Understanding the genetics and chemistry of important quality traits in rice (Oryza sativa L.)

Abstract

Rice grain quality is a determinant of rice breeding success. Although several studies have independently looked at the different traits of quality to infer the phenotype of the grain, few studies have identified the various metabolites that are produced in high quality rice. In recent years, genetic markers have proven to be effective in increasing the efficiency of breeding line selection and therefore shortening the breeding cycle. In complex traits such as rice grain quality, the availability of more specific phenotypes will increase the value of these genetic markers. Many traits of quality in rice are associated with starch; however, lipids interact with starch and have been shown to influence many of the traits of eating quality and aroma. To identify these new phenotypes and understand their genetic basis, using relevant tools that can discriminate between the phenotypes of varieties at the metabolite and genetic levels is necessary.To identify these new phenotypes, in Chapter 3, a set of 40 Cambodian rice varieties were screened for physical and texture-associated traits including apparent amylose content (AAC), gelatinisation temperature (GT), gel consistency (GelCon) and pasting properties and then a subset from this set was screened for volatile compounds using a two-dimensional gas chromatography-time-of-flight-mass spectrometer (GC×GC-TOF-MS) and fatty acids (FA) using gas chromatography-mass spectrometry (GC-MS). The FA analysis showed that unsaturated FA (UFA) oleic (C18:1n-9) and linoleic acid (C18:2n-6) were the most abundant in milled rice grains, followed by palmitic acid (C16:0). These results infer about the lipid origin of odour-active volatile compounds, which were found to be characterising the group of known fragrant, low AAC and soft-textured rice varieties, including the Jasmine-style indica variety Phka Rumduol (PRD).To identify the genetic determinants of these important traits of quality, in Chapters 4 and 5, quantitative trait loci (QTL) mapping was carried out for the texture and aroma traits in PRD using a population of about 300 F6 recombinant-inbred lines (RIL) derived from an intraspecific cross between PRD and Thmar Krem (TMK), both of which were identified as the most discriminating varieties in Chapter 3. This RIL population was genotyped for single nucleotide polymorphisms (SNPs). QTL analysis revealed large-effect QTLs for AAC, GT and all of the pasting properties. Candidate genes for these QTLs relate back to the starch biosynthesis pathways. Major QTLs were colocalised to the Waxy (Wx) gene and starch synthase IIa (SSIIa) on chromosome 6. Minor QTLs on chromosomes 3 and 8 were identified for AAC, SB, PV and HPC, and at chromosomes 4 and 7 for GT and pasting temperature (PTemp). This Chapter has shown that the rice texture is regulated by multiple genes and that although correlations do exist among the routinely measured textural traits, these correlations may not necessarily translate into the already existing molecular markers.In Chapter 5, QTLs for odour-active volatile compounds and FA were carried out on the RIL population. The presence of 2AP and its derivatives significantly characterised the fragrant lines from the non-fragrant ones and this was supported by the identification of large-effect QTLs pointing to the region of the fragrance gene (FGR). Odour-active and low odour threshold volatile compounds that can be derived from the oxidation of UFA were identified as segregating traits in the population as supported by the presence of significant SNPs on rice chromosome 6. These volatile compounds explain the differences in varieties that cannot be differentiated by the presence or absence of 2AP alone. To verify that these volatile compounds can indeed be derived from UFA, the % FA composition in the population was also determined. Here, that C18:1n-9, C18:2n-6 and C16:0 were the identified as the most abundant FA, coinciding with the results observed in Chapter 3. PRD had higher levels of C18:1n-9 and TMK had higher levels of C18:2n-6, a difference which was exhibited in a normal distribution across the population. QTL analyses revealed a common QTL on rice chromosome 7 explaining the variation in % composition of these two UFA. This QTL encompassed the regions where fatty acid biosynthesis genes were located.In order to trace back the origins of the most important FAs to their lipid molecule and to screen the most abundant lipids in the rice samples, in Chapter 6, rice lipidomics was carried out for the first time using a reversed-phase ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS). Results of this experiment showed that glycerolipids in the form of triacylglycerols (TAG) and diacylglycerols (DAG), and glycerophospholipids were the most abundant lipid features in milled rice. Statistical analyses showed strong statistical correlations between traits of quality and several lipid features, indicating a substantial influence that lipid play in determining rice quality.The information in this research will be delivered to rice improvement programs to enable breeders to select more accurately for particular traits of quality, and for novel quality traits identified in this study.