Creation of Chromosome Segment Substitution Lines libraries with wild rice introgressions
O. meridionalis[O. sativa cv. Curinga x O. meridionalis acc. OR44]
The construction of a linkage map using 123 SSR from the Universal Core Map, allowed the identification of 12 linkage groups with a distance coverage of 2005 cM. Using the BC1F1 genotyping data that allowed us to build the linkage map, a series of 130 markers were chosen with the help of the CSSL Finder software and used to score 502 BC2F2 lines. The graphical genotypes show a 100% conservation of the donor genome under the form of small introgressed chromosomal segments. The size of the segments is close to the expected one (8-10 Mb) and well distributed. In a second step, the genetic background of the candidate lines (about half the population size) will be checked for their genetic background (due by end of December).
O. rufipogon
[O. sativa cv. Curinga x O. rufipogon acc. IRGC105491]
A linkage map has been generated using 80 plants from a BC1F1 population derived from backcrossing the elite rice variety BRSMG Curinga (O. sativa ssp tropical japonica) with F1 plants from the interspecific cross Curinga x O. rufipogon acc. IRGC105491. The resulted map has a total genetic distance of 1797 cM, very similar to those reported for others maps between interspecific cross populations (Lorieux 2000). Simultaneously this genotyping work allowed the identification of the introgressed fragments from the donor parent in each line. Using the genotyping data and the program CSSL Finder (Lorieux 2005), 42 plants from the 80 BC1F1 plants were chosen to be backcrossed to the recurrent parent (Curinga) obtaining 42 BC2F1 families (representing 504 plants). It is estimated from these data that only 2% of the donor genome were lost during the process. In a second step, the genetic background of the candidate lines (about half the population size) will be checked for their genetic background (due by end of December).
O. barthii
[O. sativa cv. Curinga x O. barthii acc. IRGC101937]
The 500+ BC2F1 seeds developed at CIAT were sown at WARDA’s experimental station in Cotonou, DNAs have been extracted and are currently genotyped in the same way than for the O. meridionalis and O. rufipogon populations. The forward and background genotyping data should be available by January 2008.
O. glumaepatula
[O. sativa cv. Curinga x O. glumaepatula acc. GEN1233]
The BC2F1 lines were genotyped with the markers that flanked the wild fragments identified in each BC1F1 plant. A total of 60 fragments were tagged and the plants that showed the wild allele for the markers on the fragments were selected for the third backcross. A total of 110 plants were selected, one to two per segment, and backcrossed to Curinga one more time. Five seeds from each plant were sown and another round of selection was performed using the markers that flanked the same tagged fragments on the BC2F1 selection. A total of 153 BC3F1 plants that showed the wild allele for the markers on each fragment were selected and are now being backcrossed to Curinga to obtain BC4F1 plants. After performing two rounds of selection, the recommendation is that a higher number of plants should be selected for each fragment, even if the background is not as clear as expected. Some regions on the genetic map could not be represented by any of the SSR markers available, because either they showed extremely high segregation distortions and did not map to the correct position, or they were not polymorphic between the two genitors. In order to fill these gaps, in-del markers were screened for polymorphism using eight rice genotypes: two O. sativa cultivars (Curinga and Jefferson), three O. rufipogon accessions (IRGC106289, IRGC106286 and IRGC81802), one O. glumaepatula accession (GEN1233), one O. meridionalis accession (OR44) and one O. barthii accession (IRGC105613). PCR reactions followed standard PCR protocol as used for SSR markers and the amplification products were visualized on 1.5% agarose gels. A total of 529 markers were screened, from which 360 showed good amplification products and 128 were polymorphic between Curinga and the other genotypes in any combination. The positions of these markers are known, facilitating the choice of polymorphic markers that lie on regions that SSR markers were not able to represent. Forty-three in-del markers were polymorphic between Curinga and GEN1233 and were mapped in the BC1F1 population according to their positions on the chromosomes. Twenty-three of these markers have been genotyped so far and their mapping positions corresponded to the expected ones.