Combination of two multiplex genome-edited soybean varieties enables customization of protein functional properties

Abstract

Soybean (Glycine max) is a major source of plant-based protein for people worldwide, providing a healthy, affordable, and environmentally friendly alternative to animal-based protein (Montgomery, 2003Montgomery K.S. Soy protein.J. Perinat. Educ. 2003; 12: 42-45https://doi.org/10.1624/105812403X106946Crossref Google Scholar). It has been estimated that there are currently over 12 000 food products that contain soy protein (Montgomery, 2003Montgomery K.S. Soy protein.J. Perinat. Educ. 2003; 12: 42-45https://doi.org/10.1624/105812403X106946Crossref Google Scholar). Different food applications of soy protein require distinct properties that affect its function and sensory features. For instance, the sensory quality of soymilk is significantly affected by protein emulsibility, whereas tofu quality is determined by protein-gelling ability (Kinsella, 1979Kinsella J.E. Functional properties of soy proteins.JAOCS J. Am. Oil Chem. Soc. 1979; 56: 242-258https://doi.org/10.1007/BF02671468Crossref Scopus (1128) Google Scholar). Improving the functional properties of soybeans has long been a major aim in the soy food industry, yet it can be difficult to balance processing methods to achieve desired properties while retaining nutritional value and controlling cost (Kinsella, 1979Kinsella J.E. Functional properties of soy proteins.JAOCS J. Am. Oil Chem. Soc. 1979; 56: 242-258https://doi.org/10.1007/BF02671468Crossref Scopus (1128) Google Scholar; Rubio et al., 2020Rubio N.R. Xiang N. Kaplan D.L. Plant-based and cell-based approaches to meat production.Nat. Commun. 2020; 11: 6276https://doi.org/10.1038/s41467-020-20061-yCrossref PubMed Scopus (180) Google Scholar). The functional properties of soy protein are affected by storage protein composition, particularly β-conglycinin (7S) with high emulsibility and glycinin (11S) with high gelling abilities (Kinsella, 1979Kinsella J.E. Functional properties of soy proteins.JAOCS J. Am. Oil Chem. Soc. 1979; 56: 242-258https://doi.org/10.1007/BF02671468Crossref Scopus (1128) Google Scholar). Mutations in 7S or 11S subunits can affect the functional properties of soy protein (Poysa et al., 2006Poysa V. Woodrow L. Yu K. Effect of soy protein subunit composition on tofu quality.Food Res. Int. 2006; 39: 309-317https://doi.org/10.1016/j.foodres.2005.08.003Crossref Scopus (135) Google Scholar). However, breeding an elite variety with pyramided 7S/11S mutations would be time consuming and, with uncertain effects on seed-quality traits, is unlikely to meet the numerous and rapidly changing demands of the food industry (Montgomery, 2003Montgomery K.S. Soy protein.J. Perinat. Educ. 2003; 12: 42-45https://doi.org/10.1624/105812403X106946Crossref Google Scholar). The development of CRISPR-Cas9 technology enables rapid and precise editing of multiple genes in crops, including soybean (Cai et al., 2018Cai Y. Chen L. Liu X. Guo C. Sun S. Wu C. Jiang B. Han T. Hou W. CRISPR/Cas9-mediated targeted mutagenesis of GmFT2a delays flowering time in soya bean.Plant Biotechnol. J. 2018; 16: 176-185https://doi.org/10.1111/pbi.12758Crossref PubMed Scopus (215) Google Scholar; Li et al., 2020Li J. Li H. Chen J. Yan L. Xia L. Toward precision genome editing in crop plants.Mol. Plant. 2020; 13: 811-813https://doi.org/10.1016/j.molp.2020.04.008Abstract Full Text Full Text PDF PubMed Scopus (31) Google Scholar). In this study, we sought to combine two soybean varieties created using genome editing, one mutant lacking 7S subunits (ultra-low 7S:11S ratio) and another lacking 11S subunits (ultra-high 7S:11S ratio). We reasoned that the seeds of the two varieties could be mixed in different proportions to allow the precise custom formulation of functional properties of soy protein for food processing. The 7S globulin is a heterotrimer composed of three subunits (α, α′, and β) encoded by 7 genes, and the 11S globulin is a heterohexamer of 5 subunits (group I: A1aB1b/A1bB2 and A2B1a, group II: A3A4 and A5A4B3, and group III: Gy7) encoded by 5 genes. To facilitate efficient multiplex editing of more than 5 target loci, we developed a soybean-optimized multiplex CRISPR vector, designated pGES401, based on the single-transcript-unit system (Tang et al., 2019Tang X. Ren Q. Yang L. Bao Y. Zhong Z. He Y. Liu S. Qi C. Liu B. Wang Y. et al.Single transcript unit CRISPR 2.0 systems for robust Cas9 and Cas12a mediated plant genome editing.Plant Biotechnol. J. 2019; 17: 1431-1445https://doi.org/10.1111/pbi.13068Crossref PubMed Scopus (100) Google Scholar). We then constructed two vectors, pGES401-7S to target the seven 7S subunit genes (Figure 1A) and pGES401-11S to target the five 11S subunit genes (Figure 1B; Supplemental Figure 1). The vectors were individually transformed into an elite food-grade cultivar, QiHuang34 (QH34), by Agrobacterium-mediated transformation (Bai et al., 2020Bai M. Yuan J. Kuang H. Gong P. Li S. Zhang Z. Liu B. Sun J. Yang M. Yang L. et al.Generation of a multiplex mutagenesis population via pooled CRISPR-Cas9 in soya bean.Plant Biotechnol. J. 2020; 18: 721-731Crossref PubMed Scopus (91) Google Scholar). In the T0 transformants, the genotypes of target loci were evaluated by Hi-TOM sequencing (Liu et al., 2019Liu Q. Wang C. Jiao X. Zhang H. Song L. Li Y. Gao C. Wang K. Hi-TOM: a platform for high-throughput tracking of mutations induced by CRISPR/Cas systems.Sci. China Life Sci. 2019; 62: 1-7https://doi.org/10.1007/s11427-018-9402-9Crossref PubMed Scopus (167) Google Scholar). Simultaneous frameshift editing of all target genes was identified in 3 of the 8 pGES401-7s-null events and 5 of the 13 pGES401-11s-null events (Figures 1C and 1D). Transgene-free septuple 7S and quintuple 11s mutant lines were identified among T2 progeny, designated as 7s-null, and 11s-null, respectively (Figure 1E). qRT–PCR and SDS–PAGE showed that the expression of the targeted 7S or 11S subunits was largely abolished in the mutant seeds, and genes encoding other storage proteins accumulated due to a compensatory effect (Figures 1F and 1G). We next examined the seed yield and quality traits of the T3 mutant lines and found that seed yield per plant and oil and protein content in the 7s-null or 11s-null lines were similar to wild type (WT) (Figure 1H; Supplemental Figure 2). Therefore, seed traits other than the 7S/11S composition were apparently not affected by genome editing. We then tested whether the protein functional properties were affected in the mutants and whether custom formulation is feasible by proportional mixing 7s-null and 11s-null mutant seeds. We first monitored the emulsion stability of the soybean protein isolates of WT, 7s-null, 11s-null, or proportionally mixed samples using a LUMiSizer, in which the slopes of the curves of the integral light-transmission intensity versus time are negatively associated with the stabilities of the emulsions (Figure 1I). Compared with the QH34 WT samples, the emulsions prepared from 11s-null mutants exhibited greatly enhanced stabilities, while 7s-null emulsions had the lowest stabilities (Figure 1I). Consistent with our hypothesis, the 2:1, 1:1, or 1:2 mixtures of 7s-null:11s-null seeds have intermediate or higher emulsion stabilities (Figure 1I). Consistently, the variation in emulsions stabilities was visually distinguishable after standing for 30 days at room temperature (Figure 1J). Analysis using a laser particle-size analyzer revealed that the variations in emulsion stability were associated with differences in their particle-size distribution (Figure 1K). The gelling ability of soy protein gels was assessed using a texture analyzer to monitor the textural characteristics (Supplemental Figure 3). The gelling hardness, springiness, gumminess, and chewiness were significantly increased in the 7s-null compared with WT protein gels, while they were largely decreased in 11s-null protein gels. The 1:2, 1:1, or 2:1 seed mixtures exhibited gelling abilities that were intermediate to the 7s-null and 11s-null lines (Figu