Mapping Metabolic QTL and Constructing Metabolic Network in Perennial Ryegrass (Lolium Perenne L.)
|School||Nanjing Agricultural College|
|Course||Crop Genetics and Breeding|
|Keywords||four-way cross population metabolic QTL empirical Bayesian metabolic network perennial ryegrass|
As one important food for herbivore, perennial ryegrass is most widely planted in the temperate zone in the world. The perennial ryegrass produces alkaloid, some alkaloids are useful, i.e., peramine, and some ones are harmful, i.e., ergovaline. The the detection of quantitative trait loci (QTL) for these alkaloids (metabolites) can provide important information for marker assisted selection.In this study 136 individuals in the four-way cross population derived from the cross between two commercially heterozygous cultivars, Grasslands Impact and Grasslands Samson, in perennial ryegrass (Lolium perenne L.) was used to scan SSR marker information on the genome and measure 650 metabolites in 2005 and 2006 at AgResearch Grasslands, Palmerston North, New Zealand. The marker information is used to construct genetic linkage maps using JoinMap 3.0. The phenotypic observations for 650 metabolites along with marker information and linkage maps are used to detect QTL for each of the metabolites using empirical-Bayes-based multi-QTL joint analysis. The phenotypic values for 650 metabolites are used to construct metabolic network using empirical Bayesian approach. The main results were as follows.1. One hundred and eighty-one SSR markers were mapped into 7 linkage groups, with the 90 to 148 cM length and the 20 to 34 markers on each linkage group. All the markers were uniformly distributed on the genome. The total length of the linkage map was 814 cM, with an average marker spacing of 4.50 cM. The adjacent markers with maximum (33.9 cM) and minimum (0.1 cM) interval lengths are placed on the seventh and fourth linkage groups, respectively.2. With the critical LOD score of 2.5,1861 QTL were identified in 474 metabolites out of 650 metabolites using the multi-QTL joint analysis, including 84 environmental effects,448 main-effect QTL and 1329 environment-by-QTL interactions. The total proportion of phenotypic variance explained by all QTL detected for each metabolite ranged from 0.37% to 78.42%.At the significant level of 0.1%, there were eighteen QTL hotspots in the genome except the fifth and sixth linkage groups, including markers pps0969 and rv0654 on linkage group 1, markers pps0490, pps0395, pps0420 and pps0123 on linkage group 2, markers pps577, pps198, pps0133, pps0502.2, pps0061 and rv1152 on linkage group 3, markers pps0761, pps0312, pps0937 and rv0691 on linkage group 4, and markers pps0766.20 and pps0817 on linkage group 7.Two metabolites shared common QTL had a significant correlation.3. Six networks were constructed using empirical Bayesian approach. The environment has less impact on metabolic network, suggesting that metabolic network is conservative in different environments.