Medicinal Plant Metabolomics - Prof. Bryan Hanson - Prof. Dana Dudle

Plant Metabolomics ...

... Where Spectroscopy mets Evolution & Ecology

We - Prof. Dana Dudle (Biology) & Prof. Bryan Hanson (Chemistry & Biochemistry) - are collaborating on research in plant metabolomics.. Our current research students are Courtney Brimmer, Tanner Miller, and Kelly Summers, all members of DePauw's Science Research Fellows program (which also provides funding - thanks!). Click on the student names to download their posters from summer 2009, which describe their projects in greater detail.

Our present focus is an investigation of the role of stress on the weedy plant Portulaca oleracea (purslane, shown at right in it's favored environment - sidewalk cracks!). In plants, the notion of stress includes conditions like too much light, too little water, or toxic metals in the soil. Another kind of stress would be increasing temperatures associated with climate change. We have chosen purslane because it is easy to grow and is of interest from a medicinal/nutritional perspective - it has the most omega-3 fatty acids of any plant. Our main objective is determine whether purslane's response to stress has a genetic component which also contributes to reproductive fitness. If we can answer this affirmatively, we will investigate the specific physiological mechansms which underlie the responses we observe.

Our approach is to blend metabolomic and ecological methods. Metabolomics is the study of an organism's metabolites under a controlled set of conditions, in our case, normal versus stressful conditions. As far as possible, one tries to measure all the metabolites at once, in a holistic fashion, which is not an easy feat. Typically, this is done with NMR, MS, IR, or other forms of spectroscopy. We also supplement these instrumental techniques with more traditional single point chemical measurements such as antioxidant levels. From the ecological perspective, we record parameters of plant growth that represent measures of fitness, such as biomass produced, the number of flowers, and so forth.

To be meaningful, we need to conduct these experiments on large numbers of plants. The resulting data sets, composed of very different sorts of measurements, represent the state of the plant under the conditions tested. We use various statistical methods to figure out which treatments have produced an interesting response, and whether those responses vary with genotype. Our statistical analyses are done primarily with the ChemoSpec package for R, written locally by Bryan.