2013 Annual Report
1a.Objectives (from AD-416):
1. Determine if unintended effects are produced in transgenic crops, using fruit ripening in tomato as a model system. 1A. Determine if unintended effects are produced in transgenic crops, using gene expression analysis as a monitoring tool. 1B. Determine if unintended effects are produced in the fruit of transgenic crops that affect fruit quality or composition, through metabolomic and proteomic profiling and an examination of agronomic trait performance.
2. Genetically identify the genes affecting iron levels and bioavailability in maize seed using maize quantitative genetics and Caco-2 cell culture in vitro digestion assay. Determine Fe levels and bioavailability in genetically engineered maize seed.
1b.Approach (from AD-416):
1) Utilize genomic, metabolomic, proteomic and agronomic approaches to evaluate phenotypic difference between tomatoes. 1A) Utilize natural diversity between tomato cultivars, together with conventional breeding techniques, to capture a reasonable phenotypic range from diverse tomato germplasm. 1B) Utilize RNAi and artificial microRNA gene silencing technologies to adjust RIN gene expression levels and alter fruit ripening. Compare phenotypic effects of transgenes to the range observed with conventional cultivars..
2)Leverage research on fruit specific or ripening stage specific promoter sequences to further tailor the modulation of RIN gene expression in the target tissue. Assess the efficacy of tailored gene modulation on reducing unintended effects via genomic, metabolomic, proteomic and agronomic monitoring.
Developed a composition dataset for diverse maize grain varieties, using liquid chromatography and mass spectrometry, measuring thousands of chemicals across hundreds of varieties of maize. Our choice for which varieties was based on previous work in examining their genetic diversity. We applied statistical approaches to analyze the composition data and produce genetic explanations for many of the compounds we observed. Further, we analyzed our composition data through a lens of network modeling to estimate the regulatory pathways and processes well known from the breeding literature but hereto poorly described at the molecular/biochemical level. Chemists at Colorado State University facilitated this work, as they found a number of compounds with known biological activities (or economic importance) from our data. We established a new method for integrating large scale compositional profiling with genetic analysis. We will soon apply this method, with collaborators at Cornell University, to apple, potato, and oat.
To help evaluate our genetic analysis of grain composition, we used bioinformatics resources developed by collaborators at the University of Minnesota who study gene networks through gene expression analysis. Our premise was this: if a group of genes are identified by genetic analysis of grain composition, then we should also be able to see the same group of genes working together at another level of biology (for example, gene expression). We saw subsets of the genes identified as potential determinants for grain composition also working together in gene expression networks. We can make hypotheses about how different facets of grain composition are accomplished, due to the knowledge of cellular biochemistry and physiology collected by our collaborators at Minnesota. This combination of genetics, genomics, and metabolomics facilitated by bioinformatics is in part what led to the new work in apple, potato and oat.
To enhance our efforts, extramural funding was won to study the genetic and environmental factors that influence mineral nutritional quality. With collaborators at the Donald Danforth Plant Science Center, Minnesota, and ARS (Columbia MO and Ithaca NY), we analyzed the 5,000 varieties of the Nested Association Mapping panel grown in four different environments. We found potential genetic determinants for nutrients like iron, zinc and phosphorous. During the summer of 2013, our team is testing our first set of candidate genes using genetic research varieties obtained from a number of sources. We are also creating new research varieties to test additional genes, using new farm sites, in 2014. We are also analyzing gene expression in maize roots that are acquiring nutrients from the soil, rather than studying seedlings as most other groups have done. This will allow us to refine our gene expression networks to help identify key genes for grain composition.
Leveraging genomes to better understand crop composition. Scientists are finding it increasingly easy to collect larger datasets, while understanding these data is also an increasingly larger problem. ARS researchers at Ithaca, New York, together with ARS researchers in St. Louis, MO and scientists at Colorado State University in Fort Collins and the University of Minnesota-Twin Cities improved a procedure that combines analytical chemistry with multivariate statistics to also include gene expression analysis. This process builds on the investment in the maize genome sequence to help ascribe functionality to unknown genes and identity to unknown compounds. This discovery will help translate lab discoveries into gene/trait combinations useful to producers.
Baxter, I.R., Gustin, J.L., Settles, M.A., Hoekenga, O. 2012. Ionomic characterization of maize kernels in the Intermated B73 x Mo17 (IBM) population. Crop Science. 53:208-220.
Karuppanapandian, T., Rhee, S., Kim, E., Han, B., Hoekenga, O., Lee, G. 2012. Proteomic analysis of differentially expressed proteins in the roots of columbia-0 and landsberg erecta ecotypes of arabidopsis thaliana in response to aluminum toxicity. Canadian Journal of Plant Science. 92:1267-1282.
Shen, M., Broeckling, C., Chu, E., Ziegler, G., Baxter, I.R., Prenni, J., Hoekenga, O. 2013. Leveraging non-targeted metabolomic profiling via statistical genomics. PLoS One. 8(2):e57667. Available: http://www.plosone.org/article/info:doi/10.1371/journal.pone.0057667