Metabolomic and transcriptomic analysis of the controls on carbon partitioning into TAG reserves in oilseeds

Lead Research Organisation: University of York
Department Name: Biology


Plant oils supply about 25% of the calories in our diet. Increased consumption of oils and fats in the human diet is regarded as unhealthy, as it leads to obesity. However, the composition of plant oils influences whether they are beneficial or detrimental to human health. For example, there is evidence that consumption of long chain polyunsaturated fatty acids (LC-PUFAs), which are mainly supplied in our diet from fish oils, can improve our metabolism of fats in a beneficial way. In addition to their use as foodstuffs, plant oils are becoming increasingly important as replacements for petrochemicals in a wide range of industrial applications such as in the production of alternative fuels and lubricants. Despite the obvious uses and benefits of plant oils, we do not fully understand what controls oil yield and composition in agricultural crops. In these crops, the oils are produced in seeds. Therefore, in order to understand and optimize oil production in plants, we need to understand how oils are produced and stored in seeds. Oils in seeds are stored as triacylglycerols (TAGs), which are made up of three fatty acid molecules linked to glycerol. For many years, plant scientists have studied the biochemistry of TAG synthesis in oilseeds, with the aim to understand what metabolic pathway is responsible for TAG accumulation, and how it is regulated. However, it has become apparent that the process of TAG synthesis is only one component of overall lipid metabolism in plants, such that there are several competing pathways for the biochemical intermediates that are required to make oils. In addition, the picture is further complicated by recent work that has shown that there are multiple biochemical routes for TAG synthesis, and that TAG breakdown (catabolism) probably occurs at the same time as synthesis. Using molecular genetics approaches, researchers have identified many of the genes and metabolic intermediates that are important in lipid synthesis in plants. However, in order to understand how TAG synthesis is specifically regulated in oilseeds, we need to evaluate which genes and metabolites are primarily or specifically involved in TAG metabolism and which are involved in other areas of lipid metabolism. To answer this question, we plan to grow and harvest seeds from the model oilseed plant Arabidopsis at different developmental stages where TAG synthesis is known to be up- or down-regulated. There are hundreds of existing datasets that show how global gene expression (the transcriptome) varies over these developmental stages, and we plan to mine these data to find genes that show correlations with TAG synthesis. We also plan to generate some of our own transcriptomic data using Arabidopsis mutants where TAG synthesis is altered. Analysis of this data will uncover additional genes that may be important in regulating TAG synthesis. In order to correlate changes in gene expression with actual TAG synthesis during seed development, it is important to know how much TAG is present at any one stage, what the fatty acid composition of this TAG is, and how other lipid-related biochemical intermediates change in concentration. In addition, it is necessary to monitor how apparently unrelated biochemical pathways are changing, as some of the metabolites in these pathways may be indirectly regulating TAG biosynthesis. All these measurements can be accomplished using metabolomics, where the biochemical composition of small molecules is measured using a range of analytical techniques. Arabidopsis mutants that are deficient is specific lipid metabolism pathways will be selected for metabolome analysis, in order to understand how these pathways are linked to TAG synthesis. The results of this research will improve our understanding of lipid metabolism in plants, which will ultimately enable us to improve oil yields and the fatty acid composition of plants for dietary and industrial uses.

Technical Summary

This project aims to identify genes and metabolites that are key in regulating triacylglycerol (TAG) content in oilseeds. We will achieve this by analysing metabolome and transcriptome data from wild type and mutant Arabidopsis seed tissues sampled over six developmental stages that are representative of differential TAG synthesis and breakdown. Both global metabolomic analysis (NMR, GC-MS) and specialised targeted profiling methods for key lipid related metabolites such as acyl-CoAs, TAGs and sphingolipids will be performed. This will give rise to the first comprehensive metabolomic dataset for developing and germinating Arabidopsis seeds. Ten pre-selected mutants disrupted in biochemical pathways directly related to the acyl-CoA pool and/or in the regulation of TAG synthesis during seed maturation will also be profiled in the first phase of the work. Comprehensive transcriptomic datasets already exist for four of the selected seed stages and we will generate additional datsets for the two remaining stages and for two seed development stages of the abscisic acid insensitive abi3 regulatory mutant which is altered in seed maturation and storage reserve synthesis. This data will be combined with all public domain Affymetrix based transcriptomic data (currently amounting to 1,700 non-redundant datasets) in an in-house fully relational my SQL database. This database will be used to identify putative TAG related transcriptional modules and it will also be used to house the metabolomic datasets. Metabolite-specific temporal patterns across the developmental series of maturing and germinating seeds will be generated and dedicated pattern matching procedures employed to identify similarities between metabolite and transcriptomic data. Based on this approach a further set of mutants will be selected for the second phase metabolomic and targeted gene expression analysis. This research will provide a platform of knowledge that will facilitate the rational design of metabolic engineering of economically valuable oils in crop plants.
Description 1) We established analytical platforms and bioinformatics tools necessary to perform metabolomic analysis on developing and germinating oilseeds.
2)A key observation of the seedling developmental series in this project was a change in triacylglycerol (TAG) and fatty acid composition during lipid breakdown in beta-oxidation mutants. Our results suggest that there is a cytosolic route for TAG synthesis that is important for scavenging fatty acid substrates that are either produced in excess in membrane biosynthetic or general glycerolipid catabolic activity, or produced in a situation where catabolism is blocked.
3) We demonstrated a central role for peroxisomes in both developing and germinating seed metabolism.
Exploitation Route Finding 1) above platforms and tools developed were used for further research.
Finding 2) described above is an important discovery and there is great potential in exploring activity to enhance TAG yields in both developing seeds and green tissues.
Regarding finding 3) above targeted oxylipin profiling has revealed an alternate uptake mechanism for jasmonic acid into peroxisomes during seed development and global NMR profiling has identified a novel route for benzoic acid synthesis in peroxisomes.
Sectors Agriculture, Food and Drink,Manufacturing, including Industrial Biotechology

Description The knowledge acquired in this research programme has been used in the design of metabolic engineering strategies for altering seed oil fatty acid composition. A challenge in developing new oil crops with novel oils such as long chain polyunsaturated fatty acids is to partition specific fatty acids into and out of the lipid pool. Our work on mechanisms involved in this partitioning of fatty acids into triacylglycerols therefore underpins metabolic engineering programmes such as that carried out by BASF on PUFA production in oilseed rape.
First Year Of Impact 2010
Sector Agriculture, Food and Drink
Impact Types Economic

Title Network modelling of metabolites associated with TAG accumulation during Arabidopsis seed development 
Description Data from global analyses were compared to fatty acid yield data generated in York, and were processed and analysed by in-house pipelines of multivariate data analysis techniques, including partial-lease-squares regression (PLSR), non-parametric correlation, and rank product algorithms, all with customised statistical filtering. 
Type Of Material Biological samples 
Provided To Others? No  
Impact The pipelines described above were trialled and developed to meet two objectives: to discover metabolite signals associated with fatty acid yield, and to generate a manageable target list of a subset of these signals that could be further investigated and annotated. These methods have enabled us to produce complementary lists and network models of metabolites globally associated with oil yield across three lines, and those which differ in a mutant-dependent manner. 
Title Metabolomic dataset for selected mutants 
Description Comprehensive metabolomic dataset developed for selected mutants disrupted in synthesis and regulation of acyl CoA and TAG metabolism. Data was collected from three global platforms at MetRO (NMR, ESI-MS, GCMS) and four targeted platforms at York (fatty acids, acyl CoAs, TAGs, oxylipins). 
Type Of Material Database/Collection of data 
Year Produced 2009 
Provided To Others? Yes  
Impact In the latter part of the grant efforts were directed towards data analysis of the global metabolome and transcriptome, followed by further experimentation and analysis of selected mutant lines at key developmental stages. 
Title Metabolomic datasets for developing and germinating wild-type Arabidopsis seeds 
Description Data for ten distinct lines across six developmental stages. 
Type Of Material Database/Collection of data 
Year Produced 2009 
Provided To Others? Yes  
Impact Bioinformatics tools for metabolomic data mining were developed. 
Title Transcriptomic datasets 
Description Transcriptomic datasets were produced for wild type Arabidopsis, to provide a complete dataset for all six developmental stages. Data publicly available from NASC and TAIR were used to develop the dataset, along with data kindly supplied by E. Nambara and triplicated in-house Affymetric experiments. 
Type Of Material Database/Collection of data 
Year Produced 2009 
Provided To Others? Yes  
Impact Transcriptomic experiments were followed up by further in-depth analysis during the project. 
Description Poster presented by Dr Patrick Bueker at the International Crop Modelling Symposium (iCROPM2016) in Berlin, March 2016. 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Dr Patrick Bueker of the University of York Resilient Rice project team presented a poster titled 'Variation in rain-fed rice yields in India under a changing climate' at the International Crop Modelling (iCROPM) Symposium in Berlin. With more than 300 scientists from 47 nations, the iCROPM Symposium in 2016 brought together the major part of the international crop modellers' scene to exchange ideas on improvement and application of crop simulation models to better support agricultural production and food security under global change. His poster is available online as an open access publication here, pages 216-17:
A summary overview of the impact of the conference is available at the URL below.
Year(s) Of Engagement Activity 2016