Performing Department
(N/A)
Non Technical Summary
Corn has traditionally not been grown for specific products or uses in the US. Instead, the generic yellow dent corn has been cultivated over the last 100 years to serve every purpose as an ingredient or product. With the new tools we have in plant research, we can now try to look back at locally adapted corn that was present before the yellow dent displaced them in the early 20th century and select them for specific products that can compete with the current seed we have now.My research primarily focuses on the whiskey market and how we can make better corn for that market. To do that, I will be growing two populations of corn currently used at the small scale for whiskey making that was previously used by distillers before the onset of yellow dent corn in the US and prohibition. I will then cross the parents together to make a population and analyze that population after two rounds of self-pollunation to look at flavor diversity in this population and how it relates to the genetics of the plants. Overall, this project will help inform us how flavor traits are inherited in corn and will help begin the process of identifying and growing corn for specific uses in the American whiskey market.
Animal Health Component
0%
Research Effort Categories
Basic
50%
Applied
50%
Developmental
(N/A)
Goals / Objectives
Maize has been a staple ingredient for a diverse range of products that we have come to depend on in the modern economy. While the number of products derived from maize have been integral to the development of the agricultural economy, we have failed to acknowledge the value of diversity and a spectrum of maize lines to address modern challenges in agriculture including production needs, quality, and health. Landraces of maize grown before the genetic bottleneck of modern breeding offer exciting opportunities to look for traits that can be valuable to markets that use maize as a raw material. My research focuses on one such market: the craft distilling industry that primarily utilizes maize as the main ingredient. Using maize landraces, I hypothesize that there is more metabolomic diversity in the landraces compared to yellow-dent corn that are associated with flavor, and that there are genetic controls to these metabolites that we can select for future breeding efforts. These hypotheses will be addressed through two objectives:1. Using untargeted metabolomics approaches, I will characterize the metabolite composition of two landraces with known, distinct taste profiles in whiskey.2. Using Quantitative Trait Loci (QTL) mapping methods, I will generate a population derived from the landrace parents and map the genomic regions associated with these metabolites.Through this work, we will conduct foundational research on the identification of the genetic components of flavor metabolites in maize, all while producing lines better suited to the needs of the craft whiskey industry.
Project Methods
I will produce new-make whiskey from two landraces and explore their metabolites. The landraces will be diverse colored landraces already used for craft whiskey production: Bloody Butcher and Hopi Blue. When fermented and distilled commercially, these landraces present two dramatically different taste profiles, and I can qualitatively assess that there are flavor (and presumably metabolite) differences between the two. 100 g of seed is milled into a flour and added to a spinner flask along with water and α-amylase. After mixing, the solution (mash) cooks to gelatinize the starch. Once cooked, distiller's yeast is added to the mash with gluco-amylase. This addition allows for simultaneous conversion of dextrins to simple sugars to feed the yeast while not over-saturating the mash initially with simple sugars which can inhibit yeast. The mash ferments for 3 days and then is rapidly cooled. Once ready for distillation, the beer is added to a 1 L short path distillation column and heated to vaporize ethanol (and metabolites) and collected for analysis. This process is analogous to how distillers fractionate the volatiles and produce whiskey.To cover the full range of metabolites potentially present in the maize parent populations, I will use a two-pronged approach using both GC and LC. Using Solid Phase Micro-Extraction (SPME), which captures volatile metabolites.I will extract volatile molecules and inject the sample directly into the GC (Agilent 7890 GC QToF Mass Spectrometer (MS)) for analysis [16-17]. For LC, I will inject 1 mL of sample into the LC (Bruker Impact II QToF MS) for quantification. I will use known quantities of internal standards provided by the MU Metabolomics core to quantify each metabolite.I will use 3 technical replicates for both the GC/MS and LC/MS to ensure consistency within the batch and to eliminate outlier data. I will then use analysis of variance (ANOVA) to look for significant differences between landraces. Once I have each sample analyzed, I will compare the data to an established NIST metabolite library and look for metabolites correlated with flavors using a whiskey flavor lexicon. I will genotype each F2 individual by identifying single nucleotide polymorphisms (SNPs) throughout the genome. Genomic DNA will be extracted from leaf samples using a CTAB DNA Miniprep. I will then genotype the population using the Illumina Maize SNP50k genotyping array, which provides sufficient coverage of the maize genome to study SNP diversity while at the same time, being significantly cheaper and faster for our experiments compared to low-depth whole genome sequencing approaches. In tandem with the genotyping array, I will phenotype to identify and quantify metabolites extracted from pooled seed derived from each of the F3 families. I will produce new-make whiskey from F3 family seed (see above) and target the metabolite(s) identified previously (see above) associated with flavor. As a cheaper alternative to metabolomic assays, I may be able to use an enzymatic or spectrophotometric assay to quantify the target molecule(s).For mapping metabolite traits, I will use composite interval mapping (CIM) to map interval regions between SNP markers and look for significant correlations, while also accounting for unlinked marker effects (cofactors) to screen for false positives and increase the accuracy of QTL identification. To perform this analysis, I will first plot our phenotypic data as-is, then I will conduct a permutation analysis where I randomize the phenotypic data and repeat the same analysis 1000 times. This randomization will form a null distribution that I can then plot and apply a significant p-value cutoff for accepting false positives (0.05) and that cutoff will determine the LOD threshold that I will use to determine if a QTL is statistically significant. Once I have identified significant regions, I will use the B73 version 5 maize genome assembly to look for genes within the regions that may be promising candidates that explain the differences in metabolite quantity I see in our QTL population. Afterwards, one could utilize fine-mapping approaches to narrow down QTL regions to find genes associated with our phenotype, but this is likely due to the 3-year timeline of my proposal. For evaluation, I will work with Dr. Flint-Garcia and Dr. Braun to evaluate my objectives and goals using the IDP and weekly progress meetings. This process will provide a wide array of feedback from both my primary and collaborating mentors with different perspectives and will facilitate my research success. I will also present my findings at local and national conferences through poster presentations, the MU Plant Talks series, and at least one scientific talk to communicate the significance of my research to the scientific community. I will also give one talk to the broader public to explain the significance of my research. These events will solidify my graduate experience and education, and will prepare me for a career as a research scientist in plant breeding and quantitative genetics.