42

The objectives of this study were to investigate the expression of proteins during the induction of somatic embryogenesis and to detect marker proteins in rice. The key proteins that control developmental program during somatic embryogenesis may almost all belong to the category of extremely rare proteins which cannot be visualized with the

standard gel technology. To detect these rare proteins, some special techniques of separation and concentration are needed (Yoshida et al. 1995). Choi et al. (1987) succeeded in isolating embryoid-specific cDNA clones by immunological screening with antibodies against embryo-specific proteins. The antibodies were subtracted from an antiserum against an extract of carrot somatic embryos using nitrocellulose filters saturated with an extract of unorganized cell clusters. In this study, we demonstrated the reverse immunosubtraction method of Choi et al. (1987). To remove proteins that involved in proliferation and housekeeping enzyme function from total proteins of embryogenic calli, immunosubtraction with affinity chromatography on crude antibodies against total proteins of unorganized calli was performed.

Suspension cells of rice (var. japonica, cv. Kamenoo) were obtained as described by Yoshida et al. (1994). We used 10-day-old calli after induction of somatic embryogenesis. At this stage, we could detect only the structural change of the surface of the embryogenic calli from rough to smooth. Crude antibodies against total proteins extracted from suspension cells just before induction of embryogenesis (unorganized calli) were obtained according to Choi et al. (1987). The diagram of immunosubtraction is shown in Fig. 1. The antiserum (35 mg proteins) against the unorganized cells were bound to I g CNBr-activated Sepharose 4B (Pharmacia LKB Biotechnology AB, Uppsala, Sweden). The column was equilibrated with PBS containing 0.02 % sodium azide. Proteins (17.5 mg) that extracted from calli at 10 days after induction of

embryogenesis were loaded on the column. The column was incubated for 1 h at room temperature. The subtracted unbound proteins were harvested and concentrated by ultrafiltration with Ultracent-l0 (Tosoh Inc., Tokyo, Japan). Then, the column was eluted with 30 ml of PBS containing 2.5 M sodium iodide and the bound proteins were discarded. After the column was equilibrated with PBS containing 0.02 % sodium azide, the concentrated unbound proteins were loaded again on it. These steps repeated 5 times. The last concentrated unbound proteins were thought to be responsible for somatic embryogenesis. They were diluted with lysis buffer and 100 mg of the proteins were separated by 2D-PAGE. The procedure for separation of proteins by 2D-PAGE was followed to O'Farrell (1975) with the exception that both lysis buffer and isoelectric focusing gel contained 2 % Ampholines (Pharmacia LKB Biotechnology AB, Bromma, Sweden) comprised of 1.4 % pH range 5 to 7 and 0.6 % pH range 3.5 to 10. The second gel was stained with Silver Stain Kit (Wako Pure Chemical Industry, Ltd., Osaka, Japan).

After five rounds of immunosubtraction, 17.5 mg total proteins of embryogenic calli decreased to 2.16 mg. Comparison of 2D-PAGE profile of immunosubtracted proteins with that of total proteins of the embryogenic calli before immunosubtraction allowed us to detect 7 newly appeared proteins (Fig. 2). Because these proteins could not be detected in 2D-PAGE profile of total proteins from unorganized calli (data not shown), they are thought to be somatic embryogenesis-related proteins. The result indicated that the method of immunosubtraction with affinity chromatography is an effective method to concentrate somatic embryogenesis-specific proteins in low abundance. Though three proteins increased in amount after immunosubtraction (Fig. 2), most of them also existed in the unorganized calli (data not shown). Either because their antibodies had not been formed during immunization with unorganized calli or because specificity of the antibodies were low, they seem to escape immunosubtraction and being concentrated. We could, however, distinguish embryogenesis-specific proteins from the escape proteins by comparison of 2D-PAGE profile with unorganized calli. We may obtain the specific proteins sufficient for the analysis by protein sequencer, if we scale up the volume of affinity column and samples.

Choi, J. H., L.-S. Liu, C. Borkird and Z. R. Sung, 1987. Cloning of genes developmentally regulated during