Evolution of Plant Secondary Metabolites

Kliebenstein Lab

UC Davis Plant Sciences

NSF Plant Genome - Quantitative Genomics

Multi-locus epistasis in metabolic networks.

We are using a combination of GC-TOF metabolomics and HPLC to identify metabolomic QTL in multiple Arabidopsis and Rice populations.

Using the glucosinolates as a model system, we developed methodologies to compare metabolomics data with transcriptomics data.

We expanded this to global metabolomics analysis and identified multi-locus epistatic interactions that were the primary drivers of phenotypic variation for primary metabolism.

We have shown that this epistatic network is due to natural variation in the inputs and outputs of the circadian clock.

The goal is to better understand the quantitative genetic architecture underlying complex differences between individuals.

Project Summary

· Metabolic traits have lower heritability than transcript traits.

· Metabolic traits have more epistasis than transcript traits.

· Metabolite variation is associated with transcript variation but not absolutely.

· Secondary metabolite variation is not a primary driver of metabolomic QTLs in Arabidopsis.

· Glucosinolates show strong diversifying selection in comparison to the genome average

Most organisms contain significant variation between individuals in traits such as development, disease resistance and longevity. These traits are controlled by Quantitative Trait Loci (QTL). Recently, genomics tools such as microarrays and metabolomics have been applied to structured mapping populations in an attempt to understand the genetic architecture underlying complex differences between individuals. However, much of this work has presumed comparable variation inherent in different omics levels e.g. transcript or metabolomic. We have begun formally testing this presumption. Additionally, we are testing if quantitative genomics can enhance our ability to clone the molecular mechanism underlying QTL.

 

 We began this project using the glucosinolate secondary metabolites as a test case. This analysis showed that transcripts and metabolites have different quantitative genetic architectures. Thus requiring a modification in experiments designed to compare the two levels. Additionally, we have used this variation to clone novel glucosinolate regulators. This includes enzymes that appear to modulate the transcription factors that in turn control the enzymes, creating a circular feedback loop at the transcript level.

 

This project has been expanded to GC-TOF based metabolomics in multiple plants to test if the observations from the secondary metabolite model can be extended to primary metabolites. Potentially, this will help us to assign unknown metabolites to biosynthetic pathways and potentially identify the underlying enzymes.

 

 

Clustering of Metabolic QTLs in the Bay x Sha Arabidopsis Population

To contact us:

Phone: 530-754-7775
Fax: 530-752-9569
E-mail:
kliebenstein@ucdavis.edu

Text Box: Sample Publication
Chan, E.K.F., Rowe, H.C., Hansen, B.G, and  Kliebenstein, D.J. (2010) “The complex genetic architecture of the metabolome”. PLoS Genetics 6(11)e1001198. (Pubmed link)

Chan, E.K.F., Rowe, H.C. and  Kliebenstein, D.J. (2010) “Understanding the evolution of defense metabolites in Arabidopsis thaliana using genome-wide association mapping” Genetics 185(3)991-1007. (Pubmed link)
National Science Foundation