QTL mapping of resistance to Sclerotinia sclerotiorum (Lib.) De Bary in sunflower (Helianthus annuus L.)

摘要

Sclerotinia sclerotiorum (Lib.) de Bary is one of the most important pathogens of sunflower. Three different disease symptoms can be caused by S. sclerotiorum: Sclerotinia wilt, midstalk- and head rot. An improvement of the resistance against S. sclerotiorum would contribute to yield security and thus increase the profitability of sunflower cultivation. We investigated resistance to midstalk rot with respect to the prospects of marker-assisted selection (MAS). The bjectives were to (1) identify quantitative trait loci (QTL) involved in resistance against Sclerotinia sclerotiorum, (2) map their position in the genome, (3) characterize their gene effects, and (4) estimate their consistency across generations of the cross NDBLOSsel x CM625. Two sunflower lines with high resistance level to S. sclerotiorum and different genetic origins (NDBLOSsel and TUB-5-3234) were used as parents. They were crossed with a highly susceptible line CM625 to develop two mapping populations. A modified leaf test was used for the evaluation of midstalk-rot resistance. Three resistance traits and two morphological traits were measured. Disease resistance of 354 F3 families of the population NDBLOSsel x CM625 was screened in field trials with two different sowing times in 1999. A total 317 recombinant inbred lines (RIL) derived from F3 families were tested in 2002/2003. The 434 F3 families of cross CM625 x TUB-5-3234 were screened in 2000/2001. The field trials were conducted by using generalized lattice designs with three replications and five infected plants per replication. Highly significant genetic variation between F3 families and RIL was observed for the resistance traits in all field trials. Heritabilities ( ) were highest for stem lesion and lowest for leaf lesion for all three experiments. The resistance traits were moderately correlated with each other. For the construction of the genetic map of population NDBLOSsel x CM625, 352 F2 individuals were analyzed with 117 SSR marker loci. On the basis of results from the QTL mapping in F3 families, 41 markers were selected and genotyped in 248 RIL. A "selective genotyping" (SG) approach was used for population CM625 x TUB-5-3234. Based on the results measured in F3 families for stem lesion, the SSR genotype at 72 marker loci was determined for the 60 most resistant and 60 most susceptible F2 individuals. For QTL mapping and estimation, the method of the "composite interval of mapping" was used. For stem lesion in the population NDBLOSsel x CM625, eight QTL were detected explaining 33.7% of the genetic variance ( ). The QTL on LG8 explained 36.7% of the phenotypic variance (R2adj). All other QTL for this trait explained between 3.3 and 6.0% of R2adj. Nine QTL were detected for leaf lesion. The proportion of the phenotypic variance explained by individual QTL ranged from 3.4 to 11.3%. All detected QTL for leaf lesion explained 25.3% of the genetic variance in cross validation. For speed of fungal growth, 6 QTL were detected, which explained from 4.6 to 10.2% R2adj. Cross validation explained 24.4% of. Most QTL showed additive gene action. QTL occurring consistently across generations can be recommended for MAS and therefore, the QTL results between RIL and F3 families of population NDBLOSsel x CM625 were compared. One common QTL was identified for leaf lesion, two for stem lesion and three for speed of fungal growth. In population CM625 x TUB-5-3234, four QTL for stem lesion, three QTL for leaf lesion and three QTL for speed of fungal growth were identified. Owing to the SG approach we conjecture that not all QTL were found. The comparison of QTL results between two F3 populations showed two common QTL for stem lesion on LG4 and LG8. The QTL on LG4 originated from the susceptible parent CM625. The QTL on LG8 probably corresponds to the QTL with the largest effect determined in the population NDBLOSsel x CM625. Regarding MAS, our results indicate that two QTL detected for stem lesion and speed of fungal growth in population NDBLOSsel x CM625 are promising. They were consistent across environments, and showed no adverse correlation to leaf morphology in trials with the RIL. In mapping population CM625  TUB-5-3234, it remained unclear whether TUB-5-3234 can contribute new alleles with sufficiently large effects for resistance that were not identified in line NDBLOSsel and would be useful in MAS. The genomic region on LG10 should be analyzed in more detail with respect to its importance for resistance in multiple plant parts (head and stalk) and to verify its association with leaf morphology. Resistance breeding of sunflower against S. sclerotiorum is difficult due to the complex inheritance of the trait. This study showed that both the resistance source NDBLOSsel and the identified markers are promising in improving resistance by MAS. For a broader resistance against S. sclerotiorum, it is necessary to detect new resistance genes from different sources to pyramide them in elite lines.