Plants Metabolome Study: Emerging Tools and Techniques

Metabolomics is now considered a wide-ranging, sensitive and practical approach to acquire useful information on the composition of a metabolite pool present in any organism, including plants. Investigating metabolomic regulation in plants is essential to understand their adaptation, acclimation and defense responses to environmental stresses through the production of numerous metabolites. Moreover, metabolomics can be easily applied for the phenotyping of plants; and thus, it has great potential to be used in genome editing programs to develop superior next-generation crops. This review describes the recent analytical tools and techniques available to study plants metabolome, along with their significance of sample preparation using targeted and non-targeted methods. Advanced analytical tools, like gas chromatography-mass spectrometry (GC-MS), liquid chromatography mass-spectroscopy (LC-MS), capillary electrophoresis-mass spectrometry (CE-MS), fourier transform ion cyclotron resonance-mass spectrometry (FTICR-MS) matrix-assisted laser desorption/ionization (MALDI), ion mobility spectrometry (IMS) and nuclear magnetic resonance (NMR) have speed up precise metabolic profiling in plants. Further, we provide a complete overview of bioinformatics tools and plant metabolome database that can be utilized to advance our knowledge to plant biology.


Metabolomics: Plant Biology Perspective
Metabolomics is one of the fastest developing and attractive disciplines of the omics field, with huge potential and prospects in crop improvement programs. It is vital to review the abiotic/biotic stress tolerances and metabolomics-assisted breeding of crop plants [1]. Recent metabolomics platforms play a crucial role in exploring unknown regulatory networks that control plant growth and development [1]. Further innovative metabolomics application, called ecological metabolomics, deals with studying the biochemical interactions among plants across different temporal and spatial networks [2]. It describes the biochemical nature of various vital ecological phenomena, such as the effects of parasite load, the incidence of disease, and infection. It also helps to decode the potential impact of biotic and abiotic stresses on any critical biochemical process through the detection of metabolites [1]. Modern metabolomics platforms are being exploited to explain complex Table 1. Identification of key metabolites in various plant species using different analytical methods.

Plant Species
Class Analytical Tools Key Metabolites Reference

Secondary Metabolites
Secondary metabolites (SMs) play a crucial role in protecting plants against various environmental stresses. It has been estimated that approximately 100,000 SMs have been reported within different plant species and are classified into multiple groups, nitrogencontaining compounds, terpenes, thiols, and phenolic compounds [41]. In Scutellaria baicalensis, the major flavonoids are accumulated in the roots before the full-bloom stage [42]. Two flavonoids, kaempferol 3-(2 ,3 -diacetylrhamnoside)-7-rhamnoside and apigenin 7rhamnoside were found in all developmental stages of P. ovata [9]. The root of Achyranthes bidentata, oleanolic acid and ecdysterone levels are increased during the vegetative growth than in reproductive growth [38]. Nutraceutical flavonoids; luteolin, quercetagetin, syringetin, kaempferol, limocitrin, helilupolone and catechin/epicatechin/pavetannin B2 and were identified in leaf extract, whereas alkaloids, lunamarine and hordatine B were identified in the seed extract and pinidine was detected in the husk extract [10]. The plant growth regulators gibberellic acid (GA), indole -3-acetic acid (IAA) and 6-Benzylaminopurine (BAP) show that the main terpenes (methyl chavicol and trans-anethole) and other terpenes (eucalyptol and azulene) undergo certain changes depending on the type of the treatment of plant growth regulators in O. gratissimum [35]. The application of growth regulators enhances the production of essential oils (menthone, menthol, pulegone, and Plants 2021, 10, 2409 5 of 24 menthofuran) in Mentha piperita, which is revealed to be rich in economically important terpenes [36]. The foliar application of triacontanol significantly increased the amount of active terpenes (menthol, L-methone, isomenthone, and menthyl acetate) in Mentha arvensis [37]. Lin et al. [43] conducted a phytochemical screening of Pteris vittata and identified four flavonoids: quercetin, kaempferol, kaempferol-3-O-D-glucopyranoside and rutin [43]. Scoploletin, umbelliferone and esculetin, as well as their glycosides scopolin, skimmin, and esculin were found in Arabidopsis thaliana [39] (Table 1).

Significance of Sample Preparation in Plant Metabolites
In plant metabolomics study, plant samples are harvested, stored, metabolites extraction and quantification, followed by data interpretation. Sample preparation is a key step in plant metabolomics as it significantly changes the quantity of the metabolites. Thus, considering all the factors, harvesting and storage of plant samples should be quick as to reduce the changes of biochemical reaction in the plant cells [56]. Inappropriate handling during the sample collection is the most likely source of bias in plant metabolomic studies [57]. Sample harvesting, storage, and extract preparation should ideally follow the Metabolomics Standards Initiative (MSI) to justify plant metabolomics studies [58].

Sample Harvesting and Storage
Commonly, four major steps are involved in plant metabolomics; harvesting, storage, extraction, and sample analysis ( Figure 1). Plant sample harvesting must be carried out with caution, as the metabolome of the plant is sensitive to enzymatic reactions that can degrade different metabolites. In addition, metabolites vary with the different development stages, plant age, and time of sample harvesting [6]. Mostly, 10-100 mg of plant samples are required for each biological sample in metabolomics studies. Usually, immediately after harvesting, the plant samples are snap-frozen in liquid nitrogen to prevent metabolic changes. Similarly, various storage techniques, such as freeze-drying, oven-drying, and air-drying, are essential for the processing of metabolomics [57,59].

Sample Harvesting and Storage
Commonly, four major steps are involved in plant metabolomics; harvesting, storage, extraction, and sample analysis ( Figure 1). Plant sample harvesting must be carried out with caution, as the metabolome of the plant is sensitive to enzymatic reactions that can degrade different metabolites. In addition, metabolites vary with the different development stages, plant age, and time of sample harvesting [6]. Mostly, 10-100 mg of plant samples are required for each biological sample in metabolomics studies. Usually, immediately after harvesting, the plant samples are snap-frozen in liquid nitrogen to prevent metabolic changes. Similarly, various storage techniques, such as freeze-drying, oven-drying, and air-drying, are essential for the processing of metabolomics [57,59].  The most important step in data processing and mining includes correction of baseline shifts, background noise reduction, chromatograph alignment and peaks detection. Biological interpretation and integration include enrichment analysis, networks, and pathways analysis for a comprehensive scope of the metabolome. GC-MS, gas chromatographymass spectrometry; IMS, ion mobility spectrometry; LC-MS, liquid chromatography mass-spectroscopy; CE-MS, capillary electrophoresis-mass spectrometry; FTICR-MS, fourier transform ion cyclotron resonance-mass spectrometry; MALDI, matrix-assisted laser desorption/ionization; NMR, nuclear magnetic resonance.

Sample Preparation
Sample preparation plays a key role in metabolomic study, as it includes the extraction of metabolites using different extraction methods ( Figure 1). Among the extraction methods, quenching, mechanical and ultrasound extraction methods are promising in the metabolomic analysis [60]. In addition, high quality, yield and chemical versatility can be obtained by integrating ultrasound extraction method and mechanical grinding [61]. Apart from extraction methods, the choice of solvents is also crucial, as a single solvent cannot extract a variety of metabolites (e.g., polar or nonpolar). A wide variety of metabolites can be isolated using a solvent system composed of chloroform: methanol: water [62,63]. This solvent system is widely used for a wide variety of metabolites such as polar compounds, nonpolar compounds, and hydrophilic metabolites. Diverse solvent systems were reported for the plant metabolomics, such as extraction with pure methanol [64,65], the mixture of methanol: water [66], and methanol: methyl-tert-butyl-ether: water [67]. A specific solvent gradient extraction method was developed to recover almost all types of metabolites in a single protocol [68]. In addition, hot methanol (70% v/v) was used to extract phenolic compounds from Brassica oleracea using ultra-high-performance liquid chromatographydiode array detector-tandem mass spectrometry [69]. Various methods are used for sample preparation, such as microwave-assisted extraction [70], ultrasound-assisted extraction [71], Swiss rolling technique [72], and enzyme-assisted extraction [73].
Targeted metabolite identification and quantification are the primary approaches for metabolomics investigation [74]. Sample preparation for target metabolites extracted from plant components such as leaves, stems, roots, etc., includes enrichment for metabolites of interest and removal of contaminants such as proteins and salts that hamper the analysis. Targeted metabolomics-based quantification aims for enhanced metabolite coverage by analyzing the selected metabolites [75]. The targeted metabolites extracted using different extraction methods such as different proportion of organic solvents [67], liquid-liquid extraction [75], and solid phase extraction method [76]. To increase analytical reliability, single or multiple internal standards can be spiked into the sample mixture during sample preparation [77]. In the final step of sample preparation for LC-MS, the solvents were evaporated, followed by re-dissolving the sample with a suitable solvent for LC-MS analysis [75]. Targeted metabolite quantification has been considered as the key method because of its reliable quantification accuracy, sensitivity and stability [78]. However, this method is typically confined to measuring a small number of known pre-selected analysts and is incapable of detecting unknown and novel metabolites. LC-multiple reaction monitoring (MRM)-MS approach has been employed for targeted metabolomics quantification analysis due to its rapid scan speed and good analytic stability [79]. New techniques have been developed to broaden the choices for targeted metabolomics research, using high-resolution equipment such as parallel reaction monitoring (PRM) [78]. In plant metabolomics, new extraction methods are also developing day by day in line depending on the nature of the compounds and selection of analytical systems.

Analytical Techniques Used for Plant Metabolome
Along with sample preparation, different MS-based analytical systems are available for data acquisition. In plant metabolomics, single analytical tools cannot be used to identify all the metabolites present in a sample; instead, a set of various techniques are needed to provide the largest amount of metabolite coverage [1]. Various metabolomics tools include MS-based techniques, namely GC-MS, LC-MS CE-MS, FTICR-MS MALDI, IMS, and NMR for sensitive and specific qualitative and quantitative analyses of metabolites ( Figure 1) [6,80]. All seven mentioned analytical methods identifying metabolites in plant tissue directly or indirectly have advantages and disadvantages (Table 3). Also, the combination of analytical methods can be used to ensure the efficacy of metabolite profiling.

Gas Chromatography-Mass Spectrometry (GC-MS)
GC-MS is an ideal technique for the identification and quantification of small metabolites (~500 Daltons). These molecules include amino acids, fatty acids, hydroxyl acids, alcohols, sugars, sterols, and amines, which are identified mostly using chemical derivatization to make them volatile enough for gas chromatography [81]. Moreover, different methods of derivatization, such as alkylation, acylation, methoximation, trimethylsilylation, and silylation, can also be used. Two derivatization steps are required for the extraction and identification of metabolites using GC-MS. The first step requires the conversion of all the carbonyl groups using methoxyamine hydrochloride into corresponding oximes. The seconnhd step is followed by a trimethylsilylation reaction to increasing the volatility of the derivative metabolites using derivatizing reagents such as N-Methyl-N-(trimethylsilyl) trifluoroacetamide (MSTFA) and N,O-bis-(trimethylsilyl)-trifluoroacetamide (BSTFA) [82][83][84]. In this procedure, the hydrogen is replaced from the -NH, -SH, -OH and -COOH of specific metabolites with [-Si(CH3)3] and are converted into thermally stable, less polar and volatile trimethylsilyl (TMS)-ether, TMS-ester, TMS-amine, or TMS-sulphide groups, respectively [83]. Also, GC-MS is the preferable chromatographic technique for identifying low molecular weight compounds that are either volatile or can be converted into volatile and thermally stable metabolites by chemical derivatization prior to analysis [85]. The technique includes primary metabolites such as sugars, fatty acids, amino acids, long-chain alcohols, amines, organic acids, and sterols.
There are two major forms of ionization used in GC-MS that comprises of electron ionization (EI) and chemical ionization (CI). Till now, the majority of GC-MS methods in metabolomics utilize EI. GC with EI detector equipped with single quadrupole (Q) mass analyzer is the oldest and most advanced analytical tool with robustness, high sensitivity, resolution and reproducibility, but suffers from sluggish scanning speeds and also poor mass accuracy (~50-200 ppm). Therefore, GC with a time-of-flight mass spectrometry (TOF-MS) analyzer is more preferred for metabolic profiling as it provides higher mass accuracy, faster acquisition times, and improved deconvolution for complex mixtures [86]. Among all metabolomics techniques, GC-MS is one of the most standardized, efficient, productive technique in plant metabolomics and it is considered a most versatile platform for metabolites analysis [87]. In addition, GC-MS has the availability of the huge number of well-established libraries of both commercial and in-house metabolite databases [88][89][90]. Metabolite profiling is utilized as an essential tool for screening of GM crops with regard to quality and health requirements and in categorization to an investigation of potential changes in metabolic contents, e.g., T. aestivum [53], O. sativa [44], and Z. mays [91].

Liquid Chromatography-Mass Spectrometry (LC-MS)
LC-MS is one of the most comprehensive analytical techniques in plant metabolome research, which is used to measure a wide variety of complex metabolites. The LC-MS approach is appropriate for high molecular weight (>500 kDa) plant metabolites, heat-labile functional groups, chemically unstable functional groups, and high-vapor-point. It does not require volatilization of the metabolites. LC-MS is also quite effective techniques in profiling of SMs (e.g., alkaloids, phenolics, flavonoids and terpenes), lipids (e.g., phospholipids, sphingolipids and glycerolipids) and sterols, and steroids [19,24,92,93].
LC-MS can also be used with various ionization methods and depending on the choice of specific separating columns based on the chemical characteristics of both mobile and stationary phases [94]. Currently, reverse-phase columns such as C18 or C8 are the most widely used columns for LC gradient separation. In reverse-phase separations, organic solvent/aqueous mixed mobile phases are often used, such as water: acetonitrile or water: methanol. Atmospheric pressure ionization (API) and electron spray ionization (ESI) are the most widely used ionization tools for LC-MS [94,95]. ESI and API have provided limited structural information of the compound because they introduce less internal energy and produce only a few fragments [95]. Structural information is typically obtained by number of fragments using collision-induced dissociation (CID) on tandem MS n . Commonly, two tandem MS n analytical tool configurations are commonly available with the LC-MS-based metabolite analysis: tandem-in-time and tandem-in-space. The ion trap MS is used by tandem-in-time instruments, such as quadruple ion traps (QIT-MS), FTICR-MS and orbitrap. The tandem-in-space tool facilitates two sequential steps of mass spectrometric analysis (MS2); it includes two mass analyzers separated by a collision cell [96,97]. Although LC-MS requires standard reference compounds to identify and quantify SMs, this restricts the analysis of metabolites that are not commercially available [98,99].

Capillary Electrophoresis-Mass Spectrometry (CE-MS)
CE-MS is a strong analytical technique for evaluating a large variety of ionic metabolites based on the proportion of charge and size ratio [93]. It provides fast and highresolution of charged compounds from small injection volumes and enables the metabolites characterization based on mass fragmentation [57]. The coverage of CE-MS metabolites majorly overlaps with GC-MS, but requires no derivatization, thus this technique save time and consumables. CE is performed in a fused silica capillary tube, the ends of which are dipped in buffer solutions and across which high voltages (20-30 kV) are employed [84]. Furthermore, CE has low sensitivity and reproducibility, poor migration time and lack of reference libraries; therefore, it is the least appropriate platform for studying metabolites from complex plant samples [100,101]. However, CE has some distinct rewards over other metabolomics tools; primarily the fact that it uses low volume of separation, which is especially appropriate for the study of plant metabolome [57,102].

Fourier Transform ion Cyclotron Resonance-Mass Spectrometry (FTICR-MS)
FTICR-MS provides the highest resolving power and mass accuracy among all kinds of mass spectrometry [103]. Its specific analytical features have made FTICR an important technique for proteomics and metabolomics. The ability of FTICR-MS to provide ultimate high resolution and high mass accuracy data is now frequently used as part of metabolomics procedures [84]. It's also well compatible with multi-stage mass spectrometry (MSn) analyzers. However, the instrument associated with a high magnetic field, complex ion-ion interactions and high cost are major barriers to its widespread application and use in plant metabolomics studies [56].

Matrix-Assisted Laser Desorption/Ionization (MALDI)
Recently, the applications of MALDI-Mass Spectrometry Imaging (MSI) and other MSI tools use a non-target approach for the qualitative or quantitative imaging of a broad variety of metabolites [104]. In plants, many studies have used MALDI-MSI to assess the spatial distribution of lipids, sugars and other classes of metabolites from plant parts such as flowers, leaves and roots [105,106]. In addition, MALDI-MSI has permitted the simultaneous analysis of the distribution of many peptides and proteins actively from a plant tissue section. This method involves coating a thin film of a matrix comprising either sinapinic acid, α-Cyano-4-hydroxycinnamic acid (CHCA) and 2,5-dihdroxybenzoic acid (2,5-DHBA) on the tissue surface. At each stage, a laser beam is inserted across the matrix-coated tissue to obtain a mass spectrum. For protein/metabolites imaging, MALDI is the most used method of ionization, combined with a wide variety of different mass analyzers, namely ToF, ToF-ToF, QqToF (quadrupole time of flight), Fourier ICR transform (FT-ICR), and ion-trap (both linear and spherical). All of these have their own merits and have previously been addressed and reviewed [107]. Other different ionization techniques such as secondary ion mass spectrometry (SIMS), desorption electrospray ionization (DESI) and laser ablation electrospray ionization (LAESI) have been also investigated [108].

Ion Mobility Spectrometry (IMS)
Ion mobility spectrometry (IMS), which separates gas ions based on their size-tocharge ratio, has become a robust separation method. IMS has been widely employed in a variety of research fields ranging from environmental to pharmaceutical applications [109][110][111][112][113][114]. The use of ion mobility has gained significance in bioanalysis due to the potential improvement of the sensitivity and the ability of the technique to distinguish highly related molecules based on conformational differences of molecules [115]. The IMS-derived collision cross-section indicates the effective area for the interaction between a particular ion and gas through which it travels [116]. Initially, IMS was utilized largely as a stand-alone technique; however, in recent years, the IMS coupling with MS (IMS-MS) has developed rapidly into a robust and extensively used separation technique with applications in many fields across the biological sciences, including the glycosciences [117]. IMS-MS developed quickly into a ready-to-use technique that became commercially accessible, particularly for glycan analysis [118]. The biological applications of IMS-MS for biomolecules include the analysis of oligonucleotides carbohydrates, steroid, lipids, peptides, and proteins [119][120][121][122][123]. Furthermore, IMS-MS may be hyphenated with front-end liquid chromatography (LC) separation to increase peak capacity and separation capabilities [123]. LC-IMS-MS technique has numerous significant benefits over other technologies in terms of increased peak capacity, isomer separation, and metabolite identification [123,124].
IMS-MS derived collision cross-section (CCS) value is high reproducible characteristic of metabolite ion, allowing for metabolite identification [125]. Therefore, the most essential aspect of metabolite identification in IMS-MS is the curation of the CCS database. Many in silico CCS databases, such as LipidCCS [126], MetCCS [127], and ISiCLE [128], have been curated and include over one million CCS values. Zhou et al. [129] developed the ion mobility new CCS atlas, namely, AllCCS for metabolite annotation using known or unknown chemical structures [129]. The AllCCS atlas included a wide range of chemical structures with >5000 experimental CCS records and~12 million predicted CCS values for >1.6 million chemical molecules [129]. McCullagh et al. [130] used the TW CCSN 2 library to screen the steviol glycosides in 55 food commodities. Schroeder et al. [131] identified 146 plant natural compounds, 343 CCS values, and 29 isomers annotated (various flavonoids and isoflavonoids) in Medicago truncatula based on CCS, retention time, accurate mass, and molecular formula. The combination of a large-scale CCS database and different MS/MS spectra will assist in the discovery of new metabolites.

Nuclear Magnetic Resonance (NMR)
NMR is another popular analytical tool for investigating the varied metabolome in plants, involving the structure, content, and purity of molecules in the sample. As a result, metabolic profiling can provide qualitative and quantitative data from biological extracts [132]. The basic principle of NMR-based metabolite identification is to capture the radio frequency electromagnetic radiations emitted by atomic nuclei that have an odd atomic number ( 1 H) or an odd mass number ( 13 C) when placed in a strong magnetic field. Because there is no requirement for chromatographic separation or sample derivatization, the use of NMR has grown dramatically in recent years [94,133,134]. Furthermore, easy sample preparation procedures and excellent repeatability, non-destructive nature enables high throughput and quick analysis in NMR metabolomics but has less sensitivity than MS [135,136]. NMR is pH sensitive, buffered solutions are usually needed to keep the pH stable. A combination of methanol and aqueous phosphate buffer (pH 6.0, 1:1 v/v) or ionic liquids such as 1-butyl-3-methylimidazolium chloride has been shown to be the most effective in providing a comprehensive overview of both primary and secondary metabolites [137]. 1 H NMR is quick and easy, it has been the leading metabolites profiling technique, but it suffers from signal overlapping in the complex mixture of plant extracts during metabolites profiling. However, other advanced 2D NMR-based techniques includes two-dimensional (2D) 1 H J-resolved NMR, heteronuclear single quantum coherence spectroscopy (HSQC), heteronuclearmultiple quantum coherence (HMBC), total correlation spectroscopy (TOCSY) and nuclear overhauser effect spectroscopy (NOESY) [137]. High-resolution magic angle spinning (HRMAS)-NMR is particularly well suited for solid lyophilized tissue without the need for chemical extraction, which is essential for both MS and liquid state NMR practices [86]. The acquisition time for 2D NMR (2D J-resolved spectroscopy) is around 20 min, whereas for one-dimensional (1D) NMR it is approximately 1 min. However, due to the dispersion of the resonance peaks in a second dimension, spectral overlapping can be reduced in 2D NMR J-resolved spectroscopy to detect signals in crowded spectral regions [138]. Using advanced NMR, glycine-betaine, citric acid, trehalose and ethanol levels were higher in Cry1Ab gene transformed maize plants than non-transgenic maize plants showed [55]. Transgenic maize plants showed lower levels of pyruvic, isobutyric, succinic, lactic, and fumaric acids than non-transgenics [55]. During seed germination in chickpea, the exogenous uptake of glucose in presence of nitric oxide donor was quantified by using 1 H-NMR [33].

Metabolomic Data Processing, Annotation, Database and Bioinformatics Tools for Plants METABOLOME Analysis
GC-MS, LC-MS CE-MS, FTICR-MS MALDI, IMS and NMR are perhaps the most important techniques within the context of natural product discovery. Metabolomics generate a huge amount of metabolic data using wide range of analytical instruments. During the last decade, different software tools (web-based programs) have been designed for metabolomics raw data processing, data mining, data assessment, data interpretation, and statistical analysis as well as mathematical modelling of metabolomic networks (Figure 1).

Data Processing and Annotation
Several software programs are available for in silico data analysis of a large quantity of spectrum data of metabolites generated by various analytical instruments. The webbased programs were used for raw data processing, mining, and integration of metabolites. In general, acquired data is processed for the correction of baseline shifts, background noise reduction, peak detection and alignment, and finally, deconvolution of mass spectra ( Figure 1, Table 4). Many bioinformatic tools are designed for the data pre-processing, including XCMS (https://xcmsonline.scripps.edu, accessed on 29 June 2021), METLIN (http://metlin.scripps.edu, accessed on 29 June 2021) AMDIS (Automated Mass Spectral Deconvolution and Identification System), MeltDB, MetaboAnalys, MetAlign, MZmine 2, and AnalyzerPro for different analytical techniques (Table 1). XCMS is an online bioinformatics platform that facilitates the direct uploading of raw data and assists the user in data processing and statistical analysis [139]. For LC-MS experiments, XCMS has been developed for programmed data transfer that has reduced data processing time and improved the effectiveness of an online system [140]. METLIN is another online database, which has been used in various studies related to plant metabolic profiling of stress response. It is useful for plant metabolic profiling of specific metabolites, and it is not time-consuming for data processing, mining, and annotation [141].
MeltDB (https://meltdb.cebitec.uni-bielefeld.de, accessed on 29 June 2021) is an important web-based platform used for data assessment, processing, and statistical analysis in plant metabolomics [142]. In addition, MetaboAnalyst online platform also includes a flexible enrichment analysis tool including some topological and visualization possibilities [143]. Global natural product social molecular networking (GNPS; http://gnps.ucsd.edu, accessed on 29 June 2021) is web-based mass spectrometry (MS/MS) for processing and annotation of metabolites [144]. GNPS assists with the identification and discovery of metabolites throughout the data, from data acquisition/analysis to post-publication [144]. Finally, the MZmine 2 is a publicly accessible data processing module that supports highresolution spectral analysis. MZmine 2 is suitable for both targeted and non-targeted metabolomic studies, and it is well suited for processing large batches of data [145]. Various computational web-based, statistical and online bioinformatics tools are commonly used for data analysis in plant metabolomics (Table 4).    [139] CE-MS, capillary electrophoresis-mass spectrometry; DTIM-MS, drift tube ion mobility-mass spectrometry; EI, electrospray ionization; FTICR-MS, fourier transform ion cyclotron resonance-mass spectrometry; GC-TOF-MS, gas chromatography-time of flight-mass spectrometry; GC-MS, gas chromatography-mass spectrometry; HPLC, high-performance liquid chromatography; LC-MS, liquid chromatographymass spectrometry; MALDI-TOF, matrix-assisted laser desorption/ionization-time of flight; TWIM-MS, traveling wave ion mobilitymass spectrometry.

Conclusions
Metabolomics has achieved a prominent role in plant science research. It has wide applications ranging from investigating the stress-specific metabolites for different climatic stresses, evaluating candidate metabolic gene functions to analyzing the biological mechanism in plant cells, and dissecting the genotype-phenotype relationship in response to the various biotic and abiotic stresses. This review provides an overview of different sample collection, harvesting methods, storage, and sample preparation in the plant metabolomics experiments. Furthermore, the most widely used analytical tools in metabolomics for agriculture research viz. GC-MS, LC-MS, CE-MS, FTICR-MS, MALDI, IMS, and NMR with new development in their applications. In addition, we discussed computational software and database employed for metabolomics data processing in plant science. The integration of comprehensive bioinformatics tools with omics strategies professionally dissects novel metabolic networks for crop improvement. Metabolomics has excelled classical approach for novel metabolites discovery and simultaneously explores the complexity and enormous chemical diversity of metabolites in any crop plant. The integration of metabolomics with other "omics" technologies, e.g., genomics, transcriptomics, proteomics, can deliver novel insights into crop plants' genetic regulations in the context of their cellular function and metabolic network. The complete elucidation of physio-biochemical and molecular mechanisms underlying plant developmental and stress-responsive biology primarily depends on the comprehensive investigations using omics techniques that make metabolomics more applicable in agriculture sciences. Metabolomics has tremendous potential in plant research, as metabolites are more appropriate to the plant phenotype than DNAs, RNAs, or proteins. Therefore, studies in this field will effort on both ways, one is the systematic study of the biochemical and genetic mechanisms of metabolic variations in crop plants using both targeted and non-targeted methods; other is metabolomic platform can be used for metabolic profiling of genome-edited plants using CRISPR/Cas9 system for risk evaluation and regulatory affairs related with genetically modified crops [196]. Thus, we can say metabolomics will be able to contribute a lot to agriculture science, such as crop breeding and genome editing for crop improvement, better grain yield, and elucidating their unknown and novel metabolic pathways.

Conflicts of Interest:
The authors declare no conflict of interest.