| Plant metabolites are critically important for plant development, environmental adaptation and human nutrition. Rice is a staple food crop in the world and is also a significant model plant for functional genomic research. Identification of the genes involved in metabolism and dissection of the metabolic pathway are of great help to improve plant adaptation to environmental stresses, to improve food quality, and to increase crop yield.Analysis of metabolic quantitative trait loci(m QTLs) that treats the metabolites’ levels as quantitative traits(m-traits) to perform QTL analysis demonstrated a powerful strategy for identifying the genetic loci controlling the variation of the corresponding metabolites. Samples of different tissues and different stages from 210 RILs derived from a cross between two elite indica rice varieties, Zhenshan 97(ZS97) and Minghui 63(MH63) were collected. A genetic analysis of rice metabolome combining a newly developed widely targeted metabolic profiling method with a recombinant bin-based ultrahigh-density genetic map was performed, revealing a number of m QTLs with high resolution and large effects. Metabolic profiling of an introgression line(IL) population was carried out to evaluate the quality of m QTL results. We associated candidate genes to various m QTLs by data mining and validated the function of several genes in vivo. The corresponding pathways were updated and reconstructed by integrating previously identified and newly assigned genes. The main results are as follows: 1. Validating of the metabolic profiling methodUsing a liquid chromatography-tandem mass spectrometry(LC-MS/MS)-based, widely targeted method developed by our lab, metabolic profiling was performed in the parental lines ZS97, MH63 and other germplasm resource accessions. The results showed that the method was effective in cluster analysis of rice varieties with the high coverage, sensitivity, and accuracy. So the method could be used in large-scale detection, identification, and quantification of metabolites in rice population.2. m QTL mapping in different rice tissuesA total of 1,000 metabolites including 683 in flag leaf at heading date and 317 in seed at 72 h after germination was generated, with 100 metabolites were detected in both tissues. Based on the ultrahigh-density genetic map using 1,619 bins, m QTL mapping resulted in 1,884 and 937 m QTLs in flag leaf and germinating seed, respectively. Both tissues were mapped to some m QTLs with narrow confidence interval and/or large effects. Distinct and overlapping accumulation patterns of metabolites were observed and complex genetic regulation of metabolism was revealed in two different tissues by analysis the co-detected metabolites and their m QTLs.Genome-wide analysis of m QTLs revealed a significant deviation from random distribution across the 12 chromosomes. The m QTL enriched regions were mainly located on chromosomes 1, 6, 7, and 10 in flag leaf, but on chromosomes 5 and 6 in germinating seed. We identified 44 and 16 potential m QTL “hot spots” in flag leaf and germinating seed, respectively. The pairwise epistatic interactions between the m QTL hotspots against the average accumulation of known metabolites within the RILs were calculated, suggesting a large number of significant interactions(P < 0.01).Using an IL population in which marker-defined genomic regions of ZS97 were replaced with homologous intervals of MH63, 64 selected metabolites was detected. Fifty of the 64 major m QTLs detected in RILs were confirmed in ILs that showed the expected variation for both the direction and amplitude of the variation. The chemical structure of the metabolites, the existing knowledge of the pathway architecture, the annotated genome sequence, together with the gene expression profiling allowed the tentative assignment of a protein or protein cluster as regulating the m-traits. Candidate gene mining of m QTLs allowed the tentative assignment of function to 24 candidate genes(9 unreported genes), including ones regulating important morphological traits and biological processes. The corresponding pathways were reconstructed by updating in vivo functions of previously identified and newly assigned genes. 3. m QTL mapping in different rice stagesA widely targeted metabolic profiling was performed in three stages of rice seed. A total of 810, 836 and 855 metabolites were detected in mature, germinating, and grain-filling seeds, respectively, with 372 metabolites co-detected in three stages. For more than a half of the metabolites in each stage, the observed coefficients of variation were greater than 50%, indicating the tremendous variation of metabolomes in different stages in rice. We identified 1,506, 1,575 and 1,600 m QTLs, and detected 42, 39 and 39 m QTL “hot spots” in mature, germinating and grain-filling seed, respectively.Comparative analysis of the co-detected 372 metabolites in three stages suggested that each kind of metabolites showed different correlations in different stages. For the 372 m-traits, 746, 692 and 629 m QTLs were identified for 303, 311 and 290 metabolites in mature, germinating and grain-filling seed, respectively. However, only 57 m QTLs were coincident in three stages. A great diversity of m QTLs for the same m-trait in three stages was observed. For example, 95 metabolites were mapped to totally different m QTLs in three stages, indicating high stage specificity of rice seed metabolome. Data mining identified 33 candidate genes modulating levels of metabolites that are of potential physiological and nutritional importance. 4. Cloning of metabolite related genesWhen the two malonyltransferases genes(reported Os Ma T-2 and newly identified Os Ma T-3) were overexpressed, the content of m0723-L was substantially increased in both the Os Ma T-2 and Os Ma T-3 overexpressors compared with the control, indicating that they both function as flavone malonyltransferase in vivo. The results in overexpressors of an uridinediphosphate glucosyl transferase(UGT) gene LOC_Os11g26950 was in accordance with the m QTL result of m0760-L.An unknown metabolite m0434-L(m/z 427/397) was significantly mapped to a coincident locus in the RIL population and a natural population. Comparison of the sequences of the candidate gene LOC_Os03g25500 and the accumulation of m0434-L in gerplasm resources revealed a functional SNP(GAG for Glu in MH63 to TAG for stop codon in ZS97), which caused premature termination of LOC_Os03g25500 in ZS97 compared with MH63. Further studies are necessary for identification of the chemical structure of m0434-L and the function of LOC_Os03g25500.A minor m QTL of m0447-L had the most significant effect(P = 4.6 × 10-33) in genome-wide associated study results in a natural population. When the candidate gene LOC_Os07g32060 was over expressed, we detected up to a 200-fold increase in the levels of flavone 5-O-glucosides but no increase in flavone 7-O-glucosides, strongly supporting the notion that LOC_Os07g32060 functions as a flavone 5-O-glucosyltranferase underlying the levels of flavone 5-O-glucosides.These results enhanced our understanding of the genetic and biochemical basis of the rice metabolome, helped bridge the gap between the genome and phenome, and provided new resources and directions for rice genetic improvement. |
PDF Full Text Request