Uncovering the genetic basis of seed amino acid composition in arabidopsis using a multi-omics integrative approach

Abstract

Seeds are a vital source of protein in the diet of humans and livestock. However, protein composition in the seed is low, comprising about 10 percent of the total composition in the seed. Additionally, protein quality in the seed is poor due to low concentrations of certain essential amino acids (EAA). Since the body is unable to produce EAA, they must be consumed in the diet and failure to do so has detrimental, potentially irreversible, health implications that can result in death. In developing counties where meat and dairy are lacking, protein-energy malnutrition frequently occurs. In contrast, in developing countries large portions of seeds are used in the diet of livestock which must be supplemented with costly synthetic amino acids. Collectively seed amino acid composition of major crops are not sufficient to meet dietary requirements. Protein in the seed is comprised of free amino acids (FAA) and protein bound amino acids (PBAA) which have both been the targets of manipulation in order to create a seed with a more balanced amino acid profile. However, upon perturbations to the proteome, mutant seeds have demonstrated a rebalancing phenomenon where even large alterations to the amino acid composition activate a compensation mechanism that returns amino acid levels to a comparable composition to the wild-type. Although a lot is known about amino acid metabolic pathways, what regulates such rebalancing mechanisms is still unknown. However, despite the tight regulation, natural variation does exist in seed FAA and PBAA across Arabidopsis ecotypes with a unique composition specific to each ecotype; this suggests rebalancing has a genetic basis. Thus, the first step in seed biofortification efforts must be to first increase the fundamental understanding of the genetic basis of both FAA and PBAA composition in the seed. Chapter One of this dissertation gives a more in-depth introduction that elaborates on amino acid composition in the seed, the challenges identified in previous experimentation, and how the content of Chapter Two through Chapter Four builds upon and adds value to the area of seed amino acid research as a whole. Chapter Two focuses on uncovering the genes and biological processes that underly the regulation of free Glutamine which belongs to the Glutamate Family (Arginine, Proline, Glutamine, and Glutamate). Although Glutamine is not an EAA, it is a major nitrogencontaining amino acid that is transported to the seed; thus it's regulatory control is of particular interest. I harness the natural variation of Glutamine in a 360 Arabidopsis diversity panel to uncover key regulatory genes. Later, I validate observations from GWAS using both a quantitative trait locus (QTL) analysis and reverse genetic approaches to identify a unique, seed-specific Glutamine-glucosinolate relationship that alters nitrogen and sulfur homeostasis in the seed in the Arabidopsis 360 population. Such finds were substantial as they link primary and secondary metabolism in the seed. Chapter Three focuses on uncovering the genetic basis underlying PBAA composition in dry Arabidopsis seeds while expanding upon the work completed in Chapter Two. 576 high confidence candidate genes (HCCGs) are found through integration of GWAS using PBAA traits and transcriptomic analysis across seed development of two mutants showing active rebalancing. To reveal the underlying biological process, I further subject the HCCGs to a protein-protein interaction (PPI) network that strongly suggests that ribosomal genes and potentially other translational machinery may be in the heart of PBAA composition homeostasis and the proteomic rebalancing response. Chapter Four addresses the need of a comprehensive tool to efficiently and automatically analyze many biochemical derived-traits in GWAS, while also completing pre and post-GWAS analysis. Here, I present the R tool HAPPI GWAS, describing each step in the pipeline, and giving an example of its implementation. Lastly, Chapter Five reiterates the contributions of this dissertation to the field of seed amino acid research and provides insight into future direction and research projects. The results from this work are vital steps in understanding the complex regulatory mechanisms underlying amino acid composition in the seed which can be used in manipulating the amino acid pools in future translational crop research.