Construction Of High-density Genetic Map And Location Of QTLs For Smut In Saccharum Spontaneum L

Sugarcane (Saccharum officinarum L.) is an important sugar crop in the world. Breeding new varieties with high yield, high sucrose content and highly resistance is still the main goal in sugarcane breeding. However, genetic improvement of these traits is extremely difficult because most of traits are controlled by quantitative traits loci (QTL). In addition, conventional breeding in sugarcane is inefficient due to its complex genetic background and difficulty or rarelity in flowering. Saccharum spontaneum L. is the major wild species extensively used in sugarcane genetic improvement and cultivar development worldwide for its resistance to abiotic and biotic stresses. Genetic linkage maps construction and QTL mapping analysis in S. spontaneum provide an abundance of DNA marker-traits associations, what’s more, the molecular markers which linked these traits have a huge application potential to improve the efficiency and accuracy of conventional breeding via marker-assisted selection (MAS). In this study,157 F1 progenies derived from the cross GXS85-30 (S. spontaneum) xGXS87-16 (S. spontaneum) as mapping populations. Molecular genetic linkage maps based on SSR and AFLP markers were constructed for GXS85-30 and GXS87-16 respectively. The QTLs controlling smut were detected in the parental map by single marker analysis (SMA) and composite interval mapping (CIM) method. The main results are as follows.1. An efficient SSR and AFLP markers technical system has been established, the efficiencies for detecting polymorphism markers and single-dose markers using SSR-CE were 1.9 and 1.47 times, respectively, as much as using SSR-PAGE, and the efficiency for detecting polymorphic markers and single-dose markers using AFLP-CE was 2.5 and 3.9 times, respectively, as high as using SSR-PAGE. SSR/AFLP-CE especially the AFLP-CE are highly efficient in detecting marker locus, and will be beneficial to study high-density genetic map construction and genetic diversity.2. The molecular genetic maps of GXS85-30 (female) and GXS87-16 (male) were successfully constructed with JoinMap 4.0 software. The GXS85-30 molecular genetic map contains 72 linkage groups and 701 loci, and covers a total length of 4656.20 cM with an average distance between markers of 6.64 cM and an estimated genome size of 7141.69 cM; the map covers approximately 65.20% of the genome, and the linkage groups (LGs) include 8 homologous groups (HGs) based on the information from locus-specific SSR so the material GXS85-30 should be a homologous octoploid. The GXS87-16 molecular genetic map contains 650 loci and 72 linkage groups, and covers a total length of 4539.34 cM with an average distance between markers of 6.98 cM and an estimated genome size of 5632.69 cM; the map covers approximately 80.59% of the genome, and the LGs include 8 homologous groups (HGs) based on information from locus-specific SSR and speculate so the material GXS87-16 should be a homologous octoploid. The two maps are qualified for QTL mapping because the distance between markers is all lowere than 7 cM.3. Comparing the mapping efficiency between SSR/AFLP-CE markers and SSR/AFLP-PAGE markers, the mapping efficiency of SSR-CE markers is 2.27 times as high as that of SSR-PAGE markers, and the mapping efficiency of AFLP-CE markers is 3.72 times as high as that of AFLP-PAGE markers, while the mapping efficiency of AFLP-CE markers is 10.6% higher than SSR-CE markers. So SSR/AFLP-CE especially the AFLP-CE has much higher efficiency than SSR/AFLP-PAGE in mapping, indicating that AFLP marker is a better choice for constructing high-density genetic linkage map.4. Single marker analysis (SMA) was used for QTL analysis of smut, a total of 130 molecular markers were detected to be linked to the smut QTL (72 came from female GXS85-30, and 58 came from male GXS87-16), and five of the markers showed contribution rate over 10%, which are correlated with smut QTL at 0.1% significant level. They are Aatcga37 (12.53%), Actcaal6 (11.53%) and Attccal (10.8%) from the female parent GXS85-30, and mSSCIR17-b (12.73%) and Aatcga4 (11.29%) from the male parent GXS87-16.5. QTLs for smut were detected with software Windows QTL Cartographer 2.5 following composite interval mapping (CIM) method (threshold value of LOD>3.0). Two major QTLs and 19 micro QTLs were detected controlling smut on the GXS85-30 map. An individual QTL explained 0.43%-15.99% of the phenotypic variation. Of which,9 QTLs showed positive additive effects of QTLs which came from GXS85-30, whereas 12 QTLs showed negative additive effects of QTLs which came from GXS87-16. Two major QTLs qSmut23 and qSmut37 were found to be co-separated with Attccal and Attcaa18, respectively, and each of them explained 10.81% and 15.99% of the phenotypic variations, respectively.6. Composite interval mapping (CIM) method was used for smut QTL mapping on the GXS87-16 map. Two major QTLs and 32 micro QTLs were detected controlling smut disease, and an individual QTL explained 0.25%-12.10% of the phenotypic variation. Of which,15 QTLs showed positive additive effects of QTLs which came from GXS87-16, whereas 19 QTLs showed negative effects of QTLs which came from GXS85-30. The major QTL qSmut39-3 was found to be 0.2 cM separated with the AFLP marker Aatcga4, which explained 12.10% of the phenotypic variation. The major QTL qSmut41 was found to be co-separated with the SSR marker mSSCIR17-b, which explained 10.65% of the phenotypic variation.