Plants explore and exploit their underground environments through their root systems. They modulate the angle, rate of growth and type of individual roots to form an ensemble that constitutes the root system architecture (RSA). The complex 3D structure of a root system is determined by the activity of root tips as they grow, curve, and branch. However, much remains to be discovered as to how dynamics at the local scale result in emergent properties at the global scale (i.e., the root system) that, in turn, modify root system function. There is increasing evidence that understanding the genetic basis of root system architecture could greatly aid in developing crops with improved nutrient and water capture. Poor soil fertility and environmental stress suppress crop yields in many parts of the world, and many models predict abiotic stress will increase in coming decades with projected climate change. Intensive irrigation and fertilization are not environmentally sustainable, nor economically viable in most developing countries. Thus, identifying the genes and alleles that underlie RSA could have profound significance for agriculture and world food security.
We have pioneered the use of a root imaging platform that employs a clear gel matrix with plants grown in cylinders that are imaged through 360 degrees. The entire root system is visible, the roots don’t encounter an interface and the gel allows them to grow in three dimensions. We have also developed novel methods to quantify the 2D and 3D aspects of root architecture. Most existing platforms used 1D parameters (e.g. root mass or length) or 2D descriptors based on extracted and scanned root systems. Our team, which includes computer scientists, plant physiologists, molecular biologists, quantitative geneticists and theoretical physicists has successfully generated a set of novel quantitative descriptors able to distinguish RSA of closely related varieties. We have also built an image analysis pipeline called “Gia Roots” to automate the handling of images from our gel platform.
We have used the imaging platform to characterize the interactions between rice root systems and shown that roots of the different genotypes tend to avoid each other. Our results suggest that roots exude signals that allow them to distinguish their own versus different genotypes and to modulate their root growth accordingly. We are currently using the imaging platform to identify quantitative trait loci (QTLs) for root traits through characterization of a recombinant inbred line (RIL) population made from two rice genotypes with strikingly different RSA.