The genetic controls of plant architecture
In our paper recently published in G3, we used a Sorghum Recombinant Inbred Line (RIL) population derived from a cross between domesticated S. bicolor and a wild and weedy relative, S. propinquum, to study the genetic underpinnings of several plant traits with a particular focus on tillering (Govindarajulu et al, 2021). Tillers are new grass shoots that arise from the base of the plant after the initial parent shoot has become established. Tillering, along with several other phenotypes such as photoperiod sensitivity and height, were modified during plant domestication and improvement.
We collected trait information for 191 RILs from two field experiments (Stillwater, OK and Athens, GA) and one greenhouse experiment and performed quantitative trait locus (QTL) mapping by using low-coverage Illumina sequencing of each RIL and bin-mapping of the parental polymorphisms. We identified 30 QTL for 6 agronomic traits (height, stem diameter, biomass, flowering time, aerial branching, and tillering), many of which were penetrant across environments and co-mapped with major QTL identified in other studies. However, some QTL were environment specific.
The map position in cM is shown on the y-axis. Horizontal lines on each chromosome represent a single bin marker. Vertical colored lines show the locations of QTL, with the color for each trait indicated in the inset. From Govinadarajulu et al, 2021, G3.
We further complimented the QTL analysis with transcriptome analysis during early tiller bud elongation. We identified Dormancy Associated Protein 1 (DRM1) and various transcription factors that are differentially expressed between dormant and elongating buds and also lie within our QTL windows, suggesting that they play key regulatory roles in tiller bud elongation in sorghum.
We are currently using the S. bicolor x S. propinquum RIL population to investigate the genetic underpinnings of root system architecture. In this project, we are using high-throughput imaging in order to reduce errors in measurement related to temporal and developmental differences across a large sample size. Traits such as emerging crown root angle, length, convex area, average diameter, and surface area of whole root systems can be obtained from the image analysis software. These data will be used for QTL mapping to identify candidate loci that control the traits of interest. We will compare the results with our previous work related to aboveground architecture, described above, to identify shared and distinct genetic pathways that are involved in branching.