Dr. Collakova’s major research goal is to understand how plants make storage compounds (oil, protein, and sugar) in their seeds and to learn about how this process is regulated. Then, she can manipulate the levels of these storage compounds in the seed for food or biofuel purposes. She uses experimental and computational approaches to achieve this goal in model and crop selected legumes.
The world faces food and energy crisis and we need to find alternative energy sources and improve plant productivity in producing specific products to obtain more food or biofuel. In my laboratory, we perform basic research with a variety of plants. We are specifically trying to understand how they make seed oil, protein, and sugar. We do know the basic biochemical pathways that lead to the synthesis of these storage compounds. However, we still do not have a complete picture of what determines how much of the individual seed compounds will be formed – how seed composition is regulated.
When approaching a research question, it is best to use a model system that we know the best, at least in the beginning. In this case, we do initial research in the model plant Arabidopsis thaliana, economically unimportant weed of the mustard family, that is invaluable in research, since, due to certain advantages, it became the most studied plant. Then the knowledge can usually be easily transferred to a useful plant. Our further focus is on economically or potentially important legumes that are known to produce protein and oil as major seed components, namely soybean, peanut, and pigeonpea.
How do we do it? We take the advantage of novel experimental and computational approaches used in systems biology. Systems biology enables a global perspective of molecular processes occurring in the cell. Rather than studying a single gene or protein, one can look at the changes of tens of thousands of components in the cell (transcripts, proteins, and metabolites) and relate such changes to phenotypes (e.g., the changes in seed composition during seed development leading to seeds that produce a lot vs. little oil would be reflected as two different phenotypes). Because oil, protein, and sugar are made during seed development (when seeds are maturing in their pods on the plant), we do the experiments with developing seeds. These experiments include modern genomics and biochemical approaches including measuring the changes in transcript and metabolite levels and searching for mutants affected in seed composition and metabolism that leads to the accumulation of seed storage compounds. We also assess how fast this metabolism is in individual seeds (metabolic flux analysis or MFA in short). MFA also allows to determine how the storage compounds were made (by which metabolic pathways).
All these experiments generate a lot of data that we must make sense of. That is when the computational approaches are useful. We try to look for correlations in the spatial and temporal patterns; the idea being that if two cellular components have similar patterns and are present in the same place in the cell there is a high probability that they may be functionally connected (e.g., a regulator and its target). Therefore, we are building predictive probabilistic networks of potentially interacting genes to identify regulators of seed metabolism and composition. Potential regulators then will be tested in Arabidopsis and the proposed metabolic engineering strategies to increase oil or protein content will be implemented in legumes.