Post-translational cleavage of HMW-GS Dy10 allele improves the cookie-making quality in common wheat (Triticum aestivum)

Wheat is a major staple food crop worldwide because of the unique properties of wheat flour. High molecular weight glutenin subunits (HMW-GSs), which are among the most critical determinants of wheat flour quality, are responsible for the formation of glutenin polymeric structures via interchain disulfide bonds. We herein describe the identification of a new HMW-GS Dy10 allele (Dy10-m619SN). The amino acid substitution (serine-to-asparagine) encoded in this allele resulted in a partial post-translational cleavage that produced two new peptides. These new peptides disrupted the interactions among gluten proteins because of the associated changes to the number of available cysteine residues for interchain disulfide bonds. Consequently, Dy10-m619SN expression decreased the size of glutenin polymers and weakened glutens, which resulted in wheat dough with improved cookie-making quality, without changes to the glutenin-to-gliadin ratio. In this study, we clarified the post-translational processing of HMW-GSs and revealed a new genetic resource useful for wheat breeding.


Introduction
Common wheat (Triticum aestivum L.; AABBDD; 2n = 6x = 42) is an unusual food crop suitable for the production of various foods because of its unique processing quality, which is related to the continuous protein network in its grains (Shewry and Halford 2002;Wang et al. 2018a). The gluten proteins, which are crucial components of the protein network, are composed of gliadins and glutenins. The gliadins are single-chain polypeptides that may function as plasticizers responsible for dough extensibility (Barak et al. 2015;Qi et al. 2006). Glutenins are multi-chain polymeric proteins composed of low and high molecular weight glutenin subunits (LMW-GSs and HMW-GSs, respectively) (Jia et al. 2020;Shewry et al. 1999).
As a "backbone" for interactions with other glutenin subunits and gliadins, HMW-GSs are among the most important determinants of end-use quality, with key effects on dough strength and elasticity (Li et al. 2021;Shewry et al. 2003a;Wieser 2007;Yazar et al. 2017). The HMW-GS genes, which lack introns (Anderson and Greene 1989), have been localized to the Glu-1 loci on the long arm of chromosomes 1A (Glu-A1), 1B (Glu-B1), and 1D (Glu-D1) (Payne et al. 1982;Shewry et al. 2003b). Each locus contains two tightly linked genes encoding a smaller y-type subunit and a larger x-type subunit (Shewry et al. 1992). Normally, only 3-5 HMW-GSs are produced in common wheat (Payne et al. 1981). The HMW-GSs have a common primary structure, with a long central repetitive domain surrounded by highly conserved Nand C-terminal non-repetitive domains . The N-terminal domain generally has three or five cysteine (Cys) residues, whereas the C-terminal domain usually has one Cys and the central repetitive domain either lacks a Cys residue or has only one (Shewry and Halford 2002;Wang et al. 2018a). The sequence length, structure, and expression level of HMW-GSs are wheat quality determinants. Furthermore, the number and distribution of Cys residues are important features of HMW-GSs because disulfide bonds between these residues determine the polymeric structure and conformation of proteins, which affect the dough strength (Li et al. 2021;Shewry et al. 1992Shewry et al. , 2003aYue et al. 2019).
The Glu-D1 locus is a major genetic factor influencing the dough strength and bread-making quality of wheat. Previous studies on the effects of Glu-1 on the wheat end-use quality revealed that Dx5 and Dy10 are the alleles that produce the best subunit combination (Barro et al. 2003;Blechl et al. 2007;Lawrence et al. 1988;Shewry et al. 2003a). As expected, the Dy10 subunit is positively associated with wheat processing quality (Blechl et al. 2007; Laudencia-Chingcuanco 2012; León et al. 2009).
Storage proteins are initially synthesized on the rough endoplasmic reticulum, where the signal peptide is cleaved. Most storage proteins are subject to further processing, including post-translational modifications (Müntz 1998). The post-translational cleavage of precursor storage proteins, which has been detected in rice (Kumamaru et al. 2010;Wang et al. 2009), castor bean (Hara-Nishimura et al. 1991), soybean (Hara-Nishimura et al. 1995), and Arabidopsis (Gruis et al. 2004), is an essential part of storage protein maturation. However, there have been relatively few reports describing the post-translational cleavage of wheat seed storage proteins.
In this study, a new Dy10 allele, Dy10-m619SN, was identified by screening an ethyl methanesulfonate (EMS)-induced mutant population of the common wheat cultivar 'Shumai 482'. A further analysis revealed that Dy10-m619SN expression results in three peptides in wheat seeds because of a post-translational cleavage. Additionally, the effects of Dy10-m619SN on wheat processing quality as well as the underlying mechanism were investigated.

Materials and methods
Plant materials and growth conditions "Shumai 482" (wild type; WT) is a common wheat (Triticum aestivum) cultivar released by the Triticeae Research Institute, Sichuan Agricultural University. Its HMW-GS composition is Ax1, Bx7 + By9, and Dx5 + Dy10 (Xu et al. 2018). The mutant was isolated from "Shumai 482" seeds treated with 0.8% EMS (Sigma-Aldrich, St. Louis, MO, USA). The mutant seeds from the M 2 to the M 5 generations were analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) to verify the mutation.
To analyze wheat processing quality, the mutant was backcrossed with the WT (Fig. S1, part a). Fourteen BC 2 F 4 (seven with the WT genotype and seven with the mutant genotype) and 14 BC 3 F 4 (seven with the WT genotype and seven with the mutant genotype) homozygous lines were grown at the experimental farm of Sichuan Agricultural University (30°43′16″N, 103°52′15″E) during the 2017-2018 and 2019-2020 wheat growing seasons, respectively. Field trials were performed using a randomized block design. Each homozygous line was planted in a 2 m × 2 m area, with 20 cm between rows and 60 plants per row. A compound fertilizer [N:P:K (15:15:15)] was applied before sowing at a rate of 1 3 450 kg per hectare. Additionally, six BC 2 F 4 homozygous lines (three with the WT genotype and three with the mutant genotype) were also grown (250 plants per line) in a greenhouse with a 16-h light (23 ºC)/8-h dark (18 ºC) cycle as previously described (Zhang et al. 2017).
Mature grains were sun-dried and stored at room temperature for 60 days before analyses. The agronomic performance after backcrossing was recorded. The harvested WT and mutant seeds for each line were analyzed by SDS-PAGE and acid polyacrylamide gel electrophoresis (A-PAGE) (Fig. S1b).
The mutant was used as the female parent and crossed with the common wheat cultivar "Chinese Spring" (HMW-GS composition: Bx7 + By8 and Dx2 + Dy12). The F 1 generation was allowed to self-pollinate in the greenhouse. The HMW-GS composition was determined in the F 2 generation by SDS-PAGE.

SDS-PAGE, A-PAGE, and immunoblot analyses
Gliadins were extracted and separated by A-PAGE as previously described (Yan et al. 2003). Additionally, an SDS-PAGE analysis was completed using a published procedure (Qi et al. 2011). Specifically, the protein extraction buffer consisted of 62.5 mM Tris-HCl, pH 6.8, 10% (v/v) glycerol, 2% (w/v) SDS, 0.002% (w/v) bromophenol blue, and 1.5% (w/v) dithiothreitol (DTT). The seed storage proteins for the A-PAGE and SDS-PAGE were extracted from 10 mg ground seed powder. A urea SDS-PAGE analysis was performed under the same conditions, but urea was added to the solutions before the gel polymerized for a final concentration of 4 M urea (Wang et al. 2016).
Immature WT and mutant seeds collected at 5, 8,11,14,17,20,24, and 28 days post-anthesis (dpa) were ground to a fine powder in liquid nitrogen, after which storage proteins were extracted from 10 mg powder as described above to assess the post-translational cleavage by SDS-PAGE.
The relative proportions of each HMW-GS in mature seeds were determined using 200 WT and mutant seeds. The seed storage proteins, including the HMW-GSs, were extracted from 10 mg powder per seed as described above. The extracted HMW-GSs were separated by SDS-PAGE. The proportions of HMW-GSs (Ax1, Bx7 + By9, and Dx5 + Dy10) were determined using the Quantity One software (Bio-Rad, Hercules, CA, USA) as previously described (Wan et al. 2014).
The extra protein in the mutant was confirmed as a HMW-GS with the rabbit anti-HMW-GS polyclonal antibody (1:4,000; Denery-Papini et al. 1996) and the HRP-conjugated anti-rabbit secondary antibody (1:8,000; Sigma-Aldrich). The extracted proteins were separated by SDS-PAGE and then electrophoretically transferred to a PVDF membrane (Invitrogen, Carlsbad, CA, USA). After washing twice with TBST [Tris-buffered saline with 0.05% (v/v) Tween-20], the membrane was blocked for 1 h with 1% bovine serum albumin in TBST. The membrane was incubated with the primary antibody for 2 h followed by the secondary antibody for 1 h. After washing three times with TBST, the immunoblot was visualized with the Pierce™ ECL Plus Western Blotting Substrate (Thermo Fisher Scientific, Waltham, USA).

Mass spectrometry and C-terminal sequencing
The targeted protein bands were excised manually from the SDS-PAGE gel. For the mass spectrometry analysis, gel pieces were placed under vacuum centrifugation until completely dried, and then digested by 100 ng/μL trypsin for 16 h. The resulting tryptic peptides were identified by mass spectrometry analysis, which was completed by Qitai Biotechnology (Nanjing, China) as Shruthi et al. (2016).
Regarding the C-terminal sequencing, the targeted protein bands were washed twice with a mixture [50 mM NH 4 CO 3 and 50% (v/v) acetonitrile in Milli-Q water (Millipore)] for 30 min at 37 ºC, dehyrated with 100% acetonitrile for 30 min, and dried in a Speedvac (room temperature, 15 min). The proteins were reduced using 10 mM DTT in 7 M guanidine hydrochloride with 0.3 M Tris (pH = 9.0) for 60 min at 56 ºC. Alkylated reaction was performed by adding 55 mM iodoacetamide in 50 mM NH 4 HCO 3 followed by incubation in the dark for 60 min at room temperature. Subsequently, the samples were washed again with above mixture for 30 min at 37 ºC, dehyrated with 100% acetonitrile for 30 min, and dried in a Speedvac (room temperature, 15 min). The proteins were digested with 15 ng/μL chymotrypsin in 25 mM NH 4 HCO 3 at 37 ºC for 16 h. The resulting chymotrypsin peptides were identified by mass spectrometry analysis, which was completed by Bio-Tech Pack Technology (Beijing, China) as Samyn et al. (2005).
DNA/RNA extraction, gene cloning, and sequencing Genomic DNA was extracted from fresh leaves as previously described (Doyle and Doyle 1987). A pair of allele-specific PCR primers (F1/R1) (Table S3) were designed based on the published HMW-GS gene sequences. A high-fidelity LA-Taq polymerase (Takara, Dalian, China) was used for cloning the complete coding sequence (CDS). PCR amplifications were conducted in a 25 μL volume, consisting of genomic DNA or cDNA, 100 μM of each dNTP, 1.5 mM of Mg 2+ , 4 pmol of each primer, 0.75 U LA-Taq polymerase and 2.5 μL of 10 × buffer. The PCR reactions were run in a Mastercycler pro thermal cycler (Eppendorf, Hamburg, Germany) with the following program: an initial step of 94 ºC for 5 min; 35 cycles of 94 ºC for 45 s, 61 ºC for 30 s and 72 ºC for 2 min; then a final step of 12 min at 72 ºC. The PCR products were separated on a 1.5% agarose gel, and the expected fragment was purified and inserted into the pMD19-T vector (Takara). Positive colonies were sequenced by Sangon Biotech (Shanghai, China). Both the cloning and sequencing experiments were repeated independently at least three times.
Immature WT and mutant seeds (25 dpa) were collected and ground to a fine powder in liquid nitrogen. Total RNA was extracted from the powder using the MiniBEST Universal RNA Extraction Kit with DNase (Takara). The RNA concentration was determined with the NanoDrop One Spectrophotometer (Thermo Fisher Scientific). The RNA samples were reverse transcribed using the Prime Script™ 1st Strand cDNA Synthesis Kit (Takara). The CDS and the 3′ untranslated region (UTR) of the Dy10 allele in the WT and mutant were cloned with primers F2/R2 (Table S3) and the cDNA.

Genotyping analysis by Kompetitive Allele-Specific PCR (KASP)
The genotypes were confirmed by KASP genotyping (LGC Genomics, Teddington, Middlesex, UK). Briefly, genomic DNA samples from the WT, mutant, and their hybrid progeny (F 1 generation) were diluted to a uniform concentration. The equivalent of 50 ng DNA per sample was used for genotyping. The KASP primers (F5/F6/R5 or R6) (Table S3) were designed according to the manufacturer's protocols. The experiment was completed with the Master Mix and the CFX96™ Real-Time PCR system (Bio-Rad).

Expression of Dy10 alleles in Escherichia coli
The Dy10-m619SN gene sequence was re-amplified using the F3/R3 primer pair (Table S3) to remove the signal peptide sequence and to add NdeI and XhoI restriction sites. The PCR product was inserted into the pET-30a bacterial expression vector. The recombinant plasmid was inserted into E. coli strain BL21 (DE3) cells (TransGen Biotech, Beijing, China). Gene expression was induced by adding 1 mM isopropyl β-D-thiogalactopyranoside (IPTG) to the bacterial culture, which was incubated for 5 h until the OD 600 reached 0.6. The expressed proteins were extracted by a centrifugation at 13,800 g for 5 min and then separated by SDS-PAGE. Additionally, the nucleotide sequence encoding the N-Dy10-m619SN deduced amino acid sequence was amplified using the F4/R4 primer pair (Table S3) and then inserted into pET-30a for the subsequent expression in E. coli as described above.

Transient expression in immature endosperm
Transient expression assays were conducted using wheat endosperm at 14 dpa. The full-length Dy10 and Dy10-m619SN CDSs were amplified with the F7/ R7 primer pair (Table S3) and then inserted into the pCAMBIA1302 vector. The expression of the constructs comprising Dy10 and Dy10-m619SN fused to the gene encoding the green fluorescent protein (GFP) was under the control of the CaMV 35S promoter. The resulting recombinant plasmids (p1302-Dy10-GFP and p1302-Dy10-m619SN-GFP) were inserted into separate immature endosperm samples via particle bombardment. The transformed endosperm samples were incubated at 25 ºC for 48 h (Vicente-Carbajosa et al. 1998). The empty vector was used as a control. The expressed proteins were detected in a western blot using a rabbit anti-GFP polyclonal antibody (1:2,000; Abcam, Cambridge, UK) and an HRPconjugated anti-rabbit secondary antibody (1:5,000; Sigma-Aldrich).

Measurement of protein body sizes
Wheat caryopses were collected at 21 dpa and quickly cut into 2-mm slices starting from the center. The slices were fixed, rinsed, and dehydrated, after which they were embedded in resin at room temperature and polymerized at 55 ºC. Sections (1 µm thick) were prepared and stained for 45 s with 1% naphthol blue black dye prepared in 7% (v/v) acetic acid. Images were captured using a Zeiss Axiophot light microscope (Zeiss, Oberkochen, Germany). Protein body sizes were analyzed using ArcMap™ and ArcGIS™ software packages (ESRI, Redlands, CA, USA) as previously described (Savill et al. 2018). Analyses were completed with three biological replicates.

Scanning electron microscopy
To compare the microstructures of the WT and mutant samples, mature seeds were prepared, with three biological replicates per treatment. All samples were carefully deposited on a silicon wafer. After spraying samples with gold particles, they were analyzed using the SEM-3500 system (Hitachi High-Technologies Corp., Tokyo, Japan).

Measurement of processing quality parameters
The mature grains for each line in field trials were harvested and milled, after which the flour was prepared for analyses of typical processing trait parameters.
Seed moisture was adjusted to 16.5% and milled to white flour with the CD1 Laboratory Mill (CHO-PIN Technologies, Villeneuve-la-Garenne Cedex, France) according to the AACC Approved Method 26-70 (AACC International 2010). The grain protein content, Zeleny sedimentation value, wet gluten content, and gluten index were determined based on the standard method in GB/T 17,320-2013. Specifically, the Kjelec 8400 automatic azotometer (FOSS, Bremen, Denmark), a Zeleny analysis system (CAU-B, Beijing, China), and a Glutomatic 2200 system (Perten, Huddinge, Sweden) were used. Rheological properties were determined with a standard farinograph (Brabender GmbH & Co, KG, Germany), following the AACC Approved Method 54-21 (AACC International 2010).
The glutenin macropolymer (GMP) content was analyzed as previously described with some modifications (Jie et al. 2006). Briefly, flour samples (0.1 g) were mixed with 0.8 mL 50% (v/v) propan-1-ol for 30 min and then centrifuged for 3 min at 6,500 g. The supernatant was removed and the pellet was resuspended in 0.8 mL 50% (v/v) propan-1-ol containing 2% (w/v) DTT for 1 h at 60 ºC. The suspension was centrifuged for 10 min at 6,500 g. Each extraction step was repeated three times. The supernatant was mixed with isopyknic turbidimetry reagents [1.2% (w/v) 5-sulfosalicylic acid and 0.8% (w/v) sodium sulfate] for 20 min at 25 ºC. The absorbance was measured at 540 nm with a Microplate Reader (Thermo Scientific). The GMP content was calculated using Albumin Bovine V (Solarbio, Beijing, China) as the standard.
Glutenins and gliadins were extracted and analyzed as previously described ). Additionally, a baking test was performed using a published procedure (Xu et al. 2018). The baking procedure (standard rapid-mix-test) was completed using 36 g flour at 14% moisture content, with two replicates per sample. The loaf volume was determined with the BVM6630 volume meter (Perten, Stockholm, Sweden) following the manufacturer's instructions. Biscuits were prepared according to the AACC Approved Method 10-52 (AACC International 2010), with two replicates per flour sample. The diameter and thickness of each biscuit was measured as previously described (Mudgil et al. 2017), and spread ratio was calculated as the ratio of diameter to thickness. Biscuit hardness was determined using a texture analyzer (TA.XTC, BisinTech, Shanghai, China) following the method as previously described (Zhang et al. 2007) with some modifications. The biscuit was divided into six equal pieces using a knife, and the slice was placed on a flat metal plate and compressed to farcture using a 0.5 inch in diameter cylindrical acrylic probe (TA/0.5) at a speed of 1.0 mm/s. The peak value of fracture force was recorded as the hardness of biscuit.

Statistical analysis
The Chi-squared (χ 2 ) test was used for the segregation analysis, whereas the t-test was used to determine the 1 3 significance of any differences. Data analyses were completed with the Data Procession System software (version 12.01) (Zhejiang University, Hangzhou, China) (Tang and Zhang 2013).

Results
Identification of the Dy10-m619SN allele A mutant with six HMW-GSs, which were detected as distinct bands on an SDS-PAGE gel, was isolated after treating "Shumai 482" seeds with 0.8% EMS. No significant differences in agronomic characteristics were detected between the WT and mutant after backcrossing (Fig. 1a-d and Table S1). The SDS-PAGE analysis indicated that in addition to the five normally expressed HMW-GSs, the mutant has an extra protein band in the HMW-GS zone that is smaller than the Dy10 subunit of the WT wheat (Fig. 1e). The extra protein band was confirmed as a HMW-GS in an immunoblot involving rabbit anti-HMW-GS glutenin polyclonal antibodies, which can cross-react with both x-and y-type subunits (Fig. 1f).
Considering that the Ay subunit is not expressed in "Shumai 482," but is normally expressed in some wheat materials (Jiang et al. 2009;Roy et al. 2018;Wang et al. 2018b), we cloned the Ay gene sequence (same as GenBank No.: MF429893) and confirmed that it is silenced in both "Shumai 482" and the mutant. The Ay coding region includes four premature stop codons. Additionally, a PCR product corresponding to the expected size of the Ay coding region was not amplified when cDNA derived from young seeds was used as the template. Thus, the extra protein detected in the mutant is not the Ay subunit.  To identify the genomic loci associated with the extra protein, the mutant line was crossed with "Chinese Spring." An SDS-PAGE analysis indicated that the extra protein band was associated with Dx5 + Dy10 in the F 2 population. Moreover, the distribution of the genotypes [Dx5 + Dy10 (with the extra protein): heterozygous: Dx2 + Dy12] in 257 seeds conformed to a Mendelian ratio of 1:2:1 ( Table 1). The x-and y-type HMW-GS genes at the Glu-D1 locus are tightly linked (Payne 1987). These findings suggest that the gene encoding the extra protein is a new Dx5 or Dy10 allele.
To identify the extra protein, the corresponding band was manually cut from the SDS-PAGE gel. The subsequent analysis by mass spectrometry revealed that two peptides (QVVDQQLAGRLPWSTGLQMR and SVAVSQVAR) from the extra protein subunit are identical to Dy10, but differ from the other HMW-GSs ( Fig. S2b and S2c). This implies that the extra protein is derived from the Dy10 subunit.
The differences in the electrophoretic mobility among the HMW-GSs in the SDS-PAGE gel are mainly due to variations in the number of repeating units comprising the central repetitive domain (Shewry et al. 1995;Wang et al. 2018a) as well as amino acid substitutions (Li et al. 2014;Wang et al. 2016). A single amino acid substitution in the Dy10 subunit increased the stability of the secondary structure, protecting it from the denaturant SDS and increasing the mobility of the protein in the SDS-PAGE gel. However, the protective effect of the amino acid change may be eliminated by the addition of urea, which is a stronger denaturant than SDS, capable of thoroughly breaking down the secondary structure of proteins (Lafiandra et al. 1999;Wang et al. 2016). Thus, the SDS-PAGE analysis was repeated with a gel containing urea. In contrast to our results from an earlier study (Wang et al. 2016), the extra protein exhibited greater mobility than Dy10 in the presence of urea (Fig. 1g). Furthermore, similar to Dy10 (GenBank No.: X12929) in the WT wheat, the cloned open reading frame of Dy10 (GenBank No.: MW001617) in the mutant comprises 1,947 bp and encodes 648 residues (Fig. S2a), with no deletion of repeating units detected at the DNA and RNA levels (Fig. 1h).
Despite the sequence similarities, a missense mutation (G to A) was identified (Fig. S2a), resulting in a serine (Ser)-to-asparagine (Asn) substitution at the 619th residue within the C-terminal domain (Fig. S1e). To efficiently detect the point mutation and develop suitable markers for molecular breeding, specific KASP markers were designed (Fig.S1c). This new Dy10 allele was designated as Dy10-m619SN, and the observed protein bands on the SDS-PAGE gel corresponded to the full-length Dy10-m619SN (Full-Dy10-m619SN) and the extra protein N-Dy10-m619SN. Their predicted relative molecular masses are 67.5 kDa and 64.5 kDa, respectively. To assess the possibility of alternative splicing during transcription, cDNA samples derived from the immature seeds [i.e., 25 dpa] of the mutant were used as the templates for amplifying the CDS and the 3′ UTR. The results revealed a lack of spliced sequences ( Fig. 1h and Fig. S2d), which is consistent with a partial posttranslational cleavage of Dy10-m619SN.
To further analyze this cleavage in developing seeds, the proteins in wheat seeds collected at specific time-points after anthesis were analyzed by SDS-PAGE. The protein bands corresponding to Full-Dy10-m619SN and N-Dy10-m619SN were simultaneously observed as early as 14 dpa (Fig. 2c).
To confirm that N-Dy10-m619SN is the result of a post-translational cleavage of Full-Dy10-m619SN and to detect the downstream peptide (C-Dy10-m619SN) after the cleavage, a fusion construct comprising the Dy10-m619SN CDS and the GFPencoding gene was transiently expressed in immature wheat endosperm. The two proteins detected by the anti-GFP antibody [ Fig. 2e (lane 3)] were revealed to be Full-Dy10-m619SN (higher molecular mass) and C-Dy10-m619SN (lower molecular mass). However, N-Dy10-m619SN was not detected by the antibody because it was not tagged with GFP. In contrast, only one protein band with the same molecular mass as Full-Dy10-m619SN was detected when the WT Dy10-GFP protein was expressed [ Fig. 2e (lane 4)]. These observations indicate that Full-Dy10-m619SN is partially cleaved to produce N-Dy10-m619SN and C-Dy10-m619SN in vivo.

Effects of Dy10-m619SN on HMW-GS accumulation
To evaluate the extent of the cleavage in mature seeds, we determined the relative contents of Full-Dy10-m619SN and N-Dy10-m619SN in the mutant. The calculated proportions of Full-Dy10-m619SN and N-Dy10-m619SN were 47.4% and 52.6%, respectively (Fig. 2d), and the abundance of the latter was significantly greater than that of the former (P < 0.01). Effects of Dy10-m619SN on gluten microstructure To investigate the overall features of storage protein accumulation, the protein body content in the semithin section of the endosperm was analyzed. In the mutant section, small protein bodies (0.2-0.5 μm 2 ) represented 38-45% of the total content in each zone, which was substantially higher than the proportion in the WT Sects. (23-30%) ( Fig. 3a and b). The proportions of the other protein bodies (1.5-228.8 μm 2 ) were significantly lower in the mutant sections than in the WT sections. Next, the microstructures of mature WT and mutant seeds were compared with a scanning electron microscope. We detected more protein matrix tightly adhered to starch granules in the WT samples than in the mutant samples (Fig. S3). Additionally, the GMP content was significantly lower in the mutant than in the WT (P < 0.05) ( Table 2). These data clearly indicate that the partial post-translational cleavage decreases glutenin polymer size.
To assess the possible interference by the glutenin and gliadin contents and compositions, we analyzed the grain protein, gliadin, and glutenin contents. There was no significant difference in the grain protein contents (Table 2) and the gliadin contents (Table S2) between the WT and mutant samples. Interestingly, the glutenin content was slightly but significantly lower in the mutant samples than in the WT samples (P < 0.01), possibly because of differences in peptide solubility resulting from the posttranslational cleavage.
Effects of Dy10-m619SN on wheat processing quality To evaluate the effects of Dy10-m619SN on wheat processing quality parameters, the wet gluten content, gluten index, Zeleny sedimentation value, and dough rheological properties were compared between the WT and mutant. The data revealed that the wet gluten content of the mutant was significantly higher than that of the WT. In contrast, the Zeleny sedimentation value and gluten index were lower in the mutant than in the WT. Moreover, the dough development and stability times (i.e., rheological properties) were significantly shorter for the mutant than for the WT (Table 2). These data indicate that the post-translational cleavage decreases dough strength.
We analyzed the cookie-making quality based on the diameter, thickness, spread ratio, hardness and sensory. In contrast to those of WT line, the biscuits of mutant shown larger diameter, less thickness, higher spread ratio, similar hardness and similar sensory ( Fig. 4 and Table 2). We further analyzed the bread-making quality based on the loaf volume and sensory. Specifically, the loaf volume was significantly lower for the mutant wheat than for the WT wheat ( Fig. S4 and Table 2). There was no significant difference in loaf sensory. These results indicate that Dy10-m619SN has a positive effect on the cookiemaking quality.

Discussion
The HMW-GSs are the main determinants of end-use quality because of their distinct molecular structures, especially their interchain disulfide bonds (Li et al. 2021;Shewry et al. 1992Shewry et al. , 2003aYazar et al. 2017;Yue et al. 2019). The HMW-GS polymer backbone is established by end-to-end or head-to-tail linkages due to interchain disulfide bonds (Graveland et al. 1985;Wieser 2007). The basic molecular unit of glutenin polymers might consist of two y-type HMW-GSs, four x-type HMW-GSs, and many LMW-GSs covalently linked by interchain disulfide bonds (Wieser 2007). Consistent with the reported data, we proposed a model for glutenin polymers that explains the effect of Dy10-m619SN on wheat processing quality (Fig. 5). The Cys residue in the C-terminal domain of Dy10 (identical to the Cys in C-Dy10-m619SN) is essential for the formation of the interchain disulfide bonds that contribute to the elongation of glutenin polymers. We applied a series of approaches to demonstrate the partial post-translational cleavage at the C-terminal domain of Full-Dy10-m619SN. Additionally, Full-Dy10-m619SN (seven Cys residues) can be efficiently processed (52.6%) into N-Dy10-m619SN (six Cys) and C-Dy10-m619SN (one Cys). The change in the number of Cys residues in these two peptides leads to the production of competitive chain terminators ( Fig. 5b and c) and decreases the glutenin polymer size and dough strength, ultimately resulting in increased cookie-making quality (Table 2).
Silencing HMW-GS genes represents one strategy for breeding soft wheat. However, there are reportedly positive and negative effects associated with this approach. For example, the deletion of a single HMW-GS gene in the Ningmai 9 soft wheat variety 1 3 resulted in increased sugar snap cookie-making quality (Zhang et al. 2016). In contrast, other studies revealed that the absence of HMW-GSs Dy12 (Chen et al. 2021), Bx7 andBy9 (Chen et al. 2019), leads to decreased cake and biscuit quality, possibly because of the associated decrease in glutenin content and increased gliadin content. The polymeric glutenins are mostly responsible for dough elasticity, whereas the monomeric gliadins influence extensibility-related characteristics (Hoseney 1986). Thus, the glutenin-togliadin ratio can directly affect the balance between dough strength and extensibility, making it critical for the quality of end products (Barak et al. 2013;Wrigley et al. 2006). In this study, the weakening of gluten due to Dy10-m619SN was reflected by the gluten microstructure and certain processing quality parameters. Additionally, Dy10-m619SN did not affect the glutenin-to-gliadin ratio, but it improved the cookiemaking quality of wheat. Therefore, the post-translational cleavage of HMW-GSs is an important consideration for the breeding of soft wheat.
Several studies proved that Full-Dy10-m619SN can be efficiently cleaved after the Asn residue (Gruis et al. 2004;Hara-Nishimura et al. 1995, 2004Kumamaru et al. 2010;Wang et al. 2009). Two conserved Asn residues unique to the C-terminal domain of the By subunits result in two minor proteoforms detectable on two-dimensional electrophoresis gels (Nunes-Miranda et al. 2017). The HMW-GS CDSs encoding the C-terminal domain are highly conserved, and usually do not include a codon for Asn (Fig. S5). Therefore, on the basis of Table 2 Comparison of the major parameters related to quality traits, rheological properties, and baking quality between the WT and mutant Data are presented as the mean ± standard deviation * and ** represent significant differences at P < 0.05 and P < 0.01, respectively   Table 2 the results of the current study, we hypothesize that a modest increase in the number of Asn residues in the HMW-GS C-terminal domain via site-specific mutagenesis can improve the cookie-making quality of wheat.
In conclusion, a new Dy10 allele, Dy10-m619SN, was identified by screening an EMS-induced mutant population derived from the common wheat cultivar "Shumai 482." A subsequent analysis revealed that Dy10-m619SN expression results in three peptides in the wheat endosperm because of a post-translational cleavage. Moreover, the effects of Dy10-m619SN on wheat processing quality as well as the underlying mechanism were clarified. This study provides researchers and breeders with new genetic information relevant for enhancing wheat quality through breeding.  Full-Dy10-m619SN is cleaved after the Asn site, which changes the number of cysteine residues in the peptides. "-SH" indicates the free sulfhydryl of cysteine. (c) Model of the mutant gluten polymers. Dashed lines indicate potential interaction sites resulting from the competition involving N-Dy10-m619SN and C-Dy10-m619SN as chain terminators. "S" and "N" in red respectively indicate the serine (Ser) and asparagine (Asn) at the 619th residue