The major goal of plant production is finding the right crop that can meet our demand for food, feed and fuel without damaging the environment. Maize, the world´s most successful multi-purpose crop, is the number one summer crop in many European countries including Germany. The high increase of the maize production area is a leading current topic dominating environmental and agricultural-political discussions in Germany. Sorghum production can readily substitute maize and potentially mitigate some of the problems associated with bio-energy maize production. Sorghum is one of the hardiest plants, with an efficient C4 photosynthetic system and resistance to the maize western rootworm, a devastating pest. It has lower nitrogen and phosphorous demand and can achieve biomass yields that are competitive to maize. Especially sweet sorghum types have the potential to be used for bio-ethanol in addition to biogas production, although storability of the sugar needs to be improved.On the other hand, however, sorghum is sensitive to early stage chilling stress and is a short-day plant; these two adaptation constraints currently hinder its expansion into temperate agro-ecosystems. Fortunately, there is ample variability in sorghum for many traits including cold tolerance, especially in lines from tropical highland areas. The development of effective breeding strategies for adaptation requires a good understanding of the genetic architecture of the crucial adaptation traits. This study dissected early stage seedling development of sorghum to reveal the complex genetics underlying the slow or retarded growth of sorghum seedlings under chilling stress conditions (Bekele et al. 2013a). Controlled experiments and field trials on a recombinant inbred line (RIL) population from the cross between the sweet sorghum parent SS79 and the grain sorghum parent M71 showed contrasting segregation for pre and post emergence chilling stress. In general, chilling stress reduces emergence, root and seedling establishment. When the stress is sustained for a long time it causes reduced survival and ultimately death in genotypes that have insufficient chilling tolerance. Interrelationships/correlations among a large number of complex traits were confirmed by the co-location of QTL for multiple traits including emergence, root development and survival under prolonged chilling stress. Highly interesting QTL colocalization hubs were identified on sorghum chromosomes Sb06 and Sb01. Genome re-sequencing of the parental lines and comparison to other selected grain and sweet sorghum genotypes identified more than 1 million single nucleotide polymorphisms (SNPs). The patterns of polymorphisms identified interesting regions which might be of interest to understand the genetic changes that gave rise to sweet and grain morpho-types of sorghum. Next-generation sequencing-based bulk-segregant analysis, on 60 lines showing the highest and the lowest survival under chilling conditions, identified around 7000 SNPs that were unique to either the chilling-susceptible or chilling-tolerant phenotype group. A 3000-SNP Illumina genotyping array was developed for genetic analyses of chilling stress responses using a stringent selection of the genome-wide and trait-linked SNPs (Bekele et al. 2013b).The SNP array was used to screen a total of 564 sorghum lines, consisting of segregating mapping populations and a diversity panel. The genotype data was used for genome wide association mapping of emergence and for biparental QTL mapping of multiple traits including field biomass, chlorophyll content and brix (sugar content). This efficiently mapped major QTL to known major genes, and in other cases enabled identification of interesting candidate genes for previously unknown QTL.As a proof-of-concept, the SNP array data was used to test genome-wide prediction (genomic selection) for selected traits using the widely-used ridge-regression best linear unbiased prediction (rr-BLUP) model. Using rr-BLUP with even very small training and validation populations it was possible to detect emergence and plant height under stress and optimum conditions at good cross-validation accuracy of 0.30-0.55. This opens the possibility to use genomic prediction for recurrent selection in breeding programs for difficult traits like chilling emergence. Prediction accuracies will be improved by the use of alternative selection models and the design of breeding strategies account for the specific genetic architecture of each trait. The potential of systems biology in sorghum adaptation research for identification of key regulatory genes is discussed in the context of its potential impact on plant breeding. In the long term the integration of additional levels of data (transcriptome, metabolome) can potentially further improve the selection accuracy of genomic selection.
Verknüpfung zu Publikationen oder weiteren Datensätzen
