Agrobacterium-mediated co-transformation of rice using two selectable marker genes derived from rice genome components
- First Online:
- Cite this article as:
- Wakasa, Y., Ozawa, K. & Takaiwa, F. Plant Cell Rep (2012) 31: 2075. doi:10.1007/s00299-012-1318-9
- 1.2k Views
A method for Agrobacterium-mediated co-transformation of rice (Oryza sativa L.) was developed using rice-derived selection markers. Two T-DNAs were efficiently introduced into separate loci using selectable marker gene cassettes consisting of the mutated acetolactate synthase gene (mALS) under the control of the callus-specific promoter (CSP) (CSP:mALS) and the ferredoxin nitrite reductase gene (NiR) under the control of its own promoter (NiR P:NiR). The CSP:mALS gene cassette confers sulfonylurea herbicide resistance to transgenic rice callus. The NiR P:NiR construct complements NiR-deficient mutant cultivars such as ‘Koshihikari’, which are defective in the regulation of nitrogen metabolism. In the present study, the CaMV35S:GUS and CaMV35S:GFP gene cassettes were co-introduced into the ‘Koshihikari’ genome using our system. Approximately 5–10 independent transgenic lines expressing both the GUS and GFP reporters were obtained from 100 Agrobacterium co-inoculated calli. Furthermore, transgenic ‘Koshihikari’ rice lines with reduced content of two major seed allergen proteins, the 33 and 14–16 kDa allergens, were generated by this co-transformation system. The present results indicate that the generation of selectable antibiotic resistance marker gene-free transgenic rice is possible using our rice-derived selection marker co-transformation system.
Key message An improved rice transformation method was developed based on Agrobacterium-mediated co-transformation using two rice genome-derived selectable marker gene cassettes.
KeywordsAcetolactate synthaseAgrobacteriumCo-transformationFerredoxin nitrite reductaseGM cropTransgenic rice
Green fluorescent protein
- GM crop
Genetically modified crop
Mutated acetolactate synthase
Ferredoxin nitrite reductase
Genetically modified crops (GM crops) are at present commercially cultivated in 29 countries worldwide. By 2011, it is estimated that the total global cultivated area of transgenic crops will reach approximately 160 million hectares (James 2011). The number of species and cultivated area of GM crops will certainly increase in the future.
Transformation of crop plants has been achieved by both direct (e.g., electroporation, particle-bombardment) and indirect (Agrobacterium-mediated) methods. The selection of transformants, which is critical in this process, is mostly based on the use of antibiotic-resistant genes. However, the presence of such selection markers in commercialized GM crops is not desired, which underscores the importance of developing marker-free transformation systems to eliminate selectable markers and generate a product that is acceptable by consumers. Various marker-free plant transformation systems have been developed including the multi-auto-transformation (MAT) vector system and the Cre-lox recombination system (Ebinuma et al. 1997; Endo et al. 2002; Sugita et al. 2000; Zhang et al. 2003). Although these systems are available and have been tested using various strategies, they have not yet been established as convenient tools for the production of marker-free transgenic plants. An Agrobacterium-mediated co-transformation system has been developed to produce marker-free transgenic plants (Baisakh et al. 2006; Depicker et al. 1985; Komari et al. 1996; Parkhi et al. 2005; Tu et al. 2003). In this system, transgenic plants are generated by introducing two T-DNAs containing a selectable marker gene and the target gene. The regenerated plants are selected for the presence of both T-DNAs and the selectable marker gene is removed by segregation in the seed progeny. Recently, the efficiency of co-transformation systems has been considerably improved. Sripriya et al. (2011) described a highly efficient Agrobacterium-mediated co-transformation method based on the use of high copy number binary vectors in indica rice. Agrobacterium-mediated co-transformation systems are simple in principle and are applicable to various plant species when compared with MAT vector and Cre-lox systems.
In a previous report, we described a marker-free like transformation system based on the use of native mutated-acetolactate synthase (mALS) under the control of a callus-specific promoter as rice genome-derived selectable marker gene (Wakasa et al. 2006). ALS is the first common enzyme in the biosynthetic pathway of the branched-chain amino acids leucine, isoleucine, and valine. Sulfonylurea, imidazolinone and pyrimidinyl carboxy herbicides are competitive inhibitors of ALS (Chaleff and Mauvais 1984; Shimizu et al. 2002). Mutated-ALS (mALS) has significantly lower affinity for these herbicides, and has been used to develop transformation systems. Certain mALSs were discovered as products of somaclonal variation during tissue culture of various plant species (Shimizu et al. 2002; Okuzaki et al. 2007).
‘Koshihikari’ is the most popular cultivated variety in Japan based on its superior eating quality. ‘Koshihikari’ is cultured in not only Japan but also China, Thailand, Philippine and the USA (Ito et al. 1995). However, ‘Koshihikari’ has not been used as a transgenic host because of its low regeneration and callus propagation abilities, which have been attributed to reduced expression of the ferredoxin nitrite reductase gene (NiR) in the callus. Nishimura and co-authors reported that not only lower expression of NiR mRNA but also insufficient unspliced NiR mRNA retaining the third intron were related to the low nitrite reductase (NiR) activity in ‘Koshihikari’ cells. Furthermore, rice varieties that can easily regenerate plantlets from callus have higher NiR activity than ‘Koshihikari’ (Nishimura et al. 2005). A specific DNA deletion (31 bp) in the 5′ untranslated region (5′UTR) of the NiR gene was detected in ‘Koshihikari’ (Ozawa and Kawahigashi 2006). These findings suggest that specific culture media and culture conditions are required for ‘Koshihikari’ transformation (Ozawa and Kawahigashi 2006; Wakasa et al. 2007). The NiR gene derived from rice cultivars such as ‘Kasarath’ and ‘Nipponbare’ was therefore used as a selection marker to overcome the transformation difficulties in the low-regeneration rice ‘Koshihikari’ (Nishimura et al. 2005; Ozawa and Kawahigashi 2006).
In the present study, we developed a high-frequency co-transformation system for ‘Koshihikari’ rice using the mALS gene driven by a rice callus-specific promoter (CSP:mALS) and the rice cultivar ‘Nipponbare’ NiR gene under the control of its own promoter (NiR P:NiR) as selectable marker genes. Approximately 8 % co-transformation efficiency (number of plants integrating two T-DNAs per callus used for Agrobacterium inoculation) was achieved by this method.
The proposed system has at least two important advantages. The use of two selectable marker genes derived from the rice genome is expected to increase public acceptance regarding food safety when compared with transgenic rice seeds produced using conventional antibiotic marker selection systems. In addition, multiple desired genes can be co-introduced into the rice genome, enabling the molecular breeding of transgenic rice plants carrying desirable traits.
Materials and methods
Rice (Oryza sativa L. cv. ‘Koshihikari’) was used for transformation.
Sterilized mature ‘Koshihikari’ seeds were sown on N6D solid medium [4 g/L CHU salt mixture (Wako Pure Chemical Industries, Tokyo, Japan), 30 g/L sucrose, 2.78 g/L proline, 100 mg/L myo-inositol, 300 mg/L casamino acids, 1 mL/L 1,000× N6-vitamin, 2 mg/L 2, 4-D, 0.4 % gelrite (Wako Pure Chemical Industries, Tokyo, Japan), pH 5.8] and cultured at 32 °C for 1 week. Small calli (2–5 mm) induced from seed scutellum were transferred to modified-DKN solid medium [30 g/L sucrose, 1.15 g/L proline, 100 mg/L myo-inositol, 300 mg/L casamino acids, 50 mL/L 20× DKN-macro stock (5.46 g/L NaH2PO4, 16 g/L KNO3, 1.34 g/L (NH4)2SO4, 5 g/L MgSO4·7H2O, 3 g/L CaCl2·2H2O), 1 mL/L 1,000× DKN micro stock (1.6 g/L MnSO4·4H2O, 2.2 g/L ZnSO4·7H2O, 3 g/L H3BO3, 125 mg/L CuSO4·5H2O, 125 mg/L Na2MoO4·2H2O), 1 mL/L 1,000× R2-Iron stock (7.5 g/L Na2 EDTA, 5.5 g/L FeSO4·7H2O), 1 mL/L 1,000× modified B5-vitamin (1 g/L nicotinic acid, 10 g/L thiamin HCl, 1 g/L pyridoxine HCl, 2 g/L glycine), 660 mg/L l-aspartic acid, 731 mg/L l-glutamine, 2 mg/L 2, 4-D, pH 5.8] and cultured for 3–4 weeks at 32 °C (16 h light/8 h dark).
Agrobacterium harboring binary vectors containing either the CaMV 35S promoter:GUS and CSP:mALS gene cassettes, or the CaMV 35S promoter:GFP and NiR P:NiR gene cassettes were precultured in AB solid medium (3 g/L K2HPO4, 1.3 g/L NaH2PO4·2H2O, 1 g/L NH4Cl, 150 mg/L KCl, 10 mg/L CaCl2·2H2O, 2.5 mg/L FeSO4·7H2O, 5 g/L glucose, 500 mg/L MgSO4, 1.5 % agarose, pH 7.2) for 3 days at 28 °C. Subsequently, Agrobacterium colonies were suspended in DKN-AS liquid medium (30 g/L sucrose, 1.15 g/L proline, 100 mg/L myo-inositol, 300 mg/L casamino acids, 50 mL/L 20× DKN-macro stock, 1 mL/L 1,000× DKN micro stock, 1 mL/L 1,000× R2-Iron stock, 1 mL/L 1,000× modified B5-vitamin, 660 mg/L l-aspartic acid, 731 mg/L l-glutamine, 2 mg/L 2, 4-D, 20 mg/L acetosyringone, pH 5.2) and the concentration was adjusted to OD600 = 0.05–0.1.
After co-cultivation, Agrobacterium-inoculated calli were collected into a 50-mL tube and washed with sterilized water three times. Washed calli were transferred onto modified-DKN solid medium containing 400 mg/L carbenicillin and cultured at 32 °C for 3 days as acclimation step (16 h light/8 h dark).
Calli were collected again onto a nylon mesh (pore size 400 μm) as shown in Fig. 2b, c. Calli were passed though the nylon mesh and crushed into small pieces with a spatula. Calli pieces were collected into a conical beaker filled with 1/2 N6D liquid medium (Fig. 2d). Small calli that sunk to the bottom of the conical beaker were collected by filtration, transferred into a 50-mL tube, and then 10 mL of calli suspension were prepared by removal or addition of 1/2 N6D liquid medium (Fig. 2e).
Ten milliliters of calli suspension were spread onto 6–8 plates of N6D solid medium containing 1.0 μM pyriminobac and 400 mg/L carbenicillin (Fig. 2f). After the excess 1/2 N6D liquid medium was removed from the N6D solid medium, calli were incubated at 32 °C for 4–6 weeks (16 h light/8 h dark).
Regeneration of transformed calli
After selection, growing calli (Fig. 2g, arrowhead) were transferred onto MS regeneration medium [4 g/L MS salt mixture (Wako Pure Chemical Industries, Tokyo, Japan), 30 g/L sucrose, 30 g/L sorbitol, 2 g/L casamino acids, 0.2 mg/L NAA, 2 mg/L kinetin, 1 mL/L 1,000× B5-vitamin, 0.4 % gelrite, pH 5.8] containing 400 mg/L carbenicillin and incubated at 28 °C for shoot regeneration (16 h light/8 h dark). Regenerated shoots were transferred onto MS hormone-free medium (4 g/L MS salt mixture, 30 g/L sucrose, 1 mL/L 1,000× B5-vitamin, 0.25 % gelrite, pH 5.8) containing 400 mg/L carbenicillin and cultured at 28 °C for root regeneration (16 h light/8 h dark). The regenerated plantlets were grown in a closed glasshouse.
DNA extraction and Southern blot analysis
Genomic DNA was extracted from young leaves derived from wild type and 8 randomly selected transgenic rice lines by using the CTAB method (Murray and Thompson 1980). A total of 5 μg of genomic DNA were digested with HindIII, fractionated by electrophoresis on a 0.8 % agarose gel, and transferred onto a Hybond-N+ nylon membrane (Amersham Pharmacia Biotech, Buckinghamshire, UK). Probes were prepared from purified PCR products of the mALS coding region or the NiR terminator region. Southern hybridization was carried out with a Gene Images AlkPhos Direct Labelling and Detection System (GE Healthcare, Buckinghamshire, UK).
Protein extraction and immunoblot analysis
Mature seeds were separately ground into a fine powder. For total protein extraction, 500 μL of extraction buffer [50 mM Tris–HCl (pH 6.8), 8 M urea, 4 % SDS, 20 % glycerol, 5 % 2-mercaptoethanol, 0.01 % bromophenol blue (BPB)] was added to the seed powder and vortexed for more than 1 h at room temperature. The mixture was centrifuged at 12,000×g for 20 min at room temperature, and the crude soluble protein sample was decanted into a new tube. An aliquot of 2 μL of total protein was subjected to immunoblotting against polyclonal anti-33 kDa seed allergen or anti-14–16 kDa seed allergen antibodies after electrophoresis on 12 % SDS-PAGE gels as described by Yasuda et al. (2005). Signals were detected by the ECL detection system (GE Healthcare, Buckinghamshire, UK).
Results and discussion
Evaluation of a novel co-transformation system
Growth calli on selection medium
The key steps in this co-transformation method were the co-cultivation using moisturized filter paper and the callus crushing steps. Although in vitro co-cultivation is usually performed on a solid medium, we used filter paper moisturized with liquid medium for co-cultivation (refer to “Materials and methods”) to prevent excess growth of Agrobacterium, thus increasing the transformation efficiency of the calli (Ozawa 2009). The callus crushing step enhanced the efficacy of the NiR P:NiR selectable marker gene. Low levels of expression of the NiR gene, which does not have a null mutation in this cultivar, and decreased enzyme activity were detected in the callus (Nishimura et al. 2005). Callus selection without the callus crushing step would reduce the selection efficiency of NiR P:NiR and increase escape callus formation.
Production of transgenic rice seeds with reduced content of major rice seed allergens
The co-transformation method was used to generate transgenic rice lines with reduced contents of two seed allergens, a 33 kDa seed allergen (β-glyoxarase) and the 14–16 kDa seed allergens (α-amylase inhibitor), which react with IgE antibodies in the sera of rice-sensitive patients. The 33 kDa seed allergen is encoded by a single gene, while the 14–16 kDa seed allergens are encoded by at least nine genes in rice (Matsuda et al. 2006; Wakasa et al. 2011).
Gene integration in regenerated plants was confirmed by genomic PCR (Fig. 5b). PCR products derived from both T-DNAs were detected in seven transgenic rice lines from 13 regenerated rice plants. Total proteins from seeds derived from duplicate transgenic rice lines were analyzed by immunoblotting against anti-33 kDa seed allergen and anti-14–16 kDa seed allergen antibodies. As shown in Fig. 6c, the expression of the 33 and 14–16 kDa seed allergens was strongly suppressed in these transgenic rice lines (Fig. 6c, first row, lanes 2–5). The seeds used for the immunoblot analysis were T1 seeds, and a few segregation patterns were observed (Fig. 6c).
Conclusions and perspectives
A novel co-transformation method was used to generate stable transgenic rice plants with two traits introduced by Agrobacterium inoculation. The effectiveness of our co-transformation system was demonstrated with reporter proteins (GFP and GUS) and the down-regulation of the expression of the rice 33 kDa and 14–16 kDa seed allergens. The use of selectable marker genes derived from rice genome is of particular value for the commercial production of genetically modified rice. Furthermore, the combination of our co-transformation method and the Multi-Site Gateway system allows the introduction of multiple gene cassettes (2–6 genes) into rice. Although the present study was restricted to ‘Koshihikari’ and its varieties, co-transformation using CSP:mALS and NiR P:NiR could be applied to many japonica rice cultivars that also show low regeneration efficiency and low NiR activity such as ‘Akitakomachi’, ‘Milky-queen’, ‘Hitomebore’, ‘Norin 1’, ‘Moritawase’ and ‘Ginbouzu’ (Nishimura et al. 2005).
The introduction of antibiotic resistance genes such as NPT II and HPT using our co-transformation system should be simpler than that of the CSP:mALS and NiR P:NiR described in the present study. Furthermore, the introduction of antibiotic resistance genes enables the elimination of the callus crushing step by using a nylon mesh. In addition, there are fewer restrictions in the rice cultivars that can be used as hosts, and transformation efficiency is improved. However, the aim of the present study was to develop transgenic rice plants as GM crops. Future studies will be aimed at generating transgenic rice expressing multiple cassettes carrying proteins related to human health or secondary metabolism.
We thank Dr. Taiichi Ogawa (National Institute of Agrobiological Sciences) for providing the vector plasmid harboring the ‘Nipponbare’ NiR gene. This research was supported by the research grant “Genomics and Agricultural Innovation, GMC0003” from the Ministry of Agriculture, Forestry and Fisheries of Japan to F. Takaiwa.