Molecular mapping and genomics of soybean seed protein: a review and perspective for the future

Abstract

Genetic improvement of soybean protein meal is a complex process because of negative correlation with oil, yield, and temperature. This review describes the progress in mapping and genomics, identifies knowledge gaps, and highlights the need of integrated approaches. Meal protein derived from soybean [Glycine max (L) Merr.] seed is the primary source of protein in poultry and livestock feed. Protein is a key factor that determines the nutritional and economical value of soybean. Genetic improvement of soybean seed protein content is highly desirable, and major quantitative trait loci (QTL) for soybean protein have been detected and repeatedly mapped on chromosomes (Chr.) 20 (LG-I), and 15 (LG-E). However, practical breeding progress is challenging because of seed protein content's negative genetic correlation with seed yield, other seed components such as oil and sucrose, and interaction with environmental effects such as temperature during seed development. In this review, we discuss rate-limiting factors related to soybean protein content and nutritional quality, and potential control factors regulating seed storage protein. In addition, we describe advances in next-generation sequencing technologies for precise detection of natural variants and their integration with conventional and high-throughput genotyping technologies. A syntenic analysis of QTL on Chr. 15 and 20 was performed. Finally, we discuss comprehensive approaches for integrating protein and amino acid QTL, genome-wide association studies, whole-genome resequencing, and transcriptome data to accelerate identification of genomic hot spots for allele introgression and soybean meal protein improvement.

Fig. 1

Correlations among different seed components and yield in soybean. Positive correlation (+), negative correlation (−). Temperature (yellow high temp; blue low temp.) affects protein, oil, and sucrose concentrations and it is shown by up (increase) and down (decrease) arrows. Correlation between known sub-components such as cysteine, oleic acid, raffinose, and stachyose are shown (color figure online)

Fig. 2

Integrated genomics and breeding approaches for soybean meal improvement. This figure illustrates the systematic approach for germplasm selection, phenotyping for seed components under different environments, parent selection, and development of mapping population. Different genotyping approaches can be used to identify the QTL hotspots, haplotypes, and markers. After validation, these QTL can be subsequently tracked in marker-assisted backcrossing (MABC) and marker-assisted recurrent selection (MARS) approaches for trait introgression and cultivar development. Additionally, available genomic resources and phenotypic data (training population) can be utilized in genomic selection of progeny and crossing design

Fig. 3

Analysis of the SNPs in the genomic regions (29.8–31.6 Mb) on Chr.20 shows that the North American ancestors and elite cultivars (magenta) are different as compared to soja (red) and several Korean cultivars/landraces (black). The known US (Williams82 and Benning) and Korean (Danbaekkong) elite lines are denoted by blue dots (color figure online)

Fig. 4

Soybean production and growth trend in USA, Brazil, and Argentina from 1980 to 2014

Fig. 5

Amino acid composition of soybean protein. Red essential amino acid (EAA), black nonessential AA. Data source: Asif and Acharya (2013), Berk et al. (1992), Kuiken et al. (1949) (color figure online)

Fig. 6

Correlation of maturity (temperature effect) with seed protein and oil content in G. max accession. Number of lines in bracket