IC4R006-Metabolomics-2007-17556050

Project Title

Application of a metabolomic method combining one-dimensional and two-dimensional gas chromatography-time-of-flight/mass spectrometry to metabolic phenotyping of natural variants in rice

The Background of This Projec

• Two-dimensional gas chromatography GC × GC-TOF/MS is a novel approach for enhancing the GC resolution, and it has great advantages in increasing the resolution and peak capacity over the one dimensional separation method. The GC × GC-TOF/MS for metabolomics has been applied for com- plex metabolite profiles from mouse spleen [21]. This is the first report for the use of a technique which significantly enhances metabolite resolution. Currently, the GC × GC-TOF/MS technique has been applied in the analysis of volatile compounds [22,23] and also in the analysis of metabolite mixtures from mouse tissue, yeast cells, and human urine and serum as the comprehensive GC × GC-TOF/MS analysis [24–27].
• When non-targeted metabolic profiling data are subjected to multivariate statistical analysis such as principal component analysis (PCA) and partial least square-discriminate analysis (PLS-DA) toward the obtained data, a high throughput and high accuracy can be achieved for clustering according to the vectors of numerous metabolites.
• In order to develop a technique for studies in plant metabolomics, we have developed a combined method, namely, 1D- and GC × GC-TOF/MS. We applied this technique in the non-targeted metabolic profiling of brown rice seeds from the world rice core collection (WRC). The WRC is a representative set of Asian cultivated rice, Oryza sativa L., comprising 69 vari- eties selected from 3000 accessions stored in the Genebank of the National Institute of Agrobiological Sciences (NIAS), Japan, classified based on DNA polymorphism [28]. The WRC covers ca. 90% of the DNA polymorphisms detected in the original population and can be used as a convenient set to survey the genetic diversity of rice. Therefore, the metabolic phenotyping of the WRC varieties gives us an insight on the representative metabolite diversity of rice natural variants.
• In this project, the researchers describes a metabolomic method combining 1D-GC-TOF/MS and GC × GC-TOF/MS analyses and an example of its application to focus on the differences in the characteristics of data obtained from 1D-GC/MS and GC × GC/MS measurements of the WRC samples. This study presents the first report of the initial metabolic phenotyping in brown rice seeds of the WRC.
  
  

Figure 2. The PLS-DA score scatter plot for metabolite peaks obtained from the extracts of rice seeds of Nipponbare (brown dot), Kasalath (pale-green diamond)

Plant Materials

• Twenty-five rice seeds for each of the 68 WRC cultivars [28] and var. Dahonggu and Pokkari were sown on April 19, 2005, at a rice field in NIAS, Tsukuba, Japan. Seeds were harvested independently for each cultivar after 40 days, starting from the day on which the first panicle of rice was observed. The seeds were threshed from the panicles manually and then collected by each cultivar, after they were dried at 30 ◦ C for three days. All seeds in the husks were stored at 5 ◦ C under dark conditions until analysis. For each cultivar, 100 seeds were selected according to the average weight and length of seeds. After separating the husks from the seeds, the brown rice seeds obtained were bulked and crushed by using a Retsch mixer mill MM301 at a frequency of 20 Hz −1 for 2 min at 4 ◦ C. Successively, the obtained powder was divided into four pools for metabolic phenotyping. One hundred milligrams of each material was extracted with extraction buffer [methanol/chloroform/water (3:1:1, v/v/v)] at a concentration of 100 mg/ml and containing 10 stable isotope reference compounds. Each isotope compound was adjusted to a final concentration of 15 ng/ul for each 1-ul injection [20,29,30]. After centrifugation, a 200-ul aliquot of the supernatant was drawn and transferred into a glass insert vial. The extracts were evaporated to dryness in an SPD2010 SpeedVac ® concentrator from ThermoSavant.
  
  

Figure 3. Modulated chromatogram (above) and the corresponding two-dimensional contour plot (below) for (A) Nipponbare and those of (B) Kasalath

Research Findings

• The PLS-DA models of the two controls Nipponbare and Kasalath and three selected WRC varieties were created and analyzed as shown in Fig. 2. For example, the loading of the principal components 1 (Anjana Dhan/Nipponbare separation) and 2 (Anjana Dhan/Kasalath separation) in Fig. 2 (A) was selected according to the value of the first weight vector (w ∗ 1) and second weight vector (w ∗ 2) together with the Fig. 2. The PLS-DA score scatter plot for metabolite peaks obtained from the extracts of rice seeds of Nipponbare (brown dot), Kasalath (pale-green dia- mond), and the following selected varieties: (A) Anjana Dhan (black diamond), number of significant components: 3, R 2 X = 0.75, R 2 Y = 1.00, and Q 2 Y = 0.99. (B) Ryou Suisan Koumai (black box), number of significant components: 3, R 2 X = 0.68, R 2 Y = 0.99, and Q 2 Y = 0.95. (C) Urasan 1 (red box), number of significant components: 3, R 2 X = 0.72, R 2 Y = 0.99, and Q 2 Y = 0.95. 75 95% confidence intervals calculated using jackknifing [36,37] to extract the loadings, which is a distinguishing factor between the controls and the selected varieties. Fifty-six variables were considered as significant in the separation of Anjana Dhan and Nipponbare, and 29 variables were considered as significant in the separation of Anjana Dhan and Kasalath.
• Finally, based on the interpretation of the results of the multivariate analysis for each model as shown in the Fig. 2A–C, 10 metabolite peaks (GABA, glycerol-3-phosphate, myristate, fructose, IAA, inositol-1-phosphate, trehalose, alpha-tocopherol, cholesterol, and raffinose) were selected as representatives of metabolites, which contributed to significant differences between the controls and the three varieties in the PLS-DA models. The two con- trols and the three selected cultivars were further analyzed using the GC × GC-TOF/MS analysis for high-resolution metabolic profiling.
• The GC × GC-TOF/MS analysis enabled the detection of approximately 620 peaks from rice extracts of the WRC cultivars due to its higher resolution ability as compared to 1D-GC-TOF/MS with H-MCR method [20,29]. Fig. 3 exemplifies the separation patterns for the extracts of Nipponbare and Kasalath by using GC × GC-TOF/MS. Each total ion chromatogram (TIC) in the modulated chromatogram and the corresponding two-dimensional contour plot showed a slightly different sepa- ration pattern, indicating that both the Japonica and Indica rice varieties possess unique metabolites composition, although a majority of peaks were nearly similar with their intensities. A detailed analysis of the obtained total ion chromatogram of the three selected cultivars showed that the patterns were slightly different. For example, Ryou Suisan Koumai was classified as an Indica rice and it showed a separation pattern similar to that of Kasalath. This result suggested that rice varieties classified as closely related by analysis of DNA polymorphisms exhibit similar metabolite compositions.

Labs working on this Project

• RIKEN Plant Science Center, 1-7-22 Yokohama, Kanagawa 230-0045, Japan
• Group for Chemometrics, Organic Chemistry, Department of Chemistry, Umeå University, SE-901 87 Umeå, Sweden
• Umeå Plant Science Centre, Department of Forest Genetics and Plant Physiology, Swedish University of Agricultural Sciences, SE-901 87 Umeå, Sweden
• National Institute of Agrobiological Sciences, 2-1-2 Kannondai, Tsukuba, Ibaraki 305-8602, Japan
• Department of Molecular Biology and Biotechnology, Graduate School of Pharmaceutical Sciences, Chiba University, Chiba 263-8522, Japan

Corresponding Author

Miyako Kusano (E-mail:mkusano005@psc.riken.jp)