| The genetic linkage mapping and QTL analysis were conducted using a population ofRIL derived from“Shannong0431×Lumai21”(F7in2010). Of the original227RILs,177lines were selected at random to construct the genetic map. the main results are as followes:A total of740loci (617DArTs,118SSRs and5EST-SSRs) were mapped on21chromosomes across58linkage groups. The skeleton map consisted of606markers (483DArTs,118SSRs and5EST-SSRs), and the134co-segregating markers were all DArTs. Anumber of61SSRs of which were derived from the physical map information, showing agood colinearity to the physical maps. The final map spanned a total length of3724.5cM,with an average chromosome length of177.0cM. The map length were unequally dividedamong the three genomes:1254.7cM (33.7%),1841.3cM (49.4%) and628.5cM (16.9%)lengths and277(37.4%),307(41.5%) and156(21.1%) markers for the A, B and D genomes,respectively. The largest chromosome was1B (431.0cM), and the shortest was6D (19.2cM).The density of the markers on the maps ranged from2.3cM/marker on7D to14.6cM/markeron1D with an average density of6.1cM/marker， suggesting that it is suitable forgenome-wide QTL mapping.The field trials were conducted in six environments throughout Shandong Province,China: Heze2011(HZ11), Tai’an2010(TA10), Tai’an2011(TA11), Tai’an drought field2011(TAD11), Yantai2011(YT11), and Zibo2011(ZB11). For the RILs, the variance foreither genotype or environment effects of all of the seven investigated traits were significantat the p≤0.001level, indicating that the environments and genetic background were bothimportant in explaining the overall phenotypic variations, and continuous distributions werecommon for all of the traits in the six environments, indicating the quantitative nature ofpolygenic inheritance. The heritabilities (hB2) of the GL, GW, GLW, FFD TGW, SN, andGNS were76.9,51.0,69.6,34.6,72.0,32.1and33.5%, respectively, showing that with theexception of FFD, SN and GNS, the heritabilities were over50%. The simple correlation coefficients showed that the TGW had a significant positive correlation with the GL, GW andFFD but a significant negative correlation with the SN and GNS. The strong positivecorrelations were simultaneously obtained between the GL and GLW (r=0.756) and the GWand FFD (r=0.561). Significant negative correlations were found between the GLW andGW/FFD, GNS and GW/FFD/TGW, and the SN and the other traits (except for GLW). Thecorrelation between the GL and GW/FFD/GNS and the GLW and TGW/SN/GNS were notsignificant.A total of75QTLs were detected on19chromosomes (except for1D and3A) for all ofthe investigated traits in the six environments and AV. A single QTL in differentenvironments explained4.5–32.5%of the phenotypic variations. Of which,44QTLs showedpositive additive effects with Shannong0431increasing the effects of QTLs, whereas31QTLs were negative effects with Lumai21increasing the QTL effects. The highest LODvalue for a single QTL in the different environments was17.1for the TGW in TAD11. In thepresent study,47QTLs for grain size and28QTLs for yield components were identified.Twenty-two QTLs were detected in at least two of the six environments. Of these QTLs, five(QGl-6A, QGw-2D, QGw-5B, QGlw.2-5B and QSn-4A) were detected in three environments,two (QTgw-5B and QGns-4A) in four environments, and one (QGl-5B) in five environments,indicating that these QTLs were relatively stable.Despite the QTLs were distributed widely throughout the wheat genome, six clustersincluded three or more QTLs were identified on five chromosomes, specifically,2D,3B,4A,5B and6D. These clusters were related to all of the investigated traits and involved23QTLs(23/75×100%=30.7%). In general terms, the TGW QTLs were detected in five of the sixclusters and had significant positive correlations with the GL, GW and FFD, in agreementwith the results for the simple correlation coefficients. The strongest stable cluster, Cluster C4,in the marker region of wmc386-wPt-9467on chromosome5B involved QGl, QGlw andQTgw, which were detected in more than three environments and AV, accounting foraverages of16.2,7.8and16.1%of the phenotypic variations, respectively. The increasingeffects of the three QTLs all originated from Shannong0431, showing that Shannong0431had a locus of increased TGW, which resulted from increasing GL and GLW. The markersaround this locus should be useful for MAS to improve TGW in high-yield breeding programs.However, it is necessary to saturate the target regions by adding more molecular markers to conduct fine mapping. |
PDF Full Text Request