Genotyping BC3F2 populations of four Ethiopian sorghum varieties for Stay Green QTLs introgression through Marker assisted selection with SSRs.

Abstract

Sorghum is a subsistence and staple crop for many people living across Africa especially those in marginal arid and semi-arid tropics. While it is a staple food for millions of people in Africa, livestock feeding accounts for most of the sorghum used in the developed world especially in USA, Mexico, Australia, and South Africa. Among the major impediments to successful sorghum production is drought. This constraint is considered the most important cause of yield reduction in crop plants. Plant breeding has been continually used for crop improvement and one of its important objectives is the introgression of one or more drought resistant genes or genomic regions defined as Quantitative Trait Loci (QTL) from a donor variety into the background of a farmer-preferred or improved variety or a landrace while recovering the recurrent parent genome as rapidly as possible. Stay-green (SG), a drought tolerance trait of sorghum, has been mapped in several sorghum varieties including B35, E36-1 and SC56 and is regularly targeted in many breeding programmes. Although high yielding, several Ethiopian farmer preferred sorghum varieties (Meko, Gambella, Teshale and 76T1#23) have been identified as being susceptible to post- flowering moisture stress. Breeding for the SG trait from B35 and E36-1 donor parents into the four varieties was initiated in 2006 and had progressed to BC3F2 generation. However, selection of this population for individuals fixing the SG QTL and further those among them nearly isogenic to the recurrent parent genome was lacking. Selection through phenotypic assays would have been time consuming, cumbersome and in-precise leading to delayed release of the improved varieties. This study aimed to accelerate the development of drought tolerant Meko, Gambella, Teshale and 76T1#23 varieties through the use of marker assisted selection (MAS) to confirm the introgression of SG QTL in the BC3F2 population of 1211 plants. PCR was used to amplify SSR loci tightly linked to the SG QTL as well as SSR loci across the non- target chromosomes. The amplicons were separated at high resolution using a fluorescence based capillary sequencer system, the ABI 3730. A Genemapper (v 4.0) software was used to generate data output files. This data was sort according to various combination of alleles at each available SSR loci. A total of 119 individuals out of the 1211 screened were found to fix a single SG QTL while 4 individuals had a double SG QTL introgression. Background screening was done on 103 of these individuals to determine the level of recurrent parent homozygosity at SSR loci across all the non-target chromosomes. It was determined that 66% of the progeny from E36-1 donor parent and the four recurrent parents introgressing the SG QTL were >80% homozygous for the recurrent parent at the 19 SSR loci while 51% of the progeny from B35 donor parent and the four recurrent parents introgressing the SG QTL were >80% homozygous for the recurrent parent at the 26 SSR loci on the non-target chromosomes. Based on this information, 60 individuals were selected for field evaluation. This step is crucial since the donor parent alleles fixed in the selected progeny may not be

expressed phenotypically due to G×E interaction. Routine application of MAS in crop breeding is still lacking. The results obtained from this study demonstrated the efficiency of MAS as a complementing tool for conventional breeding.