High-molecular weight glutenin subunits (HMW-GS) have been shown to play a crucial role in determining the processing properties of the wheat grain. We have produced marker-free transgenic rice plants containing a wheat Glu-1Bx7 gene encoding the HMG-GS from the Korean wheat cultivar ‘Jokyeong’ using the Agrobacterium -mediated co-transformation method. The Glu-1Bx7 -own promoter was inserted into a binary vector for seed-specific expression of the Glu-1Bx7 gene. Two expression cassettes comprised of separate DNA fragments containing only Glu-1Bx7 and hygromycin phosphotransferase II ( HPTII ) resistance genes were introduced separately to the Agrobacterium tumefaciens EHA105 strain for co-infection. Each EHA105 strain harboring Glu-1Bx7 or HPTII was infected to rice calli at a 3:1 ratio of Glu-1Bx7 and HPTII , respectively. Then, among 216 hygromycin-resistant T ₀ plants, we obtained 24 transgenic lines with both Glu-1Bx7 and HPTII genes inserted into the rice genome. We reconfirmed integration of the Glu-1Bx7 gene into the rice genome by Southern blot analysis. Transcripts and proteins of the wheat Glu-1Bx7 were stably expressed in the rice T ₁ seeds. Finally, the marker-free plants harboring only the Glu-1Bx7 gene were successfully screened at the T ₁ generation. Rice quality has become an important consideration due to the growing demands of local and world food markets. The increasing population of rice consumers, especially in Africa, Asia, and Southeast Asia, depend solely on this cereal food. Rice flour is used in many food products and improving rice quality is of great relevance to many Asian countries. However, dough made from rice lacks extensibility and elasticity, whereas that of wheat is suitable for many food products including breads. Wheat flour is different from other cereal flours, including rice, because it contains gluten that gives it the elasticity and extensity required for bread-making [5] . Gluten consists mainly of two types of seed storage proteins, the glutenins and the gliadins. Glutenins are classified into highmolecular-weight glutenin subunits (HMW-GS) and low-molecular-weight glutenin subunits (LMW-GS). Although the HMW-GS contribute only about 5% of the total protein in mature wheat kernels [26] , the elasticity of wheat dough depends mainly on the HMW-GS, so their networking is important determinants of bread-making quality [19 , 20] . The HMW-GS are encoded by the Glu-A1, Glu-B1 and glu-D1 genes on the long arm of chromosomes 1A, 1B and 1D, respectively [21] . Each locus includes two genes linked together encoding two different types of HMW-GS, x- and y-type subunits [20 , 22] . Several HMW-GS genes have been shown to be functional when transformed into Escherichia coli [10] , tobacco [23] , wheat [1 , 2 , 5 , 6] and tritordeum [24] . Development of transgenic plants increasing bread-making quality using various HMW-GS genes is good way to solve lacks of rice extensibility and elasticity. Araki’s [4] group has developed transgenic rice using transformation-competent artificial chromosome (TAC) clones harboring a HMW-GS and LMW-GS genes from wheat genomic DNA. All transgenic lines had HMW-GS subunit proteins in the rice endosperm, and the expressed proteins are processed at the same site as the mature protein in wheat seeds. The persistence of selectable marker genes in transgenic crops destined for field cultivation and human food leads to serious public concerns about the safety of transgenic crops, even though several risk-assessment reports [14 , 25] have shown that neither the genes nor their products are harmful to human or environmental health. Moreover, generating marker-free transgenic plants responds not only to public concerns over the safety of genetically engineered crops, but supports multiple transformation cycles for transgene pyramiding. Repeated use of the same promoter and a polyadenylation signal for different selectable marker genes could result in transcriptional gene silencing [12] . Therefore, eliminating selectable marker genes is crucial for stacking multiple traits in a transgenic plant. In this study, we generated marker-free transgenic rice expressing the wheat HMW-GS protein without any herbicide or antibiotic resistance marker genes using the co-transformation method. The marker-free transgenic plant expressing Glu-1Bx7 gene is critical material for generating transgenic plant advanced quality processing of bread and noodle without antibiotic markers. ‘Jokyeong’ ( Triticum aestivum L. cv. Jokyeong) was used for cloning the Glu-1Bx7 gluenin gene. The Glu-1Bx7 gene was amplified by polymerase chain reaction (PCR) of genomic DNA using the primers Bx7-CF (primer sequences: 5'-AGGGTACCGAGATGGCTAAGCGCCTGG-3') and Bx7-CR (primer sequences: 5'-GATCTAGATCACTGCCTGGTCGACAATG-3'), which were designed from a sequence on GenBank (accession no. X13927). The PCR temperature cycling conditions were 4 min at 94℃, followed by 35 cycles of 94℃ for 30 sec, 60℃ for 30 sec, 72℃ for 2 min, and a final extension at 72℃ for 10 min. The amplification products were separated on a 1% agarose gel and visualized with EtBr. The amplified products were sub-cloned using a TOPO TA Cloning kit for sequencing (Invitrogen, Carlsbad, CA, USA). To make a marker-free vector, we first inserted Glu-Bx7 -own promoter from wheat cultivar ‘Jokyeong’ into the pBTEX binary vector, which modified from pCAMBIA1300 binary vector. The HPTII expression cassette ( CaMV 35S promoter- HPTII gene- CaMV 35S terminator) in the pBTEX binary vector was removed by Xho I and EcoR I restriction enzyme treatment. After klenow enzyme treatment for blunt ligation, the vector was self-ligated. Then, amplified the Glu-1Bx7 gene with the EcoR I and Kpn I restriction enzyme sites was constructed into pBTEX binary vectors under the control of Glu-1Bx7 -own promoter ( Fig. 1A ). The positive selectable marker cassette for co-transformation was used by an empty pBTEX binary vector ( Fig. 1B ). Competent Agrobacterium tumefaciens EHA105 was transformed with Glu-1Bx7 - cloned binary vector and an empty vector containing HPTII for the selectable marker using the freeze-thaw method [7] . T ₀ plants were selected on YEP media containing kanamycin (50 mg/l). Transformation was confirmed by PCR amplification of plasmids mini-prepped from each Agrobacterium strain [3] . Mature seeds of Oryza sativa L. subsp . japonica var. Dongjin were used to induce callus formation on callus induction (CI) medium [N ₆ salts [9] with vitamins, 2.5 g/l proline, 2 mg/l 2,4-D, 0.5 g/l casamino acid, 30 g/l sucrose and 2 g/l gelrite, pH 5.7]. After 21 days of incubation in the dark at 25℃, the scutellum-derived calli were excised and preincubated on CI medium for 1 week. Agrobacterial cells were grown on YEP solid medium containing antibiotics at 25℃ for 2 days. And then, agrobacterial cells were resuspended in suspension medium (N ₆ salts with vitamins, 2 mg/l 2,4-D, 0.5 g/l casamino acid, 30 g/l sucrose, and 10 g/l glucose, pH 5.7) with 200 µM acetosyringone as a final concentration. After two Agrobacterium cells were mixed in a 3:1 ratio of EHA105 with Glu-1Bx7 gene expressing cassette and EHA105 with HPTII gene expressing cassette, the calli were transformed by swirling in the mixture of Agrobacterium cultures for 30 min. The calli were blotted on Whatman no. 1 paper and cocultivated on the cocultivation medium (N ₆ salts with vitamins, 2 mg/l 2,4-D, 0.5 g/l casamino acid, 30 g/l sucrose, 10 g/l glucose, and 2 g/l gelrite, pH 5.2 with 200 µM acetosyringone as a final concentration). After 3 days, the calli were washed with liquid CI medium supplemented with 250 mg/l cefotaxime and 150 mg/l and placed on the selection medium (CI medium supplemented with 50 mg/l hygromycin, 250 mg/l cefotaxime). After selection and regeneration, the regenerated plantlets were acclimatized and grown in a greenhouse. PCR was performed with the GeneAmp System 9700 (Applied Biosystems, Foster City, CA, USA) with a gene-specific primer set ( Glu-1Bx7 ; forward 5'- AGGGTACCGAGATGGCTAAGCGCCTGG -3', reverse 5'- GATCTAGATCACTGCCTGGTCGACAATG -3', HPTII ; forward 5'- CGCTTCTGCGGGCGATTT -3', reverse 5'- CCCATTCGGACCGCAAGGA -3') and EF Taq DNA polymerase (Solgent Co. Seoul, South Korea). Each reaction mixture (30 µM) consisted of 10 mM Tris-HCl (pH 9.0), 1.5 mM MgCl ₂ , 40 mM KCl, 250 µM dNTPs, and 1 U Taq DNA polymerase. Amplified products were separated on a 1% agarose gel, stained with EtBr, and visualized with a UV illuminator. Rice genomic DNA was prepared by the CTAB extraction method [29] . Aliquots of 5 µg of purified DNA were digested with restriction endonuclease ( EcoRI ), size-fractionated on a 0.8% agarose gel, and the DNA was transferred to a nylon membrane through capillary blotting in 10× SSC (Gene Screen, DuPont, Wilmington, DE, USA). The blots were labeled using AlkPhos Direct (Amersham, GE Healthcare. Piscataway, NJ, USA) according to the manufacturer’s instructions. After hybridization, the filters were washed for 30 min at 55℃ to remove unlabelled probe. Subsequently, CD-star Detection Reagent (Amersham, GE Healthcare. Piscataway, NJ, USA) was used to detect and generate signals. T ₁ generation seeds were frozen in liquid nitrogen and then ground to powder using a mortar and pestle. Total RNA was extracted using a method reported previously [32] . The isolated RNA preparations were then reverse-transcribed with oligo-dT primer and a First Strand cDNA Synthesis kit for RT-PCR (Roche Co., Basel, Switzerland) with gene-specific primers. The primers were as follows: Glu-1Bx7 forward 5'- AGGGTACCGAGATGGCTAAGCGCCTGG -3', Glu-1Bx7 reverse 5' GATCTAGATCACTGCCTGGTCGACAATG -3'; OsActin primers were used as internal standards for mRNA expression profiling [17 , 31] . The PCR conditions consisted of initial denaturation at 94℃ for 4 min, followed by 30 cycles at 94℃ for 1 min, 60℃ for 1 min, 72℃ for 2 min, and a final extension at 72℃ for 10 min. The experiments were repeated three times and all produced similar results. The OsActin control primers were 5'- GGA ACT GGT ATG GTC AAG GC -3' and 5'- AGT CTC ATG GAT ACC CGC AG -3' [8] . T ₁ generation seeds were frozen in liquid nitrogen and then ground to powder using a mortar and pestle. Total storage proteins in the rice endosperm were extracted with 50 mM Tris-HCl (pH 8.0) containing 2% SDS, 50% of 1-propanol and 1% of dithiothreitol, as described [4] . Amount of extracted total proteins was measured by Nanodrop Spectrophotometer (ND-1000, Thermo Fischer Scientific, Wilmington DE, USA). Western blot analysis was performed as described [18] . To improve improved dough properties of rice flour, we cloned high-molecule weight glutenin subunit gene ( Glu-1Bx7 ) from genomic DNA of Triticum aestivum cv. Jokyeong by PCR analysis with specific primers. We tried to generate marker-free transgenic plant expressing only Glu-1Bx7 gene on rice through Agrobacterium -mediated co- transformation system. To make a marker-free vector, we first remove the HPTII expression cassette ( CaMV 35S promoter- HPTII gene- CaMV 35S terminator) by treatment of XhoI and EcoRI restriction enzymes and inserted wheat Glu-1Bx7 -own promoter. And then Glu-1Bx7 gene was constructed under the control of Glu-1Bx7 -own promoter into pBTEX vector, which was modified pCAMBIA1300 binary vector ( Fig. 1A ). And original pCAMBIA1300 binary vector harboring HPTII gene was used to select hygromycin resistant T ₀ plants ( Fig. 1B ). The two expression binary vectors were separately introduced into A. tumefaciens EHA105 strain for plant transformation. Each binary vector was rescued from the EHA105 strain harboring Glu-1Bx7 and the HPTII , and the HPTII and Glu-1Bx7 genes were validated by PCR analysis. Each EHA105 strain harboring Glu-1Bx7 expression vector or HPTII expression vector was cultured in YEP medium for plant transformation. The cultured cells were resuspended to OD ₆₀₀ =0.1 in AAM medium [11] , and each Glu-1Bx7 and HPTII cell was added at a 3:1 ratio. These mixed cells were co-infected into rice calli. The transformed calli were selected with hygromycin because we co-infected calli with the HPTII gene. We obtained 216 independent hygromycin-resistant T ₀ plants through co-infection in the Agrobacterium transformation system. Genomic DNA from 216 independent T ₀ plants was extracted and insertion of HPTII and Glu-1Bx7 genes was analyzed using PCR analysis with gene specific primers. As shown in Fig. 2 , HPTII gene in all of T ₀ plants was amplified, but no PCR products of ‘Jokyeong’ and ‘Dongjin’ used as negative controls were detected. Next, we investigate the insertion of Glu-1Bx7 gene into rice genome within T ₀ plants by PCR analysis. Among 216 independent transgenic lines, Glu-1Bx7 gene in 24 T ₀ plants was amplified PCR product which is same PCR product size of ‘Jokyeong’ and plasmind used as positive controls ( Fig. 2 ). This result means that 24 transgenic lines harbored both Glu-1Bx7 and HPTII genes. And the frequency co-transformation was about 11.1% in our experimental system ( Table 1 ). We performed Southern blot analysis with the Glu-1Bx7 -own promoter as probes to validate their insertion and guess segregation ratio of the marker-free plant in T ₁ plants. One or seven signal bands in 24 selected T ₀ plants lines were detected ( Fig. 3 ). Because Glu-1Bx7 expression in rice endosperm is important for rice flour quality and we used Glu-1Bx7 -own promoter to express Glu-1Bx7 , total RNA from one copy-inserted T ₁ generation transgenic seeds was extracted, and Glu-1Bx7 gene transcript level was examined by semi-quantitative RT-PCR. The Glu-1Bx7 transcripts were successively expressed in the T ₁ generation transgenic seeds, whereas Glu-1Bx7 expression in ‘Dongjin’ was not detected ( Fig. 4 ). OsActin expression was used as a quantitative control. And we analyzed the protein expression of Glu-1Bx7 by Western blot with an anti x-type HMW specific antibody. The six transgenic plants (4, 17, 18, 19, 20 and 22) which were shown abnormal morphologies comparing with ‘Dongjin’ were removed. After total protein extraction from wheat (‘Jokyeong’ cultivar), ‘Dongjin’ and transgenic plants, 0.5 πg of wheat and 40 πg of total protein extract of transgenic plants were used for SDS-PAGE. The immunospecificity of the anti-x-type HMW specific antibody was verified by in vivo experiment. Although the protein bands were well detected in transgenic plants, however, the level of protein expression was not depended on their inserted copy number ( Fig. 5 ). Multicopies of Glu-1Bx7 gene were inserted into genome in some transgenic plants (5, 23 and 16), no proteins were detected. We guess that this phenomenon is homology-dependent gene silencing in plants [2] . To select Glu-1Bx7 marker-free plants harboring only the Glu-1Bx7 gene, 90 T ₁ generation seeds of the transgenic plant 8 were planted in soil and genomic DNA was extracted from leaves of plantlets after 4 weeks. Insertion of the Glu-1Bx7 and HPTII genes was investigated by PCR analysis. As shown in Fig. 6 , most of the transgenic lines harbored both the Glu-1Bx7 and HPTII genes, and some inserted only the HPTII gene. However, transgenic 3, 10, 18 and 21 lines contained only the Glu-1Bx7 gene ( Fig. 6 ). This result shows that marker-free plants containing only the Glu-1Bx7 gene were successfully screened at the T ₁ generation. Finally, we produced marker-free transgenic rice plants harboring Glu-1Bx7 gene for advanced quality processing of bread and noodle. Wheat gluten proteins are classified into two broad groups based on their aggregation and functional properties, including the glutenins, which form polymers stabilized by inter-chain disulfide bonds, and gliadins, which are present as monomers and interact by non-covalent forces [27] . In particular, HMW-GSs are minor components in terms of quantity, but they are key factors during bread making because they are major determinants of gluten elasticity [28] by promoting the formation of larger glutenin polymers. In this study, we cloned Glu-1Bx7 , which is one of HMW-GS genes, and validated the insertion of the Glu-1Bx7 gene in T ₀ plants through PCR analysis with gene specific primers and Southern blot analysis ( Fig. 2 , 3 ). The Glu-1Bx7 -own promoter was introduced for seed specific expression of Glu-1Bx7 gene. The transcript and protein of Glu-1Bx7 in the transgenic plants were stably expressed in T ₁ generation rice seeds ( Fig. 4 ). This result suggested that the protein processing system was conserved between rice and wheat. However, the level of protein expression was not depended on their inserted copy number ( Fig. 5 ). Genetic engineering of plants sometimes results in transgene silencing after integration into the genome, which may relate to a defense mechanism against foreign DNA expression [15 , 30] . This phenomenon may be related to homology-dependent gene silencing in plants. Homology-dependent gene silencing has attracted considerable interest because it may be detrimental to genetic engineering and also because of its usefulness as a tool to study the mechanisms involved in detecting and inactivating exogenous DNA [15 , 16] . Multi- copies of Glu-1Bx7 gene were inserted into genome in some transgenic plants (5, 23 and 16), no proteins were detected. We guess that this phenomenon is homology-dependent gene silencing in plants [2] . The co-transformation frequency in our experimental conditions was 11.1% ( Table 1 ). In a previous report, co-transformation frequency in rice was about from 2% to 14% [13] . This result indicates that transformation efficiency is dependent on rice cultivar and the experimental conditions. Although the generating of marker-free plants based on the Agrobacterium -mediated co-transformation using two different expression cassettes was need more time consuming and effort, this method could be efficiently produce marker-free transgenic rice plants. Finally, we obtained marker-free transgenic plants containing only Glu-1Bx7 gene from each of the T ₁ plants ( Fig. 6 ). This marker-free transgenic plant harboring Glu-1Bx7 will become useful material to optimize transgenic rice plants, which has advanced quality processing of bread and noodle by crossing with genetically engineered rice plants with other gluten genes.