Uncovering the genetic architecture and metabolic basis of amino acid composition in maize kernels using multi-omics integration

Seeds are a major source of protein in human and livestock diets. Cereal grains are some of the most consumed seeds by both humans and livestock worldwide, with maize, wheat, and rice alone accounting for ~70% of the total cereal production. Maize is one of the major staple crops used for food, feed, and fuel. A mature maize kernel contains small embryo (10% of the volume) and a large endosperm (~90% of its volume). In terms of composition, majority of the kernel proportion contains around 90% of starch and around 8-10% of protein. Nine of the twenty amino acids cannot be synthesized by monogastric animals, including humans, and must be obtained through the diet and are considered essential amino acids (EAA): lysine, isoleucine, leucine, histidine, methionine, phenylalanine, threonine, tryptophan, and valine. The protein quality is poor in maize endosperm as the primary storage proteins are severely deficient in EAA such as lysine, tryptophan, and methionine. Such deficiencies can be detrimental since corn provides an important source of proteins for food in developing countries and for feed in developed countries such as the U.S. Failure to consume sufficient levels of EAA per day leads to severe malnutrition, even if one's calories requirements are met. Many attempts to increase the EAA has demonstrated only limited success since seed can rebalance their amino acids composition even when major changes are introduced in their proteome. One possible approach to solve this applied problem is by seed EAA biofortification; however, many attempts at this task fall short and strongly indicates that even though we know most of the metabolic pathways of amino acids, we know very little about their regulation especially in seed. Therefore, the first step towards efficient amino acid biofortification is to increase our fundamental understanding of their function, as well as the metabolic regulation and the biology of the plant seeds. Despite the tight regulation within any given genotype seed amino acid composition display extensive natural variation which can be utilized to uncover the genetic basis and identify new targets for seed amino acids biofortification. Hence to uncover the genetic architecture of amino acids composition in maize kernels we used Goodman-Buckler maize association panel that consists of 282 diverse maize inbred lines including stiff stalk, non-stiff stalk, tropical and subtropical, sweetcorn and popcorn lines. I performed genome wide association study (GWAS) on both the protein bound amino acids (PBAA) and free amino acids (FAA). Although, GWAS is widely used to dissect the genetic architecture of complex traits, oftentimes the GWAS outputs the extensive list of genes particularly when using multiple phenotypic traits. To overcome this, I used an integrative multi-omics approach that combines GWAS and co-expression networks modules obtained from ten seed filling stages of B73 to uncover novel key regulatory genes, characterize biological process and prioritized the candidate genes that involved in shaping the natural variation of amino acid composition. Chapter one of the dissertation is the general introduction and literature review on the seed amino acids. It briefly discuss the general introduction of PBAA and FAA, previous attempts done to improve seed PBAA and FAA composition, natural variation used to uncover the genetic architecture of complex traits including metabolic traits such as amino acids and finally discuss the multi-omics integration to uncover the genetic basis of complex traits. Chapter two elaborates the comprehensive genetic basis of PBAA in maize kernels using integrative analysis of 76 PBAA GWAS with protein co-expression network modules. Previous studies have shown that manipulation of storage proteins and amino acid pathway genes have contributed in the improvement of quality protein maize however, my study strongly suggests that in addition to the manipulation of storage protein and amino acid metabolic genes, specific ribosomal genes along with other translation machinery could be the novel target for seed amino acids biofortification. Chapter three discusses the genetic basis of FAA in maize kernels using integrative analysis of 109 FAA GWAS with protein co-expression network modules. I have presented here the comprehensive list of SNPs as well as the candidate genes and several biological processes including the translational machinery responsible for shaping the genetic architecture of FAA in seed. Chapter four includes the conclusion and future works. Maize is an important crop used for both food and feed and possesses great genotypic and phenotypic diversity. The results from my study has validated several previous characterized genes and identified novel key genes that regulate and shape the PBAA and FAA in maize kernels, which could be used further to target for amino acid biofortification.Includes bibliographical references