Fast Neutron Radiation Induced Glu-B1 Deficient Lines of an Elite Bread Wheat Variety

Five isogenic wheat lines deficient in high-molecular weight subunit (HMW-GS) proteins encoded by the B-genome were identified from a fast-neutron radiation-mutagenized population of Summit, an elite variety of bread wheat (Triticum aestivum L.). The mutant lines differ from the wild-type progenitor by the absence of the HMW-GS Bx17 and By18 proteins. Gene specific primers designed for Bx17 and By18 genes failed to generate the diagnostic amplicons in the mutant lines. Assays using the iSelect Illumina Wheat 90K SNP arrays revealed that the induced deletions in these lines ranged from 5.58 to 127.47 cM. Isogenic wheat lines deficient in the different HMW-GS proteins will facilitate the determination of the contributions of these different proteins to dough quality. This new germplasm will serve as genetic tools to generate healthier wheat varieties with altered dough performance for diverse end-uses.


Introduction
Gluten protein components glutenins and gliadins are the major determinants of dough quality. The high molecular weight glutenin subunits (HMW-GS) among the seed storage proteins have been shown to be the major determinants of dough mixing properties that are critical for the processing of food products derived from wheat including bread, pasta, cakes and cookies (Payne, 1987;Shewry et al., 2003;Lawrence et al., 1988). In hexaploid wheat the 6 HMW-GS genes are encoded by the Glu1 loci on the chromosome 1L group of the 3 genomes, namely Glu-A1, Glu-B1 and Glu-D1. Due to silencing of some genes hexaploid wheat normally expresses 3 to 5 different HMW-GS genes (Payne 1987). Each Glu1 locus contains a pair of tightly linked x-and y-type subunits, thus, a specific x-type subunit has always been associated with a defined y-type subunit. HMW-GS in the Glu1 locus are usually reported as allele pairs e.g. Bx17+By18, Dx5+Dy10. Different allele pair combinations in the HMW-GS gene loci vary in their effect on dough properties.
Surveys of commonly grown wheat varieties correlated genetic variations in the HMW-GS loci to dough quality (Lookhart et al., 1989;Graybosch, 1992;Payne et al., 1987). The Ax1+null, Bx17+By18 and Dx5 +Dy10 allele pairs were found to be superior for bread baking compared to the other alleles in the Glu1 locus in each of the corresponding genomes. The HMW-GS alleles in the D genome has been shown to contribute more significantly to dough quality compared to the Glu1 locus in the A and B genomes (Shewry et al., 2003). In particular, the Dx5+Dy10 allele pair positively correlated with good bread baking quality of the dough better than any allele pairs in the Glu-D1 locus. This paper reports the generation and identification of bread wheat lines deficient in the Glu-B1 locus genes that code for the Bx and By HMW-GS proteins. The availability of lines deficient in specific HMW-GS provides an opportunity to further determine the contribution of the individual gluten components to dough quality.
Summit is a product of hybridization between Express and Tadorna/PB775. Express is a variety from Western Plant Breeders, whereas, Proband 775 is a variety from a company previously known as Northrup King, Co. Tadorna is a Septoria leaf blotch resistant line from the University of California, Davis. Summit expresses five HMW-GS genes encoded by Glu-A1a (Ax1), Glu-B1i (Bx17+By18), Glu-D1d (Dx5+Dy10), considered to be one of the best allelic pair combinations for bread baking (Payne, 1987).
For fast neutron radiation (FNR) mutagenesis, four bags with 50 grams of seeds each (~1,250 seeds/bag) were treated with 6, 12, 18 and 24 Gray fast-neutron radiation (FNR) at the McClellan Nuclear Radiation Center in California. FNR treated seeds were grown in batches in the greenhouse. Seeds from the spike of the main stem of self-pollinated M1 plants were harvested and 15 to 20 M2 seeds per pot were planted for each line. M3 seeds were harvested from 1,000 lines for phenotyping and screening for altered seed storage protein profiles.
For increasing the seeds and for genetic crosses, lines identified with mutations were grown in the greenhouse with lighting supplemented by sodium lamps at 16-hr day length with average temperatures of ~28 o C during the day and ~18 o C at night. Plants were fertilized with Osmocote Plus 15-9-12 (Everris, USA) using an automatic drip irrigation system. Plants homozygous for the mutations were allowed to self-pollinate to at least the M4 generation before being used as parents for the initial backcross with wild type Summit line.

Mutant screening using SDS-PAGE
Three to four seeds from each M3 line were screened for altered seed storage protein patterns using one-dimensional sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Briefly, total seed protein from crushed endosperm of a half-seed was extracted with 200 ul of PAGE loading buffer [2% SDS, 50 mM DTT, 62.5 mM Tris (pH 8.5), 10% glycerol, 0.1 mg/ml pyronin]. About 4 ul of protein extracts were loaded onto 4-12% acrylamide Novex NuPAGE Bis-Tris gel (Invitrogen, Carlsbad, CA). The gels were run at 200V for 90 minutes at 4 o C and stained overnight using Brilliant Blue G (Sigma-Aldrich, St. Louis, MO). Destained gel images were documented using an EPSON Perfection V600 photo scanner (Epson, America, Long Beach CA).

DNA extraction and amplification by polymerase chain reaction (PCR)
Genomic DNA from wild type and mutant lines was extracted from young leaf tissue using CTAB as described (Granier et al. 2015). Extracted DNA quantity was measured using Qubit (Thermofisher, USA) and Nanodrop fluorometer (Thermofisher, USA) and its integrity was checked on agarose gels. Gene specific primers for Glu-B1i Bx17 and By18 genes were designed and used to detect by PCR the presence of the encoded gene fragment in wild type and absence/presence in mutated lines. Gene specific primers were also designed for the hexaploid wheat Pinb-D1 gene which was used as PCR positive control. The sequences of the primers used are shown in Table 1. PCR was carried out in 10 ul volumes using 1x Phire Hot Start II PCR master mix (Thermofisher, USA), 3 uM primers and 5-20 ng genomic DNA. Touch-down PCR profiles for Bx17/Pinb and By18/Pinb gene fragment amplification is provided in supplemental data 1. The quality of the PCR products was assessed using agarose gels and the sizes of amplicons were estimated using the DNA ladder O'Gene Ruler Express (Thermofisher, USA).

Identification of Glu-B1 deficient lines
The fast neutron radiation mutagenized population of T. aestivum cv. Summit was screened for mutations in the HMW-GS genes by total seed storage protein profiling using one-dimensional SDS-PAGE. The profile of total seed storage proteins extracted from the endosperm of M3 seeds were compared to that of the wild type progenitor. Five independent mutant lines with missing Bx17 and By18 proteins compared to wild type ( Figure 1) were identified, namely: S60E, S62D, S101A, S239F, S261A. The mutant nomenclature used includes the number that indicates the M2 line where the mutation was identified. These mutant lines were all from the batch of seeds treated with 12 Gray of fast neutron radiation.

Molecular basis of lesions in identified mutant lines
To identify the molecular basis for the loss of the Bx17 and By18 proteins, gene specific primers for Bx17 and By18 genes were designed to detect its presence or absence in the genome by PCR. As shown in Figure 2, the wild-type genome generated the diagnostic 566 bp and 652 bp amplicons for Bx17 and By18, respectively. Both were missing in the mutant lines even though the amplicon from the Pinb-D1 gene is present. The distance between the Bx17 and By18 genes in Glu-B1 locus is estimated to be 85kb (Gu et al., 2006) indicating that FNR treatment at 12 Gray generated mutations that could extend to several kilobases of DNA in length. To determine the extent of the deletions generated by FNR we assayed the genome of the mutant lines using the Illumina iSelect Wheat 90K SNP arrays. This array developed for bread wheat includes a significant fraction of common genome-wide distributed SNPs that are represented in populations of diverse geographical origin. More than half (46,977) of these SNPs have been genetically mapped using a combination of eight mapping populations (Wang et al. 2014 The gene sequences for Bx17 (AB263219.1) and By18 (KF430649.1) were used to query the mapped SNPs in the iSelect Wheat 90K SNP array (http://wheatgenomics.plantpath.ksu.edu/). The best blast hits that mapped to chromosome 1B were two SNPs that are separated by 4.66 cM, BS00084570_51 located at 79.77 cM and BS00082521_51 located at 84.43 cM. To examine more closely the blast results, the target sequences used for SNP BS00084570_51 and BS00082521_51 were used to query the NCBI non-redundant nucleotide collection database. The best hits for both SNP target sequences were similar to the HMW-GS Glu-1B x-type alleles, however, BS00084570_51 is more similar to Bx7, whereas, BS00082521_51 is more similar to Bx17. All samples including those with deletions in the Glu-B1 locus including those from Bobwhite also clustered in BS00084570_51 assay as shown in Figure 3A. In contrast, the 12 Glu-B1 Summit mutant lines formed a clustered distinct from the normal lines as shown in Figure 3C. It is still unclear why the two SNPs derived from target sequences similar to Bx genes mapped to two locations that are 4.66 cM apart. From the above data it is more likely that the Glu-B1 locus maps with BS00082521_51 at 84.43 cM. Examination of the distribution of the samples in 35 other SNPs that mapped to 84.43 cM showed that in all cases S60E clustered with the normal samples, whereas, in 12 of the 35 SNPs the other Glu-B1 deletion lines formed a distinct cluster. Examination of the distribution of the samples in the 11 other SNPs that map at 79.77 cM showed that in 6 out of the 11 SNPs the Glu-B1 deletion lines clustered with the rest of the normal samples. In 5 of the 11 SNPs at 79.77 cM the Glu-B1 deletion lines formed a distinct cluster except for 2 SNPs where S60E clustered with the normal samples, one of which is shown in Figure 3B. Further examination of the distribution of the samples in the 33 SNPs between 79.77 cM and 84.43 cM showed that in 20 of the 33 SNPs in between these two regions the Summit Glu-B1 deletion lines formed a distinct cluster from the normal samples except for S60E which clustered with the deletions lines in only 6 of the 20 SNPs. This data indicates that S60E has the smallest deletions among the Glu-B1 deletion lines.
We further examined the distribution of several SNPs located proximal and distal of 79.77 cM and 84.43 cM to estimate the length of the deletions generated by FNR in the deletion lines. As shown in Table 2 (Table 2). A deletion of a chromosome arm as large as 127.47 cM in S62D could have been lethal in a diploid organism but was made possible to be uncovered in an organism with a polyploid genome like wheat.
nd nd nd nd nd nd nd nd nd nd Note: Top row indicates the genetic distance on the long-arm of chromosome 1 of the B-genome

Inheritance of the Glu-B1 mutant phenotype
The M3 lines homozygous for the Glu-B1 mutation were allowed to self-pollinate for two more generations prior to seed increase. Seeds at M5 stage were used for the initial back cross to the wild type progenitor. The inheritance of the mutations in all lines was stable and could be easily scored both in lines allowed to self-pollinate for several generations and in out-crossed individuals. The morphological and developmental phenotype of greenhouse grown Glu-B1 deficient lines when compared to the wild type Summit progenitor and showed that the mutant lines have comparable phenotype except for the following: S60E inflorescence showed bent spikes due to extra florets ( Figure  4A), S62D and S239 exhibited shorter stature ( Figure 4B and C). The common change in height in S62D and S239F may be due to a linked mutation since both has almost half of the long arm of chromosome 1 deleted. These deviations in phenotype could also most likely be due to background mutations that are not linked to the mutation. To reduce these background mutations lines are continually being back crossed to the wild-type progenitor.

Concluding Remarks
The production of gluten protein deficient lines makes it possible to determine the effects of individual subunits on the rheological and technological properties of the wheat dough. The identification of Glu-B1 locus deficient lines identified in this work will provide tools to dissect the roles of HMW-GS encoded by the B genome i.e. Bx17 and By18 proteins, in wheat dough performance. Recent evidence indicates that glutenins are also involved in the Triticeae Genomics and Genetics 2017, Vol.8, No.1, 1-8 http://tgg.biopublisher.ca 6 development of celiac disease, a food-sensitive enteropathy caused by consumption of gluten (Tye-Din et al., 2010;Lexhaller et al., 2016). The peptide sequence QGYYPTSPQ, one of the epitopes identified to induce celiac disease  is present in 4 copies in Bx17 and 5 copies in By18 proteins (see supplemental data 2). Selection of varieties with deficiencies in specific gluten components is a tool not only to determine the molecular basis of gluten role in dough performance but also to develop low-gluten wheat products that could be safer for consumers with pre-disposition to develop gluten-sensitivities including celiac disease.