Molecular evolution of glycinin and β-conglycinin gene families in soybean (Glycine max L. Merr.)

Abstract

There are two main classes of multi-subunit seed storage proteins, glycinin (11S) and β-conglycinin (7S), which account for approximately 70% of the total protein in a typical soybean seed. The subunits of these two protein classes are encoded by a number of genes. The genomic organization of these genes follows a complex evolutionary history. This research was designed to describe the origin and maintenance of genes in each of these gene families by analyzing the synteny, phylogenies, selection pressure and duplications of the genes in each gene family. The ancestral glycinin gene initially experienced a tandem duplication event; then, the genome underwent two subsequent rounds of whole-genome duplication, thereby resulting in duplication of the glycinin genes, and finally a tandem duplication likely gave rise to the Gy1 and Gy2 genes. The β-conglycinin genes primarily originated through the more recent whole-genome duplication and several tandem duplications. Purifying selection has had a key role in the maintenance of genes in both gene families. In addition, positive selection in the glycinin genes and a large deletion in a β-conglycinin exon contribute to the diversity of the duplicate genes. In summary, our results suggest that the duplicated genes in both gene families prefer to retain similar function throughout evolution and therefore may contribute to phenotypic robustness.

Figure 1

Synteny analyses for the sequences in Table 1 among the segments on soybean chromosomes 3, 10, 13 and 19 (a); on soybean chromosome 19, M. truncatula chromosome 1, poplar chromosomes 2 and 5 (b); on soybean chromosome 3, M. truncatula chromosome 1, Arabidopsis chromosomes 4, 5 and V. vinifera chromosome 7 (c); on soybean chromosomes 2, 10 and 20 (d); on soybean chromosome 10-2, M. truncatula chromosome 1 and poplar chromosomes 5 and 10 (e); and on soybean chromosome 10-2, Arabidopsis chromosome 3 and V. vinifera chromosome 7 (f). Homologous gene pairs are connected with lines. The chromosome segment is indicated by horizontal line, and the broad line with arrowhead represents gene and its transcriptional orientation. Horizontal line is drawn roughly to scale; broken line indicates segment not drawn to scale. The text besides the gene is the locus identifier prefix. The glycinin and β-conglycinin genes and their homologs are shown in red. Figures in (a), (d) and (e) are drawn roughly to 0.2 MB resolution, and in (b), (c) and (f) to 0.1 MB. The β-conglycinin genes lie in two separate regions of soybean chromosome 10; the two regions are denoted Gl chr 10-1 and Gl chr 10-2. A full color version of this figure is available at the Heredity journal online.

Figure 2

Phylogenetic analyses of the glycinin genes and their homologs (a) and the β-conglycinin genes and their homologs (b). The trees were constructed using the neighbor joining (NJ) method implemented in MEGA 4.0. The numbers beside the branches represent bootstrap values (>50%) based on 1000 replications. Sequences Glyma10g04270, Vv7g0729, Vv7g0730 and Pt10g0706 are excluded owing to their short lengths. The same analyses were performed in Tree-Puzzle (manual delete gaps; 434 and 416 amino acids were used in (a) and (b), respectively), which yielded highly similar results (data not shown). The horizontal dashed lines were manually drawn to assist in reading the trees, and the branches that were drawn in thick lines and labeled with letters G1, G2 or S, and so on, were supposed to be under different selection pressures.

Figure 3

Phylogenetic analyses of the genes Gy1–8. The trees were constructed using the neighbor joining (NJ) method implemented in MEGA 4.0. The tree is rather crude owing to the highly divergent sequences of the pseudogenes Gy6 and Gy8.

Figure 4

The segment, without homologous genes of β-conglycinin protein, on soybean chromosome 1 is found to share strong synteny with the segments containing β-conglycinin genes. The time of the duplication of the segment on chromosome 1 and its syntenic region on chromosome 10-2 or 20 is around 50 mya, suggesting that the segments arose during the more ancient WGD (chr 1 and chr 10-2, K_S=0.64, s.d.=0.20, n=10, time=52 mya; chr 1 and chr 20, K_S=0.58, s.d.=0.10, n=5, time=48 mya).

Figure 5

Gene duplication history of glycinin (a) and β-conglycinin (b). The determined and pending directions of gene duplication events are indicated by single and double arrows at both ends, respectively. Genes and chromosomes follow the legend in Figure 1, but the genes shaded in white indicate pseudogenes and broken line boxes indicate the gene deletion events. The time 60–88 mya in (a) is the time between the poplar/soybean split event and the more ancient WGD event of soybean (Sanderson et al., 2004).