Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Improved annotation and quantification of metabolites in rice (Oryza sativa L.) seeds using two-dimensional gas chromatography–time-of-flight mass spectrometry

Improved annotation and quantification of metabolites in rice (Oryza sativa L.) seeds using... Two‑ dimensional gas chromatography coupled to time‑ of‑flight mass spectrometry (GC × GC‑ TOFMS) is a power‑ ful tool for identification of compounds in complex samples. Herein, we compared the GC × GC‑ TOFMS and GC‑ TOFMS systems for polar metabolite profiling of rice seeds. Forty‑seven and thirty‑six metabolites were detected in a Korean rice cultivar, Dongjin, using GC × GC‑ TOFMS and GC‑ TOFMS, respectively. The limits of detection of shikimic, p‑ coumaric, and sinapinic acids were 30.0‑, 1.6‑, and 2.5‑times lower, respectively, with GC × GC‑ TOFMS than with GC‑ TOFMS. The overlapped peaks of glycerol and phosphoric acid in GC‑ TOFMS were separated in GC × GC‑ TOFMS. Polar metabolite profiling of two white and six red rice cultivars was performed using GC × GC‑ TOFMS and the obtained data were subjected to principal component analysis. Remarkably, principal component 1 separated Heug‑ daegu from other cultivars, indicating that Heugdaegu has high levels of caffeic, sinapinic, and vanillic acids. Findings from this work may aid breeding programs aimed at improving the quality of rice seeds. Keywords: GC × GC‑ TOFMS, GC‑ TOFMS, Metabolic profiling, Oryza sativa L., Pigmented rice Introduction metabolites, study of biosynthetic pathway of metabo- Polar primary metabolites are essential for vital activities lites is important for understanding the plant physiology of organisms. In contrast, secondary metabolites are not and for designing new plants with enhanced nutritional directly involved in the normal growth, development, and value. Towards this end, profiling of primary metabolites reproduction, but several such metabolites are present in and analysis of the relationship between the metabolites plants for ecological purposes. Because primary metabo- in various plants has been performed using GC-TOFMS lites are used as precursors of other primary metabo- [1–4]. Researchers in the field of metabolomics have lites or as building blocks for the synthesis of secondary been striving to improve the chromatographic resolution and detection sensitivity of all metabolites present in a sample. *Correspondence: supark@cnu.ac.kr; kjkpj@inu.ac.kr Comprehensive two-dimensional gas chromatogra- Seung‑A. Baek and So Yeon Kim contributed equally to this work phy (GC × GC) is an advanced technique that improves Division of Life Sciences and Bio‑Resource and Environmental Center, the resolution of one-dimensional gas chromatography College of Life Sciences and Bioengineering, Incheon National University, Incheon 22012, Republic of Korea (GC). Two columns, with different stationary phases, Department of Crop Science and Department of Smart Agriculture are connected through a modulator (thermal or valve- Systems, Chungnam National University, 99 Daehak‑ro, Yuseong‑gu, based modulator). The eluates from the first column are Daejeon 34134, Republic of Korea Full list of author information is available at the end of the article transferred into the second column and are concentrated © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. Baek et al. Appl Biol Chem (2021) 64:65 Page 2 of 9 during the transfer in the modulator [5]. GC × GC the performance of GC-TOFMS and GC × GC-TOFMS. coupled to quadrupole mass spectrometry (qMS) or Subsequent to method validation, polar metabolites time-of-flight mass spectrometry (TOFMS) reduces the extracted from two white and six red pigmented rice cul- probability of peak overlap and increases the number tivar seeds were analysed using GC × GC-TOFMS. The of detected peaks [6–8]. GC × GC-TOFMS has higher metabolic profiling data were analysed by principal com - selectivity and greater sensitivity than GC-TOFMS [5, 8, ponent analysis (PCA) for comparing the metabolites 9], and has, therefore, been used for metabolic analysis among the eight rice cultivars. of several plants including brown rice seeds. However, to the best of our knowledge, metabolite profiling of pig - Materials and methods mented rice seeds using GC × GC-TOFMS has not been Samples reported [10–12]. Eight varieties of Korean rice seeds were categorized as In plant metabolomics, GC × GC-TOFMS has been white and red according to the colour of their pericarp. applied to metabolite fingerprinting with chemomet - Two cultivars of white rice (Dongjin, DJ; Heugdaegu, ric tools. After non-targeted metabolic profiling, com - HDG) and six of red rice (Aengmi, AM; Goryeong 8, parison of data obtained from samples (i.e., control and GR8; Hanyangjo, HYJ; Hongjinju, HJJ; Jakwangdo, JKD; treated samples, cultivars, or species) with multivari- Jeogjinju, JJJ) were used in this study. The seeds were ate statistics is performed [13–16]. In the case of rice obtained from the Agricultural Genetic Resources Center grains, GC × GC-TOFMS has been used for non-tar- at the National Academy of Agricultural Science (Suwon, geted metabolomic studies. Volatile metabolites in 10 Korea). The seeds were harvested in 2016 and manually rice varieties were assayed to understand the mechanism hulled and ground to a fine powder with a mixer mill of synthesis of aroma compounds [17]. Eight volatile (HR2860, Philips, Amsterdam, The Netherlands) and compounds were selected as key markers responsible pestle and mortar. The powdered seeds were stored at for the differences between aromatic and non-aromatic − 20 °C until they were used. rice varieties [18]. Volatiles collected from microbe- and mite-contaminated rice grain were analysed using GC Chemicals × GC-TOFMS for assessment of contamination [14]. Methanol (HPLC grade) and chloroform (HPLC grade) Non-targeted profiling of polar metabolites in five brown were purchased from Daejung Chemical & Metal cultivars was performed using GC × GC-TOFMS in (Siheung, Korea) and Burdick & Jackson (Muskegon, MI, combination with GC-TOFMS analysis for investigating USA), respectively. Water used in the experiments was the metabolite diversity of rice variants [10]. In addition, prepared using the Millipore water purification system non-targeted profiling of polar metabolites in grains of (Milli-Q Direct 8; Milford, MA, USA). Adonitol (ribitol; three rice cultivars was performed using GC-TOFMS ≥ 99%), shikimic acid (≥ 99%), sinapinic acid (≥ 98%), and the results were compared with those obtained using methoxyamine hydrochloride (MOX; 98%), pyridine GC × GC-TOFMS analysis for assessing the data pro- (HPLC grade, ≥ 99.9%), and N-methyl-N-(trimethylsilyl) cessing and analysis methods [11]. trifluoroacetamide (MSTFA) were purchased from Sigma Non-targeted metabolite profiling has been important (St. Louis, MO, USA). p-Coumaric acid was purchased in discovering biomarkers and for screening unrevealed from MP Biomedicals (Solon, OH, USA). metabolites in samples; however, in plant metabolomics the focus has been on the already known metabolites Extraction of polar metabolites related to pathways including the tricarboxylic acid cycle, Polar metabolites, including amino acids, organic acids, glycolysis, and biosynthesis of secondary metabolites. sugars, and phenolic acids, in pigmented rice (n = 3) Physiological mechanisms have been analysed by com- were extracted following a previously described method paring the levels of metabolites in samples that were [1]. Briefly, 10  mg powder was mixed with 1  mL of mutated or exposed to stimuli with those in the respec- 2.5:1:1 (v/v/v) methanol:chloroform:water containing tive control samples [19–21]. In this study, we performed 60  μL adonitol (200  μg/mL) as an internal standard (IS). targeted metabolic profiling in rice seed using GC × After shaking at 1200  rpm for 30  min at 37  °C (Eppen- GC-TOFMS for the first time. Moreover, we compared dorf Thermomixer Comfort 5355, Eppendorf, Hamburg, the GC × GC-TOFMS platform with GC-TOFMS for Germany), the mixture was centrifuged at 16,000×g for analysis of polar metabolites in rice seed. The number of 3 min at 4 °C (MX-307, TOMY, Tokyo, Japan). The upper metabolites detected using GC-TOFMS was compared layer (800 μL) was transferred to a new tube and 400 μL with those detected using GC × GC-TOFMS. Limit of water was added. The mixture was vortexed and centri - detection (LOD), limit of quantification (LOQ), accuracy, fuged at 16,000×g for 3 min at 4 °C. The methanol:water and precision of standards were measured to compare phase (900  µL) was transferred to a new tube and dried Baek  et al. Appl Biol Chem (2021) 64:65 Page 3 of 9 completely using a vacuum centrifuge dryer (VS-802F, preparation to determine linearity, precision, and accu- Visionbionex, Gyeonggi, Korea) and a freeze-dryer racy (n = 3). Ribitol (12 μg) was added as an IS to each (MCFD8512, IlShinBioBase, Gyeonggi, Korea). For deri- standard sample. The linearity was calculated as the vatisation, 80  µL 2% MOX (in pyridine, w/v) was added square of the correlation coefficient (r) of the calibration to the sample and incubated at 30  °C, with shaking at curve. LOD and LOQ of shikimic acid, sinapinic acid, 1200  rpm for 90  min. Thereafter, 80  µL MSTFA was and p-coumaric acid were estimated as the following added and the mixture was incubated at 37 °C, with shak- expressions: 3α/S (LOD) and 10α/S (LOQ), where α is the ing at 1200 rpm for 30 min. The sample was moved to a standard deviation (n = 7) and S is the slope of the cali- vial for GC-TOFMS and GC × GC-TOFMS analyses. bration curve. The precision and accuracy were defined as the relative standard deviation (RSD) and recovery, GC‑TOFMS and GC × GC‑TOFMS conditions respectively. RSD was calculated using the following Agilent 7890A GC (Agilent, Santa Clara, CA, USA) cou- equation: (α/μ) × 100, where α is the standard devia- pled to a Pegasus TOFMS 4D (LECO, St Joseph, MI, USA) tion and μ is the mean (n = 3). Recovery was estimated was used for the analysis of polar metabolites using GC-as (SS /SS ) × 100, where SS is the amount of spiked C A C TOFMS and GC × GC-TOFMS. An Rtx-5MS column standard calculated using a calibration curve and SS is (0.25  mm × 0.25  µm × 30  m; Restek, Bellefonte, PA, the amount of spiked standard. USA) was used for GC-TOFMS analysis. For GC-TOFMS analysis, the column oven temperature was maintained Multivariate statistical analysis at 80 °C for 0.5 min, was increased at 5 °C/min to 330 °C Polar metabolite analysis was performed in triplicate. and then maintained at 330 °C for 5 min. For GC × GC- PCA was performed using SIMCA-P (version 13.0; TOFMS, Rtx-5MS (0.25 mm × 0.25 µm × 30 m; Restek) Umetrics, Umeå, Sweden). The data were normalised and Rxi-17sil MS (0.15 mm × 0.15 µm × 1.2 m; Restek) with unit variance scaling. columns were used as the first and second columns, respectively. The oven conditions for the first column Results and discussion were the same as in GC-TOFMS, whereas the second Metabolic profiling of DJ using GC‑TOFMS column oven temperature program was 5  °C above that and GC × GC‑TOFMS of the first column. The two columns were connected To compare the GC-TOFMS and GC × GC-TOFMS through a cryogenic modulator and liquid nitrogen was systems, polar metabolite profiling of the same derivate used as the cryogen. The modulator temperature pro - sample (DJ) was performed using the both the instru- gram was 15  °C above the second column temperature. ments. A total 38 polar metabolites were detected in DJ The modulation period was set to 4 s, with 0.6 s hot and by GC-TOFMS analysis. Twenty one amino acids and 1.4  s cool pulse duration. Except for the column oven amino acid derivatives, six organic acids, seven sugars, condition, other conditions were the same for both GC- phosphoric acid, ferulic acid, p-hydroxybenzoic acid TOFMS and GC × GC-TOFMS analyses. One microli- and glycerol were identified (Fig.  1A). On the contrary, a tre of derivatised sample was injected in the split (25:1) total of 47 polar metabolites were detected by GC × GC- mode at 250  °C. Helium was used as the carrier gas at TOFMS analysis (Fig.  1B). Twenty-two amino acids and a flow rate of 1.2  mL/min in a constant flow mode. The amino acid derivatives, eight organic acids, seven phe- temperature of transfer line and ion source was 260 and nolic acids, eight sugars, phosphoric acid, and glycerol 230  °C, respectively. The mass electron energy was set were identified. Cysteine, methionine, pyruvic acid, shi - at −  70  eV and the mass range for scanning was in the kimic acid, caffeic acid, salicylic acid, sinapinic acid, 45–650  m/z range. The mass acquisition rate was 10 vanillic acid, p-coumaric acid, p-hydroxybenzoic acid, spectra/sec in GC-TOFMS and 100 spectra/sec in GC and glyceric acid were only detected in the GC × GC- × GC-TOFMS. The qualitative analysis was performed TOFMS analysis (Fig. 1; Additional file 1: Table S1). by comparison with the retention time of respective standards and mass spectrum in an in-house library, Comparison of the sensitivities of GC‑TOFMS Wiley9, and NIST14 using the ChromaTOF software and GC × GC‑TOFMS (V4.50, LECO). The quantitative estimation was based on Shikimic acid is a precursor of aromatic amino acids peak area ratios relative to the peak area of the IS. (phenylalanine and tyrosine) and phenylpropanoid path- way starts with phenylalanine. Flavonoids, monolignols, Analytical method validation phenolic acids, stilbenes, and coumarins are derived from Six different concentrations of shikimic acid (0.1–5.0 μg), phenylalanine through successive consecutive enzymatic p-coumaric acid (0.25–12.5  μg), and sinapinic acid reactions in most plants [22]. Kim et  al. [1] reported a (0.75–12.5  μg) standards were analysed within 1  day of positive relationship between all phenolic and shikimic Baek et al. Appl Biol Chem (2021) 64:65 Page 4 of 9 Fig. 1 GC‑ TOFMS (A) and GC × GC‑ TOFMS (B) analytical ion chromatogram (AIC) of polar metabolites extracted from Dongjin. Peak: 1, Pyruvic acid; 2, Lactic acid; 3, Alanine; 4, Oxalic acid; 5, Valine; 6, Serine‑1; 7, Ethanolamine; 8, Leucine; 9, Glycerol; 10, Phosphoric acid; 11, Isoleucine; 12, Proline; 13, Glycine; 14, Succinic acid;15, Glyceric acid; 16, Fumaric acid; 17, Serine‑2; 18, Threonine; 19, β‑Alanine; 20, Malic acid; 21, Salicylic acid; 22, Methionine; 23, Pyroglutamic acid; 24, Aspartic acid; 25, 4‑Aminobutytic acid; 26, Cysteine; 27, Threonic acid; 28, Glutamic acid; 29, p‑Hydroxybenzoic acid; 30, Phenylalanine; 31, Asparagine; 32, Xylose ‑1; 33, Xylose ‑2; 34, Ribitol (internal standard); 35, Vanillic acid; 36, Glutamine; 37, Shikimic acid; 38, Citric acid; 39, Fructose‑1; 40, Fructose ‑2; 41, Lysine; 42, p‑ Coumaric acid; 43, Tyrosine; 44, Mannitol; 45, Ferulic acid; 46, Inositol; 47, Caffeic acid; 48, Tryptophan; 49, Sinapinic acid; 50, Sucrose; 51, Raffinose acids in black rice cultivars. Several studies have reported Table 1 Linearity (r ), limit of detection (LOD), and limit of that red rice seeds contain more phytochemicals, such as quantification (LOQ) of shikimic acid, p‑ coumaric acid, and flavone (apigenin) and phenolics, than white rice [23–25]. sinapinic acid analysed using GC‑ TOFMS and GC × GC‑ TOFMS Therefore, to validate the method, a representative pre - Metabolite Calibration curve Linearity LOD (ng) LOQ (ng) cursor (shikimic acid) and phenolic acid (p-coumaric acid GC‑ TOFMS and sinapinic acid) were selected among 10 metabolites Shikimic acid y = 0.0529x − 0.9963 9.26 30.88 that were only detected by GC × GC-TOFMS analysis. 0.0054 The calibration curves of shikimic acid, p-coumaric acid, p‑ Coumaric acid y = 0.0080x − 0.9910 90.18 300.60 and sinapinic acid were produced using GC-TOFMS and 0.0083 GC × GC-TOFMS to compare the LOD and LOQ. The Sinapinic acid y = 0.0035x − 0.9920 119.58 398.59 means of three-point data for calculating the ratio rela- 0.0025 tive to the peak area of the IS were used as calibration GC × GC‑ TOFMS curves. All the calibration curves showed linearity with Shikimic acid y = 0.1486x + 0.9978 0.31 1.03 a correlation coefficient above 0.99 (Table  1). The LOD 0.0004 of shikimic acid, p-coumaric acid, and sinapinic acid was p‑ Coumaric acid y = 0.0221x − 0.9987 55.87 186.25 0.0022 9.3, 90.2, and 119.6 ng and the LOQ was 30.9, 300.6, and Sinapinic acid y = 0.0176x − 0.9966 47.22 157.41 398.6  ng, respectively, in the GC-TOFMS analysis. In 0.0078 GC × GC-TOFMS, the LOD of shikimic acid, p-cou- maric acid, and sinapinic acid was 0.3, 55.9, and 47.2 ng, Baek  et al. Appl Biol Chem (2021) 64:65 Page 5 of 9 and the LOQ was 1.0, 186.3, and 157.4  ng, respectively. [6, 27, 28]. In this study, the peaks of glycerol and phos- The LOD and LOQ of shikimic acid, sinapinic acid, and phoric acid overlapped in GC-TOFMS but could be sepa- p-coumaric acid in the GC × GC-TOFMS analysis were rated and quantified using different fragment ions (m/z) 30.0-, 1.6-, and 2.5-times lower than those of in GC- (Fig. 2A–C). However, the mass spectrum of glycerol was TOFMS, respectively. The analytes eluting from the first affected by that of phosphoric acid. The unique fragment column were trapped in the thermal modulator by cold ion, with a comparatively higher intensity, was selected as jet flow and then re-injected by hot jet flow into the sec - the quantitative fragment ion for each metabolite (Fig. 2F, ond column. Because in this step, the eluates are concen- G). The m/z 103 and 299 were chosen as quantitative trated, it improves the sensitivity of GC × GC [5]. GC fragment ions of glycerol and phosphoric acid, respec- × GC-TOFMS analyse significantly reduces the matrix tively. However, m/z 103 fragment ion was detected in effect by separation of second column. In this study, the phosphoric acid with 0.8% relative abundance. The over - LOD and LOQ of GC × GC-TOFMS were better than lapped fragment ion for two metabolites could affect those of GC-TOFMS. the area of the glycerol peak. Winnike et al. [6] reported To determine the precision and accuracy, three differ - that the peak area ratio of each metabolite (R ) was PA ent concentrations within each calibration curve range expected to be consistent. R was the calculated peak PA were measured (Table  2). The precision (%RSD) for shi - area of metabolite obtained using the GC-TOFMS analy- kimic acid, p-coumaric acid, and sinapinic acid was lower sis divided by that obtained using the GC × GC-TOFMS than 12.9% in both GC-TOFMS and GC × GC-TOFMS analysis. R of leucine, phosphoric acid, and isoleucine PA analyses. The accuracy (%Recovery) ranged from 85.1 was 0.07, 0.07, and 0.06, respectively. However, R of PA to 114.6% for the GC-TOFMS and GC × GC-TOFMS glycerol was 0.21. The peak area of glycerol was over- analyses. The precision of GC-TOFMS and GC × GC- measured with m/z 103 fragment ion of phosphoric acid TOFMS for the metabolites was lower than 15%. The and it might have increased the R . PA accuracy (%Recovery) of GC-TOFMS and GC × GC- Moreover, some of the other fragments, including TOFMS for the metabolites ranged from 85 to 115% m/z 73, 133, 147, 175, and 205, overlapped. The m /z and was in agreement with the International Guidelines 133 fragment ion was monitored in both glycerol and (FDA Guidance) [26]. Thus, the GC-TOFMS and GC phosphoric acid with a relative abundance of 19.9% × GC-TOFMS systems were found to be valid tools for and 8.2%, respectively. The prominent fragment ions polar metabolite analysis, but GC × GC-TOFMS could of glycerol, m/z 147 and 205, showed a relative abun- detect lower levels of metabolites when compared with dance of 3.9% and 1.9% in phosphoric acid, respec- GC-TOFMS. tively. This suggests that the user has to finally sort out the fragment ions from each metabolite in an over- Improved separation ability in GC × GC‑TOFMS lapped mass spectrum. In GC × GC-TOFMS, the glyc- GC × GC-TOFMS showed a better peak separation erol peak also overlapped with that of phosphoric acid ability than GC-TOFMS. The two columns with differ - in the retention time in first dimension but they were ent properties (mid-polar and non-polar) lead to a bet- separated in the second dimension (Fig.  2D, E). Thus, ter separation of co-eluted analytes from the first column the two metabolites did not affect the mass spectrum Table 2 Precision (%RSD) and accuracy (%Recovery) for determination of shikimic acid, p‑ coumaric acid, and sinapinic acid using GC‑ TOFMS and GC × GC‑ TOFMS Metabolite GC‑ TOFMS GC × GC‑ TOFMS Content (μg) Precision (%RSD) Accuracy (%Recovery) Content (μg) Precision (%RSD) Accuracy (%Recovery) Shikimic acid 0.50 1.86 85.12 ± 1.58 0.05 7.55 89.95 ± 6.79 2.50 1.70 106.85 ± 1.81 0.10 11.24 96.95 ± 10.90 5.00 12.90 108.89 ± 14.04 1.00 8.02 98.22 ± 7.87 p‑ Coumaric acid 5.00 4.17 87.13 ± 3.63 0.50 11.61 93.48 ± 10.85 10.00 8.11 105.41 ± 8.54 1.00 6.06 96.31 ± 5.83 12.50 7.85 102.49 ± 8.05 5.00 12.59 107.04 ± 13.47 Sinapinic acid 7.50 7.55 90.67 ± 6.84 0.75 3.53 114.57 ± 4.04 10.00 2.68 104.66 ± 2.80 1.00 4.33 100.61 ± 4.36 12.50 3.98 100.91 ± 4.02 5.00 8.35 94.12 ± 7.86 Baek et al. Appl Biol Chem (2021) 64:65 Page 6 of 9 Fig. 2 Partial GC‑ TOFMS total ion chromatogram ( TIC) (A), analytical ion chromatogram (AIC) (B), and selected ion chromatogram (C) of glycerol and phosphoric acid in Dongjin. Partial GC × GC‑ TOFMS TIC contour plot (D) and surface plot (E) of phosphoric acid and glycerol in Dongjin. P Phosphoric acid; G Glycerol. Mass spectrum of glycerol (F) and phosphoric acid (G) from NIST library spectrum deconvolution, resulted in better selectivity Table 3 Mass spectral similarity of phosphoric acid and glycerol in Dongjin analysed using GC‑ TOFMS and GC × GC‑ TOFMS of GC × GC-TOFMS than that of GC-TOFMS. Peak number Library GC‑ TOFMS GC × GC‑ TOFMS NIST Wiley9 NIST Wiley9 Analysis of polar metabolites in two white and six 9 Glycerol 860 860 935 927 pigmented rice cultivars with GC × GC‑TOFMS 10 Phosphoric acid 817 818 873 919 In this study, GC × GC-TOFMS showed improved sensitivity and peak separation ability compared with GC-TOFMS. Thus, polar metabolite profiling in eight rice cultivars was performed using GC × GC-TOFMS. and area of each other. In addition, the user could eas- As observed for DJ, a total 47 polar metabolites were ily determine whether the mass spectrum of a peak is detected in the seeds of all the cultivars. The obtained consistent with those of targeted metabolites by refer- data were subjected to PCA to assess the differences in ring to the similarity in a library, such as NIST and polar metabolite composition among rice seeds (Fig. 3). Wiley (Table  3). The separated peak, leading to better Baek  et al. Appl Biol Chem (2021) 64:65 Page 7 of 9 Fig. 3 Score plots (A) and loading plots (B) of principal component analysis (PCA) obtained from metabolic profiling by using GC × GC‑ TOFMS analysis. AM Aengmi; GR8 Goryeong 8; HYJ Hanyangjo; HJJ Hongjinju; JKD Jakwangdo; JJJ Jeogjinju; DJ Dongjin; HDG Heugdaegu Baek et al. Appl Biol Chem (2021) 64:65 Page 8 of 9 PCA is a good tool to obtain an overview of the com- Supplementary Information prehensive data and has been used in metabolomics The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s13765‑ 021‑ 00640‑3. studies [29]. In the score plots, each point represents an individual sample, and samples exhibiting similar Additional file 1: Table S1. Retention times (RT ), relative retention times variances are clustered together. The first two princi- (RRT ), and mass spectral data of polar metabolites as trimethylsilyl deriva‑ pal components (PCs; PC1 and PC2) had the greatest tives. Table S2. Composition and content (ratio/g) of polar metabolites in eigenvalues and captured 63.0% of the total variance 8 rice cultivars on the GC×GC‑ TOFMS. (Fig.  3A). The same cultivar was closely clustered in the score plot but was not clustered with the colour Acknowledgements of rice. The white cultivars (blue-coloured spots; DJ This work was supported by Incheon National University Research Concentra‑ tion Professors Grant in 2021. and HDG) were not clearly separated with the red cul- tivars (red-coloured spots; AM, GR8, HYJ, HJJ, JKD, Authors’ contributions and JJJ). However, clear a separation was observed Conceptualization, methodology: JKK, SUP, SYK and S‑AB. Data curation: SYK and S‑AB. Formal analysis: YJP and SYK. Writing—original draft prepara‑ between the HDG cultivar and other cultivars by tion: S‑AB, SYK and JKK. Writing—review and editing: JKK, S‑HL and TJK. PC1. Although the HDG has white-coloured pericarp, Project administration: JKK and SUP. All authors read and approved the final its hull has black colour. PC2 separated the AM cul- manuscript. tivar from other cultivars. Although the AM cultivar belongs to the same species as the cultivated rice, it is Declarations referred to as weedy rice in Korea. The corresponding Competing interests loading plot represents the metabolites responsible The authors declare that there is no competing interests. for separation on the score plots (Fig.  3B). HDG was Author details located on the left of the score plot and most metab- Division of Life Sciences and Bio‑Resource and Environmental Center, olites were located on the left of the loading plot. It College of Life Sciences and Bioengineering, Incheon National University, indicated that the HDG contained relatively higher Incheon 22012, Republic of Korea. Division of Horticultural Biotechnology, School of Biotechnology, Hankyong National University, Anseong 17579, levels of most metabolites than the other cultivars. It Republic of Korea. Department of Crop Science and Department of Smart is known that red rice seed contains relatively higher Agriculture Systems, Chungnam National University, 99 Daehak‑ro, levels of phenolics and flavonoids than white rice seed Yuseong‑gu, Daejeon 34134, Republic of Korea. Honam National Institute of Biological Resources, 58762 Mokpo, Republic of Korea. Nak donggang [24, 30]. However, Shen et  al. [25] revealed that sev- National Institute of Biological Resources, 37242 Sangju, Republic of Korea. eral white cultivars have a higher content of phenolic acids than red cultivars. In this study, relatively higher Received: 13 June 2021 Accepted: 1 September 2021 levels of caffeic acid, sinapinic acid, and vanillic acid were found in HDG, which has black-coloured hull, than in red cultivars. Phenolic acids are important References building blocks employed in the biosynthesis of fla- 1. Kim JK, Park SY, Lim SH, Yeo Y, Cho HS, Ha SH (2013) Comparative meta‑ vonoids. The level of raffinose was relatively higher in bolic profiling of pigmented rice (Oryza sativa L.) cultivars reveals primary GR8 and HYJ (Additional file 1: Table S2). The content metabolites are correlated with secondary metabolites. J Cereal Sci 57:14–20. https:// doi. org/ 10. 1016/j. jcs. 2012. 09. 012 and composition of amino acids are shifted and raffi- 2. Park SY, Lim SH, Ha SH, Yeo Y, Park WT, Kwon DY, Park SU, Kim JK (2013) nose is accumulated during seed development [31]. In Metabolite profiling approach reveals the interface of primary and this study, it was suggested that the expression pattern secondary metabolism in colored cauliflowers (Brassica oleracea L. ssp. botrytis). J Agric Food Chem 61:6999–7007. https:// doi. org/ 10. 1021/ jf401 of genes or activity of enzymes associated with amino 330e acid and raffinose metabolism might differ among the 3. Park CH, Yeo HJ, Park YE, Baek SA, Kim JK, Park SU (2019) Transcriptome cultivars. No metabolite analysis data are available analysis and metabolic profiling of Lycoris radiata. Biology 8:63. https:// doi. org/ 10. 3390/ biolo gy803 0063 except for γ-oryzanols and octacosanol in the weedy 4. Park Y, Baek SA, Choi Y, Kim JK, Park SU (2019) Metabolic profiling of nine rice, AM [32, 33]. The loading plot showed that AM Mentha species and prediction of their antioxidant properties using contained relatively higher levels of leucine, isoleu- chemometrics. Molecules 24:258. https:// doi. org/ 10. 3390/ molec ules2 40202 58 cine, methionine, mannitol, and phenylalanine than 5. Bahaghigha DH, Freye CE, Synovec RE (2019) Recent advances in modula‑ the other cultivars. Thus, chemometrics, combined tor technology for comprehensive two dimensional gas chromatography. with GC × GC-TOFMS, can facilitate a detailed met- Trac‑ Trend Anal Chem 113:379–391. https:// doi. org/ 10. 1016/j. trac. 2018. 04. 016 abolic profiling in rice for further studies in breeding 6. Winnike JH, Wei X, Knagge KJ, Colman SD, Gregory SG, Zhang X (2015) programs. Comparison of GC‑MS and GC × GC‑MS in the analysis of human serum samples for biomarker discovery. J Proteome Res 14:1810–1817. https:// doi. org/ 10. 1021/ pr501 1923 Baek  et al. Appl Biol Chem (2021) 64:65 Page 9 of 9 7. Almstetter MF, Oefner PJ, Dettmer K (2012) Comprehensive two‑ fruit maturity. Food Chem 331:127286. https:// doi. org/ 10. 1016/j. foodc dimensional gas chromatography in metabolomics. Anal Bioanal Chem hem. 2020. 127286 402:1993–2013. https:// doi. org/ 10. 1007/ s00216‑ 011‑ 5630‑y 21. You MK, Lee YJ, Kim JK, Baek SA, Jeon YA, Lim SH, Ha SH (2020) The 8. Prebihalo SE, Berrier KL, Freye CE, Bahaghighat HD, Moore NR, David PK, organ‑specific differential roles of rice DXS and DXR, the first two Synovec RE (2018) Multidimensional gas chromatography: advances in enzymes of the MEP pathway, in carotenoid metabolism in Oryza instrumentation, chemometrics, and applications. Anal Chem 90:505– sativa leaves and seeds. BMC Plant Biol 20:167. https:// doi. org/ 10. 1186/ 532. https:// doi. org/ 10. 1021/ acs. analc hem. 7b042 26s12870‑ 020‑ 02357‑9 9. Pierce KM, Hoggard JC, Mohler RE, Synovec RE (2008) Recent advance‑ 22. Deng Y, Lu S (2017) Biosynthesis and regulation of phenylpropanoids in ments in comprehensive two‑ dimensional separations with chemomet‑ plants. CRC Crit Rev Plant Sci 36:257–290. https:// doi. org/ 10. 1080/ 07352 rics. J Chromatogr A 1184:341–352. https:// doi. org/ 10. 1016/j. chroma. 689. 2017. 14028 52 2007. 07. 059 23. Kim JK, Lee SY, Chu SM, Lim SH, Suh SC, Lee Y T, Cho HS, Ha SH (2010) 10. Kusano M, Fukushima A, Kobayashi M, Hayashi N, Jonsson P, Moritz T, Variation and correlation analysis of flavonoids and carotenoids in Korean Ebana K, Saito K (2007) Application of a metabolomic method combining pigmented rice (Oryza sativa L.) cultivars. J Agric Food Chem 58:12804– one‑ dimensional and two‑ dimensional gas chromatography‑time ‑ of‑ 12809. https:// doi. org/ 10. 1021/ jf103 277g flight/mass spectrometry to metabolic phenotyping of natural variants in 24. Jun HI, Song GS, Yang EI, Youn Y, Kim YS (2012) Antioxidant activities rice. J Chromatogr B 855:71–79. https:// doi. org/ 10. 1016/j. jchro mb. 2007. and phenolic compounds of pigmented rice bran extracts. J Food Sci 05. 002 77:C759–C764. https:// doi. org/ 10. 1111/j. 1750‑ 3841. 2012. 02763.x 11. Allwood WJ, Erban A, de Koning S, Dunn WB, Luedemann A, Lommen A, 25. Shen Y, Jin L, Xiao P, Lu Y, Bao J (2009) Total phenolics, flavonoids, anti‑ Kay L, Löscher R, Kopka J, Goodacre R (2009) Inter‑laboratory reproduc‑ oxidant capacity in rice grain and their relations to grain color, size and ibility of fast gas chromatography–electron impact–time of flight mass weight. J Cereal Sci 49:106–111. https:// doi. org/ 10. 1016/j. jcs. 2008. 07. 010 spectrometry (GC–EI–TOF/MS) based plant metabolomics. Metabolomics 26. Food and drug administration (2018) Bioanalytical method validation 5:479–496. https:// doi. org/ 10. 1007/ s11306‑ 009‑ 0169‑z guidance for industry. https:// www. fda. gov/ files/ drugs/ publi shed/ Bioan 12. Xu B, Zhang L, Ma F, Zhang W, Wang X, Zhang Q, Luo D, Ma H, Li P (2018) alyti cal‑ Method V‑alid ation‑ Guida nce‑ for‑ Indus try. pdf . Accessed 13 Jan Determination of free steroidal compounds in vegetable oils by compre‑ 2021 hensive two‑ dimensional gas chromatography coupled to time‑ of‑flight 27. Welthagen W, Shellie RA, Spranger J, Ristow M, Zimmermann R, Fiehn O mass spectrometry. Food Chem 245:415–425. https:// doi. org/ 10. 1016/j. (2005) Comprehensive two‑ dimensional gas chromatography–time‑ of‑ foodc hem. 2017. 10. 114 flight mass spectrometry (GC × GC‑ TOF) for high resolution metabo‑ 13. Pierce KM, Hope JL, Hoggard JC, Synovec RE (2006) A principal compo‑ lomics: biomarker discovery on spleen tissue extracts of obese NZO nent analysis based method to discover chemical differences in com‑ compared to lean C57BL/6 mice. Metabolomics 1:65–73. https:// doi. org/ prehensive two‑ dimensional gas chromatography with time‑ of‑flight 10. 1007/ s11306‑ 005‑ 1108‑2 mass spectrometry (GC × GC‑ TOFMS) separations of metabolites in plant 28. Ahn YG, Jeon SH, Lim HB, Choi NR, Hwang GS, Kim YP, Lee JY (2018) samples. Talanta 70:797–804. https:// doi. org/ 10. 1016/j. talan ta. 2006. 01. Analysis of polycyclic aromatic hydrocarbons in ambient aerosols by 038 using one‑ dimensional and comprehensive two‑ dimensional gas chro‑ 14. Salvador ÂC, Baptista I, Barros AS, Gomes NC, Cunha Â, Almeida A, Rocha matography combined with mass spectrometric method: a comparative SM (2013) Can volatile organic metabolites be used to simultaneously study. J Anal Methods Chem. https:// doi. org/ 10. 1155/ 2018/ 83416 30 assess microbial and mite contamination level in cereal grains and coffee 29. Kim TJ, Park JG, Ahn SK, Kim KW, Choi J, Kim HY, Ha SH, Seo WD, Kim beans? PLoS ONE 8:e59338. https:// doi. org/ 10. 1371/ journ al. pone. 00593 JK (2020) Discrimination of adzuki bean (Vigna angularis) geographi‑ 38 cal origin by targeted and non‑targeted metabolite profiling with gas 15. Hurtado C, Parastar H, Matamoros V, Piña B, Tauler R, Bayona JM (2017) chromatography time‑ of‑flight mass spectrometry. Metabolites 10:112. Linking the morphological and metabolomic response of Lactuca sativa L https:// doi. org/ 10. 3390/ metab o1003 0112 exposed to emerging contaminants using GC × GC‑MS and chemomet ‑ 30. Kim GR, Jung ES, Lee S, Lim SH, Ha SH, Lee CH (2014) Combined mass ric tools. Sci Rep 7:1–13. https:// doi. org/ 10. 1038/ s41598‑ 017‑ 06773‑0 spectrometry‑based metabolite profiling of different pigmented rice 16. Lim V, Gorji SG, Daygon VD, Melissa F (2020) Untargeted and targeted (Oryza sativa L.) seeds and correlation with antioxidant activities. Mol‑ metabolomic profiling of Australian indigenous fruits. Metabolites 10:114. ecules 19:15679–15686. https:// doi. org/ 10. 3390/ molec ules1 91015 673 https:// doi. org/ 10. 3390/ metab o1003 0114 31. Rosental L, Nonogaki H, Fait A (2014) Activation and regulation of primary 17. Daygon VD, Calingacion M, Forster LC, De Voss JJ, Schwartz BD, metabolism during seed germination. Seed Sci Res 24:1–15. https:// doi. Ovenden B, Alonso DE, McCouch SR, Garson MJ, Fitzgerald MA (2017) org/ 10. 1017/ S0960 25851 30003 91 Metabolomics and genomics combine to unravel the pathway for the 32. Lee MC, Lee DJ, Lee GA, Park HJ, Lee JR, Choi YM, Lee SK, Jung Y, Cho YG, presence of fragrance in rice. Sci Rep 7:8767. https:// doi. org/ 10. 1038/ Song JY (2013) Analysis and comparison of the γ‑ oryzanol content based s41598‑ 017‑ 07693‑9 on phylogenetic groups in Korean landraces of rice (Oryza sativa L.). Plant 18. Setyaningsih W, Majchrzak T, Dymerski T, Namieśnik J, Palma M (2019) Breed Biotechnol 1:58–69. https:// doi. org/ 10. 9787/ PBB. 2013.1. 1. 058 Key‑marker volatile compounds in aromatic rice (Oryza Sativa) grains: an 33. Cho YH, Farhoudi R, Farooq M, Lee DJ (2017) Evaluating Korean rice HS‑SPME extraction method combined with GC × GC‑ TOFMS. Molecules genotypes and landraces for octacosanol contents and antioxidant activ‑ 24:4180. https:// doi. org/ 10. 3390/ molec ules2 42241 80 ity. Nat Prod Res 31:2778–2782. https:// doi. org/ 10. 1080/ 14786 419. 2017. 19. Shen S, Weng R, Li L, Xu X, Bai Y, Liu H (2016) Metabolomic analysis of 12922 71 mouse embryonic fibroblast cells in response to autophagy induced by acute starvation. Sci Rep 6:34075. https:// doi. org/ 10. 1038/ srep3 4075 Publisher’s Note 20. Kim TJ, Hyeon HJ, Park NI, Yi TG, Lim SH, Park SY, Ha SH, Kim JK (2020) A Springer Nature remains neutral with regard to jurisdictional claims in pub‑ high‑throughput platform for interpretation of metabolite profile data lished maps and institutional affiliations. from pepper (Capsicum) fruits of 13 phenotypes associated with different http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Applied Biological Chemistry Springer Journals

Improved annotation and quantification of metabolites in rice (Oryza sativa L.) seeds using two-dimensional gas chromatography–time-of-flight mass spectrometry

Loading next page...
 
/lp/springer-journals/improved-annotation-and-quantification-of-metabolites-in-rice-oryza-P2hajFRnLU
Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2021
ISSN
2468-0834
eISSN
2468-0842
DOI
10.1186/s13765-021-00640-3
Publisher site
See Article on Publisher Site

Abstract

Two‑ dimensional gas chromatography coupled to time‑ of‑flight mass spectrometry (GC × GC‑ TOFMS) is a power‑ ful tool for identification of compounds in complex samples. Herein, we compared the GC × GC‑ TOFMS and GC‑ TOFMS systems for polar metabolite profiling of rice seeds. Forty‑seven and thirty‑six metabolites were detected in a Korean rice cultivar, Dongjin, using GC × GC‑ TOFMS and GC‑ TOFMS, respectively. The limits of detection of shikimic, p‑ coumaric, and sinapinic acids were 30.0‑, 1.6‑, and 2.5‑times lower, respectively, with GC × GC‑ TOFMS than with GC‑ TOFMS. The overlapped peaks of glycerol and phosphoric acid in GC‑ TOFMS were separated in GC × GC‑ TOFMS. Polar metabolite profiling of two white and six red rice cultivars was performed using GC × GC‑ TOFMS and the obtained data were subjected to principal component analysis. Remarkably, principal component 1 separated Heug‑ daegu from other cultivars, indicating that Heugdaegu has high levels of caffeic, sinapinic, and vanillic acids. Findings from this work may aid breeding programs aimed at improving the quality of rice seeds. Keywords: GC × GC‑ TOFMS, GC‑ TOFMS, Metabolic profiling, Oryza sativa L., Pigmented rice Introduction metabolites, study of biosynthetic pathway of metabo- Polar primary metabolites are essential for vital activities lites is important for understanding the plant physiology of organisms. In contrast, secondary metabolites are not and for designing new plants with enhanced nutritional directly involved in the normal growth, development, and value. Towards this end, profiling of primary metabolites reproduction, but several such metabolites are present in and analysis of the relationship between the metabolites plants for ecological purposes. Because primary metabo- in various plants has been performed using GC-TOFMS lites are used as precursors of other primary metabo- [1–4]. Researchers in the field of metabolomics have lites or as building blocks for the synthesis of secondary been striving to improve the chromatographic resolution and detection sensitivity of all metabolites present in a sample. *Correspondence: supark@cnu.ac.kr; kjkpj@inu.ac.kr Comprehensive two-dimensional gas chromatogra- Seung‑A. Baek and So Yeon Kim contributed equally to this work phy (GC × GC) is an advanced technique that improves Division of Life Sciences and Bio‑Resource and Environmental Center, the resolution of one-dimensional gas chromatography College of Life Sciences and Bioengineering, Incheon National University, Incheon 22012, Republic of Korea (GC). Two columns, with different stationary phases, Department of Crop Science and Department of Smart Agriculture are connected through a modulator (thermal or valve- Systems, Chungnam National University, 99 Daehak‑ro, Yuseong‑gu, based modulator). The eluates from the first column are Daejeon 34134, Republic of Korea Full list of author information is available at the end of the article transferred into the second column and are concentrated © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/. Baek et al. Appl Biol Chem (2021) 64:65 Page 2 of 9 during the transfer in the modulator [5]. GC × GC the performance of GC-TOFMS and GC × GC-TOFMS. coupled to quadrupole mass spectrometry (qMS) or Subsequent to method validation, polar metabolites time-of-flight mass spectrometry (TOFMS) reduces the extracted from two white and six red pigmented rice cul- probability of peak overlap and increases the number tivar seeds were analysed using GC × GC-TOFMS. The of detected peaks [6–8]. GC × GC-TOFMS has higher metabolic profiling data were analysed by principal com - selectivity and greater sensitivity than GC-TOFMS [5, 8, ponent analysis (PCA) for comparing the metabolites 9], and has, therefore, been used for metabolic analysis among the eight rice cultivars. of several plants including brown rice seeds. However, to the best of our knowledge, metabolite profiling of pig - Materials and methods mented rice seeds using GC × GC-TOFMS has not been Samples reported [10–12]. Eight varieties of Korean rice seeds were categorized as In plant metabolomics, GC × GC-TOFMS has been white and red according to the colour of their pericarp. applied to metabolite fingerprinting with chemomet - Two cultivars of white rice (Dongjin, DJ; Heugdaegu, ric tools. After non-targeted metabolic profiling, com - HDG) and six of red rice (Aengmi, AM; Goryeong 8, parison of data obtained from samples (i.e., control and GR8; Hanyangjo, HYJ; Hongjinju, HJJ; Jakwangdo, JKD; treated samples, cultivars, or species) with multivari- Jeogjinju, JJJ) were used in this study. The seeds were ate statistics is performed [13–16]. In the case of rice obtained from the Agricultural Genetic Resources Center grains, GC × GC-TOFMS has been used for non-tar- at the National Academy of Agricultural Science (Suwon, geted metabolomic studies. Volatile metabolites in 10 Korea). The seeds were harvested in 2016 and manually rice varieties were assayed to understand the mechanism hulled and ground to a fine powder with a mixer mill of synthesis of aroma compounds [17]. Eight volatile (HR2860, Philips, Amsterdam, The Netherlands) and compounds were selected as key markers responsible pestle and mortar. The powdered seeds were stored at for the differences between aromatic and non-aromatic − 20 °C until they were used. rice varieties [18]. Volatiles collected from microbe- and mite-contaminated rice grain were analysed using GC Chemicals × GC-TOFMS for assessment of contamination [14]. Methanol (HPLC grade) and chloroform (HPLC grade) Non-targeted profiling of polar metabolites in five brown were purchased from Daejung Chemical & Metal cultivars was performed using GC × GC-TOFMS in (Siheung, Korea) and Burdick & Jackson (Muskegon, MI, combination with GC-TOFMS analysis for investigating USA), respectively. Water used in the experiments was the metabolite diversity of rice variants [10]. In addition, prepared using the Millipore water purification system non-targeted profiling of polar metabolites in grains of (Milli-Q Direct 8; Milford, MA, USA). Adonitol (ribitol; three rice cultivars was performed using GC-TOFMS ≥ 99%), shikimic acid (≥ 99%), sinapinic acid (≥ 98%), and the results were compared with those obtained using methoxyamine hydrochloride (MOX; 98%), pyridine GC × GC-TOFMS analysis for assessing the data pro- (HPLC grade, ≥ 99.9%), and N-methyl-N-(trimethylsilyl) cessing and analysis methods [11]. trifluoroacetamide (MSTFA) were purchased from Sigma Non-targeted metabolite profiling has been important (St. Louis, MO, USA). p-Coumaric acid was purchased in discovering biomarkers and for screening unrevealed from MP Biomedicals (Solon, OH, USA). metabolites in samples; however, in plant metabolomics the focus has been on the already known metabolites Extraction of polar metabolites related to pathways including the tricarboxylic acid cycle, Polar metabolites, including amino acids, organic acids, glycolysis, and biosynthesis of secondary metabolites. sugars, and phenolic acids, in pigmented rice (n = 3) Physiological mechanisms have been analysed by com- were extracted following a previously described method paring the levels of metabolites in samples that were [1]. Briefly, 10  mg powder was mixed with 1  mL of mutated or exposed to stimuli with those in the respec- 2.5:1:1 (v/v/v) methanol:chloroform:water containing tive control samples [19–21]. In this study, we performed 60  μL adonitol (200  μg/mL) as an internal standard (IS). targeted metabolic profiling in rice seed using GC × After shaking at 1200  rpm for 30  min at 37  °C (Eppen- GC-TOFMS for the first time. Moreover, we compared dorf Thermomixer Comfort 5355, Eppendorf, Hamburg, the GC × GC-TOFMS platform with GC-TOFMS for Germany), the mixture was centrifuged at 16,000×g for analysis of polar metabolites in rice seed. The number of 3 min at 4 °C (MX-307, TOMY, Tokyo, Japan). The upper metabolites detected using GC-TOFMS was compared layer (800 μL) was transferred to a new tube and 400 μL with those detected using GC × GC-TOFMS. Limit of water was added. The mixture was vortexed and centri - detection (LOD), limit of quantification (LOQ), accuracy, fuged at 16,000×g for 3 min at 4 °C. The methanol:water and precision of standards were measured to compare phase (900  µL) was transferred to a new tube and dried Baek  et al. Appl Biol Chem (2021) 64:65 Page 3 of 9 completely using a vacuum centrifuge dryer (VS-802F, preparation to determine linearity, precision, and accu- Visionbionex, Gyeonggi, Korea) and a freeze-dryer racy (n = 3). Ribitol (12 μg) was added as an IS to each (MCFD8512, IlShinBioBase, Gyeonggi, Korea). For deri- standard sample. The linearity was calculated as the vatisation, 80  µL 2% MOX (in pyridine, w/v) was added square of the correlation coefficient (r) of the calibration to the sample and incubated at 30  °C, with shaking at curve. LOD and LOQ of shikimic acid, sinapinic acid, 1200  rpm for 90  min. Thereafter, 80  µL MSTFA was and p-coumaric acid were estimated as the following added and the mixture was incubated at 37 °C, with shak- expressions: 3α/S (LOD) and 10α/S (LOQ), where α is the ing at 1200 rpm for 30 min. The sample was moved to a standard deviation (n = 7) and S is the slope of the cali- vial for GC-TOFMS and GC × GC-TOFMS analyses. bration curve. The precision and accuracy were defined as the relative standard deviation (RSD) and recovery, GC‑TOFMS and GC × GC‑TOFMS conditions respectively. RSD was calculated using the following Agilent 7890A GC (Agilent, Santa Clara, CA, USA) cou- equation: (α/μ) × 100, where α is the standard devia- pled to a Pegasus TOFMS 4D (LECO, St Joseph, MI, USA) tion and μ is the mean (n = 3). Recovery was estimated was used for the analysis of polar metabolites using GC-as (SS /SS ) × 100, where SS is the amount of spiked C A C TOFMS and GC × GC-TOFMS. An Rtx-5MS column standard calculated using a calibration curve and SS is (0.25  mm × 0.25  µm × 30  m; Restek, Bellefonte, PA, the amount of spiked standard. USA) was used for GC-TOFMS analysis. For GC-TOFMS analysis, the column oven temperature was maintained Multivariate statistical analysis at 80 °C for 0.5 min, was increased at 5 °C/min to 330 °C Polar metabolite analysis was performed in triplicate. and then maintained at 330 °C for 5 min. For GC × GC- PCA was performed using SIMCA-P (version 13.0; TOFMS, Rtx-5MS (0.25 mm × 0.25 µm × 30 m; Restek) Umetrics, Umeå, Sweden). The data were normalised and Rxi-17sil MS (0.15 mm × 0.15 µm × 1.2 m; Restek) with unit variance scaling. columns were used as the first and second columns, respectively. The oven conditions for the first column Results and discussion were the same as in GC-TOFMS, whereas the second Metabolic profiling of DJ using GC‑TOFMS column oven temperature program was 5  °C above that and GC × GC‑TOFMS of the first column. The two columns were connected To compare the GC-TOFMS and GC × GC-TOFMS through a cryogenic modulator and liquid nitrogen was systems, polar metabolite profiling of the same derivate used as the cryogen. The modulator temperature pro - sample (DJ) was performed using the both the instru- gram was 15  °C above the second column temperature. ments. A total 38 polar metabolites were detected in DJ The modulation period was set to 4 s, with 0.6 s hot and by GC-TOFMS analysis. Twenty one amino acids and 1.4  s cool pulse duration. Except for the column oven amino acid derivatives, six organic acids, seven sugars, condition, other conditions were the same for both GC- phosphoric acid, ferulic acid, p-hydroxybenzoic acid TOFMS and GC × GC-TOFMS analyses. One microli- and glycerol were identified (Fig.  1A). On the contrary, a tre of derivatised sample was injected in the split (25:1) total of 47 polar metabolites were detected by GC × GC- mode at 250  °C. Helium was used as the carrier gas at TOFMS analysis (Fig.  1B). Twenty-two amino acids and a flow rate of 1.2  mL/min in a constant flow mode. The amino acid derivatives, eight organic acids, seven phe- temperature of transfer line and ion source was 260 and nolic acids, eight sugars, phosphoric acid, and glycerol 230  °C, respectively. The mass electron energy was set were identified. Cysteine, methionine, pyruvic acid, shi - at −  70  eV and the mass range for scanning was in the kimic acid, caffeic acid, salicylic acid, sinapinic acid, 45–650  m/z range. The mass acquisition rate was 10 vanillic acid, p-coumaric acid, p-hydroxybenzoic acid, spectra/sec in GC-TOFMS and 100 spectra/sec in GC and glyceric acid were only detected in the GC × GC- × GC-TOFMS. The qualitative analysis was performed TOFMS analysis (Fig. 1; Additional file 1: Table S1). by comparison with the retention time of respective standards and mass spectrum in an in-house library, Comparison of the sensitivities of GC‑TOFMS Wiley9, and NIST14 using the ChromaTOF software and GC × GC‑TOFMS (V4.50, LECO). The quantitative estimation was based on Shikimic acid is a precursor of aromatic amino acids peak area ratios relative to the peak area of the IS. (phenylalanine and tyrosine) and phenylpropanoid path- way starts with phenylalanine. Flavonoids, monolignols, Analytical method validation phenolic acids, stilbenes, and coumarins are derived from Six different concentrations of shikimic acid (0.1–5.0 μg), phenylalanine through successive consecutive enzymatic p-coumaric acid (0.25–12.5  μg), and sinapinic acid reactions in most plants [22]. Kim et  al. [1] reported a (0.75–12.5  μg) standards were analysed within 1  day of positive relationship between all phenolic and shikimic Baek et al. Appl Biol Chem (2021) 64:65 Page 4 of 9 Fig. 1 GC‑ TOFMS (A) and GC × GC‑ TOFMS (B) analytical ion chromatogram (AIC) of polar metabolites extracted from Dongjin. Peak: 1, Pyruvic acid; 2, Lactic acid; 3, Alanine; 4, Oxalic acid; 5, Valine; 6, Serine‑1; 7, Ethanolamine; 8, Leucine; 9, Glycerol; 10, Phosphoric acid; 11, Isoleucine; 12, Proline; 13, Glycine; 14, Succinic acid;15, Glyceric acid; 16, Fumaric acid; 17, Serine‑2; 18, Threonine; 19, β‑Alanine; 20, Malic acid; 21, Salicylic acid; 22, Methionine; 23, Pyroglutamic acid; 24, Aspartic acid; 25, 4‑Aminobutytic acid; 26, Cysteine; 27, Threonic acid; 28, Glutamic acid; 29, p‑Hydroxybenzoic acid; 30, Phenylalanine; 31, Asparagine; 32, Xylose ‑1; 33, Xylose ‑2; 34, Ribitol (internal standard); 35, Vanillic acid; 36, Glutamine; 37, Shikimic acid; 38, Citric acid; 39, Fructose‑1; 40, Fructose ‑2; 41, Lysine; 42, p‑ Coumaric acid; 43, Tyrosine; 44, Mannitol; 45, Ferulic acid; 46, Inositol; 47, Caffeic acid; 48, Tryptophan; 49, Sinapinic acid; 50, Sucrose; 51, Raffinose acids in black rice cultivars. Several studies have reported Table 1 Linearity (r ), limit of detection (LOD), and limit of that red rice seeds contain more phytochemicals, such as quantification (LOQ) of shikimic acid, p‑ coumaric acid, and flavone (apigenin) and phenolics, than white rice [23–25]. sinapinic acid analysed using GC‑ TOFMS and GC × GC‑ TOFMS Therefore, to validate the method, a representative pre - Metabolite Calibration curve Linearity LOD (ng) LOQ (ng) cursor (shikimic acid) and phenolic acid (p-coumaric acid GC‑ TOFMS and sinapinic acid) were selected among 10 metabolites Shikimic acid y = 0.0529x − 0.9963 9.26 30.88 that were only detected by GC × GC-TOFMS analysis. 0.0054 The calibration curves of shikimic acid, p-coumaric acid, p‑ Coumaric acid y = 0.0080x − 0.9910 90.18 300.60 and sinapinic acid were produced using GC-TOFMS and 0.0083 GC × GC-TOFMS to compare the LOD and LOQ. The Sinapinic acid y = 0.0035x − 0.9920 119.58 398.59 means of three-point data for calculating the ratio rela- 0.0025 tive to the peak area of the IS were used as calibration GC × GC‑ TOFMS curves. All the calibration curves showed linearity with Shikimic acid y = 0.1486x + 0.9978 0.31 1.03 a correlation coefficient above 0.99 (Table  1). The LOD 0.0004 of shikimic acid, p-coumaric acid, and sinapinic acid was p‑ Coumaric acid y = 0.0221x − 0.9987 55.87 186.25 0.0022 9.3, 90.2, and 119.6 ng and the LOQ was 30.9, 300.6, and Sinapinic acid y = 0.0176x − 0.9966 47.22 157.41 398.6  ng, respectively, in the GC-TOFMS analysis. In 0.0078 GC × GC-TOFMS, the LOD of shikimic acid, p-cou- maric acid, and sinapinic acid was 0.3, 55.9, and 47.2 ng, Baek  et al. Appl Biol Chem (2021) 64:65 Page 5 of 9 and the LOQ was 1.0, 186.3, and 157.4  ng, respectively. [6, 27, 28]. In this study, the peaks of glycerol and phos- The LOD and LOQ of shikimic acid, sinapinic acid, and phoric acid overlapped in GC-TOFMS but could be sepa- p-coumaric acid in the GC × GC-TOFMS analysis were rated and quantified using different fragment ions (m/z) 30.0-, 1.6-, and 2.5-times lower than those of in GC- (Fig. 2A–C). However, the mass spectrum of glycerol was TOFMS, respectively. The analytes eluting from the first affected by that of phosphoric acid. The unique fragment column were trapped in the thermal modulator by cold ion, with a comparatively higher intensity, was selected as jet flow and then re-injected by hot jet flow into the sec - the quantitative fragment ion for each metabolite (Fig. 2F, ond column. Because in this step, the eluates are concen- G). The m/z 103 and 299 were chosen as quantitative trated, it improves the sensitivity of GC × GC [5]. GC fragment ions of glycerol and phosphoric acid, respec- × GC-TOFMS analyse significantly reduces the matrix tively. However, m/z 103 fragment ion was detected in effect by separation of second column. In this study, the phosphoric acid with 0.8% relative abundance. The over - LOD and LOQ of GC × GC-TOFMS were better than lapped fragment ion for two metabolites could affect those of GC-TOFMS. the area of the glycerol peak. Winnike et al. [6] reported To determine the precision and accuracy, three differ - that the peak area ratio of each metabolite (R ) was PA ent concentrations within each calibration curve range expected to be consistent. R was the calculated peak PA were measured (Table  2). The precision (%RSD) for shi - area of metabolite obtained using the GC-TOFMS analy- kimic acid, p-coumaric acid, and sinapinic acid was lower sis divided by that obtained using the GC × GC-TOFMS than 12.9% in both GC-TOFMS and GC × GC-TOFMS analysis. R of leucine, phosphoric acid, and isoleucine PA analyses. The accuracy (%Recovery) ranged from 85.1 was 0.07, 0.07, and 0.06, respectively. However, R of PA to 114.6% for the GC-TOFMS and GC × GC-TOFMS glycerol was 0.21. The peak area of glycerol was over- analyses. The precision of GC-TOFMS and GC × GC- measured with m/z 103 fragment ion of phosphoric acid TOFMS for the metabolites was lower than 15%. The and it might have increased the R . PA accuracy (%Recovery) of GC-TOFMS and GC × GC- Moreover, some of the other fragments, including TOFMS for the metabolites ranged from 85 to 115% m/z 73, 133, 147, 175, and 205, overlapped. The m /z and was in agreement with the International Guidelines 133 fragment ion was monitored in both glycerol and (FDA Guidance) [26]. Thus, the GC-TOFMS and GC phosphoric acid with a relative abundance of 19.9% × GC-TOFMS systems were found to be valid tools for and 8.2%, respectively. The prominent fragment ions polar metabolite analysis, but GC × GC-TOFMS could of glycerol, m/z 147 and 205, showed a relative abun- detect lower levels of metabolites when compared with dance of 3.9% and 1.9% in phosphoric acid, respec- GC-TOFMS. tively. This suggests that the user has to finally sort out the fragment ions from each metabolite in an over- Improved separation ability in GC × GC‑TOFMS lapped mass spectrum. In GC × GC-TOFMS, the glyc- GC × GC-TOFMS showed a better peak separation erol peak also overlapped with that of phosphoric acid ability than GC-TOFMS. The two columns with differ - in the retention time in first dimension but they were ent properties (mid-polar and non-polar) lead to a bet- separated in the second dimension (Fig.  2D, E). Thus, ter separation of co-eluted analytes from the first column the two metabolites did not affect the mass spectrum Table 2 Precision (%RSD) and accuracy (%Recovery) for determination of shikimic acid, p‑ coumaric acid, and sinapinic acid using GC‑ TOFMS and GC × GC‑ TOFMS Metabolite GC‑ TOFMS GC × GC‑ TOFMS Content (μg) Precision (%RSD) Accuracy (%Recovery) Content (μg) Precision (%RSD) Accuracy (%Recovery) Shikimic acid 0.50 1.86 85.12 ± 1.58 0.05 7.55 89.95 ± 6.79 2.50 1.70 106.85 ± 1.81 0.10 11.24 96.95 ± 10.90 5.00 12.90 108.89 ± 14.04 1.00 8.02 98.22 ± 7.87 p‑ Coumaric acid 5.00 4.17 87.13 ± 3.63 0.50 11.61 93.48 ± 10.85 10.00 8.11 105.41 ± 8.54 1.00 6.06 96.31 ± 5.83 12.50 7.85 102.49 ± 8.05 5.00 12.59 107.04 ± 13.47 Sinapinic acid 7.50 7.55 90.67 ± 6.84 0.75 3.53 114.57 ± 4.04 10.00 2.68 104.66 ± 2.80 1.00 4.33 100.61 ± 4.36 12.50 3.98 100.91 ± 4.02 5.00 8.35 94.12 ± 7.86 Baek et al. Appl Biol Chem (2021) 64:65 Page 6 of 9 Fig. 2 Partial GC‑ TOFMS total ion chromatogram ( TIC) (A), analytical ion chromatogram (AIC) (B), and selected ion chromatogram (C) of glycerol and phosphoric acid in Dongjin. Partial GC × GC‑ TOFMS TIC contour plot (D) and surface plot (E) of phosphoric acid and glycerol in Dongjin. P Phosphoric acid; G Glycerol. Mass spectrum of glycerol (F) and phosphoric acid (G) from NIST library spectrum deconvolution, resulted in better selectivity Table 3 Mass spectral similarity of phosphoric acid and glycerol in Dongjin analysed using GC‑ TOFMS and GC × GC‑ TOFMS of GC × GC-TOFMS than that of GC-TOFMS. Peak number Library GC‑ TOFMS GC × GC‑ TOFMS NIST Wiley9 NIST Wiley9 Analysis of polar metabolites in two white and six 9 Glycerol 860 860 935 927 pigmented rice cultivars with GC × GC‑TOFMS 10 Phosphoric acid 817 818 873 919 In this study, GC × GC-TOFMS showed improved sensitivity and peak separation ability compared with GC-TOFMS. Thus, polar metabolite profiling in eight rice cultivars was performed using GC × GC-TOFMS. and area of each other. In addition, the user could eas- As observed for DJ, a total 47 polar metabolites were ily determine whether the mass spectrum of a peak is detected in the seeds of all the cultivars. The obtained consistent with those of targeted metabolites by refer- data were subjected to PCA to assess the differences in ring to the similarity in a library, such as NIST and polar metabolite composition among rice seeds (Fig. 3). Wiley (Table  3). The separated peak, leading to better Baek  et al. Appl Biol Chem (2021) 64:65 Page 7 of 9 Fig. 3 Score plots (A) and loading plots (B) of principal component analysis (PCA) obtained from metabolic profiling by using GC × GC‑ TOFMS analysis. AM Aengmi; GR8 Goryeong 8; HYJ Hanyangjo; HJJ Hongjinju; JKD Jakwangdo; JJJ Jeogjinju; DJ Dongjin; HDG Heugdaegu Baek et al. Appl Biol Chem (2021) 64:65 Page 8 of 9 PCA is a good tool to obtain an overview of the com- Supplementary Information prehensive data and has been used in metabolomics The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s13765‑ 021‑ 00640‑3. studies [29]. In the score plots, each point represents an individual sample, and samples exhibiting similar Additional file 1: Table S1. Retention times (RT ), relative retention times variances are clustered together. The first two princi- (RRT ), and mass spectral data of polar metabolites as trimethylsilyl deriva‑ pal components (PCs; PC1 and PC2) had the greatest tives. Table S2. Composition and content (ratio/g) of polar metabolites in eigenvalues and captured 63.0% of the total variance 8 rice cultivars on the GC×GC‑ TOFMS. (Fig.  3A). The same cultivar was closely clustered in the score plot but was not clustered with the colour Acknowledgements of rice. The white cultivars (blue-coloured spots; DJ This work was supported by Incheon National University Research Concentra‑ tion Professors Grant in 2021. and HDG) were not clearly separated with the red cul- tivars (red-coloured spots; AM, GR8, HYJ, HJJ, JKD, Authors’ contributions and JJJ). However, clear a separation was observed Conceptualization, methodology: JKK, SUP, SYK and S‑AB. Data curation: SYK and S‑AB. Formal analysis: YJP and SYK. Writing—original draft prepara‑ between the HDG cultivar and other cultivars by tion: S‑AB, SYK and JKK. Writing—review and editing: JKK, S‑HL and TJK. PC1. Although the HDG has white-coloured pericarp, Project administration: JKK and SUP. All authors read and approved the final its hull has black colour. PC2 separated the AM cul- manuscript. tivar from other cultivars. Although the AM cultivar belongs to the same species as the cultivated rice, it is Declarations referred to as weedy rice in Korea. The corresponding Competing interests loading plot represents the metabolites responsible The authors declare that there is no competing interests. for separation on the score plots (Fig.  3B). HDG was Author details located on the left of the score plot and most metab- Division of Life Sciences and Bio‑Resource and Environmental Center, olites were located on the left of the loading plot. It College of Life Sciences and Bioengineering, Incheon National University, indicated that the HDG contained relatively higher Incheon 22012, Republic of Korea. Division of Horticultural Biotechnology, School of Biotechnology, Hankyong National University, Anseong 17579, levels of most metabolites than the other cultivars. It Republic of Korea. Department of Crop Science and Department of Smart is known that red rice seed contains relatively higher Agriculture Systems, Chungnam National University, 99 Daehak‑ro, levels of phenolics and flavonoids than white rice seed Yuseong‑gu, Daejeon 34134, Republic of Korea. Honam National Institute of Biological Resources, 58762 Mokpo, Republic of Korea. Nak donggang [24, 30]. However, Shen et  al. [25] revealed that sev- National Institute of Biological Resources, 37242 Sangju, Republic of Korea. eral white cultivars have a higher content of phenolic acids than red cultivars. In this study, relatively higher Received: 13 June 2021 Accepted: 1 September 2021 levels of caffeic acid, sinapinic acid, and vanillic acid were found in HDG, which has black-coloured hull, than in red cultivars. Phenolic acids are important References building blocks employed in the biosynthesis of fla- 1. Kim JK, Park SY, Lim SH, Yeo Y, Cho HS, Ha SH (2013) Comparative meta‑ vonoids. The level of raffinose was relatively higher in bolic profiling of pigmented rice (Oryza sativa L.) cultivars reveals primary GR8 and HYJ (Additional file 1: Table S2). The content metabolites are correlated with secondary metabolites. J Cereal Sci 57:14–20. https:// doi. org/ 10. 1016/j. jcs. 2012. 09. 012 and composition of amino acids are shifted and raffi- 2. Park SY, Lim SH, Ha SH, Yeo Y, Park WT, Kwon DY, Park SU, Kim JK (2013) nose is accumulated during seed development [31]. In Metabolite profiling approach reveals the interface of primary and this study, it was suggested that the expression pattern secondary metabolism in colored cauliflowers (Brassica oleracea L. ssp. botrytis). J Agric Food Chem 61:6999–7007. https:// doi. org/ 10. 1021/ jf401 of genes or activity of enzymes associated with amino 330e acid and raffinose metabolism might differ among the 3. Park CH, Yeo HJ, Park YE, Baek SA, Kim JK, Park SU (2019) Transcriptome cultivars. No metabolite analysis data are available analysis and metabolic profiling of Lycoris radiata. Biology 8:63. https:// doi. org/ 10. 3390/ biolo gy803 0063 except for γ-oryzanols and octacosanol in the weedy 4. Park Y, Baek SA, Choi Y, Kim JK, Park SU (2019) Metabolic profiling of nine rice, AM [32, 33]. The loading plot showed that AM Mentha species and prediction of their antioxidant properties using contained relatively higher levels of leucine, isoleu- chemometrics. Molecules 24:258. https:// doi. org/ 10. 3390/ molec ules2 40202 58 cine, methionine, mannitol, and phenylalanine than 5. Bahaghigha DH, Freye CE, Synovec RE (2019) Recent advances in modula‑ the other cultivars. Thus, chemometrics, combined tor technology for comprehensive two dimensional gas chromatography. with GC × GC-TOFMS, can facilitate a detailed met- Trac‑ Trend Anal Chem 113:379–391. https:// doi. org/ 10. 1016/j. trac. 2018. 04. 016 abolic profiling in rice for further studies in breeding 6. Winnike JH, Wei X, Knagge KJ, Colman SD, Gregory SG, Zhang X (2015) programs. Comparison of GC‑MS and GC × GC‑MS in the analysis of human serum samples for biomarker discovery. J Proteome Res 14:1810–1817. https:// doi. org/ 10. 1021/ pr501 1923 Baek  et al. Appl Biol Chem (2021) 64:65 Page 9 of 9 7. Almstetter MF, Oefner PJ, Dettmer K (2012) Comprehensive two‑ fruit maturity. Food Chem 331:127286. https:// doi. org/ 10. 1016/j. foodc dimensional gas chromatography in metabolomics. Anal Bioanal Chem hem. 2020. 127286 402:1993–2013. https:// doi. org/ 10. 1007/ s00216‑ 011‑ 5630‑y 21. You MK, Lee YJ, Kim JK, Baek SA, Jeon YA, Lim SH, Ha SH (2020) The 8. Prebihalo SE, Berrier KL, Freye CE, Bahaghighat HD, Moore NR, David PK, organ‑specific differential roles of rice DXS and DXR, the first two Synovec RE (2018) Multidimensional gas chromatography: advances in enzymes of the MEP pathway, in carotenoid metabolism in Oryza instrumentation, chemometrics, and applications. Anal Chem 90:505– sativa leaves and seeds. BMC Plant Biol 20:167. https:// doi. org/ 10. 1186/ 532. https:// doi. org/ 10. 1021/ acs. analc hem. 7b042 26s12870‑ 020‑ 02357‑9 9. Pierce KM, Hoggard JC, Mohler RE, Synovec RE (2008) Recent advance‑ 22. Deng Y, Lu S (2017) Biosynthesis and regulation of phenylpropanoids in ments in comprehensive two‑ dimensional separations with chemomet‑ plants. CRC Crit Rev Plant Sci 36:257–290. https:// doi. org/ 10. 1080/ 07352 rics. J Chromatogr A 1184:341–352. https:// doi. org/ 10. 1016/j. chroma. 689. 2017. 14028 52 2007. 07. 059 23. Kim JK, Lee SY, Chu SM, Lim SH, Suh SC, Lee Y T, Cho HS, Ha SH (2010) 10. Kusano M, Fukushima A, Kobayashi M, Hayashi N, Jonsson P, Moritz T, Variation and correlation analysis of flavonoids and carotenoids in Korean Ebana K, Saito K (2007) Application of a metabolomic method combining pigmented rice (Oryza sativa L.) cultivars. J Agric Food Chem 58:12804– one‑ dimensional and two‑ dimensional gas chromatography‑time ‑ of‑ 12809. https:// doi. org/ 10. 1021/ jf103 277g flight/mass spectrometry to metabolic phenotyping of natural variants in 24. Jun HI, Song GS, Yang EI, Youn Y, Kim YS (2012) Antioxidant activities rice. J Chromatogr B 855:71–79. https:// doi. org/ 10. 1016/j. jchro mb. 2007. and phenolic compounds of pigmented rice bran extracts. J Food Sci 05. 002 77:C759–C764. https:// doi. org/ 10. 1111/j. 1750‑ 3841. 2012. 02763.x 11. Allwood WJ, Erban A, de Koning S, Dunn WB, Luedemann A, Lommen A, 25. Shen Y, Jin L, Xiao P, Lu Y, Bao J (2009) Total phenolics, flavonoids, anti‑ Kay L, Löscher R, Kopka J, Goodacre R (2009) Inter‑laboratory reproduc‑ oxidant capacity in rice grain and their relations to grain color, size and ibility of fast gas chromatography–electron impact–time of flight mass weight. J Cereal Sci 49:106–111. https:// doi. org/ 10. 1016/j. jcs. 2008. 07. 010 spectrometry (GC–EI–TOF/MS) based plant metabolomics. Metabolomics 26. Food and drug administration (2018) Bioanalytical method validation 5:479–496. https:// doi. org/ 10. 1007/ s11306‑ 009‑ 0169‑z guidance for industry. https:// www. fda. gov/ files/ drugs/ publi shed/ Bioan 12. Xu B, Zhang L, Ma F, Zhang W, Wang X, Zhang Q, Luo D, Ma H, Li P (2018) alyti cal‑ Method V‑alid ation‑ Guida nce‑ for‑ Indus try. pdf . Accessed 13 Jan Determination of free steroidal compounds in vegetable oils by compre‑ 2021 hensive two‑ dimensional gas chromatography coupled to time‑ of‑flight 27. Welthagen W, Shellie RA, Spranger J, Ristow M, Zimmermann R, Fiehn O mass spectrometry. Food Chem 245:415–425. https:// doi. org/ 10. 1016/j. (2005) Comprehensive two‑ dimensional gas chromatography–time‑ of‑ foodc hem. 2017. 10. 114 flight mass spectrometry (GC × GC‑ TOF) for high resolution metabo‑ 13. Pierce KM, Hope JL, Hoggard JC, Synovec RE (2006) A principal compo‑ lomics: biomarker discovery on spleen tissue extracts of obese NZO nent analysis based method to discover chemical differences in com‑ compared to lean C57BL/6 mice. Metabolomics 1:65–73. https:// doi. org/ prehensive two‑ dimensional gas chromatography with time‑ of‑flight 10. 1007/ s11306‑ 005‑ 1108‑2 mass spectrometry (GC × GC‑ TOFMS) separations of metabolites in plant 28. Ahn YG, Jeon SH, Lim HB, Choi NR, Hwang GS, Kim YP, Lee JY (2018) samples. Talanta 70:797–804. https:// doi. org/ 10. 1016/j. talan ta. 2006. 01. Analysis of polycyclic aromatic hydrocarbons in ambient aerosols by 038 using one‑ dimensional and comprehensive two‑ dimensional gas chro‑ 14. Salvador ÂC, Baptista I, Barros AS, Gomes NC, Cunha Â, Almeida A, Rocha matography combined with mass spectrometric method: a comparative SM (2013) Can volatile organic metabolites be used to simultaneously study. J Anal Methods Chem. https:// doi. org/ 10. 1155/ 2018/ 83416 30 assess microbial and mite contamination level in cereal grains and coffee 29. Kim TJ, Park JG, Ahn SK, Kim KW, Choi J, Kim HY, Ha SH, Seo WD, Kim beans? PLoS ONE 8:e59338. https:// doi. org/ 10. 1371/ journ al. pone. 00593 JK (2020) Discrimination of adzuki bean (Vigna angularis) geographi‑ 38 cal origin by targeted and non‑targeted metabolite profiling with gas 15. Hurtado C, Parastar H, Matamoros V, Piña B, Tauler R, Bayona JM (2017) chromatography time‑ of‑flight mass spectrometry. Metabolites 10:112. Linking the morphological and metabolomic response of Lactuca sativa L https:// doi. org/ 10. 3390/ metab o1003 0112 exposed to emerging contaminants using GC × GC‑MS and chemomet ‑ 30. Kim GR, Jung ES, Lee S, Lim SH, Ha SH, Lee CH (2014) Combined mass ric tools. Sci Rep 7:1–13. https:// doi. org/ 10. 1038/ s41598‑ 017‑ 06773‑0 spectrometry‑based metabolite profiling of different pigmented rice 16. Lim V, Gorji SG, Daygon VD, Melissa F (2020) Untargeted and targeted (Oryza sativa L.) seeds and correlation with antioxidant activities. Mol‑ metabolomic profiling of Australian indigenous fruits. Metabolites 10:114. ecules 19:15679–15686. https:// doi. org/ 10. 3390/ molec ules1 91015 673 https:// doi. org/ 10. 3390/ metab o1003 0114 31. Rosental L, Nonogaki H, Fait A (2014) Activation and regulation of primary 17. Daygon VD, Calingacion M, Forster LC, De Voss JJ, Schwartz BD, metabolism during seed germination. Seed Sci Res 24:1–15. https:// doi. Ovenden B, Alonso DE, McCouch SR, Garson MJ, Fitzgerald MA (2017) org/ 10. 1017/ S0960 25851 30003 91 Metabolomics and genomics combine to unravel the pathway for the 32. Lee MC, Lee DJ, Lee GA, Park HJ, Lee JR, Choi YM, Lee SK, Jung Y, Cho YG, presence of fragrance in rice. Sci Rep 7:8767. https:// doi. org/ 10. 1038/ Song JY (2013) Analysis and comparison of the γ‑ oryzanol content based s41598‑ 017‑ 07693‑9 on phylogenetic groups in Korean landraces of rice (Oryza sativa L.). Plant 18. Setyaningsih W, Majchrzak T, Dymerski T, Namieśnik J, Palma M (2019) Breed Biotechnol 1:58–69. https:// doi. org/ 10. 9787/ PBB. 2013.1. 1. 058 Key‑marker volatile compounds in aromatic rice (Oryza Sativa) grains: an 33. Cho YH, Farhoudi R, Farooq M, Lee DJ (2017) Evaluating Korean rice HS‑SPME extraction method combined with GC × GC‑ TOFMS. Molecules genotypes and landraces for octacosanol contents and antioxidant activ‑ 24:4180. https:// doi. org/ 10. 3390/ molec ules2 42241 80 ity. Nat Prod Res 31:2778–2782. https:// doi. org/ 10. 1080/ 14786 419. 2017. 19. Shen S, Weng R, Li L, Xu X, Bai Y, Liu H (2016) Metabolomic analysis of 12922 71 mouse embryonic fibroblast cells in response to autophagy induced by acute starvation. Sci Rep 6:34075. https:// doi. org/ 10. 1038/ srep3 4075 Publisher’s Note 20. Kim TJ, Hyeon HJ, Park NI, Yi TG, Lim SH, Park SY, Ha SH, Kim JK (2020) A Springer Nature remains neutral with regard to jurisdictional claims in pub‑ high‑throughput platform for interpretation of metabolite profile data lished maps and institutional affiliations. from pepper (Capsicum) fruits of 13 phenotypes associated with different

Journal

Applied Biological ChemistrySpringer Journals

Published: Dec 1, 2021

Keywords: GC  ×  GC-TOFMS; GC-TOFMS; Metabolic profiling; Oryza sativa L.; Pigmented rice

References