| Common wheat (Triticum aestivum L.) is one of the most important and the oldest crops in the world. The vascular bundle plays an important role in transportation of photosynthetic products, mineral nutrients, and water. The number, size, and capacity of vascular bundle influence the transportation ability, especially for the photosynthetic products. The morphological and anatomical traits of wheat were correlated with many grain yield and spike related traits, generally controlled by multiple genes. With conventional breeding methods,the selection efficiency of these traits was low. The advent of molecular markers and QTL (quantitative trait loci) mapping has made it possible to localize individual genes for this type trait. With the development of molecular genetics and genomics, location, separation and exploring QTL have become an important and essential component element in crop genetic improvement. The work will not only prepare gene resources for molecular marker assisted selection and molecular breeding by design, but also lay a foundation for fine mapping and clone some important genes. In the present study, we report a new genetic linkage map developed from an F1-derived doubled haploid (DH) population of 168 lines, which was generated from the cross between two elite Chinese common wheat varieties Huapei 3 and Yumai 57. QTL analyses were performed using the software of QTLNetwork 2.0 based on the mixed linear model approach. QTLs for 8 traits including uppermost internode length (UIL), uppermost internode diameter (UID), culm wall thickness (CWT), culm wall area (CWA), the number of total vascular bundles (TVB), the number of large vascular bundles (LVB), the number of small vascular bundles (SVB), and the ratio of large vs. small vascular bundles (L/S) were analyzed. The results were as the following:1. A population of 168 DH lines was produced by anther culture from the cross between two Chinese common wheat varieties Huapei 3 (female parent)×Yumai 57 (male parent). Genetics analysis of quantitative traits were carried out by using the investigated results in four environment: normal irrigated and topdressed urea in 2007 (environment I), normal irrigated and topdressed urea in 2008 (environment II); normal irrigated but no topdressed urea in 2008 (environment III); no irrigated but topdressed urea in 2008 (environment IV). Analysis the variance of difference between the parental groups on 8 traits in the experiments showed significant difference among the uppermost internode length, the number of total vascular bundles, and the number of smal vascular bundles. Analysis the frequency distribution of each phenotype trait in DH population by SPSS,the test showed that all traits approximately fit normal distributions. It was shown that the DH population can be used for the construction of the genetic map and QTL mapping.2. A genetic linkage map containing 324 SSR markers, using MAPMAKER/Exp version 3.0 software, was finally developed. The map covered a total length of 2485.7 cM with an average distance of 7.67 cM between adjacent markers in the map, which resulted in 24 linkage groups comprising 3 to 24 loci. Each of the linkage groups could be assigned to one of the 21 chromosomes based on the micro-satellite consensus map. This work found some large bins of co-segregating markers near the centromeric regions, which are known to have suppressed recombination. These unresolved regions may provide access to a high density of markers and thereby increase the chance of finding polymorphism among different germplasm materials. It will be interesting to see if the markers in the centromeric regions found in our study would be useful in detecting and tagging QTLs in further research. A further agreement between our work and similar studies is the observation of segregation distortion of certain markers.3. Based on the genetic map established from the DH population, the QTLs were detected using the software QTLNetwork version 2.0 with the composite interval mapping of the mixture linear model. QTLs for 8 traits including uppermost internode length, vascular bundles and related morphological traits and anatomical traits were analyzed in 2 croping seasons and 4 environments. A total of 20 additive QTLs and 1 pair of epistatic effects were detected and distributed on 10 chromosomes1A, 1B, 2A, 2D, 3D, 4D, 5D, 6A, 6D and 7D. Among them, 12 additive QTLs were major genes, while 8 additive QTLs were minor genes. The multigenic effects for vascular bundle system and correlative traits of uppermost internode obviously exhibited on chromosomes 1A, 2D, 4D, 5D and 7D, and then additive effects and epistatic interactions were also sometimes subject to environmental susceptibility:3.1 The major QTLs control TVB and SVB were mapped on chromosomes 1A and 2D under environment III and IV. It may be indicate that the genes has a common hereditary background with resistance;3.2 It was observed that the QTLs for uppermost internode length, plant height (explaining 20.22% of the phenotypic variance), and lodging resistance of minor genes all mapped in the same interval XBARC334-XWMC331 on chromosome 4D;3.3 The interval of XWMC215-XBARC345 on chromosome 5D gathered UID, CWA, TVB, LVB, the ratio of L/S in environment I, near the interval with the same direction of additive effects for grain yield, grains per spike, spikelets per spike, compactness, and fertile spikelets per spike;3.4 The interval of XGWM676-XGWM437 on chromosome 7D for UIL, TVB, and SVB under four different environments, which means the main effect QTLs are stable and can be used in marker assistant selection (MAS).3.5 One pair of significant epistasis was identified for TVB under environment III. The QTL pair, one located on chromosome 2A and another on chromosome 6D, accounted for 10.81% of the observed phenotypic variation and it has not any significant additive effects of their own but involved in epistatic interactions. Such loci might play the role of modifying agents that tend to activate other loci or modify the action of other loci. |
