Developing and applying transient expression systems to identify and understand gene regulatory elements in plants

Abstract

In this dissertation, I develop transient expression systems to enable massively parallel reporter assays (MPRAs) in crop models. I apply these transient assay systems to identify, characterize and design plant gene regulatory elements. To allow for the scale of MPRAs, I develop a high-efficiency transformation protocol for maize mesophyll protoplasts. Using this protocol, I can transform millions of maize mesophyll protoplasts without losing viability. I use transient transformation of maize protoplasts along with Agrobacterium-mediated transient transformation in tobacco to develop a plant-specific MPRA, plant STARR-seq, to ascertain the activity of regulatory elements such as core promoters and enhancers. My contribution was essential in assuaging reviewer concerns about prior studies using a different, animal-specific assay design. Using plant STARR-seq, I and my co-author Tobias Jores identify and characterize 75,000 core promoters from maize, sorghum and Arabidopsis. We identify features required for the function of core promoters in both maize and tobacco, and we find differences in the effect of GC content between maize and tobacco elements corresponding to the GC content of their respective genomes. We use machine learning and in silico evolution to design synthetic core promoters that rival the viral Cauliflower Mosaic Virus 35S core promoter in activity. Finally, I use plant STARR-seq to dissect the activity of three known light-responsive enhancers. I perform deep mutational scans of all three enhancers and identify regions in which mutations affect their function in the dark and light. I combine these regions to create synthetic enhancers with a wide range of transcriptional responses including enhancers that show greater light response than any of the original enhancers. I show that most of the observed enhancer activity is explained by an additive model, albeit there are rare exceptions.