Item No. 1 of 1

INVESTIGATOR: Rhodes, D. H.

PERFORMING INSTITUTION:
AGRICULTURAL RESEARCH SERVICE
BROOKINGS, SOUTH DAKOTA 57006

DEVELOPMENT OF MOLECULAR BREEDING RESOURCES FOR INCREASED GRAIN CAROTENOIDS IN SORGHUM

NON-TECHNICAL SUMMARY: Carotenoids are red, yellow, and orange plant pigments-such as the orange beta-carotene in sweet potatoes and the red lycopene in tomatoes-that are involved in photosynthesis and protection against oxidative stress. In humans, dietary carotenoids act as precursors to vitamin A, as well as antioxidant and anti-inflammatory compounds, protecting against a myriad of diseases, including cancer, cardiovascular disease, and age-related macular degeneration. Cereals are generally poor sources of carotenoids compared to fruits and vegetables, but since cereals make up the majority of the human diet an increase in cereal carotenoids could have a significant impact on human nutrition. Sorghum is one of the world's largest produced cereal crops and a staple food for millions of people in semi-arid regions of sub-Saharan Africa and Asia. In the United States, it is primarily used in animal feed, but is increasingly used in specialty food products, especially those that are gluten free. Efforts to improve the nutritional quality of sorghum grain have relied primarily on traditional plant breeding, a process that can take years before a line is ready to be released. Marker-assisted breeding (the use of genetic markers to select for a trait of interest in crop breeding) can vastly accelerate the development of biofortified crops.The ultimate goal of this research is to provide genetic tools that will help breeders rapidly develop high-carotenoid sorghum varieties. To this end, we will quantify grain carotenoid concentrations in hundreds of sorghum lines using high-performance liquid chromatography (HPLC), a gold-standard technique for quantifying compounds. Since HPLC is a time-consuming and expensive technique, we will develop a high-throughput method for measuring carotenoid concentrations using near-infrared spectroscopy, a technique that predicts the concentration of a compound by reflecting near-infrared light on the whole grain. We will identify the genomic regions containing genes that control carotenoid concentrations using a genome-wide association study (GWAS), a method that scans the genome for associations between carotenoid concentrations and variations in single nucleotides. Finally, we will identify the nucleotide variations responsible for carotenoid variation by sequencing candidate genes that were identified through GWAS. The allelic variants can then be used as markers for marker-assisted breeding to develop high-carotenoid sorghum varieties that will benefit human health.

OBJECTIVES: The major goals of this project are to find new sources of high carotenoid sorghum and develop carotenoid genetic markers for usein marker-assisted breeding. Objectives are as follows:Characterize the natural variation in sorghum grain carotenoids in a largediverse panel using HPLC and colorimetry.Develop a high-throughput method of measuring sorghum grain carotenoids bycreating NIRS equations to predict carotenoid content.Identify QTL underlying natural variation of sorghum grain carotenoids usingGWAS.Validate NIRS calibration equations by comparing HPLC and NIRS on abiparental breeding population.Confirm GWAS QTL by conducting linkage mapping on a biparental breedingpopulation.Identify functional allelic variants at candidate genes.

APPROACH: To characterize the natural variation in sorghum grain carotenoids, I will use the Sorghum Association Panel (SAP) and additional accessions added to ensure there will be enough carotenoid-containing sorghums in the study. The SAP is comprised of ~400 lines assembled with the intent for use in association mapping and is designed to include all major cultivated races and important U.S. breeding lines. Using the Germplasm Resources Information Network (GRIN), I selected 467 additional sorghum accessions based on presence of a yellow endosperm (i.e. carotenoid-containing) or yellow pericarp in the grain. There are 94 accessions that are listed in GRIN as photoperiod insensitive (will flower in temperate latitudes) and so can set seed within the growing season in the shorter days in Manhattan, KS. The remaining accessions are listed as either photoperiod sensitive (which will not flower in temperate latitudes) or there is no data on their response to photoperiod, so will be grown in longer days in a winter nursery in Puerto Rico. Six carotenoids--apocarotenal, zeaxanthin, beta-cryptoxanthin, beta-carotene, and alpha-carotene--will be identified and quantified using a high-performance liquid chromatography (HPLC) method already developed for sorghum carotenoids. Pericarp and endosperm color will be quantified by colorimetry and compared to HPLC values to determine if color is a good indicator of carotenoid content.To develop a high-throughput method of measuring sorghum grain carotenoids, prediction equations will be developed using software that correlates the HPLC and near-infrared spectroscopy (NIRS) carotenoid measurements and converts the NIRS reflectance spectra to estimates of carotenoid concentrations.To identify quantitative trait loci (QTL) underlying natural variation of sorghum grain carotenoids, I will perform genome-wide association studies (GWAS) to find candidate genes for further characterization. The 94 additional photoperiod insensitive accessions added to the SAP to create the carotenoid panel will be sequenced using a genotyping-by-sequencing (GBS) system for the Illumina next-generation sequencing platform to generate large numbers of SNPs for use in the GWAS. Genotypes for the accessions that are part of the SAP are already available. GWAS will be carried out using the statistical genetics package Genome Association and Prediction Integrated Tool (GAPIT) using mixed linear models with kinship to control for relatedness of the accessions in the panel.To determine the effectiveness of the NIRS calibration equations, HPLC and NIRS will be conducted on a biparental breeding population with far less genetic diversity than the sorghum association panel.This population was developed through a cross between the pollinator RTx430, a common breeding line developed in Texas, and Macia, a food-grade sorghum that is used in parts of sub-Saharan Africa as a staple food. This population is one family from the sorghum Nested Association Mapping (NAM) population, a resource being developed at Kansas State University for connecting genetics and breeding in sorghum. RTx430 has a yellow endosperm while Macia does not, so the biparental population is segregating for the carotenoid trait.To confirm GWAS QTL, I will conduct linkage mapping on the biparental breeding population. The SAP was developed to capture all the genetic diversity found in sorghum. To determine if the same QTL can be identified in a population that is similar to what a breeder would use, I will use the R/qtl package to conduct linkage mapping on the biparental population.To identify putative functional alleles for the high or low levels of carotenoids in sorghum seeds, allele specific sequencing of candidate genes will be performed. For the QTL that colocalize witha prioricandidate genes, I expect to design primers for 8-12 candidate genes within those peaks of association. I will use PCR to amplify the targeted regions and then sequence single-band PCR products to identify possible allelic variants.Milestones that will signify objectives are being met:completion of HPLC phenotyping for sorghum grain carotenoidssubmission of carotenoid genotype data to genotype databasecompletion of GWAS on carotenoid diversity panelcompletion of NIRS equation developmentcompletion of NIRS on biparental populationcompletion of biparental population linkage mappingidentification of allelic variantssubmission of manuscript(s)External evaluation of this project will be conducted by a plant breeding and genetics scientist at Cornell University. Internal evaluation of progress towards achieving the objectives will be conducted through weekly meetings with my mentors.

PROGRESS: 2017/02 TO 2019/01
Target Audience: Nothing Reported Changes/Problems:I have just accepted a faculty position, so must terminate the Fellowship early. I will continue to work on this project in my new position. What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest? Nothing Reported What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

IMPACT: 2017/02 TO 2019/01
What was accomplished under these goals? This project aimed to developgenetic tools to help breeders more rapidly develop high-carotenoid (pro-Vitamin A) sorghum varieties in order to alleviateVitamin A deficiency in semi-arid regions of sub-Saharan Africa and Asia where sorghum is a staple food. Although this project was terminated shortly after it was initiated due to a career advancement to a faculty position, Imade substantial progress on mymajor goals and amclose to developing a high-throughput method of measuring sorghum carotenoids and identifying genetic markers associated with carotenoid variation. The work that I have accomplished shows that the variation in sorghum grain carotenoids is significant enough to be able to develop tools to quickly predict the concentration of sorghum carotenoids (using NIRS) and to identify regions in the genomethat can be used to assist in breeding high carotenoid sorghum. The major goals of this project were to find new sources of high carotenoid sorghum and develop carotenoid genetic markers for usein marker-assisted breeding. Objectives were as follows: Characterize the natural variation in sorghum grain carotenoids in a largediverse panel using high performance liquid chromatography (HPLC) and colorimetry. Carotenoids have been quantitatively measured by HPLC in the whole grain flour of nearly 300 sorghum accessions from the sorghum diversity panel. There was significant variation in carotenoids between the accessions, with a 42-fold difference between the highest and lowest concentrations oflutein (0.27- 11.48ug/g), a 43-fold difference ofzeaxanthin (0.21- 9.08ug/g), a 68-fold difference ofbeta-cryptoxanthin (0.01- 0.81ug/g), and a 57-fold difference of beta-carotene (0.02- 0.87ug/g). The concentrations are low, but breeding could increase the concentrations to an amount that could improve health. Additionally, we grew 300 carotenoid-containing accessions in winter nursery, and measuring the carotenoid concentrations in these will give us a fuller picture of the carotenoid diversity in sorghum. Color differences in the whole grain flour of 263 samples have beencalculated using colorimetry. There was a positive significant correlation between yellowness and all carotenoids (data below), although the correlations are not high enough to be able to use colorimetry as a proxy for measuring carotenoid concentrations. It could be used as a binary indicator of high vs low carotenoid concentrations. Additional colorimetry testing on decorticated grain (to measure just the yellowness of the endosperm) would likely result in higher correlations with carotenoid concentrations. Lutein: r = 0.39; p < 0.001 Zeaxanthin: r = 0.37; p =0.002 beta-carotene: r = 0.33; p = 0.005 alpha-carotene: r = 0.43; p < 0.001 beta-cryptoxanthin: r = 0.34; p =0.003 Develop a high-throughput method of measuring sorghum grain carotenoids bycreating NIRS equations to predict carotenoid content. We are currently incorporating the newest HPLC data into the NIR equations Identify QTL underlying natural variation of sorghum grain carotenoids usingGWAS. Using the HPLC data from the first100 accessions that were quantified, GWAS identified several significant QTL with minor allele frequencies greater than 5%. Incorporating the subsequent accessions with HPLC data will improve the statistical power to identify QTL. Validate NIRS calibration equations by comparing HPLC and NIRS on abiparental breeding population. The biparental breeding population has been harvested, and their carotenoid concentrations will be predicted oncethe NIRS equations are complete. Confirm GWAS QTL by conducting linkage mapping on a biparental breedingpopulation. This will be done once we are able to gather the NIRS data on the biparental population. Identify functional allelic variants at candidate genes. Based on the initial GWAS, a gene candidate for zeaxanthin variation has been identified. Further experiments will be needed to confirm this.

PUBLICATIONS (not previously reported): 2017/02 TO 2019/01
No publications reported this period.