Plant metabolomics coming of age

Abstract

The metabolome is defined as the total small-molecule complement of a cell, and metabolomics is therefore the study of all of an organism’s low-molecular-weight molecules or metabolites (Oliver et al. 1998, Tweeddale et al. 1998). The closely allied field of metabonomics is defined as ‘the quantitative measurement of the dynamic multiparametric metabolic response of living systems to pathophysiological stimuli or genetic modification’ (Nicholson et al. 1999, 2002). Interest in metabolomics and/or metabolite profiling in plant biology continues to grow following the development and use of non-targeted multivariate approaches to analyze plant metabolite composition (Roessner et al. 2000). To illustrate the growth of interest in the application of metabolomic approaches in plant biology, a PubMed search was conducted on 14 October 2007 using the Boolean search string of [plant AND metabolomics OR metabolome OR metabolite profiling OR metabolite profile]. A total of 994 documents were retrieved by the search, and 343 were deemed to have a focus on or some facet relating to plants. A breakdown of articles by year shows a dramatic increase in reports on plant metabolomics and/or metabolite profiling since 2000 (Fig. 1). Using this particular search string, only one paper prior to 2000 was retrieved (Graham 1991). Two papers published in 2000 by Roessner et al. (2000) and by Fiehn et al. (2000) are instrumental in demonstrating the importance of metabolite profiling for plant functional genetics; they would be cited 192 and 382 times, respectively. In 2001, papers by Raamsdonk et al. (2001) and Roessner et al. (2001) illustrate the ability of metabolite profiling to reveal biochemical phenotypes for mutant plants with no visually obvious phenotypes and for plants growing under different environmental conditions. These papers would garner 310 and 218 citations, respectively. By early 2002, Fiehn (2002) published a review on the linkage between genotypes and phenotypes in plant metabolomics. This paper would be cited 316 times. Later in 2002 and 2003, Hall et al. (2002) and Sumner et al. (2003) described how important metabolomics would become in plant functional genomics. These papers would be cited 210 and 196 times, respectively. Thus, by 2003, the foundation that plant metabolomics would prove to be quite valuable in the advancement and acceleration of plant biology studies for a post-genomics era was well established (Sumner et al. 2003). From this point on, metabolite profiling and metabolomic approaches would become widely used in plant biology (Fig. 1). Plant-related published journal articles by year of publication as retrieved from PubMed using the Boolean search string [plant AND metabolomics OR metabolome OR metabolite profiling OR metabolite profile] on 14 October 2007. The value for 2007 is a projection of the number of articles for the entire year based on the number of published articles (73) as of 14 October 2007. Plant metabolomics and metabolite profiling approaches are being employed in an ever widening array of uses, from efforts to better understand secondary metabolism (Goossens et al. 2003) to making quality assessments of green tea (Le Gall et al. 2004) to discriminating taxonomic relationships (Frederich et al. 2004) and plant natural products (Pauli et al. 2005), in revealing metabolic features of source to sink transitions (Jeong et al. 2004), illuminating genotypic differences controlling tomato fruit volatiles (Tikunov et al. 2005), uncovering regulators of secondary cell wall biosynthesis (Andersson-Gunnerås et al. 2006), linking metabolite factors with desiccation tolerance (Avelange-Macherel et al. 2006) or measuring grape flavonol and anthocyanin content in food chemistry (Mattivi et al. 2006). In response to the growing use of metabolomics in plant biology research, ArMet (architecture for metabolomics) has been proposed (Jenkins et al. 2004). Its purpose is to advance discussion and community adoption of experimental standards for metabolomics and promote proper collection, storage and transmission of experiment data. As intended, ArMet is designed to provide a template to capture a complete formal data description for plant metabolomics that supports data sets and the necessary metadata to provide robust experimental context. This special issue of Physiologia Plantarum is dedicated to the rising field of plant metabolomics. In this issue, you will find nine review articles that focus on various aspects of contemporary plant metabolomics; research approaches, methods, technology, training and analytical applications that are being used in the pursuit of a better understanding of plant metabolism and how plants function. The paper by Allwood et al. (2007) presents a very lucid account of some of the key instrumentations, technologies and methods of data processing and data mining used in metabolomic studies. Following the introduction to hardware and software for metabolomic studies is an integrated discussion of some of the metabolomic approaches employed to address aspects of plant biology over the past 7 years. Then Allwood et al. (2007) discuss the use of metabolomics in the analysis of the metabolic interplay between host and pathogen and how metabolomics can help to advance a better understanding of the metabolites that may functionally contribute to plant susceptibility and resistance to pathogens. Important features in the expansion of metabolomic approaches in plant biology research is the development of robust technologies and methodologies and the prerequisite expertise in their proper use for experimental purposes. In the paper by Böttcher et al. (2007), a retrospective and analysis of the 2006 PlantMetaNet European Training Network Activity (ETNA) Metabolomic Research School is presented to highlight both the need and the educational resources that can help to advance the utilization of metabolomics in plant research. The 10-day summer school course included a mixture of lectures, hands-on experiments and computational analyses. As such, the information in this paper can serve as a template for future schools and workshops on metabolite profiling and metabolomics. Of particular general interest, readers will find slide sets for many of the lectures given during the summer school, which should be invaluable for those seeking to know more about plant metabolomics. An important and essential ingredient of metabolomic approaches is the analysis of raw data that gives fundamental insights into important biological processes. The paper by Steinfath et al. (2007) outlines the complexity of data output from metabolomic studies that include data generation to missing value estimation, from data normalization to metabolic correlation networks and from multivariate analysis of combinations of metabolite levels and dimensionality reduction to higher level regression and correlative relationships. An important application of plant metabolomics is the characterization of plant metabolites of nutritional importance and significance in human health. The paper by Hall et al. (2007) outlines how metabolomics can serve to provide deeper insights into the qualitative and quantitative variation of a wide variety of crucial plant metabolites that will provide key information needed to help improve human nutrition and enhance the nutritional status of the plant foods that we depend on for sustenance and good health. For example, using metabolomic approaches will accelerate and advance our ability to identify key genes of metabolism, when coupled with quantitative trait locus (QTL) analyses, that function to influence the steady-state concentration of nutritionally important metabolites. A grand design of the -omics era is the simultaneous acquisition of data sets for a range of components involved in information conveyance and the regulation of information flow from gene to RNA to protein to enzyme activity to metabolites and metabolism. The paper by Weckwerth (2007) describes efforts to advance functional genomics through the combination of proteomic and metabolomic data sets. While straightforward in concept, the integration of data sets for different molecular and biochemical constituents of plant cells in efforts to elucidate mechanisms of biochemical regulation from the integration of metabolomics and proteomics remains a great challenge. Still, the outlook for realizing the overarching goal of integrating diverse data sets to uncover biological meaning is promising, given the progress that has already been made. Recognizing the integrated nature of metabolism in complex biological systems, the paper by Höefgen and Nikiforova (2007) addresses the question of integrating transcriptomic and metabolomic data sets to appraise the responses of Arabidopsis to sulfur deficiency. They describe results of time-series experiments during the imposition a sulfur deficiency, which suggest that plants first attempt physiological and morphological adjustment to acquire additional sulfur to relieve the deficiency status, and when that is not possible, the plant then reacts to a long-term starvation with a metabolic conservation mode to maximize reproductive capacity. Metabolite profiling using a number of analytical platforms to capture an array of S-containing metabolites has made it possible to undertake a meaningful parallel assessment of metabolite and transcript profiles. One of the most effective applications of plant metabolomics has been to advance our understanding of the dynamics of metabolite changes during stress. Perhaps this is because abiotic stresses have such a profound impact on plant metabolism. Shulaev et al. (2007) discuss how metabolomic approaches have helped to shed light on some of the many roles that low-molecular-weight metabolites play in plants during abiotic stress and during combinations of abiotic stresses. Their paper highlights a range of studies that have been conducted with plants subjected to a variety of abiotic stresses. The ability of non-targeted metabolite profiling to provide qualitative and quantitative data on a wide range of metabolites makes it a powerful method to delve into the highly complex metabolic and molecular responses of plants to different forms of stress. The paper of Sanchez et al. (2007) describes results of metabolite profiling analyses of plants subjected to high salinity levels. In one way or the other, nearly all abiotic stresses elicit a set of metabolic and molecular responses that are unique to each type of stress and those that are shared or are conserved. Sanchez et al. (2007) show by comparing results of studies with several plant species during salt stress that some of the changes in primary metabolism are well conserved across divergent taxa and that such conservation suggests future avenues of investigation, which could lead to information important in crop improvement toward increased salt tolerance. Continuing the focus on abiotic stress, Guy et al. (2007) summarize studies that have employed metabolomic approaches to investigate how temperature stress affects the metabolite composition of plants. Long-standing issues in the field of plant temperature stress have been to determine the role(s) that metabolism and metabolite composition may have during the induction of acquired temperature stress tolerance and reveal how metabolism must be reconfigured to restore the stressed plant to metabolic homeostasis. Metabolite-profiling studies have been able to demonstrate the range and magnitude of metabolic adjustments made by plants during temperature stress and to suggest new metabolic targets for more intensive investigation. Together, it is our hope that this special issue on plant metabolomics will provide readers with an introduction, and an enhanced understanding of the state of the art in the field of plant metabolomics, and that it will help to advance our common journey to understand the role of metabolism and the metabolome in all its many ways in plant function.