good dough properties and may be new candidate gene for wheat quality improvement.
In the present study, all LMW-GS proteins belonged to typical LMW-m-type subunits. Wang et al. (Reference Wang, Li, Wang, Wang, Li, Zhang, Guo, Zeller, Hsam and Yan2011a) identified a LMW-i-type gene (GQ870242) from Ae. umbellulata PI573516, while this gene is not active due to the presence of premature stop codon in the ORFs. To our best knowledge, the active LMW-GS genes cloned in Ae. umbellulata so far all belonged to LMW-m-type subunits. We have found considerable SNPs variations at the Glu-3U loci, which could be produced by unequal crossing over, point mutations and illegitimate recombination (Anderson and Greene, Reference Anderson and Greene1989; An et al., Reference An, Zhang, Yan, Li, Zhang, Wang, Pei, Tian, Wang, Hsam and Zeller2006; Zhang et al., Reference Zhang, Li, Yan, Zheng, An, Xiao, Wang, Wang, Hsam and Zeller2006; Li et al., Reference Li, Ma, Gao, Zhang, Wang, Ji, Wang, Appels and Yan2008). The extensive allelic variations at Glu-3U could facilitate the development of genome-specific molecular markers and marker-assisted selection (MAS) in wheat quality breeding. The novel LMW-GS genes could be served as valuable genetic sources for wheat quality improvement.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1479262122000016
Acknowledgements
This work was financially supported by grants from the National Natural Science Foundation of China (31901493), the Science & Technology Department of Sichuan Province (2021YJ0505 and 2021YFH0110).
