29. Proteomic identification of RuBisCO/LS gene mutations in radiation mutant of rice by two-dimensional gel electrophoresis, mass spectrometry, single strand conformation polymorphism, and nucleotide sequencing

1) Department of Biochemistry, National Institute of Agrobiological Sciences (NIAS), Tsukuba, Ibaraki 305-8602, Japan
2) Institute of Radiation Breeding, NIAS, Ohmiya, Naka, Ibaraki 319-2293, Japan
3) Chemical Analysis Research Center, National Institute for Agro-Environmental Sciences, Tsukuba, Ibaraki, 305-8602, Japan

Genome analyses have been conducted in higher organisms including rice (Yu et al. 2002). The post-genome era has already begun for rice. We analyzed differential electrophoresis of rice using radiation mutants which have been obtained since 1962 (Tanaka 1967) because expressed protein pattern of mutants was expected to be different from normal cultivar. For the identification of mutant gene, the conventional two-dimensional electrophoresis and mass spectrometry (MS), a proteomic approach, was employed to the identification of mutant gene. In this paper, we attempted mutant analysis of the ribulose bisphosphate carboxylase oxygenase large subunit (RuBisCO/LS) genes, the key enzyme of photosynthesis, which were identified by a combination of 2D-PAGE, MS, single strand conformation polymorphism (SSCP), and nucleotide sequencing.

Electrophoresis of rice leaf proteins has sometimes been hindered by unknown substances which were thought to be polyphenol and salts. Thus an acetone precipitation was applied prior to the separation of the first dimension on 2D-PAGE. The lysis buffer used in this experiment did not contain any detergents (Kajiwara and Tomooka 1999) to avoid the overlapping of proteins on 2D-gel and the RuBisCO/LS that existed in the soluble fraction (Kajiwara and Kaneko 2002). Prominent spots were analyzed by MS/MS (Kaneko et al 2002) to identify the gene encoding the protein by Mascot searching (Perkins et al. 1999).

Many differences were observed in 250 radiation mutant rice lines which had yellowish leaf against normal cultivars, Nihonmasari and Norin-22. They showed differences in pIs, molecular weights, and/or the amounts of expressed proteins which might be caused by mutations in the gene or post-translational modifications on the protein (Data not shown). Another possibility was that the expression might be changed through the effects of protein-protein interaction networks. We thus focused on the RuBisCO/LS because it is the key enzyme in photosynthesis.

Based on the first screening of radiation mutant lines by 2D-PAGE, five lines, i.e. Nos. 29, 169, 180, 238, and 255, were assumed to be undergoing mutation on RuBisCO/LS (Data not shown). The changes in pI might have been caused by the replacement of amino acids caused by mutations of the gene, or by post-translational modifications of the proteins. We hypothesized that the changed pI of the proteins was related to the mutations.

Theoretically, the amplified DNA of normal rice RuBisCO/LS should be 1,926 bp (Shimada and Sugiura 1991). The amplified DNA from both normal cultivars and mutant lines was 1.9 kb in agarose gel electrophoresis; no difference was observed (Data not shown).

Some of the five analyzed mutant lines showed differences in the SSCP analysis (Fig. 1). All of the PCR products were digested by the sets of AccII, AluI, HhaI, HaeIII or Cfr13I, MboI, MspI, XspI restriction enzymes. Polymorphism in the SSCP analysis digested by the AccII series was observed in lines 180, 205, and 238 in the electrophoresis at 23C and 8C. SSCP analysis also showed pattern changes in the digested PCR product of the Cfr13I series (Data not shown). These were assumed to be changes in restriction sites or in the secondary structure of ssDNA caused by the mutations.

We identified some mutations by DNA sequencing of RuBisCO/LS gene in mutant line No. 180 (Data not shown) in detail. In the upstream and downstream regions of the noncoding sequence, there are several insertions, deletions, and substitutions of nucleotides. Though their meanings are not known, these mutations in the non-coding regions might be causing the decrease of RuBisCO/LS gene product in mutant rice line No. 180.

The nucleotide sequence of the coding region of the RuBisCO/LS gene of mutant rice line

180 was conserved more than that of the non-coding region. In all, 14 substitutions of nucleotides were found in the coding region of the No. 180 RuBisCO/LS gene. Six amino acids in RuBisCO/LS were predicted, based on the changes from normal cultivars; they were K14Q, D94E, P176Q, L178W, K183I, and S228A. These mutational substitutions which caused changes in charges, and thus in the electrophoresis pattern were considered to be changed in the first dimension. Though we do not have direct evidence of the functional changes caused by these substituted amino acids in RuBisCO/LS activities, some mutations, i.e. P176Q, L178W, and K183I, accumulated in the region of the RuBisCO activase binding region.

RuBisCO/LS makes a complex with RuBisCO/SS for the fixation of carbon dioxide. Additionally, RuBisCO is required for the Bsd2 gene product (Brutnell et al. 1999) and RuBisCO activase (Ott et al. 2000) for full activity. Though further analysis is required, these mutations in the RuBisCO activase binding region might affect characters, including leaf color, through interaction with Bsd2 gene product and the RuBisCO activase.

We could identify the mutant gene from protein analysis to nucleotide sequencing using RuBisCO gene as an example. The approach used here was considered to be one of the effective proteomic approaches to identify the mutant gene.


Brutnell, T.P., R.J.H. Sawers, A. Mant, and J.A. Langdale, 1999. Bundle sheath defectives, a novel protein required for post-translational regulation of the rbcL gene of maize. Plant Cell 11: 849-864.

Kajiwara, H. and N. Tomooka, 1999. Comparative analysis of genus Vigna seeds using antiserum against a synthesized multiple antigenic peptide. Electrophoresis 19: 3110-3113.

Kajiwara, H. and T. Kaneko, 2002. Utility of single-strand conformation polymorphism analysis was improved by low-cross-linking polyacrylamide gel and quick low-background silver staining. Anal. Biochem. 300: 273-274.

Kaneko, T., M. Ishizaka, and H. Kajiwara, 2002. Proteomic analysis of rice endosperm proteins. RGN 18: 100-102.

Ott, C.M., B.D. Smith, A.R. Portis, and R.J. Spreitzer, 2000. Activase region on chloroplast ribulose-1,5-bisphosphate carboxylase/ oxygenase - Nonconservative substitution in the large subunit alters species specificity of protein interaction. J. Biol. Chem. 275: 26241-26244.

Perkins, D.N., D.J. Pappin, D.M.Creasy, and J.S. Cottrell, 1999. Probability-based protein identification by searching database using mass spectrometry data. Electrophoresis 20: 3551-3567.

Shimada, H. and M. Sugiura, 1991. Fine structural features of the chloroplast genome: Comparison of the sequenced chloroplast genomes. Nuc. Acids Res. 19: 983-995.

Tanaka, S., 1967. Studies on recurrent irradiation of rice plants in their successive generations. Gamma Field Symposia 6: 97-106.

Yu, J., S.N. Hu, J. Wang, et al., 2002. A draft sequence of the rice genome (Oryza sativa L. ssp Indica). Science 296: 79-92.