Novel Soybean Variety Lacking Raffinose Synthase 2 Activity

Abstract

Variation in the raffinose family oligosaccharide (RFO) content in soybean is advantageous for livestock farming and health science. In this study, a soybean variety (GmJMC172) with a significantly low stachyose content in its seeds was identified in the NARO Genebank core collection. The results of the single-nucleotide polymorphism (SNP) analysis suggested that this phenomenon was related to a single-base deletion, inducing a frameshift mutation in raffinose synthase 2 (RS2), rather than the polymorphisms in the RS3, RS4, and stachyose synthase (STS) sequences. Differences in the enzymatic properties between the native RS2 and truncated RS2 were examined by using a three-dimensional model predicted using Alphafold2. In addition to revealing the missing active pocket in truncated RS2, the modeled structure explained the catalytic role of W331* and suggested a sufficient space to bind both sucrose and raffinose in the ligand-binding pocket. The soybean line, with seeds available from the NARO Genebank, could serve as breeding materials for manipulating the RFO content.

Figure 1

Biosynthesis of RFOs.

Figure 2

Structures of sucrose, raffinose, and stachyose.

Figure 3

High-performance liquid chromatography (HPLC) charts of (a) GmWMC036, (b) GmJMC055, and (c) GmJMC172. HPLC was performed using a Shodex NH2P-50 4E column at 30 °C. The fluorescent responses of sugars were enhanced using a fluorescent developing reagent (50 mM guanidine-HCl, 1.5 mM periodic acid, 100 mM boric, and 125 mM KOH), excited at 325 nm, and detected at 420 nm. Sucrose, raffinose, and stachyose were identified based on the retention times of the standard compounds.

Figure 4

Saccharide content in soybean seeds cultured in 2010 and 2021. The correlation coefficients (r values) for 2010 and 2021 are displayed on each chart. The solid and dashed lines indicate the borders of the mean ± 3SD and mean ± 2SD of each distribution, respectively. The numbers near closed circles correspond to the accession numbers of the mini-core collection described in the text.

Figure 5

Comparison of WT-RS2 and GmJMC172-RS2. The noted residues are displayed as the space-filling model of CPK color with the names of yellow characters. (a) Pairwise alignment of WT-RS2 and GmJMC172-RS2 sequences. Because the residue numbers do not correspond between the two sequences, the residue numbers of WT are indicated as asterisks. The cyan-highlighted residues are found on the surface of the estimated active pocket in the Alphafold2-predicted modeled structure. The underlined residues indicate the chain following the frameshift mutation. (b) Superposed structures of predicted WT-RS2 (white/green tube) and GmJMC172-RS2 (orange tube). The green-colored structure is lacking in GmJMC172-RS2. The atoms drawn by the space-filling model indicate W331*. (c) Positions of T107*, S150*, and W331* on the predicted 3D structure. An α-helix from G485* to N494* fixed by T107* is colored red, and the continuing flexible loop consisting of the pocket is colored white (undefined secondary structure) and blue (turn).

Figure 6

Comparison between WT-RS2 and STS. (a) Pairwise alignment of WT-RS2 and STS. The red lines under the sequences indicate conserved positions. (b) Superposed structures of the predicted WT-RS2 (tube of pale pink) and STS (light green). The yellow arrows indicate the two unique α-helices in the STS sequence.

Figure 7

(a) Magnified view of the predicted RS2 pocket with galactinol (space-filling model) and W331* (pink ball-and-stick model). (b) Contact map of the modeled complex of galactinol and WT-RS2.