Effects of selective genetic introgression from wild soybean to soybean

Authors: AKPERTEY, ABRAHAM - University Of Illinois; BELAFFIF, MOHAMMAD - University Of Illinois; GRAEF, GEORGE - University Of Nebraska; Mian, Rouf; SHANNON, J - University Of Missouri; Cregan, Perry; HUDSON, MATTHEW - University Of Illinois; DIERS, BRIAN - University Of Illinois; Nelson, Randall

Submitted to: Crop Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/13/2014
Publication Date: 5/15/2015
Publication URL: http://handle.nal.usda.gov/10113/61862
Citation: Akpertey, A., Belaffif, M., Graef, G., Mian, R.M., Shannon, J.G., Cregan, P.B., Hudson, M.E., Diers, B.W., Nelson, R.L. 2015. Effects of selective genetic introgression from wild soybean to soybean. Crop Science. 54:2683-2695. doi:10.2135/cropsci2014.03.0189.

Interpretive Summary: Wild relatives of crop plants are an important source of genetic diversity. This is especially true for soybean because it is one of the least diverse crop species. Wild soybean crosses easily with soybean but has very poor agronomic characteristics and progeny from these crosses have generally not been very productive. However, limited research has been done using wild soybean to improve soybean yield and even less research has been done to quantify the amount of wild soybean DNA in lines selected for good agronomic traits derived from soybean by wild soybean crosses. We developed high yielding experimental lines derived from backcrossing 5 different wild soybean lines with soybean varieties. Most of these experimental lines should theoretically have had 12% of the DNA from wild soybean. Data collected in 10 evironments identified lines that yielded similarly to the highest yielding soybean parent. Based on analysis of DNA markers, some of these lines had twice as much wild soybean DNA as theoretically predicted without selection even though these lines were very intensely selected to eliminate the wild soybean plant type. Each wild soybean parent consistently contributed some specific chromosomal segments to the high yielding lines indicating that genes in these segments may be making a positive contribution. However, there was no consistency of these choromosomal segments among wild soybean parents. That could mean that each wild soybean parent was making unique contributions. Among the highest yielding lines the percent of DNA markers contributed by the wild soybean parent ranged from 6 to 24% and was not a good predictor of seed yield. This research demonstrates the potential value of wild soybean in improving soybean productivity and will be of interest to soybean breeders and geneticists.

Technical Abstract: Commercial soybean [Glycine max (L.) Merr.] breeding in the U.S. currently relies on a narrow genetic base in which more than half of the genetic contribution, calculated by pedigree analysis, comes from only 5 ancestral lines. For decades, but more intensely in recent years, efforts have been made to incorporate exotic soybean germplasm into the breeding pool. Although wild soybean (G. soja Seib. & Zucc.) is genetically much more diverse than soybean, much less effort has been devoted to utilizing wild soybean in soybean breeding. The objectives of this research are to identify high yielding lines derived from crosses between 5 wild soybean accessions and soybean cultivars; and determine if there are differences in the genetic contributions of each wild soybean parent. Each wild soybean was crossed to Williams 82 and the F1 plants were backcrossed to Williams 82. The BC2 parent lines were developed through intensive family selection and backcrossed to Williams 82. Family selection beginning in the F2 generation was used to develop lines from PI 507807 and PI 549046. The lines from PI 479767 and PI 483461 were selected by early generation testing through yield testing F2 lines in the F3 and F4 generation. The lines derived from PI 65549 were developed from an SSD population. Field evaluation of the derived lines in 6 environments in 2011 and 4 environments in 2012 identified lines that are not significantly different from their recurrent parents. Genotyping the experimental lines and all the parental lines using 1536 SNP markers with the Illumina GoldenGate assay, we found unique contributions being made by the G. soja parents. Also, alleles that are consistently introgressed from the G. soja parents into all derived lines were identified. Despite intense selection pressure to recover good agronomic types, an average of 13% of SNP alleles in the derived lines came from the G. soja parents.