35. Antinecrotic substances improved regeneration frequency of transgenic rice
  M. DEY, H. JIANG and R. WU

Dept. of Molecular Biology and Genetics, Cornell University, Ithaca, NY14853

Rice is the most important cereal of the developing world and a model crop plant. A relatively efficient method to transform rice using Agrobacterium has already been widely used to generate transgenic plants (Hiei et al., 1994; Datta et al., 1996; Roy and Wu, 2001). This method has also been used in functional genomics studies. Recently, draft sequences of the entire rice genome have been published. To assign functions to the assumed 50,000 rice genes, a rice gene-disruption library needs to be constructed. To achieve a 99% probability of tagging all the genes in rice genome, with an average gene size of 3 kb, 660,000 randomly-produced gene-disruption plant lines are required (Jeon et al., 2000).

We therefore tested several modifications of the standard protocol in an effort to improve the efficiency of the Agrobacterium-mediated transformation procedure for rice, in terms of reducing the effort and the time required to produce a large number of transgenic plants. Enriquez-Obregon et al. (1999) used three antinecrotic compounds to treat rice explants and successfully produced transgenic plants. Their system uses a non-supervirulent Agrobacterium strain and can also omit the use of acetosyringone, which induces vir genes. Their aim was to show that the antinecrotic compounds could reduce the difficulty of Agrobacterium infection in monocots by preventing necrosis of target cells. In our case, however, we used these compounds (40 mg/L L-cysteine, 5 mg/L silver nitrate and 15 mg/L ascorbic acid) in addition to a supervirulent Agrobacterium strain carrying a binary vector and acetosyringone to enhance the frequency of transgenic plant recovery using calli. We made use of these three antinecrotic compounds at a minimal level as optimized by Enriquez-Obregon et al. (1999) during co-cultivation of the embryogenic calli with Agrobacterium cells. We also altered the duration for selection and regeneration.

Four-week-old actively-growing mature-seed-derived calli of the japonica variety Nipponbare were infiltrated under vacuum for 15-20 minutes with Agrobacterium cells resuspended in liquid MS medium supplemented with 500 mg/L casein hydrolysate, 68.5 g/L sucrose, 36 g/L glucose and 200 microM acetosyringone. This was followed by a 3-day co-cultivation in the dark of the infiltrated calli on solid MS medium (pH 5.2) supplemented with 2 mg/L 2,4-D, 10 g/L glucose, 30 g/L sucrose, 100 microM acetosyringone, 40 mg/L L-cysteine, 15 mg/L ascorbic acid, 5 mg/L silver nitrate. The treated calli were then rinsed with sterile distilled water containing 250 mg/L cefotaxime, and the excess water was soaked onto sterile filter papers before plating the calli for the first selection cycle in the dark using the same medium used for callus induction but supplemented with 50 mg/L hygromycin and 250 mg/L cefotaxime. After 15-18 days, calli were transferred to regeneration selection medium, which is MS medium supplemented with 2 mg/L kinetin, 0.5 mg/L alpha-naphthalene acetic acid (NAA) and same concentrations of the antibiotics. A total of three 3-week regeneration selection cycles were carried out under 14/10 hours light/dark photoperiods. Cefotaxime was omitted in the second cycle of regeneration and onwards. Thus the total time to regenerate transgenic plants was 11 weeks after co-cultivation with Agrobacterium cells. Next, the plantlets were grown in tap water for 7 days for acclimatization and for fast rooting, before transferring them to soil under greenhouse conditions. In our previous protocol, we used 3 selection cycles in the dark (2-3 weeks each) and 2 regeneration cycles (3-4 weeks each) with the total time of approximately 15 weeks.

For transformation, we used a plasmid carrying the GUS reporter gene and the hygromycin phosphotransferase gene as the selectable marker, both driven by the constitutive CaMV 35S promoter. The plasmid was mobilized into Agrobacterium strain LBA4404. We repeated each experiment 3 times, each time regenerating approximately 100-150 plants. In the different experiments, 90-95% of the regenerated plants tested positive for GUS in a histochemical assay. The results on the presence of the transgene were confirmed by Southern blot hybridization.

In the preliminary experiments, we tried three different schedules for treating the calli with the antinecrotic compounds. We used a 30-minute pre-treatment with the antinecrotic compounds for one batch of calli, treatment only during co-cultivation with Agrobacterium cells for the second batch of calli and a combination of the two for a third batch of calli. A fourth control batch of calli was transformed at the same time without any treatment by antinecrotic compounds. The preliminary results showed that using the antinecrotic compounds only during co-cultivation gave the best results and it did not require an additional step of pretreatment. Thus, we used this schedule for the next two experiments.

Overall, we found that the number of putative independent transgenic regenerants was increased by 32-42% when calli were treated with antinecrotic compounds as compared to the corresponding control set (Table 1). It is likely that these compounds, with known anti-oxi-

dant activity, helped to minimize the damage due to oxidative bursts during plant-microorganism interaction and to prevent necrosis of rice cells. These treatments resulted in a better survival rate of the calli, thereby increasing the overall frequency of transgenic plant regeneration.

For our experiments, approximately 90 GUS-positive plants were randomly chosen to perform Southern blot analyses, 40% of which were found to carry a single copy of the transgene. All the plants regenerated by this method had normal seed setting and grain filling as compared to the control set of transgenic plants. As shown in Table 1, the transformation frequencies were 54 and 70% in two independent experiments. The regeneration frequency is much higher even in the control plants (53 and 38%, Table 1) that were not treated with antinecrotic compounds as compared to the previous reports of 3-8% in a japonica variety TNG67 (Roy and Wu, 2001). It appears that the difference in regeneration frequency is also attributed to the alterations in the conditions of the selection cycles. In addition, variations in the rice cultivar used and minor differences in cell culture conditions can affect the overall regeneration efficiency between independent experiments. In conclusion, in the present experiment, we have achieved a large increase in the regeneration frequency of transgenic rice in a shorter amount of time. Thus, the improved method would greatly favor any experiment that requires large-scale transgenic rice production such as those carried out in functional genomics studies.

References

Datta, K., N. Oliva, L. Torrizo, E. Abrigo, G. S. Khush and S. K. Datta, 1996. Genetic transformation of Indica and Japonica rice by Agrobacterium. Rice Genetics Newsletter 13: 136-139.

Enriquez-Obregon, G. A., D. L. Prieto-Samsonov, G. A. de la Riva, M. Perez, G. Selman-Housein and R. I. Vazquez-Padron, 1999. Agrobacterium-mediated japonica rice transformation: a procedure assisted by antinecrotic treatment. Plant Cell, Tissue and Organ Culture 59: 159-168.

Hiei, Y., S. Ohta, T. Komari and T. Kumashiro, 1994. Efficient transformation of rice (Oryza sativa L.) mediated by Agrobacterium and sequence analysis of boundaries of T-DNA. Plant J. 6: 271-282.

Jeon, J. S. et al., 2000. T-DNA insertional mutagenesis for functional genomics in rice. Plant J. 22: 561-570.

Roy, M. and R. Wu, 2001. Arginine decarboxylase transgene expression and analysis of environmental stress tolerance in transgenic rice. Plant Science 160: 869-875.