Notes on Plant Metabolomics

Read this article to learn about Plant Metabolomics. After reading this article you will learn about: 1. Meaning of Plant Metabolomics 2. Metabolomics Technologies and Appli­cations 3. Mass Spectrometry 4. Nuclear Magnetic Spectrophotometry 5. Fourier Transform-Infrared Spectrosco­py (FT-IR) 6. Gas Chromatography in Metabolomic Study and other details.

Meaning of Plant Metabolomics:

Metabolomics aims to quantify and identify all metabolites in a given biological context an aim that admittedly is as challenging as full-scale proteomics.

It is thus a manifestation of the endpoint of metabolic and physiological processes. Taken individually, measurements of the metabolome in different physiological conditions are likely to be more indicative for the purposes of systems biolo­gy studies than transcriptome profiling’s.

The metabolome is a complete set of metab­olites in a cell or tissue consists of low-molecular weight chem­ical intermediates which are the end products of gene expression.

It is understood that any change in gene (protein) expression levels will have only small effects on metabolic fluxes, they must have large effects on metabolite concentra­tions.

Metabolomic approaches seek to profile me­tabolites in a non-targeted way, i.e. to effectively separate and detect as many metabolites as possi­ble in a single analysis.

Eventually, the informa­tion on transcript and protein abundance was com­bined and this would ideally lead to a nearly com­plete molecular picture of the state of a particular biological system at a given time. Plant metabo­lomics is essentially a comprehensive, high-throughput analyses of complex metabolite mix­tures typical of plant extracts.

To achieve the overview of metabolic com­position it is essential to establish a multifaceted, fully integrated strategy for optimal sample extrac­tion, metabolite separation, detection, identifica­tion, automated data gathering systems, handling and analysis system and finally quantification.

Both analytical and computational developments are essential to achieve this aspect. And moreover these data also require proper storage in suitably configured databases.

It is very apparent that large quantities of me­tabolomic or metabolic profiling data if generat­ed will help to open up many previously inacces­sible areas of plant drug research. The objective of metabolomics targets at assessing metabolic changes in a comprehensive and global manner in order to check the biological functions or pro­vide detailed biochemical responses of cellular systems.

The Metabolomics Society (2005) is to pro­mote the international growth and development of the field of metabolomics, to provide the op­portunity for collaboration among researchers in metabolomics, including connections between academia, government and industry, regulators and vendors in the field of metabolomics, and to provide opportunities for dissemination of research achievements in workshops, conferences, and journal publications.

The First International Congress on Plant Me­tabolomics was held in Wageningen, The Neth­erlands, in April 2002, with the primary goal of bringing together those scientists who are active­ly involved in this field and those who soon plan to be in future.

In such instances, opportunities are created for collaboration and joint strategies can be determined to meet the metabolomics chal­lenge. Selected abstracts from the oral and poster presentations at the meeting are now accessible at www.metabolomics.nl and this site will con­tinue to be used as an aid for information exchange and collaboration.

Although microbes prove to be the richest overall source of metabolites, plants are the source of the most complex individual mixtures.

Mariet van der Werf (TNOFood, Zeist, The Netherlands) reported that it has been predicted that bacterial genomes, which were already been sequenced can support the biosynthesis of just a few hundred metabolites (e.g. 580 for Bacillus subtilis and 800 for Escherichia coli), but for individual plants, this value is likely to be in the tens of thousands.

Approximately 50,000 different phytoconstituents have been elucidated in plants and it is pre­dicted that the final figure for the plant kingdom will approach or even exceed 200,000 (Fiehn, 2001) Thus, metabolomics exerts a considerable challenge for plant scientists.

Metabolomics information not only will as­sist in the deeper insights in understanding of the complex interactive nature of plant metabolic networks and their responses to environmental and genetic change but also will provide information regarding the fundamental nature of plant phenotypes in relation to development, physiology, tis­sue identity, resistance, biodiversity, etc.

Fur­thermore, the coupling of electrospray ionization (ESI) MS with CE and hydrophilic interaction chromatography has been successfully used to anal­yse metabolomics problems.

Metabolome is a complex mixture of chemical compounds with broadly differing chemical and physical proper­ties which presents limitations to all analytical technologies.

Instrumentation utilized in measure­ment of the metabolome includes mass spectrom­etry, Fourier transform infra-red spectros­copy nuclear magnetic resonance spectroscopy (NMR).

A num­ber of analytical strategies are employed ranging from the study of a few metabolites com­mon to a specific metabolic pathway or chemical class such as amino acids to the comprehensive analysis of hundreds of chemically diverse me­tabolites (metabolite profiling, metabolic finger , printing, metabolomics, metabolomics).

Currently no single technique can provide non-biased de­tection of all compounds present in a metabolome and in future applications it is expected that a number of different technologies will be employed to fulfil this objective.

Metabolite profiling involves separation of metabolites, using gas chromatog­raphy, liquid chromatography or capillary electrophoresis, prior to detection of individual compounds and with metabolite identification by comparison to already available mass spectral/retention index libraries.

However, for large studies sample preparation and analysis are both time consuming and laborious. Typical runtimes of 10-20 min gives detection of more than 900 metabolite peaks. To overcome problems of long runtimes when analysing large sample sets metabolic fingerprinting can be employed.

This involves analysis at the expense of lower sensitiv­ity and a reduced ability to identify metabolites. The objective is to provide high-throughput (1-3 min analysis time) analyses with the objective of screening samples to enable discrimination be­tween samples of differing biological origin or sta­tus.

Plant metabolomics aims at providing com­prehensive information that could help to optimize metabolic engineering strategies by providing more information about biosynthetic pathways and the interactions between them.

Metabolomics is the study of all the metabolites of a given biological sample but since no single technique provides such comprehensive detection, numerous analytical tools such as gas chromatography (GC), liquid chromatography (LC), mass spectrometry (MS) and nuclear mag­netic resonance (NMR) have been used to investi­gate the metabolome.

Notable advances have been made with single techniques such as GC/MS. Studies that involve multiple techniques such as NMR, LC/UV and LC/ MS have also appeared.

Estimating and monitoring the metabolome are perhaps two of the most important analytical challenges of the twenty-first century. While meth­odologies for genomic and proteomic analyses have stabilized into several well-characterized ap­proaches, analytical methods for metabolomic analysis remain in a considerable flux.

The prob­lem rests with the entire diversity of the metabo­lome. Estimates for the number of metabolites in the human and plant metabolomics range from 2,766 to about 200,000.

The polarity of these metabolites varies from nonpolar lipids to extremely polar inorganic ions. Functional groups occurs in metabolites include alcohols, amines, organic acids, aromatic alco­hols, esters, aldehydes, ethers, phosphates and many others.

Some metabolic sub-groups are so large they have their own classification such as the lipidome and the glycome. Also, because the metabolome encompasses a relatively narrow band of molecular weights numerous isomers ex­ist.

Compound chirality is also a concern and a challenge to analysis. In addition to the chemical complexity of the metabolome, physiological con­centrations of metabolites in the plant tissues range between millimolar to picomolar levels.

Metabolomics Technologies and Appli­cations:

The development of atomic theory at the begin­ning of the twentieth century served as a platform for the early forms of spectrometry.

The construc­tion of the first mass spectrometer by J.J. Thom­son in 1912 was soon followed by a series of tech­nological approaches, leading to the first applica­tion of ICR-MS by 1949 and the first spectrome­ters coupled with gas chromatographs was per­formed in the late 1950s.

Although the phenomenon of NMR was not described until the mid-1940s, it was not long before this property was being used for the structural elucidation and determination of molecules. The application of these analytical methods to metabolomics is more recent, and has been accompanied by rapid and extensive im­provements in data-processing technologies.

Mass Spectrometry in Metabolomics Study:

Instrumental methods in different fields (especial­ly plants and microbes versus animals) varies, largely for historical reasons, but there is increas­ing convergence to use as many as possible for all samples due to their complementarity.

As well as increasingly refined gas chromatography-mass spectrometry (GC-MS) methods, especially those using gas chromatography time-of-flight mass spectrometry (GCTOF) instruments that allow much better de-convolution than by most GC-MS instruments because they can record spectra, and thus sample, very quickly.

Fourier transform ion cyclotron resonance (FT-1CR) mass spectrometry is a very high-resolution mass spectral method (10⁵-10⁶), with mass accuracy better than 1 ppm), which allows separation and empirical formula measurement of potentially thousands of metabolites.

The potentially holistic approach to metabo­lome analysis is done primarily by recent advanc­es in mass spectrometry (MS) technology and by the efforts of functional genomics research. Most technology for metabolomics is based on MS. Gas chromatography- MS (GC-MS) and HPLC-photodiode array-MS remain the methods of choice for quantitative and qualitative metabolite profil­ing.

The ultimate goal of metabolomics— the abil­ity to efficiently detect and quantify every metab­olite in a plant extract—is difficult to be attained by any single analytical method available at present. Some metabolite selection occurs in all methodologies, from initial different solvent ex­traction process through chromatography to MS ionization.

Oliver Fiehn report­ed the use of rapid scanning time-of-flight (TOF) MS coupled with GC separa­tion and also integrated with peak de-convolution software.

This technique increased the number of metabolites detectable by GC-MS in crude plant extracts to 500 to 1000.

Direct infusion of extracts into MS instruments using “soft” electrospray or atmospheric pressure chemical ionization sources is an attractive way to obtain “fingerprints” of the molecular ions of the metabolites present in plant extracts. Such use of FT-MS was demonstrated in presentations by Asaph Aharoni and Dayan Goodenowe (Phenomenome Discoveries, Saskatoon, Canada) in the same conference.

This powerful, and relatively expensive, mass analyser has the ability to gener­ate mass data of sufficient accuracy for the deter­mination definitive empirical formula. Several hundred ions were observed in this for each ion­ization method in both positive and negative modes.

Metabolomic analysis is supposed to avoid bias against certain compound classes and to al­low for the analysis of every metabolite individu­ally. Although this aim has not been reached but clear progress in this direction was reported.

For example, a combination of different metabolite- profiling tools was used by Ric de Vos for the analysis of metabolic effects in high-flavonoid genetically modified tomatoes.

Using LC/photodiode array detection as well as LC-QTOF-MS and direct infusion- QTOF-MS, it was found that flavonoid contents were up to 70- fold more when compared with those of common cultivars, and glycosidic structures were present­ed for all aglycone intermediates that accumulat­ed.

These analyses were performed in tandem with GC-MS profiling of volatile compounds us­ing headspace solid-phase micro extraction. In this, in addition to an increase in flavonoids, the methyl-salicylate a volatile compound was found to be increased in transgenic plants over expressing the Lc/C1 transcription factor.

Sumner, in the same conference provided specific examples, includ­ing LCUV- MS of flavonoids and saponins, GCMS of polar and lipophilic extracts, and capillary electrophoresis-MS of anions and amino acids.

Nuclear Magnetic Spectrophotometry in Metabolomic Probing:

In addition to MS, NMR also plays an important role in recording the metabolic profiling data’s. For the more concentrated metabolites, nuclear mag­netic resonance (NMR) has provided both quali­tative and quantitative information however; NMR is not sufficiently sensitive for use with metabo­lites at low concentrations.

Marianne Defernez (Institute of Food Re­search, Norwich; Mike Beale, Institute of Arable Crops Research-Long Ashton; Nigel Bailey, Im­perial College, London) and Adrian Charlton (Cen­tral Science Laboratory, York) presented the use of proton NMR of crude plant extracts, followed by multivariate analysis, to cluster data sets to high­light differences.

This method of analysis gives a comprehensive summation fingerprint of all (hy­drogen-containing) metabolites extracted and can give direct structural information regarding indi­vidual metabolites in the mixture, particularly when two-dimensional techniques are used.

Therefore, it is suitable for high-throughput, rap­id, first-pass screening. Subtraction of these data sets generates virtual NMR spectra and hence important structural data on compound(s) contrib­uting to differences between samples.

Fourier Transform-Infrared Spectrosco­py (FT-IR) in Metabolomic Study:

Fourier transform-infrared spectroscopy (FT-IR) has also been used to spectroscopically analyse the metabolome. This method detects even small phenotypic differences, which cannot be measured by oth­er methods.

Because FT-IR is chemically nonse­lective, it is very much suited for the rapid finger­printing of a metabolome rather than the identifi­cation of specific metabolites.

Gas Chromatography in Metabolomic Study:

At lower concentrations in metabolomic study, sequential approaches of concentration, separa­tion, and detection methods are commonly em­ployed.

Those metabolites which have vapor pres­sures or can be converted into volatile products can be separated by gas chromatography (GC). The major disadvantage of GC for metabolic pro­filing is that non-volatile compounds cannot be probed unless time-consuming sample derivatization is employed.

HPLC in Metabolomic Study:

Separation of non-volatile metabolites, which make up most of the metabolome concentration, can be achieved by liquid chromatographic tech­niques. Reverse phase high performance liquid chromatography (HPLC) has been used for the separation of non-volatile metabolites at low con­centrations and varying polarities.

HPLC, how­ever, is not ideal for a rapid screening process because of its low resolution and slow analysis time as compared to gas phase separations.

Recently ultra-performance liquid chromatography (UPLC) has been intro­duced for metabolic probing. Due to the small diameter (2 µm) of the packing material in UPLC and the novel silicoethylboded phase, linear mobile phase velocities of up to 10 mm/s can be achieved without much change in resolution.

Capillary Electrophoresis (CE) in Metab­olomic Profiling:

Capillary electrophoresis (CE) can also be used in metabolomic profiling especially to separate cat­ions and anions in the metabolome but is slow, in some cases requiring up to 16 h for a complete analysis.

After chromatograph­ic or electrophoresis separation, identification of metabolites is normally achieved by mass spec­trometry. Time-of-flight, quadrupole, and ion trap mass spectrometers have also been used after chro­matographic separation to identify metabolites.

Electrospray lonisation Time-of-Flight Mass Spectrometry (ESI-TOF-MS) in Metabolomic Studies:

Electrospray lonisation Time-of-Flight Mass Spec­trometry (ESI-TOF-MS) can be employed in me­tabolomic studies and is generally referred to as Direct Injection (or infusion) Mass Spectrometry (DIMS). Electrospray ionisation is applied to me­tabolites containing polar moieties in their mo­lecular structure.

lonisation provides minimal frag­mentation of molecular ion and a less complex mass spectrum when compared to the mass spec­tra of electron impact ionisation of multi-compo­nent samples. Although the techniques of mass spectra are still complex the presence of molecu­lar ions enhances the ability to identify metabo­lites. Therefore chromatographic separation of complex samples is not required.

Commercially available Time-of-Flight (TOF) in­struments can provide a greatly improved mass resolution (FWHM = 5000+ at mass 500 Da) com­pared to additional quadrupole instruments. Plant metabolomics studies have also been useful for the characterization of olive oils and of vegetable oils used in its adulteration.

Also separation and characteriza­tion of medicinal plant extracts by detection of a wide range of secondary metabolites have been reported, with subsequent metabolite identification by tandem mass spec­trometry.

It is now known that Fourier Transform Ion Cyclotron Resonance Mass Spectrometry (FT- ICR-MS) provides many advantages over other mass spectrometers in terms of sensitivity, resolv­ing power and metabolite identification.

A metabolome analysis of strawberry fruits and tobacco flowers has been conducted and showed its potential by identification of metabo­lites differing between developmental stages of strawberry fruit.

Experimental and instrumental parameters highly influence the mass spectra of biological samples. Reliable metabolic fingerprints therefore require reliable standardisation and verification.

Thus high standards now employed for analysis of microarray experiments (Minimum Information about a Microarray Experiment – MIAME) required for publication in peer-reviewed journals should be established for metabolomics.

Hyphenation Techniques of Chromatog­raphy and Spectroscopy in Metabolite Profiling:

Through the years many analytical methods have been developed that are not restricted to the se­lective analysis of one or a few compounds. With the improved performance of chromatography methods in the late 1960s and early 1970s (im­provements in reliability, selectivity and resolu­tion), peak identification and detection in com­plex matrices were made possible based purely on retention times.

These separation techniques were then coupled to highly sensitive detectors that also had high effective ranges of quantifica­tion, such as flame ionization coupled to gas chromatography (GC/FID) or fluorescence and UV detectors to liquid chromatography (LC).

These analytical methods were soon applied to plant tis­sues to profile important compound classes such as amino acids. Also in the late 1960s, mass spectrometry (MS) was improved to make it as universal as flame ionization, in addi­tion of offering a completely independent method for compound identification and classification when coupled to gas chromatography.

Halket et al. (1999) utilized mass spectral de-convolution software to increase reli­ability of metabolite detections. For the first time, peak identification of 68 pre-selected target com­pounds was based on both purified mass spectra and retention time indices in an automated and robust way.

In the plant field, less effort has been put into profiling compound classes. Sauter et al. (1991) chose peaks that apparently detect major com­pounds in GC/MS chromatograms in order to get an overview of major steps in metabolism before and after pesticide spraying.

Derivatization con­ditions for metabolite profiling were optimized by selecting 12 compounds representing plant pri­mary metabolism (Adams et al., 1999), and were applied to profile polar organics (sugars, polyols, acids and amino acids) in apricots.

Another step in elucidating metabolism was conducted by Christensen and Nielsen (2000), who used GC/MS to profile the fractional enrich­ment of 13C-labelled substrates in order to study biochemical pathways for production of second­ary metabolites.

Metabolite profiling was extend­ed to cellular by combining GC/MS, HPLC analy­sis of nucleotides, enzyme assays and pyrophos­phate target analysis solvents fractionation.

Such compartmental analysis is clear­ly needed for understanding plant metabolism. Using LC/MS it was possible to detect and identi­fy more secondary metabolites from plants than by NMR. These were semi-quantified by LC/MS for sample to sample variation.

GC-MS/ LC-MS:

In general GC-MS is effective for the profiling of relatively low molecular weight, hydrophobic compounds such as essential oils, hydrocarbons, esters and metabolite derivatives with less polari­ty.

LC-MS equipped with atmospheric pressure chemical ionization (APCI) is useful for the profil­ing of polar and low to moderate molecular weight (i.e. MW\2000) metabolites, those of sterols, fat­ty, organic and amino acids, metal ions, and al­kaloids.

Polar molecules within a molecular weight range (i.e. from 10 to 300,000), including proteins and DNA, can be probed by electrospray ionization (ESI)-LC-MS. An alternative ionization method for the analysis of high molecular weight molecules and other compounds like carotenoid is matrix-assisted laser desorption/ionization (MAL- Dl) coupled with time-off light (TOF) MS.

GC-MS, which has been enumerated as the ‘gold standard’ of metabolomics, involves the separation of volatile, thermally stable compounds by GC and subsequent detection by electron ionization (El) MS. An attractive feature of GC-MS is the direct measurement of volatile compounds.

The situa­tion for LC-MS is more complex since the pres­ence of acidic groups or electronegative atoms, such oxygen and nitrogen, can facilitate the for­mation of negative ions. However, many com­pounds in plant extracts will not fully ionize, if at all, under the conditions used for a single analy­sis.

For this reason, LC-MS is better suited to the targeted profiling of specific types of compounds that shows similar ionizing behavior (e.g. alkaloids) and is generally not useful for broad-scope me­tabolomics studies. However, the combination of LC-MS and GC-MS data certainly provides a more complete picture of a plant metabolome.

Metabolomic Sample Preparation:

When attempting for simultaneous detection of all metabolites in plant tissues, the applied methods cannot be restricted to the technical question which type of data acquisition might be most suit­able but also must seriously consider adequate methods for sample preparation.

As an initial step for such methods, plant physiologists were long been aware of the importance of readily stopping the inherent enzymatic activity of biological sam­ples. This has been achieved by freeze clamping, immediate freezing in liquid nitrogen, or by acid­ic treatments using per chloric or nitric acid.

Although advantageous for extraction of amines acidic treatments renders severe problems for many analytical methods that follow, and harvest­ing and keeping plant tissues into tubes for liquid nitrogen freezing may take up to 15 s.

Compared to freeze clamping techniques, freezing samples in liquid nitrogen is a slower process that could potentially produce artefacts caused by patholog­ical wound responses, rapid activation of touch- inducible genes, etc.

This might occur in case if samples are weighed before freezing, and if no attention is paid to this special problem. Since freeze clamping is not easily applicable in cases were lots of samples have to be harvested, freez­ing in liquid nitrogen remains a reasonable way to stop enzymatic activity.

Certainly, great care must be also taken to avoid incomplete thawing tissues before extract­ing metabolites. One way to do this is to freeze-dry biological samples resulting in completely dried samples.

In the absence of water in the sam­ples, enzymes or transporters are inactive. If stored before sample extraction, samples have to be maintained in dry environments like evacuated desiccators, since tissues are highly hygroscopic.

Alternatively, frozen tissues can be directly ex­tracted by adding organic solvents and applying heat, thereby also inhibiting the enzymatic activ­ity.

Extracting frozen tissues that still contain the original amount of water in them might be advan­tageous for metabolomic studies when compared to extracting freeze-dried samples, since freeze- drying may cause irreversible adsorption of me­tabolites on cell walls and membranes.

If metab­olomic analysis sets out to distinguish between metabolite levels in different compartments, sam­ples need to be freeze-dried prior to non-aqueous fractionation methods.

An alternative method to non­-aqueous fractionation is the use of nuclear mag­netic resonance analysis (NMR) to distinguish steady-state concentrations of metabolites in dif­ferent compartments in vivo.

Such approaches will become increasingly important as it is recognized that gene function cannot be assigned without understanding the essential role of biochemical pathway and without the help of the plant physiology.

Prior to sample extraction, different types of sample homogenization is done, depending on the number of samples to be treated, and on the type of tissue. Leaf tissues may be ground under liquid nitrogen using mortar and pestle, or using a ball mill with pre-chilled holders, or together with the extraction solvent by ultraturrax devices.

Other plant organs such as roots, sometimes to be too hard to use ball mills, whereas potato tubers are too soft for homogenization. After ho­mogenization, different methods of metabolite extraction can be used but, again, no systematic study is available that directly compares the re­sults of these techniques.

Most frequently, polar organic solvents like ethanol, methanol, methanol-water mixtures are directly added to freshly frozen tissues, with an additional step of using non-polar solvents such as chloroform to exhaus­tively extract lipophilic components.

In order to increase the extraction efficiency, additional en­ergy is put into the system either directly by heat (e.g. 70°C), or by other techniques such as pres­surized liquid extraction, supercritical fluid extraction, sonication, sub- critical water extraction, microwave techniques, or pervaporation.

Rarely data are available that compare these techniques and few systematic studies have been carried out on the occurrence of possible metabolite breakdown reactions caused by oxidation. It is similar for sample storage, although it is generally assumed that alterations of metabo­lite content can be excluded during storage at -80°C.

Tentative Metabolite Probe:

In order to provide preliminary identification for metabolites, a list of over 700 metabolites with their metabolic pathways are known to be present in plant and microbial metabolomes, was com­piled which were available from several data­bases including the KEGG database (www(dot)genome(dot)ad(dot)jp/kegg/kegg2(dot)html).

The monoisotopic masses of molecular ions are present in plant extracts as [M+H]+, [(M)HJ, [M+NH4]+, [M+Na]+ or [M+K]+ ions and were calculated and tabulated by metabolite name vs. monoisotopic mass for each molecular ion; for example lactic acid, [M+HJ+, 91.0395.

The metabolite identifi­cation table is available on request from various sources. Metabolites were identified by search­ing, the “lookup” function in Microsoft Excel, of determined accurate masses against monoisotopic masses of known metabolites. W.B. Dunn et al./ conducted and recorded Automated electrospray- TOF mass spectrometry for metabolic fingerprint­ing of the plant metabolome.

Data Processing:

Metabolite target analysis is a combination of tech­niques to prepare and analyse samples for one or a small number of compounds from a complex mixture.

Often, the sample preparation techniques aim at pre-concentration and purification of the metabolite under study, before analysing it with a hyphenation of chromatography and a selective detector (such as liquid chromatography coupled to fluorescence detection, or gas chromatography coupled to sulfur chemiluminescence detector).

Metabolite target analysis is clearly the most wide­ly used technique, and it is applied in all areas of plant research such as phytohormone analysis.

Data Analysis:

Irrespective of the analytical technique used, the analysis of the data is usually performed in three stages. Firstly the raw data need to be preprocessed to convert them to a suitable form.

Secondly these modified data’s may be subjected to data reduc­tion so that only the most relevant input variables are used in the subsequent data analysis. Thus normalizing to a con­stant total signal introduces dependencies between the variables that would not exist without this step.

Similarly the missing variables can have signifi­cant effects on the position of individual samples in clustering diagrams. Missing values may arise because they are below the limit of detection or because they were not collected.

De-convolution and further processing of hyphenated data to es­tablish the contribution of each eluting compo­nent is a very difficult, active and vital area, which needs to begin by “registering” or aligning datasets.

Automating this step reliably is a high priority for metabolomics. The objective of the third stage of the data analysis is to record patterns within the data which give useful biological information that may be useful to generate hypotheses that can be further tested and refined.

The methods available for metabolome analysis can be placed in four main (partly overlapping) categories – univariate and multivariate statistical, unsupervised learning (which looks at the overall pattern or structure of the data), supervised learning (which uses known information to help to study the classification of the data, and system-based analyses which use theories such as MCA (Fell, 1996) to help interpret the data in terms of the biological networks.

Many unsupervised learning methods are equiva­lent to clumping methods and are often statisti­cally based, while supervised methods are from many varieties, including statisti­cal, neural, rule-based, evolutionary and so on.

Conclusion:

The combination of analytical results from all lev­els of gene products (transcriptome, proteome, and metabolome) remains more vision than reality. Metabolomics is a rapidly developing area in plant biology and has become an integral part of many functional genomics programs.

Metabolomic pro­filing of plant is helpful not only in identification and quantification of the various metabolites present in the plant tissues but also helps in un­derstanding the biosynthetic pathways in the pro­duction of these metabolites and also their re­sponse to genetic and environmental change.

The plant metabolome is large and diverse, and al­though the technologies reviewed here are suffi­cient to generate extensive datasets, the ability to identify individual metabolites remains a general limitation. Even though metabolomic analysis is comparably fast and cheap, reliable and precise, the unambiguous and simultaneous identification of all metabolites in a biological system is a chal­lenge.

Finally, if metabolomic profiling is to be used to its fullest, it is imperative that publicly available metabolomic databases be created. Metabolomic data are rich in information, and there is considerable interest in re-assessing pre­viously acquired data under different perspectives. 

Through this progress, the responses of plants to genetic manipulations and environmental pertur­bations will become increasingly predictable. This will make systems biology attractive as a tool for the creation of hypotheses.