In plants regenerated from cell and tissue cultures, it is well known that genetic variations occur at high frequency and it is called as somaclonal variation. Somaclonal variation has been reported in a large number of plants, including potato, sugarcane, tobacco, wheat, rice and so on. In addition, it has been found in various agronomic traits such as disease resistance, plant height, tiller number, maturity, and various physiological and biochemical traits (Henry et al. 1998). It is also known that somaclonal variation may involve polyploidy, aneuploidy, abnormal structural changes of chromosomes, polymorphism of isozymes, reactivity of transposons, methylation, homologous recombination (Hirochika et al. 1996, Liu et al. 2004, Miyao et al. 2003). The work on cell and tissue culture and plant regeneration is indispensable to genetic engineerings such as somatic cell fusion and transformation. However, the somaclonal variation gives rise to either a barrier to development or an acceleration of development of new varieties by genetic engineerings (Karp et al. 1995). Although there are many reports on phenomena of somaclonal variation, we are still not able to figure out how somaclonal variation occurs. Rice (Oryza sativa L.) is one of the most extensively studied crops in the fields of genetics, plant breeding, morphology, physiology, cell technology and so on. Since the rice plant is a self-pollinating crop, cultivated rice plants are considered to be homozygous for the most genes. Therefore, it is possible to find easily somaclonal variation by using rice plants regenerated from calli.

We obtained 128 regenerated plants from callus of cv. Mutsuhomare and cv. Tsugaruotome cultured for six months. We analyzed the nuclear genomic DNAs by Southern blot hybridization using 73 landmarker clones of genomic DNA as probes after digestion with Hind III. As the results, only 4 out of 73 probes showed somaclonal variation as RFLP and the other probes did not show any polymorphisms (Table1). In total, 4 and 15 regenerated plants in cv. Mutsuhomare and Tsugaruotome, respectively, showed four types of polymorphic patterns. The sizes of all bands showed in Table1 were deduced by single logarithmic chart. In all polymorphic plants, the sizes of additional bands were larger than those of control bands. Polymorphic patterns observed by using the same probe were all identical even in different cultivars of Mutsuhomare and Tsugaruotome. Further Southern blot analyses were carried out in three polymorphic plants by using BamH I and a G232 clone as a probe. Polymorphic patterns of three plants digested with Hind III were all the same, but BamH I showed different patterns

among plants (Fig. 1). Because the deduced size of additional bands were 4.3 kb larger than that of the biggest control bands in Southern blot analysis digested with Hind III, we estimated that the additional bands are caused by a retrotransposon of Tos17, the size of which is 4.3 kb, and that the transposed location of Tos17 may be around or at inside of G232 locus. Furthermore, Southern blot analysis by using Tos17 as a probe proved that the additional bands are caused indeed by Tos17. In order to determine accurate locations of Tos17 in the additional bands, we amplified the DNA and sequenced around and the inside of G232 locus. As a result, it was clarified that in three plants Tos17 transposed into different region around G232 or at inside of G232 locus (Fig. 2). This result may provide a full account of different polymorphic patterns among plants in Southern blot analysis by using DNAs digested with BamH I (Fig. 1). Tos17 has a BamH I site at 2.3 kb from the 5' end but not any HindIII site. In the case of Mu-247 plant, Tos17 transposed into the edge of G232 locus. Thus, we could not probably detect a small additional band. In the case of Tu-107 plant, Tos17 transposed into the middle of G232 locus and we could detect two additional bands. In the Tu-143 plant, Tos17 transposed at outside of G232 locus and there was no additional band.

By this study we could indicate that the transposition of Tos17 is one of the causes of

somaclonal variation. Transpositional sites of Tos17 were different in three polymorphic plants, but they located mutually at quite near region. Therefore, it is considered that there is a hot spot for transposition of Tos17 around and at inside of G232 locus. However, in this study we could not find any features of hot spot so far.

Acknowledgement

We thank the Group for Rice Genome Research Program of the National Institute of Agrobiological Resources (NIAR) and the Institute of the Society of Techno-Innovation in Agriculturte, Forestry and Fisheries (STAFF) in Japan for providing landmarkers of genomic DNA clones.

References

Henry, R. J., 1998. Molecular and biochemical characterization of somaclonal variation. In somaclonal variation and induced mutation in crop improvement. Kluwer Academic Publishers, Dordrecht. Printed in Great Britain. p. 485-499.

Hirochika, H., K. Sugimoto, Y. Otsuki, H. Tsugawa and M. Kanda, 1996. Retrotransposon of rice involved in mutations induced by tissue culture. Proc. Natl. Acid. Sci. USA. 93: 7783-7788.

Karp, A., 1995. Somaclonal variation as a tool for crop improvement. Euphytica. 85: 295-302.

Linacero, R., E.A. Freitas and A.M.Vazquez, 2000. Hot spots of DNA instability revealed through the study of somaclonal variation in rye. Theor. Appl. Genet. 100: 506-511.

Miyao, A., K. Tanaka and H. Hirochika, 2003. Target specificity of the Tos17 retrotransposon shows a preference for insertion within genes and against insertion in retrotransposon-rich region of the genome. Plant Cell.15: 1771-1780.

Z.L., Liu, F.P. Ham and B. Liu, 2004. Activation of a rice endogenous retrotransposon Tos17 in tissue culture is accompanied by cytosine demethylation and causes heritable alteration in methylation pattern of flanking genomic regions. Theol. Appl. Genet. 109: 200-209.