1. Production of Transgenic Rice Plants

Report of Committee on Genetic Engineering

transgenic rice plants are produced in each set of experiments (Xu et al. 1996; Zhang et al. 1996).

Several investigators have found that the use of immature embryos yields a higher number of fertile plants. However, initial experiments in transforming rice immature embryos by the biolistic method resulted in a high percentage of chimeric plants (Christou et al. 1991). To lower the percentage of chimeric plants produced, after transformed immature rice embryos began to produce calli, fast-growing protuberances on the surface of calli were excised and transferred to new plates containing a fresh selective medium after two and four weeks of selection (Christou et al. 1992; Li et al. 1993). C) Transformation of intact rice cells using Agrobacterium

Early attempts at using Agrobacterium to transform rice did not give reproducible results, or produced very few transgenic plants (for a review, see Ayres and Park 1994).

A more efficient method for transforming rice by the Agrobacterium-mediated system was reported by Hiei et al. (1994). By constructing super-binary vectors that included several vir genes and the use of more virulent Agrobacterium strains, they were able to produce a large number of transgenic plants. More recently, Komari et al. (1996) transformed rice with an Agrobacterium-based vector that carried two separate T-DNAs flanking two different genes. A large number of rice transformants were produced. DNA blot hybridization and analysis based on the polymerase chain reaction (PCR) confirmed integration and segregation of the T-DNAs in Ro and R1 plants.

Surprisingly, close to 50% of the r1 plants showed independent segregation of the two input genes. D) General trends in generating successful results

A general trend can be seen in using the above-described methods; that is, the initial attempts often resulted in marginal success, such as the production of only a few transgenic plants. In addition, adequate testing was often not done to completely confirm transformation. Follow-up experiments have been needed, as well as improvements to the methods, in order to produce a larger number of transgenic plants and to prove, with different types of analyses, that transformation was indeed achieved. Only after this follow-up work has been carried out and the results reproduced in at least two independent groups, can a new method be considered real and valid. This is a necessary requirement, because it has been a problem in the past in that certain claims for the production of transgenic plants could not be confirmed by other laboratories (for more details, see Potrykus 1990; 1991).

II. Useful Methods to Analyze Transgenic Plants

Transformed plants may be analyzed in different ways to determine whether or not they are transgenic. The types of analyses described here are more specific and more comprehensive than the criteria used to prove integrative transformation suggested by Potrykus (1990; 1991). When putative transgenic plants are produced, the following analyses are usually carried out in the same order as those presented in sections A, B and

38 Rice Genetics Newsletter Vol. 13

D below.

A) Analysis based on gain of function in the plants

For example, putative transgenic plants derived from transformation with a plasmid that contains a gene encoding for the phosphinothricin acetyltransferase (bar) can be rapidly screened for resistance to phosphinothricin (PPT)-containing herbicides. Next, the PPT-resistant plants are analyzed for the presence of a new enzyme activity, encoded by the foreign gene, which usually can be determined by assaying for the PPT-acetyltransferase activity.

B) Analysis to show the presence of the foreign gene based on hybridization results

A comprehensive set of DNA gel blot (Southern blot) hybridization analyses is necessary to show that the foreign DNA is integrated into the plant genome. This can be done by digesting the plant genomic DNA in several ways, using different restriction enzymes, followed by gel electrophoresis and hybridization. The choice of restriction enzymes should include those that give two, one, or no cuts in the plasmid used for transformation (Jenes et al. 1993).

The presence of the foreign gene can be shown by using an enzyme that cuts twice within the plasmid, followed by gel blot analysis. This analysis is more informative than an analysis based on the PCR. However, these analyses can only show the presence of the foreign gene in the sample, but not its integration into the rice genome.

In order to indicate integration of the foreign gene, one can use an enzyme that cuts only once within the inserted plasmid. In such an analysis, restriction fragments of varying sizes are expected because a restriction site in the host DNA must also be cut in order to release a DNA fragment that can be detected by hybridization. In this analysis, the number of hybridizing bands may reflect the copy number of the integrated plasmid in the genome. Positive results in this analysis prove that integration of the foreign gene has occurred.

Additional analysis to indicate integration of the foreign gene can be done by using an enzyme that does not cut within the plasmid. In this analysis, the restriction fragment from digestion of the genomic DNA is expected to be larger than the size of the foreign plasmid. The interpretation of the result is the same as discussed in the above paragraph, and an approximate copy number can be inferred. This analysis is more informative than carrying out DNA blot analysis without digesting the genomic DNA, which also should show high molecular weight hybridizing bands, but a copy number can not be inferred by the result.

C) Analyses to detect the flanking DNA sequences to indicate integration

A more definitive analysis to indicate integration of the foreign gene is to use inverse PCR analysis (Ochman et al. 1993). In this analysis, one can isolate the DNA that flanks the foreign gene (plasmid). Next, one of the flanking DNA can be used as a probe for genomic blot analysis after cutting the genomic DNA with an enzyme that does not cut within the plasmid. Then, the hybridization pattern of the transformed and non-transformed plants are compared. If the hybridizing band from the transformed plant is larger than the non-transformed plant by the size corresponding to that of the

Report of Committee on Genetic Engineering 39

foreign gene (plasmid), this proves integration of the foreign gene. However, in all the DNA blot analyses described so far, one can not tell whether integration of the foreign gene occurred in the plant genome or the genome of a contaminating organism or an endophyte.

In order to show integration of the foreign DNA into the rice genome, one can carry out the following experiment. To determine the origin of the hybridizing DNA in the transgenic and non-transgenic plants described in the above paragraph, one can analyze the DNA from several batches of rice seeds of non-transformed plants from diverse sources. After germinating these seeds, genomic DNA is isolated and digested with the same enzyme as previously described, and DNA blot analysis is carried out. If the size of the hybridizing band in most or all batches of seeds is the same as that in the non-transformed plants used for the original experiment to produce transgenic plants. then it is likely that this band truly represents rice DNA, and is not due to a contaminating organism. It would be unlikely that this organism would be present in different batches of rice seeds from diverse sources.

D) Analysis based on genetic segregation

True integration of the foreign gene into the host plant genome can also be proven by genetic analysis of the R1 and R2 populations. Until recently, most investigators who have reported genetic analysis used only R1 plants, but not R2 populations.

For statistically significant results, one needs to analyze at least 20 R1 plants grown from seeds of each of the Ro generation plants. For a self-pollinated plant with the foreign gene integrated into one genetic locus in a given line of transgenic plants, if the segregation ratio is 3:1, then this line is most likely transgenic. In general, the result needs to be confirmed by segregation analysis of the R2 plants. This is necessary, especially in cases where only one or two lines out of a dozen show a ratio close to 3:1 because this ratio may have occurred by chance.

For analyzing the second generation transgenic plants (R2), at least thirty R2 plants grown from each of nine R1) plants derived from a given Ro plant need to be analyzed to give good statistical values. If an Ro plant is truly transgenic, then in most cases, one would expect to find that all 30 R2 plants derived from each of three R1 lines contain the foreign gene by both phenotypic analysis and DNA blot analysis (these lines have become homozygous), and the progenies from the remaining six R1 lines to segregate at a ratio of 3:1.

E) Interpretation of results

By using a combination of most or all of the analyses described in the above sections, one can determine whether or not the transfer of the foreign gene into the rice genome was truly successful. In general, it is sufficient to use only the complete DNA blot analysis (but without performing inverse PCR and the associated analysis) and genetic segregation analysis of both R1 and R2 plants. In earlier years (1988-1989), the analysis of transgenic rice plants from all four laboratories mentioned previously (see Section I.A) mainly involved analysis based on gaining an enzyme activity and a simple DNA-blot analysis. The work by Shimamoto el at. (1989) used these two analyses, as

40 Rice Genetics Newsletter Vol. 13

well as genetic segregation analysis of first generation (r1) plants. However, the segregation ratios of different plant lines were not close to the theoretical ratios, and more extensive DNA blot analysis and segregation analysis of the R2 plants were not done.

Thus, strictly speaking, conclusive proof for true transformation of the rice plants could not be claimed by any one of the four groups.

Since 1993, an increasing amount of work has been reported by different authors, such as the work of Duan et al. (1996) and Komari et al. (1996), who used several of the analyses described in this article, including genetic segregation analysis of R2 plants. Thus, their claims of transformation of rice plants are conclusive.

III. Concluding Remarks

Many of the past claims for obtaining transgenic cereal plants have suffered from a degree of uncertainty by today's standards because only a minimal amount of evidence was presented. This may be because insufficient types of analyses were carried out, or too few transgenic plants were obtained. Comprehensive analyses are needed, especially when one tests a new method of transformation, or when one obtains transgenic plants of a certain recalcitrant species or variety for the first time. Thus, in order for someone to claim that they have obtained transgenic rice plants, they must perform the proper types of analyses to prove this is true. Then, only after other laboratories can repeat their work and get similar results, can their claim be considered as reliable.

Ayres, N.M. and W.D. Park, 1994. Genetic transformation of rice. Critical Reviews in Plant Science 13:

Christou, P., T.L. Ford and M. Kofron, 1992. The development of a variety-independent gene transfer method

Duan, X., X. Li, Q. Xue, M. Abo-EI-Saad, D. Xu and R. Wu, 1996. Transgenic rice plants harboring an

introduced potato proteinase inhibitor II gene are insect resistant. Nature Biotechnology 14: 494-498.

Jenes, B., H. Moore, J. Cao, W. Zhang and R. Wu, 1993. Techniques for gene transfer, In Transgenic Plants

Klein, T.M., E.D. Wolf, R. Wu and J.C. Sanford, 1987. High-velocity microprojectiles deliver nucleic acids

Komari, T., Y. Hiei, Y. Saito, N. Murai and T. Kumashiro, 1996. Vectors carrying two separate T-DNAs for

Report of Committee on Genetic Engineering 41

Li, L., R. Qu, A. de Kochko, C. Fauquet and R.N. Beachy, 1993. An improved rice transformation system

Ochman. h.. f.j. Ayala and D.L. Hartl, 1993. Use of polymerase chain reaction to amplify segments outside

Potrykus. I., 1990. Gene transfers to cereals: an assessment. Bio/Technology 8: 535-542.

Potrykus, I., 1991. Gene transfer to plants: assessment of published approaches and results. Annu. Rev.

Shimamoto, K., R. Terada, T. Izawa and H. Fujimoto, 1989. Fertile transgenic rice plants regenerated from

Toriyama, K., Y. Arimoto, H. Uchimiya and K. Hinata, 1988. Transgenic rice plants after direct gene transfer

Wang, Y.C., T.M. Klein, M. Fromm, J. Cao, J.C. Sanford and R. Wu, 1988. Transient expression of foreign

genes in rice, wheat and soybean cells following particle bombardment. Plant Mol. Biol. II: 433-439.

Zhang, W. and R. Wu, 1988. Efficient regeneration of transgenic plants from rice protoplasts and correctly