A Genome-wide Association Study of the Quantitative Resistance to Striga hermonthica and Plant Architecture of Sorghum bicolor in Northwestern Ethiopia

2019-11-20T16:23:53Z (GMT) by Megan E Khangura

Sorghum (Sorghum bicolor) is a well-known agronomic crop of global importance. The demand for sorghum as a food crop makes it the fifth most important cereal in the world. The grain of sorghum is utilized for food and feed, whereas the sorghum biomass may have many other uses such as for fodder, bioenergy or even for construction. Globally, sorghum is consumed as a food crop and used for home construction primarily in the developing world. The grain and biomass yield of sorghum is drastically reduced by the parasitic plant Striga hermonthica which is endemic to Sub-Saharan Africa. To date, only one sorghum gene, LGS1, has been characterized as a genetic mechanism that reduces S. hermonthica parasitism by altering the strigolactone composition of the host root exudates which results in a reduction of the parasites ability to germinate. To establish more durable resistance additional genetic variation needs to be identified that reduces the S. hermonthica parasitism in sorghum, but also reduces the parasitic weed seed bank by promoting suicidal germination. To that end, the PP37 multi-parent advanced generation inter-cross (MAGIC) population was developed, originally as a recurrent selection population that was developed to recombine sorghum accessions with different putative resistance mechanisms to S. hermonthica. Whole genome sequences were developed for approximately 1,006 individuals of the PP37 MAGIC population. The population was phenotyped for S. hermonthica resistance during the 2016 and 2017 growing season in Northwestern Ethiopia. There was significant spatial variation in the S. hermonthica natural infestations that were partially attenuated for with artificial inoculation. The data was used to conduct a genome-wide association study that detected several subthreshold peaks, including the previously mapped LGS1. The highly quantitative nature of S. hermonthica resistance confounded with the complex spatial variation in the parasite infestations across a given location make it difficult to detect highly heritable variation across years and environments.

In addition to S. hermonthica resistance, the plant architecture of the PP37 MAGIC was also assessed at a location in Northwestern Ethiopia that is free of the parasite, as it significantly reduces plant height. To asses plant architecture the total plant height, the height of the panicle base, flag leaf height, and pre-flag leaf height were collected using a relatively high-throughput barcoded measurement system. Sorghum head exertion and panicle length were derived from this data. The actual measures of plant architecture and the derived traits were used to conduct a genome-wide association study. The high heritability of this trait demonstrated the statistical power of the PP37 mapping population. Highly significant peaks were detected that resolved the dwarf3 locus and an uncharacterized qHT7.1 that had only been previously resolved using a recombinant inbred line population. Furthermore, a novel significant locus was associated with exertion on chromosome 1. The random mating that was utilized to develop the PP37 MAGIC has broken the population structure that when present can hinder our ability associate regions of the genome to a given phenotype. As a result, novel candidate gene lists have been developed as an outcome of this research that refined the potential genes that need to be explored to validate qHT7.1 and the novel association on chromosome 1.

This research demonstrated the power of MAGIC populations in determining the genomic regions that influence complex phenotypes, that facilitates future work in sorghum genetic improvement through plant breeding. This research however also demonstrates a large international research effort. The nuisances and lessons learned while conducting this international research project are also discussed to help facilitate and guide similar research projects in the future. The broader impacts of this research on the society at large are also discussed, to highlight the unique potential broader impacts of international research in the plant sciences. The broader impacts of this research include germplasm development and extensive human capacity building in plant breeding genetics for developing country students and aspiring scientists. Overall this research attempts to serve as a model for highlighting the interdisciplinary nature and complexity of conducting international plant science research, while also making significant strides in improving our understanding the genetic architecture of quantitative traits of agronomic importance in sorghum.