Rice Genetics Newsletter, Vol. 18

Rgn18

44.	Proteomic analysis of rice endosperm proteins
	T. KANEKO¹, M. ISHIZAKA², and H. KAJIWARA¹ 1)Department of Biochemistry, National Institute of Agrobiological Science, Tsukuba, Ibaraki 305-8602, Japan. 2)Chemical Analysis Research Center, National Institute of Agrobiological Science, Kannondai 2-1-2, Tsukuba, Ibaraki 305-8602, Japan.

Proteomic analysis usually consists of the separation of a complex mixture of proteins using two-dimensional polyacrylamide gel electrophoresis followed by identification of the proteins by mass spectrometry (MS). Proteomic analysis benefits greatly from DNA sequence information provided by genomic projects such as that for rice. Though rice endosperm proteins have been well analyzed in terms of molecular genetics, biochemical analyses have been lacking because of the insolubilities of seed proteins. Some proteins, such as prolamine, could not be dissolved in lysis buffer for two-dimensional polyacrylamide gel electrophoresis. We have therefore directly applied MS to rice seed proteins separated by one-dimensional SDS-polyacrylamide gel electrophoresis. Tentatively, the genes encoding the rice seed proteins were identified by comparing peptide sequence data derived from MS with publicly available DNA sequences.

Figure 1 shows the SDS-polyacrylamide gel electrophoresis (SDS-PAGE) pattern of endosperm proteins. Prominent bands were cut out from the gel and each one was cleaved by

trypsin in gel after modification by 4-vinylpyridine (Cavins and Friedman, 1970). MS analysis was performed by ion trap mass spectrometer (LCQ-Deca, Thermoquest) with a nanoelectrospray system (Kawakami et al., 2000). Database search was done by Mascot, which is available through the internet (http://www.matrixscience.com/home.html).

Nineteen prominent bands were observed on SDS-PAGE gel stained with Coomassie brilliant blue. Each band was used for trypsin digestion after the modification by 4-vinylpyridine and applied to MS/MS analysis. According to MS analysis and database searching, 13 of the 19 bands were related to known genes. The candidate gene encoding each analyzed protein fulfilled two requirements: a statistically significant score and at least two identified amino acid sequences. Identified proteins included starch debranching enzymes, glutelin, and prolamine.

There are several isoforms of certain rice seed proteins. Though 2D-PAGE was useful for separating these isoforms, it could not be applied to rice seed proteins because of

insolubility. Isoforms and modified proteins were observed as one band in SDS-PAGE. Modifications such as glycosylation were ignored in the MS database searching.

Though the separation efficiency of SDS-PAGE was lower than that of 2D-PAGE, SDS sample buffers could extract many insoluble proteins. On 2D-PAGE, only 5-10 spots were observed in gel because of the solubility and limitations of the pH range.

Though most seed proteins in rice are considered to be cloned and registered in the DNA bank, some bands were not identified in this experiment. The first reason for this incompleteness of identification is the incompleteness of the database for MS, which has not fully covered newly identified nucleotide sequence data. The second reason is the post-translational modification of amino acids. In this analysis, we did not consider amino acid modifications such as acetylation and glycosylation. If there was a modification of the proteins, the tryptic peptide was different from the computer simulated tryptic peptide based on DNA sequence information. Further detailed analysis must be undertaken for amino acids that have undergone post-translational modifications. These modifications and post-translational cleavage should be analyzed precisely. Finally, we limited the samples for MS analysis to know the usefulness and limitation of MS analysis. Faint bands of the 19 bands were not identified by MS analysis and database searching. Usually, amino acid sequence analysis is applied to protein sequencer using 5-10 bands after the electroblotting onto the membrane, and 10-15 amino acids can be identified by protein sequencer, if the N-terminal amino acid has not been modified. We could obtain better proteomic data using an MS system that had been obtained by a protein sequencer.

Acknowledgement

This work has been supported in part by a grant of the Rice Genome Project.

References

Cavins, J.F. and Friedman, M. (1970) An internal standard for amino acid analyses: S-beta-(4-pyridylethyl)-L-cysteine. Anal. Biochem. 35: 489-493.

Kawakami, T., F. Usui, and T. Nishimura (2000) A system for on-line highly sensitive analysis of expressed proteome using nanoLC-nanoES/MS/MS and two-dimensional electrophoresis: Towards large-scale and identification of disease-related proteins. Jpn. J. Electroph. 41: 185-190.

H. Kajiwara and N. Tomooka (1998) Comparative analysis of Vigna species by antiserum against multiple antigenic peptide. Electrophoresis 19: 3110-3113.