Screening the IRRI mutant rice population for mutations that have altered the structure of endosperm starch

Abstract

The International Rice Research Institute (IRRI) maintains a population of mutants based on IR64 one of the most popular varieties throughout Asia. The popularity of IR64 is due to its quality. IR64 is a low yielding relative to newer varieties, and its yield is further compromised by its susceptibility to many biotic stresses. Nevertheless, it has maintained its fame because the grain it produces is of very high quality, and it is difficult to capture this in higher-yielding backgrounds. IR64 produces long, slender, translucent grains, it has intermediate amylase and gelatinisation temperature, and it does not become firm on cooling. These are the traits that are valued by almost every consumer and trader of indica rice. The key to understanding the quality of IR64, and genetic factors of intermediate amylase and gelatinisation temperature will be to find the genes that confer that quality and develop ways to select for them. The mutant population provides an excellent resource to achieve this. Unravelling genetic processes defining the quality of IR64 is useful to rice improvement programs around the world, since all select for different metrics of the same traits of quality, and many select for intermediate amylase and gelatinisation temperature. The aims of this project are to (i) screen a subset of the collection of IR64 mutants to determine the structures of starch that associate with the quality of IR64; (ii) identify candidate genes that synthesise directly, or indirectly, the particular structures of starch; and (iii) search for allelic variation in the genes that could explain the phenotype of the mutation. One IR64 mutant was found with decreased amylase content and gelatinisation temperature, and other cooking properties were significantly different from the IR64 wildtype. The mutant was severely depleted in the long chains of amylase that leach out of the granule in hot water, and was depleted in the long chains of amylopectin that strengthen the molecule by spanning several clusters. Differences in amylase and amylopectin between the various phenotypes of the mutant studied and the wildtype suggest that this mutant carries a mutation in two genes. Candidates for each phenotype are granule bound starch synthase (GBSSI) and starch synthase Illa (SSIIIa). The conserved region of SSIIIa, and exon 2 of GBSSI, where transcription begins, did not carry any mutation, but other regions of each gene were not genotyped. Paradoxically, a high level of GBSSI protein accumulated in the mutant. The other starch synthases are post-translationally regulated by starch phosphorylase. Starch phosphorylase could also be involved in post-translational regulation of GBSSI and SSIIIa activity. In the mutant, starch phosphorylase was found in the soluble fraction, along with many other granule proteins, but it and the other proteins were absent from the soluble fraction of granule proteins, thus the mutation could reside in a gene encoding a protein that regulates the activity of these two starch synthases, both of which elongate relatively long chains on their respective molecules. This mutation changes eating quality enormously, and the structural studies here demonstrate the huge effect of long chains of amylase and long chains of amylopectin on the excellent quality of IR64.