Rice (Oryza sativa L.) is one of the most important cereal crops and the staple food for over half of the world’s population.Under the efforts of breeders, the quality of rice has been becoming better in recent years, and basically meets the growing life requirements of people. However, the major micronutrients in seeds are gradually decreasing in many rice cultivars, which is highly relevant to chronic diseases and micronutrient deficiencies and influences the health of children and women in the developing countries (Wahlqvist et al, 1999;Zhang et al, 2005). Therefore, how to increase the nutritional quality in rice has attracted growing attention in recent years.
Brown rice, consisting of pericarp, seed coat, aleurone layer,embryo and endosperm, contains a lot of nutritional components,such as oil, iron, amino acid, vitamin, fat, protein and starch(Zhang et al, 2005). The lysine, lipid, protein and γ-aminobutyric acid (GABA) are concentrated in rice embryo, and their contents in giant embryo rice is higher than those in conventional rice (Park et al, 2009). For example, the giant embryo rice Keunnunbyeo contains more lipid and mineral contents than the normal embryo rice Ilmibyeo (Choi et al, 2006). The black waxy rice Milyang 263 with a giant embryo phenotype produces high micronutrients, such as GABA, essential amino acids and cyanidin-3-glycoside (Kim et al, 2013). The giant embryo mutant YR23517Acp79 exhibits increased protein, lipid,γ-oryzanol, vitamin B1, tocopherols and mineral contents (Seo et al, 2011). The four giant embryo rice mutants originated from Mizuhochikara contain high oil content (Sakata et al,2016). Therefore, breeding of giant embryo rice is of interest for crop breeders.
Some studies show that the rice embryo size is controlled by a single recessive nuclear gene, GE, located on the rice chromosome 7, which encodes a cytochrome P450 monooxygenase (Satoh and Omura, 1981; Qian et al, 1996). To the best of our knowledge, several GE alleles in different mutants have been identified, including ge-1 to ge-10 (Nagasawa et al,2013), get(Park et al, 2009), ges(Koh et al, 1996), ge (Yang et al,2013), OsGE/CYP78B5 (Chen et al, 2014), MGE12 and MGE13(Sakata et al, 2016). All the allelic variations of GE are the substitution of single nucleotide at different locations. The substitution of single nucleotide in ge-1, ge-2 and ge-5 results in premature termination, and the others result in the substitution of single amino acid (Nagasawa et al, 2013). The getmutant has a point mutation on G1185 to T1185 and alters the corresponding tryptophan at the 395th amino acid to leucine(Park et al, 2009). OsGE/CYP78B5 mutant has a single nucleotide transition from G1482 to A1482 and results in premature termination at the 493th amino acid (Chen et al,2014). A substitution of T to A in the first exon of ge mutant changes the phenylalanine to isoleucine (Yang et al, 2013). In the above GE mutants, the size of embryo increases, and the volume of endosperm becomes small.
Mature seeds of GE mutants have high nutritional components.However, the endosperm is the main component of rice mature seed, and physicochemical properties of seed determine its applications. Though Chung et al (2017) have compared the physicochemical properties of two giant embryo rice and one normal embryo rice, it is unclear whether the embryo enlargement influences the physicochemical properties of seeds.In this study, we isolated two giant embryo mutants from a japonica rice cultivar Kitaake. The morphology characteristics and seed physicochemical properties of giant embryo mutants and their wild type (WT) Kitaake were investigated.
Numerous mutants with defective seeds were screened from a60Co-induced mutant pool, and two giant embryo mutants were isolated and named M12 and M13, respectively. The seed-setting rate of M13 (70.1%) was significantly lower than WT (86.6%) and M12 (88.3%), but there were no significant differences in other agronomic traits among WT, M12 and M13(Supplemental Table 1). The mature grains (with hulls) of M12 and M13 were phenotypically similar to those of WT (Fig. 1-A),while the brown rice (dehusked) of M12 and M13 showed a giant embryo, and the embryo of M13 was larger than that of M12 (Fig. 1-B). Brown rice length and width were largely comparable among the WT, M12 and M13, while the brown rice thickness and 1000-brown rice weight reduced significantly in M12 and M13 (Fig. 1-C to -F). The embryo ratios of M12(18.0%) and M13 (27.4%) were higher than that of WT (7.5%)(Fig. 1-G). The 100-embryo weights of M12 (0.12 g) and M13(0.18 g) were higher than that of WT (0.08 g) (Fig. 1-H). As shown in Fig. 1-I and -J, the semi-thin sections of WT and two mutant developing seeds at 18 d after fertilization (DAF)showed different embryo sizes in the order: M13 (2.95 mm2) >M12 (2.02 mm2) > WT (0.96 mm2).
For genetic analysis, M12 and M13 were hybridized with a wide-compatibility indica variety Dular. All the F1hybrid seeds showed normal embryo. The mutant phenotype of F2populations co-segregated at a ratio of approximately 3:1 (χ2(M12) = 0.76 <= 3.84, χ2(M13) = 0.63 <= 3.84).Those results indicated that the giant embryo phenotype in both M12 and M13 is controlled by a single recessive nuclear gene.Two F1populations were produced from reciprocal crosses between M12 and M13. All the F2seeds from the crosses showed giant embryo compared with the WT (Supplemental Fig. 1). Therefore, M12 and M13 were two allele mutants. To identify the mutation loci of M12 and M13, map-based cloning was adopted. Ten F2giant embryo and ten normal embryo individuals were selected to produce two DNA pools for bulk-segregant analysis (BSA). The two DNA pools were genotyped with 170 selected InDel markers for genetic linkage analysis. The M12 mutation locus was initially mapped between InDel markers Z7-9 and Z7-11 on the long arm of chromosome 7. Meanwhile, fine mapping based on 586 and 469 recessive individuals narrowed the M12 and M13 loci to an 84.3-kb region between InDel markers Z7-13 and Z7-18 on the BAC clone OSJNb0040H10 and OSJNb0039M16 (Fig. 2-A).The 84.3-kb region contains 14 candidate genes according to the annotation of the Rice Genome Annotation Project(http://rice.plantbiology.msu.edu) (Supplemental Table 2). The mapping region contained a previously reported giant embryo locus GE (LOC_Os07g41240) (Yang et al, 2013).
The GE gene encodes a cytochrome P450 protein composed of 525 amino acid residues and contains two exons and one intron (Yang et al, 2013). The GE genomic sequences of M12 and M13 were amplified by PCR and sequenced. Compared with the sequences of WT, the M12 allele carried a single nucleotide substitution of guanine (G) to adenine (A) in the first exon, leading to a non-synonymous mutation from adenine to threonine (T) at the 113th amino acid (Fig. 2-A, -B and -D).
Amino acid sequence analysis showed that A-residue was highly conserved in the GE-related proteins among different plant species (Nagasawa et al, 2013). While the M13 allele carried a single nucleotide substitution of G1111 to A1111 in the second exon, which caused the tryptophan (W) at the 340th amino acid replacement by a premature termination stop codon(Fig. 2-A, -C and -D). Thus, M13 was a nonsense mutant in GE.In summary, we concluded that the mutations of the LOC_OS07g41240 are responsible for M12 and M13 phenotypes.Consistent with the phenotype differentiation, M12 might be a leaky mutant, while M13 most likely corresponded to a loss-of-function mutant of GE gene (Figs. 1 and 2). Thus, M12 and M13 were two new GE alleles compared with the mutation sites that have been reported.
The contents of GABA and amino acids in brown rice flour were significantly different among WT, M12 and M13. The GABA contents of M13 (0.165 mg/g) and M12 (0.110 mg/g)were significantly higher than that of WT (0.065 mg/g) (Fig.3-A). The GABA contents of M12 and M13 were significantly higher than those of other giant embryo mutants (Zhang et al,2005). GABA is widely recognized as a functional nutrient to reduce blood pressure symptoms (Zhao et al, 2017). Meanwhile,GABA can be used to improve the menopause and mental disorders in the conditions that 26.4 mg of GABA per day is taken in the diet (Kim et al, 2013). According to this standard,an effective amount of GABA can be taken by eating about 240 g of M12 or 160 g of M13 brown rice, which is very achievable for the people eating rice as a staple food. Glutamic acid and aspartic acid were two amino acids with higher content, while cysteine was the limiting amino acid among the WT, M12 and M13 (Table 1). The present results are in accordance with the previous report (Chung et al, 2017). Except cysteine, the contents of the other 16 amino acids and total amino acids increased significantly in M12 and M13 (Table 1). Isoleucine, leucine,lysine, methionine, phenylalanine, threonine, tryptophan and valine are the eight essential amino acids for the human body and must be ingested from food. The essential amino acids of M12 and M13 were about 1.2 times higher than that of the WT. The essential amino acids of giant embryo rice Milyang 263 have two times higher than those of normal rice Ilmibyeo (Kim et al, 2013).
The protein contents in brown rice flour of M12 (10.8%) and M13 (12.1%) were higher than that of WT (9.3%) (Fig. 3-B).The brown rice of M12 and M13 had the total starch contents of 68.1% and 66.1%, respectively, which were significantly lower than that of WT (72.6%) (Fig. 3-C). Additionally, the amylose contents of M12 (12.4%) and M13 (14.5%) had a significant decrease compared with WT (16.3%) (Fig. 3-D). These results are in agreement with the results of Chung et al (2017).
The thermal properties of brown rice flour were also investigated (Table 2). The gelatinization onset, peak and conclusion temperatures among the three brown rice flour ranged from 60.3 °C to 61.9 °C, 68.4 °C to 70.0 °C, and 75.7 °C to 77.2 °C, respectively. Except gelatinization enthalpy, the gelatinization onset, peak and conclusion temperatures of M12 and M13 were lower than those of WT. These differences might be due to the different genetic backgrounds and mutation sites.In this study, the lower gelatinization temperature in twomutants might be attributed to the lower amylose content. The peak, hot, breakdown, final and setback viscosities among the three brown rice flour ranged from 684 to 1 314 MPa/s, 461 to 955 MPa/s, 223 to 359 MPa/s, 1 239 to 2 002 MPa/s, and 778 to 1 047 MPa/s, respectively (Table 2). Except breakdown viscosity,the peak, hot, final and setback viscosities of M12 and M13 were lower than those of WT. The pasting viscosity is affected by many factors, such as protein, lipid and total starch contents in flour (Martin and Fitzgerald, 2002; Noda et al, 2004; Kaur et al,2016). The high protein and lipid contents can inhibit the binding of water and flour and reduce the pasting viscosity of flour (Pelissari et al, 2012). Starch is mainly responsible for pasting viscosity, and the flour with low starch content has low pasting viscosity. Amylose content is positively correlated with hot, final and setback viscosities (Yangcheng et al, 2016; Zhang et al, 2017). The low pasting viscosities of M12 and M13 mutants might be due to the high protein and the low starch and amylose contents in brown rice flour.
Table 1. Amino acid contents in brown rice.mg/g
Table 2. Thermal properties and pasting properties of brown rice flour.
It is important to clarify how a developing embryo is genetically regulated because the final volume of the endosperm as a storage organ of starch and protein is affected by embryo size (Du et al, 2019). The brown rice of M12 and M13 displayed higher embryo ratio and lower 1000-brown rice weight, starch content and amylose content than those of WT(Figs. 1 and 3). Therefore, the enlarged embryos in M12 and M13 affected the endosperm development and starch accumulation. The cytochrome P450 monooxygenase catalyzes the formation of indole-3-acetaldoxime from tryptophan, which is involved in indole-3-acetic acid (IAA) biosynthesis(Mikkelsen et al, 2000). Chen et al (2014) reported that the IAA levels are one-third of those of WT and the expression level of auxin-responsive genes is downregulated dramatically in the OsGE mutant seeds. We speculated that the mutations in GE gene of M12 and M13 hindered the auxin synthesis, which in turn affected the endosperm development and starch synthesis.
Allelic variations of rice genes are often used in breeding programs. The seeds of M12 and M13 showed higher contents of GABA and free amino acids than those of WT, while M12 contained higher 1000-brown rice weight, starch content and seed-setting rate than those of M13 (Figs. 1 and 3, Table 1 and Supplemental Table 1). Therefore, the new GE allele in M12 is more suitable for developing giant embryo variety by marker-assisted selection in rice breeding programs.
This study was supported by grants from the Natural Science Foundation of Jiangsu Province (Grant No. BK20160461), the China Postdoctoral Science Foundation (Grant No. 2018T110561),the Innovation Program for Graduates of Jiangsu Province (Grant No. XKYCX19_145), the Qing Lan Project of Jiangsu Province,the Talent Project of Yangzhou University, and the Priority Academic Program Development of Jiangsu Higher Education Institutions.
The following materials are available in the online version of this article at http://www.sciencedirect.com/science/journal/16726308; http://www.ricescience.org.
Supplemental File 1. Materials and methods used in this study.Supplemental Table 1. Comparison of agronomic traits among wild type (WT), M12 and M13 mutants.
Supplemental Table 2. Fine mapping region contained 14 predicted genes.
Supplemental Table 3. Primers used in mapping GE gene.
Supplemental Fig. 1. The allelic test of M12 and M13.