Skip to main content
NCBI home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2023 May 1;120(19):e2220622120. doi: 10.1073/pnas.2220622120

Loss of function of SSIIIa and SSIIIb coordinately confers high RS content in cooked rice

Anqi Wang a,1, Yanhui Jing a,1, Qiao Cheng a,b,1, Hongju Zhou a, Lijun Wang a, Wanxin Gong c, Liquan Kou a, Guifu Liu a, Xiangbing Meng a, Mingjiang Chen a, Haiyan Ma a, Xiaoli Shu c, Hong Yu a,b,2, Dianxing Wu c,2, Jiayang Li a,b,d,2
PMCID: PMC10175802  PMID: 37126676

Significance

Resistant starch potentially reduces the postprandial glucose and insulin responses to help control type 2 diabetes, obesity, and other related diseases. Here, we studied a high-RS mutant rs4 (~10.8% in cooked rice) and revealed that deficiency in SSIIIb could further increase RS contents dependent on loss of function of SSIIIa. The duplication of SSIII and neofunctionalization of SSIIIa with high expression levels in the endosperm was associated with the reduced RS contents in the cooked grains of tested cereals, whereas the dicots without this neofunctionalization of SSIII showed high RS contents in their cooked seeds. These findings shed light on the molecular mechanism of RS biosynthesis in rice and provided evolutionary evidence to help breed high-RS varieties in different cereals.

Keywords: resistant starch, soluble starch synthase, cereals, SSIIIb, Oryza sativa

Abstract

The sedentary lifestyle and refined food consumption significantly lead to obesity, type 2 diabetes, and related complications, which have become one of the major threats to global health. This incidence could be potentially reduced by daily foods rich in resistant starch (RS). However, it remains a challenge to breed high-RS rice varieties. Here, we reported a high-RS mutant rs4 with an RS content of ~10.8% in cooked rice. The genetic study revealed that the loss-of-function SSIIIb and SSIIIa together with a strong Wx allele in the background collaboratively contributed to the high-RS phenotype of the rs4 mutant. The increased RS contents in ssIIIa and ssIIIa ssIIIb mutants were associated with the increased amylose and lipid contents. SSIIIb and SSIIIa proteins were functionally redundant, whereas SSIIIb mainly functioned in leaves and SSIIIa largely in endosperm owing to their divergent tissue-specific expression patterns. Furthermore, we found that SSIII experienced duplication in different cereals, of which one SSIII paralog was mainly expressed in leaves and another in the endosperm. SSII but not SSIV showed a similar evolutionary pattern to SSIII. The copies of endosperm-expressed SSIII and SSII were associated with high total starch contents and low RS levels in the seeds of tested cereals, compared with low starch contents and high RS levels in tested dicots. These results provided critical genetic resources for breeding high-RS rice cultivars, and the evolutionary features of these genes may facilitate to generate high-RS varieties in different cereals.

The prevalence of diabetes has become one of the biggest public health problems in recent years (1). Type 2 diabetes (T2D) was closely related to blood glucose abnormalities, and the dietary approaches were considered as an optimal strategy to manage and prevent T2D by regulating glucose homeostasis (2–4). Starch provided ~50% of human daily energy uptake, but the rapid digestion and absorption of starch could lead to glucose abnormalities and has become a serious health issue (5). Besides the rapidly digested starch, resistant starch (RS) is one kind of special starch that cannot be digested in the small intestine and passed on to the large intestine for slow fermentation (6, 7). Foods with high RS contents can suppress the acute elevation of blood glucose, which was considered as an effective dietary approach for human health management (8, 9). Rice, its kernel composed of more than 80% starch, is a major staple food for about half of the world’s population (10, 11). However, most cooked rice only contain low RS (< 2%), far below the criteria (> 10%) for the recommended healthy diet (12, 13). Therefore, developing high-RS rice varieties will provide effective ways to improve public health (14).

Starch is composed of two types of glucose polymers, amylose and amylopectin (15). Amylose consists of linear α-1,4-linked glucose chains with few branches, the synthesis of which is mainly controlled by the Waxy (Wx) gene that encodes granule-bound starch synthase (GBSS) (16–19). Whereas, amylopectin consists of highly branched chains of α-1,4- and α-1,6-linked glucose units, and its synthesis involves multiple processes mediated by soluble starch synthases (SSs), starch branching enzymes (BEs), and debranching enzymes (DBEs) (20–22). Although the molecular structure of RS in cooked rice remains elusive, amylose contents were found critical to RS formation (23, 24). The RS contents in indica rice cultivars with strong functional alleles of Wxa could reach 2%, which was higher than those of japonica rice cultivars harboring weak functional Wxb alleles (25, 26). Although Wx could contribute to RS formation, further elevating RS to significant levels requires other high-RS genes.

SSs play key roles in elongating amylopectin chains through the formation of α-1,4-glycoside bonds (27). The rice genome encodes nine SSs, which can be grouped into five classes, SSI to SSV. SSI, SSIIa, and SSIIIa could elongate amylopectin chains with different degrees of polymerization (DP) (28–31), while SSIVb and SSV functioned in the starch granule initiation process (32, 33), but the functions of other SSs, including SSIIb, SSIIc, SSIIIb, and SSIVa remained largely unknown. In our previous work, we found that SSIIIa was a high-RS gene and the RS content in the loss-of-function ssIIIa mutant in the indica rice background with Wxa allele was increased to ~5.5% in cooked rice (25). However, it is still critically required to identify other high-RS genes to further increase RS contents to meet the criteria for the recommended healthy diet. Here, we report the characterization of high-RS mutant rs4 containing ~10.8% RS content in the cooked rice and uncovered that the deficiency in both SSIIIa and SSIIIb in rs4 led to increased amylose, lipids, and RS contents. Furthermore, we found that SSIII was duplicated in different cereals with one paralog mainly expressed in the endosperm, while the other one mainly in leaves, and the duplication of SSIII was associated with low RS levels in these cereal grains compared with tested dicots. Our results shed light on generating high-RS varieties in rice and possibly other cereals.

Results

Characterization of High-RS Mutant rs4.

A high-RS mutant rs4 was generated from gamma-radiated indica rice R7954 (34). Compared with 1.8% RS content in the cooked rice of R7954, the RS content of rs4 reached 10.8% (Fig. 1A). rs4 displayed a normal plant morphology compared with R7954, despite floury seeds (Fig. 1 B and C). Meanwhile, the RS content in the raw rice of rs4 was also increased to ~2.8% (SI Appendix, Fig. S1). We then crossed rs4 with another indica rice ZF253 (RS content ~0.8%) to generate an F2 population. The genetic analysis of RS contents of 308 F2 plants showed a ratio of high (>6.0%)-to-low RS levels in 1:15 (24:284; χ2 = 0.19; P = 0.65), suggesting that rs4 may contain two recessive high-RS genes (SI Appendix, Fig. S2A and Table S1). We then sequenced four previously reported high-RS-related genes, SSIIIa, BEIIb, Wx, and BEI (25, 35–37), and found that SSIIIa harbored a G-to-A mutation at the 3′ splice site of intron 5 (SI Appendix, Table S2), which was identical to the allele in previously reported high-RS (5.8%) mutant b10, an ssIIIa mutant in the R7954 background (25). We further crossed rs4 with b10 to generate an F2 population of 208 plants, in which the segregation ratio was 1:3 (48:160; χ2 = 0.41; P = 0.52) for relative high-RS contents (>10.5%) and relative low (5 to 10.5%), suggesting that one recessive gene was responsible for the differences of RS levels between rs4 and b10 (SI Appendix, Figs. S2B and S3 and Table S3). We therefore used this F2 population to clone this recessive gene by MutMap (38, 39). The single-nucleotide polymorphism (SNP) index between two extreme pools revealed a peak at 1.1 Mb region located in the long arm of chromosome 4 (Fig. 1D), which contained seven indels or SNPs in four genes (Fig. 1E and SI Appendix, Table S4). We then genotyped these genes in individual plants, and found a 2-bp deletion was associated with the RS levels, which located in the 15th exon of SSIIIb (LOC_Os04g53310), leading to a premature stop codon (Fig. 1F).

Fig. 1.

Fig. 1.

Open in a new tab

Characterization of the high-RS mutant rs4 and cloning of SSIIIb. (A) Pedigree and RS contents of wild-type R7954, high-RS mutant b10 and rs4. Values are means ± SD (n = 3), and different letters at top of each column indicate a significant difference at P < 0.05 determined by Tukey’s HSD test. (B) Plant morphologies of R7954, b10, and rs4, bar = 20 cm. (C) Seed morphologies of R7954, b10, and rs4, bar = 0.5 cm. (D) SNP index between extreme low- and high-RS pools in an F2 population generated from the cross between rs4 and b10. (E) Mapping SSIIIb with extreme low- and high-RS plants in high SNP index region. Blue box, the homozygous b10 genotype; yellow box, the homozygous rs4 genotype; green box, the heterozygous genotype. (F) Gene structure of SSIIIb and mutation sites in b10 and rs4.

The wild-type SSIIIb encompasses almost 9 kb, comprising 16 exons and 3,651-bp cDNA. To validate whether the mutation of SSIIIb was responsible for the rs4 phenotype, we transformed the wild-type genomic SSIIIb containing a 2-kb upstream region into rs4, and found that the RS content of SSIIIb:gSSIIIb/rs4 was rescued to 5.7%, similar to that of the b10 mutant (Fig. 2A). Consistent with this, the SEMs of the endosperm of SSIIIb:gSSIIIb/rs4 were partially rescued from small spherical starch granules in rs4, which contained spherical starch granules variable in size similar to b10 (Fig. 2B). Taken together, the mutation in SSIIIb was responsible for the high RS in rs4.

Fig. 2.

Fig. 2.

Open in a new tab

SSIIIb is responsible for the high-RS phenotype in the rs4 mutant. (A) RS contents of R7954, b10, rs4, and SSIIIb:gSSIIIb/rs4. Values are means ± SD (n = 3), and different letters at the top of each column indicate a significant difference at P < 0.05 determined by Tukey’s HSD test. (B) Morphologies of seed and endosperm of R7954, b10, rs4, or SSIIIb:gSSIIIb/rs4. Intact seeds and transverse sections revealed by light microscopy (first column), and scanning electron micrographs of the endosperm in transverse sections with different magnifications (second to the fourth column) are shown. (C) RT-qPCR analyses of SSIIIa and SSIIIb at different developmental stages and tissues of rice. Expression levels are normalized to rice Actin. Endosperm and embryo are harvested at 10 days after flowering (DAF). Values are means ± SD (n = 3). (D) GUS staining of ProSSIIIb:GUS and ProSSIIIa:GUS. Bars = 5 mm.

SSIIIa and SSIIIb Are Paralogous Genes with Different Expression Patterns.

Rice SSIII family has two paralogous genes, SSIIIa and SSIIIb, and their protein sequences share more than 41% identity, especially at the C-terminal regions (SI Appendix, Fig. S4A). Moreover, SSIIIa and SSIIIb both contain seven conserved domains, including three carbohydrate-binding module 25 (CBM25), two coiled-coil (CC), a glycosyltransferase 5 (GT5), and a glycosyltransferase 1 (GT1) domain, while the orientation of CBM25 and CC was different in two proteins (SI Appendix, Fig. S4B).

To study the function of SSIIIb and SSIIIa, we first examined their expression patterns in different organs and tissues by qPCR. The results showed that SSIIIb was highly expressed in leaves, sheathes, and embryo; moderately in endosperm and stems; and lowly in roots and panicles, whereas SSIIIa was specifically expressed in the endosperm and embryo with undetectable expression levels in other tissues (Fig. 2C). The GUS staining analysis of the SSIIIa and SSIIIb promoters showed similar results as qPCR data that SSIIIa was mainly expressed in the endosperm and SSIIIb in leaves and stems (Fig. 2D). Moreover, we in-depth examined the expression levels of SSIIIa and SSIIIb at different developmental stages of endosperm and embryo and found that SSIIIa was predominantly expressed in the endosperm and moderately at early stages of 10 DAF and 15 DAF in the embryo (SI Appendix, Fig. S5 A and B). Interestingly, we found that SSIIIb was predominantly expressed in the embryo much higher than SSIIIa but moderately expressed in the endosperm apparently lower than SSIIIa (Fig. 2C and SI Appendix, Fig. S5 A and B). All these data suggested that SSIIIa and SSIIIb have divergent expression patterns in different organs and tissues.

Loss of Function of SSIIIb Elevates RS Content Dependent on Deficiency in SSIIIa.

To study the gene function of SSIIIb, we generated a homozygous ssIIIb single mutant (named as b20) by crossing R7954 with rs4. Surprisingly, we found that b20 showed no significant changes in the RS content (1.7%) compared with R7954 (Fig. 3A). Consistent with this, we further found that the SEM micrographs of the transverse section of the b20 endosperm had a similar morphology to that of R7954, including similar polygonal and compound starch granules with sharp edges and smooth flat surfaces (Fig. 3B). In addition, we examined whether the mutation of SSIIIa and SSIIIb affected the expression levels of other starch biosynthetic genes in the 15 DAF endosperm of R7954, b10, b20, and rs4 (SI Appendix, Fig. S6). The results showed that majority of 28 genes involved in starch synthesis had little differences, whereas SSI, SSIIIb in b10, and SSI in b20 were up-regulated significantly (P < 0.01 and fold change > 2) and AGPL2, AGPL3, AGPL4, and SSIIc in rs4 were down-regulated significantly (P < 0.01 and fold change < 0.5), suggesting that starch synthase activity might be compromised in the SSIII single mutant and ADP glucose pyrophosphorylase activity might decrease in the SSIII double mutant.

Fig. 3.

Fig. 3.

Open in a new tab

Increase in RS contents by loss of function of SSIIIb depends on deficiency in SSIIIa. (A) RS contents of R7954, b10, b20, and rs4. (B) Morphologies of seed and endosperm of R7954, b10, b20, or rs4. (C) RS contents of ZH11, ssIIIaCR-1, ssIIIbCR-1, and ssIIIaCR ssIIIbCR-1. (D) Morphologies of seed and endosperm of ZH11, ssIIIaCR-1, ssIIIbCR-1, or ssIIIaCR ssIIIbCR-1. (E) Genotypes and AAC contents of Wxa, Wxb, and Wxmp alleles. Int1-1, the first bp in the first intron. Ex4-53, the 53rd bp in the fourth exon. (F) RS contents of Jia58, ssIIIaCR-2, ssIIIbCR-2, and ssIIIa ssIIIbCR-2. In A, C, and F, values are means ± SD (n = 3), and different letters at the top of each column indicate a significant difference at P < 0.05 determined by Tukey’s HSD test.

Moreover, we applied CRISPR/Cas9 to edit SSIIIa and SSIIIb in japonica rice cultivar ZH11 with the Wxb allele and obtained an ssIIIaCR-1 mutant harboring 1-bp frameshift deletion in exon 1 and an ssIIIbCR-1 mutant harboring 1-bp frameshift deletion in exon 3 (SI Appendix, Fig. S7 A and B). Furthermore, the ssIIIaCR ssIIIbCR-1 homozygous double mutant was generated by crossing ssIIIaCR-1 and ssIIIbCR-1. Similar to the mutants in the indica background, the RS content of ssIIIaCR-1 was increased to 4.9% compared with 0.5% in ZH11 and further elevated to 8.5% in the ssIIIaCR ssIIIbCR-1 double mutant, whereas the ssIIIbCR-1 displayed no significant differences from ZH11 on the RS content (Fig. 3C). SEM revealed that ssIIIbCR-1 had regular polyhedral starch granules as ZH11, but the granules of ssIIIaCR-1 and ssIIIaCR ssIIIbCR-1 were spherical and variable in size and shape with irregular surfaces (Fig. 3D). All these data demonstrated that the loss of function of SSIIIb had no effect on RS content when SSIIIa functioned normally, but could further elevate the RS content in the ssIIIa mutant background.

GBSSI Activity Level Stringently Controls RS Contents of ssIIIa ssIIIb Mutants.

Since increases in RS contents mediated by SSIIIa required a high expression level of Wx, we therefore tested whether RS levels in the ssIIIa ssIIIb double mutant were also regulated by Wx expression levels. The Wxa allele was reported to have a significantly higher expression level and protein activity than those of Wxb allele (40), whereas the encoded protein of the Wxb allele showed significantly higher activity than that of Wxmp with similar expression levels (41). Therefore, in addition to R7954 with the Wxa allele and ZH11 with the Wxb allele, we used a Jia58 japonica cultivar with the Wxmp allele to generate ssIIIaCR-2, ssIIIbCR-2, and ssIIIa ssIIIbCR-2 mutants by CRISPR/Cas9 (Fig. 3E and SI Appendix, Fig. S7 A, C, and D). We found that ssIIIa ssIIIbCR-2 in the Jia58 background showed the lowest RS content, ssIIIaCR ssIIIbCR-1 in the ZH11 background a middle RS content, and rs4 in the R7954 background the highest RS content (Fig. 3 A, C, and F). The RS contents of these ssIIIa ssIIIb double mutants were highly correlated with the GBSSI activity levels in different backgrounds. Similarly, the RS contents of ssIIIa mutants were also correlated with the GBSSI activities in different backgrounds (Fig. 3 A, C, and F). All these results indicated that the GBSSI activity levels stringently influence RS biosynthesis of ssIIIa and ssIIIa ssIIIb mutants.

Increased Contents of Amylose and Amylose–Lipid Complex Are Associated with High RS Contents in ssIIIa ssIIIb.

To study the changes in chemical composition in ssIIIa- and ssIIIb-related mutants, we analyzed their physical–chemical properties in both ZH11 and R7954 backgrounds. For the total starch content, ssIIIaCR ssIIIbCR-1 showed no significant differences from ZH11, whereas rs4 was slightly but significantly lower than R7954 (Fig. 4A). The apparent amylose content (AAC) and pasting temperature of ssIIIa ssIIIb mutants in both backgrounds were higher than those of the corresponding ssIIIa mutants and wild types (Fig. 4 B and C), but the peak viscosity of ssIIIa ssIIIb mutants was lower than that of the related ssIIIa mutants and wild types (Fig. 4D). The chain-length distribution of amylopectin revealed that the portion of chain DP ≥ 36 declined in ssIIIa ssIIIb mutants (SI Appendix, Fig. S8 A–D).

Fig. 4.

Fig. 4.

Open in a new tab

Physical–chemical properties in ssIIIa- and ssIIIb-related mutants. (A) Total starch contents, (B) AAC, (C) pasting temperature, (D) peak viscosity, (E) total lipid contents, (F) crystallinity, and (G) amylose–lipid complex contents of R7954, b10, b20, rs4, ZH11, ssIIIaCR-1, ssIIIbCR-1, and ssIIIaCR ssIIIbCR-1. (A–G) Values are means ± SD (n = 3), and different letters at the top of each column indicate a significant difference at P < 0.05 determined by Tukey’s HSD test.

Total lipid contents of ssIIIa ssIIIb mutants in both ZH11 and R7954 backgrounds were also increased compared with the corresponding ssIIIa mutants and wild types (Fig. 4E). Furthermore, the X-ray diffraction showed that ssIIIa and ssIIIa ssIIIb had lower crystallinity and a significant peak for amylose–lipid complex in both backgrounds (Fig. 4F and SI Appendix, Fig. S8 E and F). The contents of amylose–lipid complex of ssIIIa and ssIIIa ssIIIb exhibited two-fold and three-fold increases, respectively, compared with their wild types, which was consistent with their total lipid contents (Fig. 4 E and G). Taken together, deficiencies in SSIIIa and SSIIIb in the rs4 mutant elevated the contents of apparent amylose, total lipid, and amylose–lipid complex, which may lead to a high level of RS.

Divergent Expression Patterns of SSIIIa and SSIIIb Are Critical to Their Differential Functions in Seeds and Leaves.

To fully understand the function of SSIIIa and SSIIIb in regulating RS biosynthesis, we tested whether they have a redundant enzymatic function in starch biosynthesis through overexpressing SSIIIb driven by the ubiquitin promoter or SSIIIa by its native promoter in rs4 and found that the RS contents of both Ubi:cSSIIIb/rs4 and SSIIIa:gSSIIIa/rs4 were reduced from 11.4% in rs4 to 1.1% and 1.7%, respectively, which were similar to the wild-type R7954 (1.8%) (Fig. 5A). Consistent with this, the SEM of both Ubi:cSSIIIb/rs4 and SSIIIa:gSSIIIa/rs4 showed regular starch granules (Fig. 5B). These results indicated that SSIIIa and SSIIIb were functionally redundant in starch biosynthesis.

Fig. 5.

Fig. 5.

Open in a new tab

Divergent functions of SSIIIa and SSIIIb in different tissues. (A) RS contents of R7954, rs4, Ubi:cSSIIIa/rs4, and SSIIIa:gSSIIIa/rs4. (B) Morphologies of seed and endosperm of R7954, rs4, Ubi:cSSIIIa/rs4, or SSIIIa:gSSIIIa/rs4. (C) RS contents of R7954, b10, SSIIIb:cSSIIIb/b10, and Ubi:cSSIIIb/b10. (D) Morphologies of seed and endosperm of R7954, b10, SSIIIb:cSSIIIb/b10, or Ubi:cSSIIIb/b10. (E) Total starch contents in leaves of ZH11, ssIIIaCR-1, ssIIIbCR-1, and ssIIIaCR ssIIIbCR-1 at the end of the day and night. (F) Starch levels in leaves of ZH11, ssIIIaCR-1, ssIIIbCR-1, and ssIIIaCR ssIIIbCR-1 detected by iodine staining at the end of the day and night. Bar = 0.1 cm. (G) Seed setting rate and (H) thousand-seed weight of ZH11, ssIIIaCR-1, ssIIIbCR-1, and ssIIIaCR ssIIIbCR-1. In A, C, and F, values are means ± SD (n = 3), in G and H, values are means ± SD (n = 10), and different letters at the top of each column indicate a significant difference at P < 0.05 determined by Tukey’s HSD test.

SSIIIa and SSIIIb were functionally redundant; however, ssIIIa but not ssIIIb showed an increase in RS contents (Fig. 3 A, C, and F). We therefore wondered whether the promoter of SSIIIb was responsible and overexpressed SSIIIb driven by either its native promoter or the ubiquitin promoter in b10. The results showed that only Ubi:cSSIIIb/b10 but not SSIIIb:cSSIIIb/b10 could reduce the RS content and rescue the starch granule morphology to the wild-type level (Fig. 5 C and D), suggesting that the SSIIIb promoter had a weak activity in the tissues for RS biosynthesis, possibly in the endosperm.

It was reported that the starch content was not altered in the flag leaf of T-DNA insertion mutant ssIIIa/flo5 in the japonica cultivars of the Dongjin background (30). As SSIIIb was highly expressed in leaves (Fig. 2C), we analyzed the starch contents in leaves using 5-week-old plants grown in field conditions. The results showed that compared with ZH11 the total leaf starch content was significantly reduced in ssIIIbCR-1 and ssIIIaCR ssIIIbCR-1 but had no significant differences in ssIIIaCR-1 at both the end of day and night (Fig. 5E). Consistent with this, ssIIIbCR-1 and ssIIIaCR ssIIIbCR-1 leaves showed less iodine staining than ssIIIaCR-1 and ZH11 at both the end of day and night, indicating that starch contents were reduced in ssIIIbCR-1 but not in ssIIIaCR-1 (Fig. 5F). All these results demonstrated that SSIIIa and SSIIIb played divergent roles in starch biosynthesis of different tissues owing to their differential tissue-specific expression patterns.

In addition, to test whether the inhibited starch biosynthesis in ssIIIa- and ssIIIb-related mutants also affected yield-related traits, we examined the plant morphology and seed phenotype, including plant height, tiller number, grain number per panicle, seed setting rate, and thousand-seed weight. The results showed that in both ZH11 and R7954 backgrounds, the plant height, tiller number, and grain number per panicle were not affected in ssIIIa and ssIIIb single mutants or in the ssIIIa ssIIIb double mutants. The seed setting rates in ssIIIbCR-1, ssIIIaCR-1, and ssIIIaCR ssIIIbCR-1 were all significantly decreased compared with ZH11, whereas only ssIIIaCR ssIIIbCR-1 showed a slightly but significantly decreased thousand-seed weight (Fig. 5 G and H). In contrast, in the R7954 background, the thousand-seed weight in b20 and b10 was all significantly lower than that of R7954, and the double mutant rs4 showed an even larger decrease in thousand-seed weight. However, the seed setting rate of b20 or b10 showed a slightly but not significantly decrease compared with that in R7954, but rs4 showed a significantly decreased seed setting rate (SI Appendix, Fig. S9). These results indicated that the loss of function of SSIIIa and SSIIIb has little effect on plant morphology, but showed different levels of negative effects on seed setting rates and thousand-seed weight in japonica and indica backgrounds.

Duplication and Divergent Expression Patterns of SSII and SSIII in Cereals Are Associated with Low Grain RS Contents.

Duplicated copies of SSII and SSIII were observed in many cereals but barely in dicots (27, 42, 43). As SSIIIa was predominantly expressed in seeds while SSIIIb mainly in leaves, we therefore studied whether this was a general phenomenon in different cereals. We downloaded the protein sequences of SSIII homologous genes in four cereals including rice, wheat (B genome), corn, and millet plus three dicots including soybean, Arabidopsis, and tomato and collected their expression levels in different tissues (44–50). The rice and wheat B genomes contain two SSIII genes, corn and soybean contain three, and other two dicots have only one SSIII gene. The phylogenetic analysis clearly clustered these proteins into three groups of SSIIIa and SSIIIb in cereals and SSIII in dicots (Fig. 6A). Interestingly, the genes in the three groups showed obviously divergent expression patterns. All SSIIIa in cereals were predominantly expressed in grains; in contrast, all SSIIIb in cereals were highly expressed in leaves. In dicots, SSIII genes were highly expressed in leaves of tomato, leaves or seeds of Arabidopsis, and leaves or roots of soybean (Fig. 6A). Moreover, we performed a similar analysis on SSII homologous genes and found that SSII showed a similar phenomenon (Fig. 6B). SSII proteins were clustered into four groups, including SSIIa, SSIIb, and SSIIc in cereals and SSII in dicots. SSIIa in cereals were predominantly expressed in seeds, and most of the SSIIb and SSIIc were highly expressed in leaves or inflorescences. In dicots, SSII genes were highly expressed in leaves and inflorescences of Arabidopsis, leaves and seeds of tomato, and leaves of soybean (Fig. 6B). Different from SSII and SSIII, SSI and SSV were not duplicated in rice (51), and SSIV was only duplicated in rice but not in wheat, corn, millet, soybean, Arabidopsis, and tomato, and SSIV homologous genes showed no conserved expression profiles (SI Appendix, Fig. S10). All these data showed that SSIIIb, SSIIb, and SSIIc in cereals showed similar expression patterns to SSIII and SSII in dicots, while SSIIIa and SSIIa may have experienced neofunctionalization of expression profile in cereals.

Fig. 6.

Fig. 6.

Open in a new tab

SSII and SSIII homologous genes and their expression patterns and RS contents in different cereals and dicots. (A) Phylogenetic tree and expression patterns of the SSIII gene family. (Left) The phylogenetic tree of SSIII family of cereals including Oryza sativa, Triticum aestivum (B genome), Zea mays, and Setaria italica and dicots including Glycine max, Arabidopsis thaliana, and Solanum lycopersicum. (Right) Z-score normalized expression levels of SSIII genes were shown in the heatmap. (B) Phylogenetic tree and expression patterns of SSII gene family. (Left) The phylogenetic tree of SSII family of cereals and dicots as listed in A. (Right) Z-score normalized expression levels of SSII genes were shown in the heatmap. Gray boxes indicate that the gene expression was not detected (ND). (C) RS contents of total starch in the cooked seeds of five dicots including soybean, mung bean, almond, lemon, and sunflower and four cereals including rice, wheat, corn, and millet. (D) Total starch contents of nine species as listed in C. In C and D, values are means ± SD (n = 3) and different letters at the top of each column indicate a significant difference at P < 0.05 determined by Tukey’s HSD test.

Since the disruption of SSIIIa high expression in endosperm was more critical in increasing RS contents, we therefore investigated whether the duplicated copies of SSIII were correlated with the RS levels in the seeds of different cereals and dicots. We collected the seeds of four cereals including rice, wheat, corn, millet and five dicots including soybean, mung bean, almond, lemon, and sunflower, and measured their RS contents. The results showed that all the four cereals have a low RS content less than 2% of total starch in their cooked grains, whereas the five dicots have high RS contents ranging from 3.3 to 10.6% in their cooked seeds (Fig. 6C). Meanwhile, the total starch contents were high in cereals but low in dicots (Fig. 6D). All these data suggested that the duplication and neofunctionalization of expression profiles in SSII and SSIII may be associated with the reduction of RS contents in cereal grains.

Discussion

The RS contents in indica and japonica rice varieties were below 2% (12, 26). Therefore, it is extremely hard to increase RS content in rice breeding by utilizing current rice varieties. To solve this challenging problem, we applied gamma radiation to mutate a low RS (~2%) hybrid-rice restorer line R7954 and generated high-RS mutants b10 (5.8%) and rs4 (10.8%) (Fig. 1A). Through a map-based cloning approach in our current and previous studies (25), we showed that the high-RS phenotype in rs4 was caused by the deficiencies in two paralogous genes SSIIIa and SSIIIb (Fig. 1 and SI Appendix, Fig. S3). Compared with R7954, SSIIIa harbored a G-to-A point mutation that leads to an alternative splicing and SSIIIb contained a 2-bp deletion leading to a premature stop codon in the rs4 mutant (Fig. 1F). In indica backgrounds, the ssIIIa single mutant b10 could increase the wild-type RS content from 1.7 to 5.8% in cooked rice (Fig. 1A and SI Appendix, Table S2). In contrast, the ssIIIb single mutation had no effect on the RS content, but when it pyramided with ssIIIa to form a double mutant ssIIIa ssIIIb (rs4), its RS contents was further increased to 10.8% (Fig. 3A). The complementation of rs4 by SSIIIa:gSSIIIa could completely reduce the RS content to a wild-type level (Fig. 5A), suggesting that SSIIIa and SSIIIb proteins were functionally redundant for RS biosynthesis. Meanwhile, the activity levels of GBSSI stringently controlled RS contents in ssIIIa ssIIIb (Fig. 3). Therefore, SSIIIa, SSIIIb, and Wx are all critical in creating high-RS rice varieties, whereas SSIIIa and SSIIIb negatively regulate RS contents and Wx positively regulates RS contents.

Based on the physical structure and digestive resistance mechanism, RS can be divided into five types: physically inaccessible starch (RS1), nongelatinized native starch granules (RS2), retrograded starch (RS3), chemically modified starch (RS4), and amylose–lipid complex (RS5) (52). RS3 and RS5 are the main components in cooked rice, whereas RS2 mainly exists in raw rice (53). In the rs4 mutant, the RS content in cooked rice and the AAC, lipids contents, and amylose–lipids complex were all significantly increased (Fig. 4 B, E, and G), suggesting that both RS3 and RS5 contributed to high RS levels in rs4. Meanwhile, the RS content in raw rice of rs4 was also increased (SI Appendix, Fig. S1). It appears that enhancing amylose biosynthesis and inhibiting amylopectin biosynthesis are two practical ways to elevate RS contents (52). Previous reports showed that suppression of BEIIb could increase RS levels (54), and both beIIb beI and ssIIIa beIIb double mutants could further increase RS contents remarkably (23, 35, 36). But, only specific combinations of mutants in starch synthase and branch enzymes could lead to high RS contents (33, 55, 56). Meanwhile, the loss-of-function mutations of these genes usually lead to irregular starch granules with a rounded shape, increased variation in size, and irregular surfaces, leading to undesired phenotypes such as poor eating and cooking quality, white core floury endosperm, and reduced grain yield. Here, we found that mutations of SSIIIa and SSIIIb adversely affected the yield and quality-related traits, but the levels were different in japonica and indica backgrounds (Figs. 4 and 5 G and H and SI Appendix, Fig. S9). In future, genome editing technology will allow us to systematically study different combinations of amylopectin biosynthetic genes to increase cereal RS contents and further to create unique beneficial alleles to optimize RS contents, grain quality, and yield.

Gene duplication is a major source of raw genetic material that drives the evolution of flowering plants (57). Among the five SS subclasses, SSII and SSIII were found duplicated in cereals before the grasses’ divergence (Fig. 6 A and B). Other studies also showed that AGPL3, AGPS, GBSS, SSII, SSIII, and SBEII gene families were duplicated before the grasses’ divergence (58, 59). Genes in the rice starch biosynthetic pathway can be classified into two groups due to their preferential expression in the endosperm or leaves (43, 60). However, the genes preferentially expressed in leaves were less studied. The main evolution consequences of duplicated genes are pseudogenization, neofunctionalization, and subfunctionalization (61). Here, based on the functional study of two isoforms of SSIII in rice and the evolutionary studies between typical cereals and other dicot species, we proposed that the duplication of SSIII and SSII in cereals may undergo a process of neofunctionalization in the gene expression of SSIIIa and SSIIa. Among the four tested cereals, all SSIIIa and SSIIa were predominately expressed in grains, whereas SSIIIb and SSIIb were predominately expressed in leaves (Fig. 6 A and B). Consistent with these expression patterns, OsSSIIIa functions predominantly in the endosperm and OsSSIIIb in leaves. Moreover, the SSII and SSIII proteins in the four examined dicots were all clustered in same clade and were preferentially expressed in leaves similar to SSIIb and SSIIIb in cereals, suggesting that SSIIa and SSIIIa might be neofunctionalized alleles by the changes of expression patterns possibly caused by their promoters.

Seeds in the tested dicot species without SSII and SSIII duplication all showed high RS levels and low starch contents; however, grains in the tested cereals all contained low RS levels and high starch contents (Fig. 6 C and D). Taken together with the findings that ssIIIa and ssIIIa ssIIIb mutants showed irregular starch granules with increased RS contents in rice, we thus proposed a model to show the relationship between SSIII neofunctionalization and RS contents in cereals and other dicot species (Fig. 7), and that the duplication of starch biosynthesis genes and neofunctionalization of endosperm-preferentially expressed alleles are associated with and possibly prefer a decreased RS content and densely packed starch granules in cereal grains. Moreover, the duplicated starch synthesis genes with their endosperm-preferential expression may be necessary for high-order starch structure. Although SSIIIb is lowly expressed in seeds and the ssIIIb single mutant shows no increase in RS levels, SSIIIb is critical in the ssIIIa mutant background to further elevate RS contents and breed high-RS rice varieties. Further functional study of their homologous genes may also provide practical ways in breeding high-RS varieties in other cereals.

Fig. 7.

Fig. 7.

Open in a new tab

Proposed model of SSIII duplication and reduced RS contents in cereals. SSII and SSIII were duplicated in cereals with the low RS content in seeds and the dicots contained high RS contents without neofunctionalization of SSII and SSIII. SSIIa and SSIIIa in cereals were preferentially expressed in the endosperm, whereas SSIIb, SSIIc, and SSIIIb in cereals and SSII and SSIII in dicots were preferentially expressed in leaves. To breed high-RS rice varieties, the deficiencies in SSIIIa and SSIIIb together could elevate both amylose and lipid contents to form amylose–lipid complex and increase the RS contents.

Materials and Methods

A high-RS rice mutant rs4 was isolated after 60-Cobalt gamma-ray mutagenesis of the wild-type R7954 (Oryza sativa L. ssp. indica) (34). The japonica cultivar ZH11 was used to generate CRISPR/Cas9 genome-editing mutants, including ssIIIaCR-1, ssIIIbCR-1, and ssIIIaCR ssIIIbCR-1 mutants, and the japonica variety Jia58 was also used to produce ssIIIaCR-2, ssIIIbCR-2, and ssIIIa ssIIIbCR-2 mutants through the CRISPR/Cas9 genome-editing. Details about plant growth conditions, plasmid construction and transformation, and GUS staining are described in SI Appendix, Materials and Methods. Physicochemical properties and starch analysis, SEM, real-time qRT-PCR analysis, sequence alignment, and expression data analysis were carried out according to protocols described in SI Appendix, Materials and Methods.

Supplementary Material

Appendix 01 (PDF)

Click here for additional data file. (905.3KB, pdf)

Acknowledgments

We thank Prof. Yaoguang Liu (South China Agricultural University) for providing the pYLCRISPR/Cas9Pubi-H vector and National Center for Gene Research (http://www.ncgr.ac.cn/scientific_databases.asp) and Chinese Academy of Sciences for providing BAC plasmid OSIGBa0150p03. This work was supported by grants from the National Key R&D Program of China (2021YFF1000202), the Hainan Excellent Talent Team, China Agriculture Research System (CARS-01-4), and Functional Rice Breeding (2022C02011).

Author contributions

A.W., H.Y., and J.L. designed research; A.W., Y.J., Q.C., H.Z., L.W., W.G., L.K., G.L., M.C., and H.M. performed research; A.W., L.K., X.M., X.S., H.Y., and D.W. analyzed data; and A.W. and J.L. wrote the paper.

Competing interests

Reviewer J.-K.Z. has coauthored one comment with J.L. in 2020.

Footnotes

Reviewers: J.-S.J., Kyung Hee University; Q.L., Yangzhou University; and J.-K.Z., Institute of Advanced Biotechnology and School of Life Sciences, Southern University of Science and Technology.

Contributor Information

Hong Yu, Email: hyu@genetics.ac.cn.

Dianxing Wu, Email: dxwu@zju.edu.cn.

Jiayang Li, Email: jyli@genetics.ac.cn.

Data, Materials, and Software Availability

All study data are included in the article and/or SI Appendix. The raw sequence data reported in this paper have been deposited in the Genome Sequence Archive at the National Genomics Data Center, China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences, under accession number CRA010048 (62).

Supporting Information

References

  • 1.Zimmet P., Alberti K. G., Shaw J., Global and societal implications of the diabetes epidemic. Nature 414, 782–787 (2001). [DOI] [PubMed] [Google Scholar]
  • 2.Wolever T. M. S., Carbohydrate and the regulation of blood glucose and metabolism. Nutr. Rev. 61, S40–S48 (2003). [DOI] [PubMed] [Google Scholar]
  • 3.Forouhi N. G., Misra A., Mohan V., Taylor R., Yancy W., Dietary and nutritional approaches for prevention and management of type 2 diabetes. BMJ. 361, k2234 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Chen L., et al. , Regulation of glucose and lipid metabolism in health and disease. Sci. China Life Sci. 62, 1420–1458 (2019). [DOI] [PubMed] [Google Scholar]
  • 5.Wee M. S. M., Henry C. J., Reducing the glycemic impact of carbohydrates on foods and meals: Strategies for the food industry and consumers with special focus on Asia. Compr. Rev. Food Sci. Food Saf. 19, 670–702 (2020). [DOI] [PubMed] [Google Scholar]
  • 6.Englyst H. N., Kingman S. M., Cummings J. H., Classification and measurement of nutritionally important starch fractions. Eur. J. Clin. Nutr. 46 (suppl. 2), S33–S50 (1992). [PubMed] [Google Scholar]
  • 7.Asp N. G., Resistant starch. Proceedings for the 2nd plenary meeting of EURESTA: European FLAIR Concerted Action No. 11 on physiological implications of the consumption of resistant starch in man. Eur. J. Clin. Nutr. 46, S1–S148 (1992). [PubMed] [Google Scholar]
  • 8.Raben A., et al. , Resistant starch: The effect on postprandial glycemia, hormonal response, and satiety. Am. J. Clin. Nutr. 60, 544–551 (1994). [DOI] [PubMed] [Google Scholar]
  • 9.Birt D. F., et al. , Resistant starch: Promise for improving human health. Adv. Nutr. 4, 587–601 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Qian Q., Guo L., Smith S. M., Li J., Breeding high-yield superior quality hybrid super rice by rational design. Natl. Sci. Rev. 3, 283–294 (2016). [Google Scholar]
  • 11.Chen R., et al. , Rice functional genomics: Decades’ efforts and roads ahead. Sci. China Life Sci. 65, 33–92 (2022). [DOI] [PubMed] [Google Scholar]
  • 12.Patterson M. A., Maiya M., Stewart M. L., Resistant starch content in foods commonly consumed in the United States: A narrative review. J. Acad. Nutr. Diet. 120, 230–244 (2020). [DOI] [PubMed] [Google Scholar]
  • 13.Maki K. C., et al. , Resistant starch from high-amylose maize increases insulin sensitivity in overweight and obese men. J. Nutr. 142, 717–723 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Yu H., Li J., Short- and long-term challenges in crop breeding. Natl. Sci. Rev. 8, nwab002 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Zeeman S. C., Kossmann J., Smith A. M., Starch: Its metabolism, evolution, and biotechnological modification in plants. Annu. Rev. Plant Biol. 61, 209–234 (2010). [DOI] [PubMed] [Google Scholar]
  • 16.Shure M., Wessler S., Fedoroff N., Molecular identification and isolation of the Waxy locus in maize. Cell 35, 225–233 (1983). [DOI] [PubMed] [Google Scholar]
  • 17.Sano Y., Differential regulation of waxy gene expression in rice endosperm. Theor. Appl. Genet. 68, 467–473 (1984). [DOI] [PubMed] [Google Scholar]
  • 18.Wang Z., et al. , The amylose content in rice endosperm is related to the post-transcriptional regulation of the waxy gene. Plant J. 7, 613–622 (1995). [DOI] [PubMed] [Google Scholar]
  • 19.Zhang J., Zhang H., Botella J. R., Zhu J. K., Generation of new glutinous rice by CRISPR/Cas9-targeted mutagenesis of the Waxy gene in elite rice varieties. J. Integr. Plant Biol. 60, 369–375 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Tian Z., et al. , Allelic diversities in rice starch biosynthesis lead to a diverse array of rice eating and cooking qualities. Proc. Natl. Acad. Sci. U.S.A. 106, 21760–21765 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Nakamura Y., Rice starch biotechnology: Rice endosperm as a model of cereal endosperms. Staerke 70, 1600375 (2018). [Google Scholar]
  • 22.Jeon J. S., Ryoo N., Hahn T. R., Walia H., Nakamura Y., Starch biosynthesis in cereal endosperm. Plant Physiol. Biochem. 48, 383–392 (2010). [DOI] [PubMed] [Google Scholar]
  • 23.Tsuiki K., Fujisawa H., Itoh A., Sato M., Fujita N., Alterations of starch structure lead to increased resistant starch of steamed rice: Identification of high resistant starch rice lines. J. Cereal Sci. 68, 88–92 (2016). [Google Scholar]
  • 24.Chen M. H., Bergman C. J., McClung A. M., Everette J. D., Tabien R. E., Resistant starch: Variation among high amylose rice varieties and its relationship with apparent amylose content, pasting properties and cooking methods. Food Chem. 234, 180–189 (2017). [DOI] [PubMed] [Google Scholar]
  • 25.Zhou H., et al. , Critical roles of soluble starch synthase SSIIIa and granule-bound starch synthase Waxy in synthesizing resistant starch in rice. Proc. Natl. Acad. Sci. U.S.A. 113, 12844–12849 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Parween S., et al. , Balancing the double-edged sword effect of increased resistant starch content and its impact on rice texture: Its genetics and molecular physiological mechanisms. Plant Biotechnol. J. 18, 1763–1777 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Pfister B., Zeeman S. C., Formation of starch in plant cells. Cell. Mol. Life Sci. 73, 2781–2807 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Fujita N., et al. , Function and characterization of Starch Synthase I using mutants in rice. Plant Physiol. 140, 1070–1084 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Fujita N., et al. , Characterization of SSIIIa-deficient mutants of rice: The function of SSIIIa and pleiotropic effects by SSIIIa deficiency in the rice endosperm. Plant Physiol. 144, 2009–2023 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Ryoo N., et al. , Knockout of a starch synthase gene OsSSIIIa/Flo5 causes white-core floury endosperm in rice (Oryza sativa L.). Plant Cell Rep. 26, 1083–1095 (2007). [DOI] [PubMed] [Google Scholar]
  • 31.Nakamura Y., et al. , Essential amino acids of Starch Synthase IIa differentiate amylopectin structure and starch quality between japonica and indica rice varieties. Plant Mol. Biol. 58, 213–227 (2005). [DOI] [PubMed] [Google Scholar]
  • 32.Abt M. R., et al. , STARCH SYNTHASE5, a noncanonical starch synthase-like protein, promotes starch granule initiation in Arabidopsis. Plant Cell 32, 2543–2565 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Toyosawa Y., et al. , Deficiency of Starch Synthase IIIa and IVb alters starch granule morphology from polyhedral to spherical in rice endosperm. Plant Physiol. 170, 1255–1270 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Shu X., Jia L., Ye H., Li C., Wu D., Slow digestion properties of rice different in resistant starch. J. Agric. Food Chem. 57, 7552–7559 (2009). [DOI] [PubMed] [Google Scholar]
  • 35.Zhu L., et al. , High-amylose rice improves indices of animal health in normal and diabetic rats. Plant Biotechnol. J. 10, 353–362 (2012). [DOI] [PubMed] [Google Scholar]
  • 36.Miura S., et al. , Generation and starch characterization of non-transgenic BEI and BEIIb double mutant rice (Oryza sativa) with ultra-high level of resistant starch. Rice (N Y). 14, 3 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Zhang Y., Li J., Gao C., Generation of stable transgenic rice (Oryza sativa L.) by Agrobacterium-mediated transformation. Curr. Protoc. Plant Biol. 1, 235–246 (2016). [DOI] [PubMed] [Google Scholar]
  • 38.Abe A., et al. , Genome sequencing reveals agronomically important loci in rice using MutMap. Nat. Biotechnol. 30, 174–178 (2012). [DOI] [PubMed] [Google Scholar]
  • 39.Lin T., et al. , A truncated F-box protein confers the dwarfism in cucumber. J. Genet. Genomics 43, 223–226 (2016). [DOI] [PubMed] [Google Scholar]
  • 40.Zhang C., et al. , Wx(lv), the ancestral allele of rice Waxy gene. Mol. Plant 12, 1157–1166 (2019). [DOI] [PubMed] [Google Scholar]
  • 41.Zhang C., et al. , A rare Waxy allele coordinately improves rice eating and cooking quality and grain transparency. J. Integr. Plant Biol. 63, 889–901 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Zhang X., Myers A. M., James M. G., Mutations affecting starch synthase III in Arabidopsis alter leaf starch structure and increase the rate of starch synthesis. Plant Physiol. 138, 663–674 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Ohdan T., et al. , Expression profiling of genes involved in starch synthesis in sink and source organs of rice. J. Exp. Bot. 56, 3229–3244 (2005). [DOI] [PubMed] [Google Scholar]
  • 44.Sato Y., et al. , RiceXPro version 3.0: Expanding the informatics resource for rice transcriptome. Nucleic Acids Res. 41, D1206–D1213 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Ma S., et al. , WheatOmics: A platform combining multiple omics data to accelerate functional genomics studies in wheat. Mol. Plant 14, 1965–1968 (2021). [DOI] [PubMed] [Google Scholar]
  • 46.Lawrence C. J., et al. , MaizeGDB: The maize model organism database for basic, translational, and applied research. Int. J. Plant Genomics 2008, 496957 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Yang Z., et al. , A mini foxtail millet with an Arabidopsis-like life cycle as a C4 model system. Nat. Plants 6, 1167–1178 (2020). [DOI] [PubMed] [Google Scholar]
  • 48.Sullivan A., et al. , An "eFP-Seq Browser" for visualizing and exploring RNA sequencing data. Plant J. 100, 641–654 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Grant D., Nelson R. T., Cannon S. B., Shoemaker R. C., SoyBase, the USDA-ARS soybean genetics and genomics database. Nucleic Acids Res. 38, D843–D846 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Garcia-Hernandez M., et al. , TAIR: A resource for integrated Arabidopsis data. Funct. Integr. Genomics 2, 239–253 (2002). [DOI] [PubMed] [Google Scholar]
  • 51.Huang L., Tan H., Zhang C., Li Q., Liu Q., Starch biosynthesis in cereal endosperms: An updated review over the last decade. Plant Commun. 2, 100237 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Shen L., Li J., Li Y., Resistant starch formation in rice: Genetic regulation and beyond. Plant Commun. 3, 100329 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Yi X., Li C., Main controllers for improving the resistant starch content in cooked white rice. Food Hydrocoll. 122, 107083 (2022). [Google Scholar]
  • 54.Sun Y., et al. , Generation of high-amylose rice through CRISPR/Cas9-mediated targeted mutagenesis of starch branching enzymes. Front. Plant Sci. 8, 298 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Itoh Y., Crofts N., Abe M., Hosaka Y., Fujita N., Characterization of the endosperm starch and the pleiotropic effects of biosynthetic enzymes on their properties in novel mutant rice lines with high resistant starch and amylose content. Plant Sci. 258, 52–60 (2017). [DOI] [PubMed] [Google Scholar]
  • 56.Zhang G., et al. , Double repression of soluble starch synthase genes SSIIa and SSIIIa in rice (Oryza sativa L.) uncovers interactive effects on the physicochemical properties of starch. Genome 54, 448–459 (2011). [DOI] [PubMed] [Google Scholar]
  • 57.McKain M. R., et al. , A phylogenomic assessment of ancient polyploidy and genome evolution across the Poales. Genome Biol. Evol. 8, 1150–1164 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Li C., Li Q., Dunwell J. M., Zhang Y., Divergent evolutionary pattern of starch biosynthetic pathway genes in grasses and dicots. Mol. Biol. Evol. 29, 3227–3236 (2012). [DOI] [PubMed] [Google Scholar]
  • 59.Lee S. K., et al. , Identification of the ADP-glucose pyrophosphorylase isoforms essential for starch synthesis in the leaf and seed endosperm of rice (Oryza sativa L.). Plant Mol. Biol. 65, 531–546 (2007). [DOI] [PubMed] [Google Scholar]
  • 60.Dian W., Jiang H., Wu P., Evolution and expression analysis of Starch Synthase III and IV in rice. J. Exp. Bot. 56, 623–632 (2005). [DOI] [PubMed] [Google Scholar]
  • 61.Zhang J., Evolution by gene duplication: An update. Trends Ecol. Evol. 18, 292–298 (2003). [Google Scholar]
  • 62.Wang A., Loss-of-function of SSIIIa and SSIIIb coordinately defines high RS content in rice. Genome Sequence Archive. https://bigd.big.ac.cn/gsa/browse/CRA010048. Accessed 2 March 2023.

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Appendix 01 (PDF)

Click here for additional data file. (905.3KB, pdf)

Data Availability Statement

All study data are included in the article and/or SI Appendix. The raw sequence data reported in this paper have been deposited in the Genome Sequence Archive at the National Genomics Data Center, China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences, under accession number CRA010048 (62).

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES