Panzea Project Summary

Scientific Objectives and Approaches

In our previous grant, we determined the genetic relationships among maize and teosinte germplasm, examined how average loci evolve, and developed methods for relating nucleotide diversity to phenotypic effects. Our objective for this new proposal is to understand how selection has shaped molecular diversity in maize and then relate molecular diversity to functional phenotypic variation. How has selection shaped molecular diversity? To address this, 4000 loci will be screened for selection evidence, and then 1000 loci will be studied extensively by doing SNP surveys across diverse maize and teosinte. A range of tests of selection will be used to identify genes showing positive, diversifying and purifying selection. The identified genes will be those involved in domestication, agronomic improvement, and local adaptation. How does this molecular diversity relate to functional trait variation? A wide range of maize and maize-teosinte linkage and association mapping populations will be created that capture a tremendous range of diversity. These populations will be genotyped for SNPs and candidate genes (many of the selected genes noted above) and phenotyped for domestication, agronomic and developmental traits. This will permit high-power and high-resolution dissection of a wide range of traits, and relate the molecular diversity to functional variation.

Broader Impacts

In 2001, maize became the number one production crop in the world, and current US maize production is almost four times wheat and rice production combined. Maize is also the most diverse known crop at the molecular and phenotypic levels. This project's diversity dissection will enable genomics and breeding to help improve maize yield and quality and to adapt maize to a wide range of biotic and abiotic conditions. This project will provide insights into the ways selection modifies a genome. Additionally, this will produce two major resources: (1) a large database of validated SNPs throughout the genome, and (2) a set of maize and teosinte mapping populations that could be used to map almost any trait. These results will be made accessible to two audiences. A strong informatics program will be used to make these results available to a wide range of researchers through public databases such as Gramene and Genbank. Additionally, we will make this research accessible to a wider range of people through outreach programs. We will focus on transmitting the excitement of science and knowledge of maize to students in rural schools and through mentoring programs in our labs.

Introduction

Maize is the most diverse crop species in the world. This diversity is manifested at both the molecular and phenotypic levels. Considering nucleotide polymorphism in genes, two maize lines are on average as diverged from one another as humans are from chimpanzees. The phenotypic diversity we see in maize today is the product of a long tradition of plant breeding practiced by Native Americans who first converted the wild grass, teosinte, into maize and then proceeded to adapt this crop to virtually every habitable environment in the Americas including deserts, tropical rainforest, high mountains (3500 m above sea level), and the short growing season of Canada. Most maize diversity remains undescribed, poorly understood and under utilized in modern plant improvement largely because of the difficult of identifying useful genetic variants hidden in the background of low yielding local varieties or lines (Tanksley and McCouch 1997). Thus, there is a compelling opportunity to apply genomic tools to sift through the countless allelic variants in the maize gene pool and understand how they impact phenotypes of agronomic and evolutionary importance.

In our past project, we determined the structure of genetic diversity in the maize and teosinte gene pools, examined how genetic loci evolve, and developed methods for relating nucleotide diversity to phenotypic effects. In our new project, we want to extend this work to understand two things: (1) Which genes have been under selection during maize domestication and improvement? By addressing this question, we can identify specific genes that have experienced past selection and thereby improve our understanding of how evolution modulates diversity across the genome. Our group's prior research suggests that although only 1% of random maize genes are under selection (Vigouroux et al. 2002b), two-thirds of the genes involved in the agronomically important starch pathway have been modified by selection (Whitt et al. 2002). The identification of numerous genes under selection will provide insights into how evolution modifies pathways and which genes control agronomic phenotypes. (2) Which genes and alleles control key traits? Although finding agronomically relevant genes is important, the improvement of maize requires finding better alleles at these genes. We will develop a high- throughput, high-resolution platform for finding and evaluating allelic variation throughout the genome and the breadth of the gene pool. Combining these two questions, our overall goal is to turn the identification and evaluation of functional and evolutionarily important allelic variation into a comprehensive (genomics) activity.

Specific Aims

The proposal has two sections. Section 1 will examine the impact of past selection on molecular diversity in maize genes. Section 2 will investigate a large number of candidate genes for their associated effects on agronomic or evolutionarily important phenotypes. These two sections are interrelated. If a gene is identified in Section 1 as having experienced selection, this implies that it controls an important phenotype and thus would be an appropriate candidate for functional analysis in Section 2. Similarly, genes shown to possess functional allelic diversity in Section 2 need to be assayed for molecular diversity to reveal how selection has patterned diversity across the gene pool and to identify where interesting allelic diversity occurs. Our specific aims are:

Section 1. Molecular Diversity:

1. Perform SNP discovery for 1000 genes in diverse maize inbreds and teosinte to identify an unbiased set of SNP for molecular diversity analysis.
2. Perform tests of selection during maize domestication and improvement on 4000 random and phenotype-candidate genes.

Section 2. Functional Diversity:

1. Develop maize and maize-teosinte mapping populations to capture diverse alleles and dissect a wide range of quantitative traits.
2. Conduct large-scale QTL mapping experiments in maize and maize-teosinte populations to identify, localize and characterize the genomic regions harboring QTL.
3. Dissect domestication, agronomic, and plant development QTL using candidate gene association analyses.

In addition, we will develop a database that has user interfaces for (1) presenting our maize QTL maps in the context of the existing genetic maps for maize and other cereals and (2) viewing SNP and phenotypic diversity. These features will be integrated into the Gramene database in a way that makes it easy for researchers to perform analyses across diverse germplasm and to gain access to the underlying biological reagents. Finally, we will work with students in rural and under resourced communities to enhance science education and to help bring underrepresented groups into plant science.