Tomato quality — more than a question of taste
The quest to develop new and improved tomato varieties is benefiting from a systems biology approach to understand the ripening process.
Dr Charlie Baxter. Image: Sygenta
In this article Dr Charlie Baxter from Syngenta reports on how the results from a BBSRC-funded collaboration with scientists at the University of Nottingham and Royal Holloway, University of London are feeding directly into breeding programmes to increase the efficiency of selection for commercially important traits.
The tomato is a major fruit crop which is grown in practically every country of the world. Over 141M tonnes are produced globally each year (2009 FAOSTAT). As well as being incredibly tasty, tomatoes are packed with vitamins and minerals and frequent consumption of tomatoes has been associated with a reduced risk from certain types of cancer and heart disease.
In addition to better consumer satisfaction and nutrition, the introduction of varieties with enhanced flavour, texture and shelf life has the potential to increase food security through harvest flexibility, lower wastage in the food chain and a reduction in the use of energy inputs. However, these important traits are controlled by complex interactions between many environmental signals and developmental processes, which make targeted plant breeding (through the identification of useful molecular markers) time consuming and costly. As a consequence Syngenta has been exploring the potential of systems biology to highlight the pathways and genes responsible for controlling key traits.
Tomatoes in a hothouse. Image: iStockphoto ThinkStock 2011
In 2008, a BBSRC-funded Exploiting Systems Biology LINK project brought together expertise in systems biology (Professor Charlie Hodgman) and tomato genetics and fruit ripening (Professor Graham Seymour) at the University of Nottingham with metabolomics expertise at Royal Holloway, University of London (Dr Paul Fraser), combined with plant breeding expertise at Syngenta to characterise the regulatory network controlling tomato ripening.
Bridging the gap between gene and function
Using known tomato ripening mutants to pinpoint perturbations in developmental processes, we compared metabolite and gene expression data, which has revealed many interesting insights.
Shifts in the metabolic profiles of tomato fruit with different ripening phenotypes have revealed important links between fruit ripening and primary and secondary metabolism. These metabolic changes provide important insights around the regulation of sugar and amino acid synthesis through the fruit development.
In addition, network analysis of gene expression changes has highlighted a number of genetic factors that regulate changes in fruit development. Researchers at Nottingham University have identified a series of transcription factors that play a role in regulating the ripening process. Such discoveries can be used to develop gene specific molecular markers that allow plant breeder to directly track ripening traits in an efficient way. This information is being fed directly into Syngenta's breeding programs to aid the selection of key traits.
Harvesting model data
The initial collaboration has formed the basis of an ongoing Syngenta-sponsored systems biology initiative. In 2009 Syngenta signed an agreement with Imperial College London to form the Systems Biology Innovation Centre. This has brought together scientists from the tomato project at Nottingham with computational modellers led by Professor Stephen Muggleton at Imperial.
The data generated in the BBSRC ESB-LINK project is now being used to develop and test predictive models designed to generate a deeper understanding of the regulation of metabolic processes. Using the Ondex data integration software, developed with BBSRC funding and led by Professor Chris Rawlings group at Rothamsted Research in collaboration with the Universities of Manchester and Newcastle, we have been able to quickly develop the background knowledge required to develop and run machine learning algorithms. In addition the Ondex software has allowed visualisation of the output from this process. In silico experiments are ongoing at Imperial College that will further elucidate the links between tomato fruit development and key metabolic processes to provide new molecular-genetic targets for plant breeding.
Work continues on the tomato Systems Biology project at Imperial College and Nottingham University, whilst the wider potential of machine learning in Systems Biology is being evaluated in further Sygenta-Imperial initiatives, looking at the biomarkers for cancer and the development of ecosystems.