RT Dissertation/Thesis
T1 Genomics-assisted breeding strategies for quantitative resistances to Northern corn leaf blight in maize (Zea mays L.) and Fusarium diseases in maize and in triticale (× Triticosecale Wittm.)
A1 Galiano Carneiro,Ana Luísa
WP 2021/06/22
AB Fusarium head blight (FHB) in triticale (× Triticosecale Wittm.), Gibberella ear rot (GER) and Northern corn leaf blight (NCLB) in maize (Zea mays L.) are devastating crop diseases causing yield losses and/or reducing grain quality worldwide. Resistance breeding is the most efficient and sustainable approach to reduce the damages caused by these diseases. For all three pathosystems, a quantitative inheritance based on many genes with small effects has been described in previous studies. Hence, this thesis aimed to assess the potential of genomics-assisted breeding strategies to reduce FHB, GER and NCLB in applied breeding programs. In particular, the objectives were to: (i) Dissect the genetic architecture underlying quantitative variation for FHB, GER and NCLB through different quantitative trait loci (QTL) and association mapping approaches; (ii) assess the potential of genomics-assisted selection to select superior triticale genotypes harboring FHB resistance; (iii) phenotype and characterize Brazilian resistance donors conferring resistance to GER and NCLB in multi-environment trials in Brazil and in Europe; and (iv) evaluate approaches for the introgression and integration of NCLB and GER resistances from tropical to adapted germplasm.
The genome-wide association study (GWAS) conducted for FHB resistance in triticale revealed six QTL that reduced damages by 5 to 8%. The most prominent QTL identified in our study was mapped on chromosome 5B and explained 30% of the genotypic variance. To evaluate the potential of genomic selection (GS), we performed a five-fold cross-validation study. Here, weighted genomic selection increased the prediction accuracy from 0.55 to 0.78 compared to the non-weighted GS model, indicating the high potential of the weighted genomic selection approach. The successful application of GS requires large training sets to develop robust models. However, large training sets based on the target trait deoxynivalenol (DON) are usually not available. Due to the rather moderate correlation between FHB and DON, we recommend a negative selection based on genomic estimated breeding values (GEBVs) for FHB severity in early breeding stages. In the long-run, however, we encourage breeders to build and test GS calibrations for DON content in triticale.
The genetic architecture of GER caused by Fusarium graminearum in maize was investigated in Brazilian tropical germplasm in multi-environment trials. We observed high genotype-by-environment interactions which requires trials in many environments for the identification of stable QTL. We identified four QTL that explained between 5 to 22% of the genotypic variance. Most of the resistance alleles identified in our study originated from the Brazilian tropical parents indicating the potential of this exotic germplasm as resistance source. The QTL located on chromosome bin 1.02 was identified both in Brazilian and in European trials, and across all six biparental populations. This QTL is likely stable, an important feature for its successful employment across different genetic backgrounds and environments. This stable QTL is a great candidate for validation and fine mapping, and subsequent introgression in European germplasm but possible negative linkage drag should be tackled.
NCLB is another economically important disease in maize and the most devastating leaf disease in maize grown in Europe. Virulent races have already overcome the majority of known qualitative resistances. Therefore, a constant monitoring of S. turcica races is necessary to assist breeders on the choice of effective resistances in each target environment. We investigated the genetic architecture of NCLB in Brazilian tropical germplasm and identified 17 QTL distributed along the ten chromosomes of maize explaining 4 to 31% of the trait genotypic variance each. Most of the alleles reducing the infections originated from Brazilian germplasm and reduced NCLB between 0.3 to 2.5 scores in the 1-9 severity scale, showing the potential of Brazilian germplasm to reduce not only GER but also NCLB severity in maize. These QTL were identified across a wide range of environments comprising different S. turcica race compositions indicating race non-specific resistance and most likely stability. Indeed, QTL 7.03 and 9.03/9.04 were identified both in Brazil and in Europe being promising candidates for trait introgression.
These major and stable QTL identified for GER and NCLB can be introgressed into elite germplasm by marker-assisted selection. Subsequently, an integration step is necessary to account for possible negative linkage drag. A rapid genomics-assisted breeding approach for the introgression and integration of exotic into adapted germplasm has been proposed in this thesis.
Jointly, our results demonstrate the high potential of genomics-assisted breeding strategies to efficiently increase the quantitative resistance levels of NCLB in maize and Fusarium diseases in maize and in triticale. We identified favorable QTL to increase resistance levels in both crops. In addition, we successfully characterized Brazilian germplasm for GER and NCLB resistances. After validation and fine mapping, the introgression and integration of the QTL identified in this study might contribute to the release of resistant cultivars, an important pillar to cope with global food security.
K1 Pflanzenzüchtung
K1 Resistenzzüchtung
K1 Getreide
K1 Mais
K1 Fusarium
K1 Genomik
PP Hohenheim
PB Kommunikations-, Informations- und Medienzentrum der Universität Hohenheim
UL http://opus.uni-hohenheim.de/volltexte/2021/1898