Structure and Properties of Starch in Rice Double Mutants Lacking Starch Synthase (SS) IIa and Starch Branching Enzyme (BE) IIb

Abstract

Starch biosynthetic enzymes form multi-protein complexes consisting of starch synthase (SS) I, SSIIa, and starch branching enzyme (BE) IIb, which synthesize amylopectin clusters. This study analyzed the starch properties in two double mutant rice lines lacking SSIIa and BEIIb, one of which expressed an inactive BEIIb protein. The ss2a be2b lines showed similar or greater seed weight than the be2b lines, and plant growth was not affected. The ss2a line showed increased short amylopectin chains resulting in a lower gelatinization temperature. Starch granule morphology and A-type crystallinity were similar between the ss2a line and the wild type, except for a mild chalky seed phenotype in the ss2a line. However, the starch phenotype of the ss2a be2b lines, which was similar to that of be2b but not ss2a, was characterized by increased long amylopectin chains, abnormal starch granules, and B-type crystallinity. The similarity in phenotype between the ss2a be2b and be2b lines may be attributed to the inability of the be2b mutants to generate short amylopectin branches, which serve as primers for SSIIa. Therefore, the presence or absence of SSIIa hardly affected the amylopectin structure under the be2b background. The amylose content was significantly higher in the ss2a be2b lines than in the be2b lines. Starch crystallinity was greater in ss2a be2b lines than in be2b lines, despite the fact that starch crystallinity is generally negatively correlated with amylose content. This suggests that the formation of a double helix between long amylopectin chains and amylose affects starch crystallinity in the ss2a be2b mutants.

Abbreviations

BE, starch branching enzyme; DAF, days after flowering; dCAPS, derived cleaved amplified polymorphic sequence; DP, degree of polymerization; GBSS, granule-bound starch synthase; RSC, relative starch crystallinity; SS, starch synthase; WT, wild-type.

INTRODUCTION

Starch is composed of two glucose polymers, essentially linear amylose and highly branched amylopectin, the latter of which is the major component of starch. Amylopectin forms a tandem cluster structure by repeating cluster regions and non-cluster regions. The cluster regions form double helices with adjacent glucose chains, while the non-cluster regions are enriched with branch points.¹⁾

Amylose is solely synthesized by granule-bound starch synthase (GBSS) I, which elongates α-1,4 glucans using ADP-glucose generated by ADP-glucose pyrophosphorylase.²⁾ Whereas, amylopectin is synthesized by the orchestrated actions of several enzymes. Starch synthase (SS) elongates α-1,4 linked glucans using ADP-glucose as the substrate. Starch branching enzyme (BE) introduces α-1,6 linked branches to the glucans generated by SS. Excess branches are trimmed off by starch debranching enzymes such as isoamylase and pullulanase. In addition, phosphorylase is involved in the initiation step of starch synthesis.²⁾ Isozymes of SS and BE have preferred primer or substrate structures and specific end product structures, and ultimately determine the structure of amylopectin. The generation of amylopectin clusters is initiated by the elongation of long glucans by SSIIIa and the generation of branches by BEI.²⁾ Then, BEIIb adds multiple short branches to the long branches generated by BEI. Short branches generated by BEIIb with a degree of polymerization (DP) 6–7 are elongated to DP 8–12 by SSI,³⁾ and further elongated to DP 13–24 by SSIIa.²⁾

The enzymes involved in amylopectin biosynthesis form a multi-protein complex that was first identified in wheat,⁴⁾⁵⁾ and later in other cereals such as maize^6)7)8)9) and rice.¹⁰⁾ The protein complex found in maize consists of SSI, SSIIa, and BEIIb, and SSIIa is located at the center.¹¹⁾ These isozymes are responsible for generating cluster regions of amylopectin enriched with short and medium chains, and the formation of hetero-trimeric protein complexes is thought to facilitate the efficient synthesis of amylopectin clusters.⁹⁾¹²⁾

In rice, the presence of the hetero-trimeric protein complex SSI-SSIIa-BEIIb was also suggested,¹⁰⁾ although the activity of SSIIa in japonica cultivars is one-tenth of that in indica cultivars because of amino acid substitutions.¹³⁾ Starch structure and properties, as well as protein complex formation, have been analyzed using single mutants of japonica rice lacking SSI, SSIIa, or BEIIb. The absence or mutation of a component of the hetero-trimeric protein complex often induces the recruitment of isozymes to form an alternative protein complex, which results in altered starch structure and properties.¹⁴⁾¹⁵⁾

The rice ss2a null mutant EM204 shows a 15 % reduction in seed weight, as well as increased short amylopectin chains and reduced gelatinization temperature. This is caused by the inability of short amylopectin chains (DP ≤ 11) to be elongated to DP ≥ 12 because of the loss of SSIIa.¹⁴⁾ The protein complex found in the ss2a mutant consists of SSI-SSI-BEIIb, where loss of SSIIa is complemented by SSI.¹⁴⁾

Protein complex formation has been analyzed in two be2b rice mutant lines: a null be2b mutant designated as be2b (-) or EM10, and a line expressing an inactive BEIIb protein designated as be2b (+) or ssg3. These be2b mutants indicate that BEIIa is recruited to form an alternative protein complex.¹⁵⁾ The be2a single mutant does not show changes in starch structure,²⁾ suggesting that BEIIa is not essential in the presence of BEIIb, whereas it plays an important role in amylopectin synthesis in the absence of BEIIb.¹⁶⁾ Inactive BEIIb protein associates with several starch biosynthetic enzymes to form high molecular weight protein complexes of >700 kDa.¹⁵⁾ However, seed weight is lower in be2b (+) than in be2b (-), suggesting that formation of a large protein complex in be2b (+) inhibits amylopectin biosynthesis.¹⁵⁾

To understand how the hetero-trimeric complexes are complemented and how alteration of the protein complex affects starch structure, we isolated double mutants lacking ss2a and be2b among the three components of the hetero-trimeric protein complex. Endosperm starch structure and characteristics, as well as seed morphology, were analyzed using ss2a be2b (-), named #1520, and ss2a be2b (+), named #1522 in this study.

MATERIALS AND METHODS

Plant materials. Two double mutant lines, ss2a be2b (-) (#1520) and ss2a be2b (+) (#1522), were generated by crossing a ss2a null mutant (EM204)¹⁴⁾ with a be2b null mutant [be2b (-), EM10]¹⁵⁾¹⁶⁾ or a mutant expressing an inactive BEIIb protein [be2b (+), ssg3],¹⁵⁾¹⁷⁾ respectively. The parental cultivars Kinmaze, the parent of EM204 and EM10, and Nipponbare, the parent of ssg3, were used as wild-type (WT) lines. All rice plants were grown in an experimental paddy field at Akita Prefecture University under natural conditions during the summer season.

Genotype analyses by molecular marker. Genomic DNA was isolated from the seedlings following Crofts et al. (2015)¹⁰⁾ and analyzed using a derived cleaved amplified polymorphic sequence (dCAPS) marker. PCR amplification was performed using the following pairs of primers: EM204 specific primers 5′-CAGACAGGTGAAGCTTCTATCTG-3′ and 5′-CAAAACAGAATCATGCGCTTCATGAGATC-3′; EM10 specific primers 5′-GACTATGCAGGAGAAGTACATATTCAAGC-3′ and 5′-CTGTGAACAACATCCATGAGCAC-3′; and ssg3 specific primers 5′-TGAATTTTAATGAATACTGGCA-3′ and 5′-TGAATCCAGGGAATGCAGAAC-3′, followed by a second round of PCR with 5′-CATTGCATAAAATGATTAGACTTATCACCATG-3′ and 5′-CTAGGTGACAAGAACCATTGTCATC-3′. PCR reactions were performed using QuickTaq following the manufacturer’s instructions (TOYOBO Co., Ltd., Osaka, Japan). PCR products were digested using the following restriction enzymes: NlaIV (New England Biolabs Inc. (NEB), Ipswich, MA, USA) for EM204 and EM10, and NcoI (NEB) for ssg3. Digested samples were separated using 7.5 % acrylamide gels for EM204 and ssg3, and 2 % agarose gels for EM10. Gels were stained with ethidium bromide.

SEM observation. Scanning electron microscopy observation of purified starch granules was performed (TM3030Plus Miniscope^®; Hitachi High-Tech Corporation, Tokyo, Japan) according to the method described by Fujita et al. (2009)¹⁸⁾ and Fujita et al. (2011).¹⁹⁾

Analysis of starch structure. Gel filtration chromatography of debranched starch was performed as described by Toyosawa et al. (2016).²⁰⁾ The chain length distribution of endosperm α-glucans was analyzed by capillary electrophoresis (P/ACE MDQ carbohydrate system; AB Sciex, Framingham, MA, USA), according to the method described by O’Shea and Morell (1996),²¹⁾ and Fujita et al. (2001).²²⁾

Thermal properties of endosperm starch. The thermal properties of endosperm starch were determined by differential scanning calorimetry (DSC) (DSC6100; Seiko Instruments Inc., Chiba, Japan), according to the method described by Fujita et al. (2009).¹⁸⁾

X-ray diffraction. X-ray diffraction analysis was performed, and relative starch crystallinity (RSC) was determined using an internal standard as described by Hamanishi et al. (2000),²³⁾ Kodama et al. (2011),²⁴⁾ and Abe et al. (2013).³⁾

RESULTS AND DISCUSSION

Selection of double mutants using molecular markers.

The absence of SSIIa was analyzed using the dCAPS marker in genomic DNA isolated from seedlings (Fig. 1A, B). An undigested DNA band was detected at 141 bp in the WT, whereas ss2a (EM204), ss2a be2b (-) (#1520), and ss2a be2b (+) (#1522) showed a DNA band at 111 bp after NlaIV restriction enzyme digestion (Fig. 1A, B). The genotype of be2b (-) was analyzed using the dCAPS marker in genomic DNA (Fig. 1C). The PCR product from the WT was digested into 579 bp and 447 bp bands by the NlaIV restriction enzyme, whereas be2b (-) (EM10) and ss2a be2b (-) were not cleaved by the restriction enzyme and were detected at 1,026 bp (Fig. 1C). Analysis of the genotype of be2b (+) using the dCAPS marker in genomic DNA (Fig. 1D) detected a band at 145 bp digested by NcoI in the WT, whereas be2b (+) (ssg3) and ss2a be2b (+) (#1522) showed an undigested DNA band at 173 bp (Fig. 1D). The double mutant lines with the correct genotype were grown and used for further study.

Fig. 1. Genotype analyses using the dCAPS marker in genomic DNA isolated from young leaves.

　(A) and (B) show the ss2a genotype; (C) shows the be2b (-) genotype; and (D) shows the be2b (+) genotype. EM204, ss2a; EM10, be2b (-); ssg3, be2b (+); #1520, ss2a be2b (-); and #1522, ss2a be2b (+).

　

Seed morphology and grain weight.

Seed morphology was observed by illuminating the dehulled seeds from above or below (Fig. 2). Both the ss2a be2b double mutant lines and be2b lines had an opaque seed phenotype, and a dark shadow was observed when the seeds were illuminated from below. Seed weight was greater in the ss2a be2b double mutants than in the parental be2b lines from two different harvest years (Table 1), despite the fact that they lacked SSIIa in addition to BEIIb. The seeds were heavier in 2019 than in 2018 in most lines except in the ss2a (EM204) line (Table 1). This is likely because the temperature during seed development was higher in 2019 than in 2018 (Fig. 3), and the relatively warmer temperature may have enhanced starch biosynthesis (Table 1). The WT lines, Kinmaze and Nipponbare, produced translucent seeds in both harvest years, whereas the ss2a (EM204) line produced a greater amount of mildly chalky seeds in 2019 (data not shown) than in 2018 (Fig. 2). The proportion of partially opaque seeds increases and the seed weight decreases when BEIIb expression levels are decreased in WT japonica rice cultivars (‘Akitakomachi’) in the presence of high temperatures during seed development, which leads to abnormal starch synthesis.²⁵⁾ The weather in July 2019 was quite different from 2018 in Akita area (data not shown). For example, the average temperature up to August 20^th was much higher in 2019. This likely promoted the flowering date of EM204 by two weeks. The cumulative temperature between 5 and 30 days after flowering (DAF) in 2018 was 540.3 ˚C, while that in 2019 was approximately 100 ˚C higher and was 638.1 ˚C (Fig. 3). Therefore, it is possible that the high temperature in 2019 affected ss2a (EM204), increasing the rate of partially opaque or chalky seeds and decreasing seed weight.

Fig. 2. Seed morphology observed by stereo-microscope.

　The top row images were taken with an overhead light, and the bottom row images were taken with light from beneath. Dehulled seeds harvested in 2018 were used. Scale bars, 5 mm.

Table 1. Genotype of rice lines used in this study and dehulled grain weight in two different harvest years.

Line name	Genotype	Background	Grain weight (mg)
			'2018	'2019
#1520	ss2a be2b (-)	Kin	13.5±0.2c (72.3)	14.0±0.2d (64.8)
#1522	ss2a be2b (+)	Kin/Nip	14.8±0.3c (79.5)	17.2±0.3c (79.6)
EM204	ss2a	Kin	18.1±0.4a (97.4)	17.2±0.4c (79.8)
EM10	be2b (-)	Kin	11.3±0.4d (60.6)	12.1±0.3e (56.1)
ssg3	be2b (+)	Nip	7.7±0.5e (41.2)	13.1±0.4de (60.5)
Kinmaze (Kin)	WT	Kin	18.6±0.5a (100)	21.6±0.3a (100)
Nipponbare (Nip)	WT	Nip	17.8±0.5b (95.5)	19.1±0.5b (88.5)

Values are expressed as the mean ± SE (n = 20). Values in parentheses are the percentages of the wild type (WT, Kinmaze). Samples annotated with different letters (a–e) are significantly different from one another as determined using the Tukey–Kramer method (p < 0.05).

Fig. 3. Temperature change in Akita, Japan from August to September in 2018 and 2019.

　The line graphs show the average temperature of 5 days using the data released from Japan Meteorological Agency (https://www.data.jma.go.jp/obd/stats/etrn/index.php). Flowering date (*) and cumulative temperature (**) during seed development from 5 days after flowering (DAF) to 30 DAF in EM204 (ss2a) and ssg3 (be2b (+)) are as indicated.

By contrast, the seeds of be2b (+) (ssg3) showed a 1.7-fold greater weight in 2019 than in 2018, with 60.5 % of the weight of the WT, Kinmaze, in 2019 and 41.2 % in 2018. This is also likely due to differences in temperature during seed development (Fig. 3). Flowering in the be2b (+) (ssg3) mutant occurs in early September, which is later than in other lines (Fig. 3). The cumulative temperature during be2b (+) (ssg3) seed development (DAF 5–30) in 2019 (626.9 ˚C) was much higher than in 2018 (534.4 ˚C; Fig. 3). This likely promoted starch biosynthesis and production of heavier seeds in 2019 than in 2018 (Fig. 2). However, we cannot exclude the possibility that be2b (+) (ssg3) expresses genes that increase the susceptibility to extreme growth conditions because the effect of temperature differences was greater for be2b (+) (ssg3) than for other lines. There were no significant differences in seed weight between ss2a be2b (-) (#1520) and ss2a be2b (+) (#1522) in 2018; however, seed weight was significantly greater in ss2a be2b (+) (#1522) than in ss2a be2b (-) (#1520) lines in 2019. This suggests that ss2a be2b (+) (#1522) inherited genes related to the susceptibility to environmental factors from be2b (+) (ssg3) (Table 1).

　

Morphology of starch granules.

Starch granule morphology in ss2a (EM204) was polygonal and similar to that of the WT lines Kinmaze and Nipponbare. However, the ss2a be2b double mutant lines #1520 and #1522 had round starch granules of various sizes similar to those of the be2b lines EM10 and ssg3 (Fig. 4). This is a typical characteristic associated with loss of BEIIb¹⁷⁾²⁶⁾, which results in the production of large starch grains within amyloplast that do not easily break into individual starch granules. This may be caused by reduced amylopectin biosynthesis and loose packaging of starch due to loss of BEIIb. These results indicate that loss of BEIIb led to the production of opaque seeds and altered starch granule morphology in the ss2a be2b double mutant lines.

Fig. 4. Scanning electron microscopy observation of purified starch granules in various mutant and wild-type lines.

　Scale bars, 10 or 20 μm as indicated.

　

Apparent amylose content.

The apparent amylose content of the be2b (+) (ssg3) mutant was 20.7 %, which was similar to the levels of the WT (19.2–20.7 %)²⁷⁾. The apparent amylose content of ss2a (EM204)¹⁴⁾ and be2b (-) (EM10)²⁷⁾ was 24.1 % and 27.4 %, respectively, which was higher than that of ssg3 and the WT lines. The apparent amylose content of the ss2a be2b double mutant lines was 29.8–32.2 %, which was higher than that of the parental lines (Table 2).

Previous studies reported an increase of amylose in some mutant rice lines. For example, the apparent amylose content of ss3a be2b²⁶⁾ and be1 be2b²⁹⁾ is 45.1 % and 51.7 %, respectively, which is higher than that of the be2b single mutant²⁶⁾ (28.1 %). The reason for this is that loss of SS and BE increases GBSSI expression levels.²⁶⁾ The ss2a (EM204) mutant also showed an increase in GBSSI expression levels.¹⁴⁾ The increased amylose content in these mutant rice lines can be explained by the reduction of amylopectin biosynthesis resulting in leftover ADP-glucose as a substrate, and GBSSI was able to use ADP-glucose for amylose synthesis.¹⁴⁾

Table 2. Carbohydrate content (weight %) in debranched endosperm starch fractions separated by gel filtration chromatography (Toyopearl HW55S/HW50S×3).

Line name	Genotype	Fr. I (%)*	Fr. II (%)**	Fr. III (%)***	III/II
#1520	ss2a be2b (-)	29.8±0.2ab	34.7±0.6	35.5±0.6	1.0±0.0b
#1522	ss2a be2b (+)	32.2±0.4a	34.5±0.4	33.3±0.6	1.0±0.0b
EM204X	ss2a	24.1±0.5cd	17.8±0.2	58.1±0.7	3.3±0.1a
EM10Y	be2b (-)	27.4±0.8bc	37.5±0.6	35.2±0.3	0.9±0.0b
ssg3	be2b (+)	20.7±0.7de	39.1±1.0	40.2±0.3	1.0±0.0b
KinmazeY	WT	20.7±0.2de	19.8±0.2	60.4±0.2	3.2±0.1a
Nipponbare	WT	19.2±0.9e	21.4+1.6	59.4±2.5	2.8±0.3a

^*Apparent amylose content. ^**Long amylopectin chains. ^***Short amylopectin chains. Values represent the mean ± SE (n = 3). Samples annotated with different letters (a–e) are significantly different from one another as determined by the Tukey–Kramer method (p < 0.05). ^XData from Miura et al. (2018).^{14) Y}Data from Itoh et al. (2017).²⁷⁾

The seed weight of double mutant lines lacking SS and BE isozymes, such as ss3a be2b,²⁶⁾ ss1^L be2b,²⁸⁾ and be1 be2b,²⁹⁾ is greater than that of the be2b single mutant line. Consistent with previous studies, the seed weight of the two ss2a be2b lines analyzed in this study was also increased.

Based on previous reports, there are two possible explanations for the increase of apparent amylose content in the ss2a be2b mutant lines analyzed in this study. One of these is that the loss of two enzymes involved in amylopectin biosynthesis (SSIIa and BEIIb) decreased the content of amylopectin, and the proportion of amylose increased relative to the amylopectin content. The other possibility is that loss of BEIIb decreased the number of non-reducing ends of the amylopectin molecule, and loss of SSIIa further decreased the amount of ADP-glucose available for amylopectin biosynthesis. Excess ADP-glucose can be used to enhance amylose synthesis. Future studies should compare the expression levels of GBSSI between ss2a be2b and be2b mutant lines.

The apparent amylose content of be2b (+) (ssg3) harvested in 2019 was 20.7 %, which was lower than that of ss2a (EM204; 24.1 %) and be2b (-) (EM10; 27.4 %) (Table 2). Because high temperature decreases the expression levels of GBSSI,²⁵⁾ it is likely that the low apparent amylose content in be2b (+) (ssg3) harvested in 2019 was due to high temperature during seed development. In addition, amylopectin biosynthesis increases under high temperature, resulting in a higher amylopectin content relative to the amylose content.

　

Analysis of amylopectin structure.

The ratio of short to long amylopectin chains was calculated by measuring Fr. III/Fr. II of gel filtration chromatography. The III/II value of the ss2a be2b double mutant lines was similar to that of the be2b single mutant lines (1.0), and considerably lower than that of the WT (2.8–3.2) and the ss2a single mutant (3.3) (Table 2). These results suggest that the absence of BEIIb decreased the content of short amylopectin chains and increased that of long amylopectin chains.

To compare the structure of amylopectin in detail, subtraction curves were generated by subtracting the pattern of the WT from that of the ss2a be2b and parental mutants (Fig. 5). The subtraction curves of ss2a be2b (-) (#1520) and ss2a be2b (+) (#1522) were similar to those of the parental mutants be2b (-) (EM10) and be2b (+) (ssg3) (Figs. 5A and B). This indicated that regardless of the presence or absence of inactive BEIIb proteins or the absence of SSIIa, loss of BEIIb activity had a considerable effect on amylopectin structure. However, small differences were observed between the ss2a be2b and be2b lines: the ss2a be2b lines had more short amylopectin chains (DP 6 and DP 7–10) and fewer intermediate to long amylopectin chains (DP 11–16 and DP 19–34) (Fig. 5C). These differences resembled the characteristics of loss of SSIIa, although the differences were minimal (< 1 % in ΔMolar %) (Fig. 5C).

Fig. 5. Differences in chain length distribution patterns of amylopectin from the endosperm. (A) and (B).

　The chain length distribution patterns of single and double mutant lines were subtracted from that of the WT to show differences in amylopectin branch structure. (C) The chain length distribution patterns of ss2a be2b double mutant lines were subtracted from that of the parental be2b single mutant line. Regions that contributed to decreased and increased gelatinization temperature are shown in blue and red, respectively. Line names and genotypes are as follows: Kin, Kinmaze, WT; EM204, ss2a; EM10, be2b (-); ssg3, be2b (+); #1520, ss2a be2b (-); and #1522, ss2a be2b (+).

The activity of SSIIa in japonica rice cultivars is one-tenth of that in indica rice cultivars.¹³⁾ A previous study analyzed the effects of these two SSIIa enzymes in the absence of BEIIb using ss2a^L be2b and SS2a be2b rice lines. The results showed that the be2b mutant lines (SS2a be2b) had more intermediate amylopectin chains (DP 11–18), whereas only a small difference was observed in DP 6–12 and DP 13–24 chains, although SSIIa activity was markedly different between the lines. This indicates that SSIIa cannot elongate amylopectin chains because of the lack of amylopectin branches with DP 6–12, which serve as primers in the absence of BEIIb.²⁷⁾

The present results are consistent with those reported previously. Loss of BEIIb reduced the number of short amylopectin chains with DP 6–12, resulting in a similar amylopectin structure between ss2a and ss2a^L because ss2a^L could not function in the absence of BEIIb.

　

Thermal properties of endosperm starch.

To examine the thermal properties of purified endosperm starch, onset gelatinization temperature (T_o), peak gelatinization temperature (T_p), conclusion gelatinization temperature (T_c), and gelatinization enthalpy (ΔH) were measured by differential scanning calorimetry. T_o, T_p, and T_c were 3–11 ˚C lower in ss2a (EM204) than in the WT, Kinmaze (Table 3), which is consistent with previous results.¹⁴⁾ T_o, T_p, and T_c were 10–18 ˚C higher in be2b (-) (EM10) and be2b (+) (ssg3) than in the parental WT line (Table 3), which is also consistent with previous results.²⁷⁾ T_o, T_p, T_c, and gelatinization enthalpy (ΔH) were higher in the two ss2a be2b mutant lines, ss2a be2b (-) (#1520) and ss2a be2b (+) (#1522), than in the WT lines, but lower than that in the parental be2b mutant lines.

Gelatinization temperature is closely correlated with chain length distribution of amylopectin.^2)14)30) In particular, amylopectin chains with DP < 24 greatly affect gelatinization temperature because they account for > 95 % amylopectin chains.³¹⁾ A higher number of short amylopectin chains with DP < 13 or a lower number of long amylopectin chains with 13 ≤ DP < 25 result in a lower gelatinization temperature. Conversely, a lower number of short amylopectin chains with DP < 13 or a higher number of long amylopectin chains with 13 ≤ DP < 25 result in a higher gelatinization temperature.³²⁾³³⁾ The reason why ss2a be2b mutant lines had a lower gelatinization temperature than be2b single mutant lines may be related to the subtraction curve of amylopectin chain length distribution: “#1520-EM10” and “#1522-ssg3” both had a greater number of short chains with DP < 13 and a smaller number of long chains with 13 ≤ DP < 25 (Fig. 5C).

Table 3. Thermal properties of purified endosperm starch in mutant and wild-type lines.

Line name	Genotype	To (˚C)*	Tp (˚C)**	Tc (˚C)***	ΔH (mJ/mg)****
#1520	ss2a be2b (-)	51.5±0.4ab	64.4±0.4ab	74.6±0.1a	11.7±0.7abc
#1522	ss2a be2b (+)	53.5±0.7ab	65.7±0.2ab	75.6±0.2a	11.4±0.7abc
EM204X	ss2a	33.9±0.5b	47.2±0.4d	56.5±0.2c	7.1±0.0c
EM10	be2b (-)	57.4±0.2ab	68.1±0.2ab	78.0±0.4a	14.4±0.2ab
ssg3	be2b (+)	60.9±2.2a	72.5±0.1a	79.6±0.3a	15.8±0.5a
KinmazeX	WT	45.4±0.2ab	52.8±0.0cd	59.6±0.2bc	9.5±0.4bc
Nipponbare	WT	50.2±0.1ab	57.1±0.1c	63.7±0.2b	15.7±1.3a

^*Onset temperature. ^**Peak temperature. ^***Conclusion temperature. ^****Gelatinization enthalpy of starch. Values are expressed as the mean ± SE (n = 3). Samples annotated with different letters (a–d) are significantly different from one another as determined by the Tukey–Kramer method (p < 0.05). ^XData from Miura et al. (2018).¹⁴⁾

　

Crystallinity of starch granules.

The crystallinity of endosperm starch granules was analyzed by X-ray diffraction (Fig. S1 ; see J. Appl. Glycosci. Web site). The WT lines Kinmaze and Nipponbare, and the ss2a (EM204) line showed typical A-type crystallinity, whereas the be2b single mutants, be2b (-) (EM10) and be2b (+) (ssg3), and two ss2a be2b mutant lines, ss2a be2b (-) (#1520) and ss2a be2b (+) (#1522), showed typical B-type crystallinity (Fig. S1 ; see J. Appl. Glycosci. Web site). The starch of WT rice shows A-type crystallinity,³⁴⁾³⁵⁾ whereas the starch of the be2b mutant displays B-type crystallinity in rice and maize.^3)34)35)

To determine the degree of crystallinity, starch was exposed to calcium fluoride as an internal standard, and relative starch crystallinity (RSC) was calculated using three different methods (Table 4). In the first method, in which the influence of starch moisture content is minimal,²³⁾ the height of 3b and 4b peaks represented starch with A-type crystallinity, whereas the height of 3b peak represented starch with B-type crystallinity.³⁶⁾ The peak height ratio was calculated relative to the peak height of calcium fluoride. In the second method, the peak area of 3b and 4b indicated starch with A-type crystallinity, whereas that of 3b indicated starch with B-type crystallinity. The ratio was calculated relative to the peak area of calcium fluoride. In the third method, a total peak area of 4°< 2θ < 27.5° was divided by the peak area of calcium fluoride.³⁾

Table 4. Type of crystallinity (A- or B-type) and relative starch crystallinity (RSC) of purified starch by X-ray diffraction analysis.

Line name	Genotype	A- or B-type crystallinity	RSC (%)
			Height			Area
			3b/CaF2	4b/CaF2	3b/CaF2	4b/CaF2	4˚ < 2θ < 27.5˚
#1520	ss2a be2b (-)	B	5.1±0.9b	-	17.5±3.9cd	-	2.14±0.1 a
#1522	ss2a be2b (+)	B	5.3±3.6b	-	16.9±8.1cd	-	2.11±0.1a
EM204	ss2a	A	9.0±0.1ab	8.0±0.5b	22.4±1.4bc	18.8±1.8b	1.48±0.1e
EM10	be2b (-)	B	3.4±0.9b	-	12.5±10.1d	-	1.87±0.1c
ssg3	be2b (+)	B	4.2±0.2b	-	11.3±0.9d	-	1.62±0.0de
Kinmaze	WT	A	11.5±0.3a	10.9±1.1ab	29.1±0.3ab	25.9±1.2ab	1.81±0.0c
Nipponbare	WT	A	13.1±3.7a	14.1±6.6a	31.2±2.8a	31.0±18.6a	1.93±0.0b

Values are the average of at least three replicates (mean ± SE). Samples annotated with different letters (a–e) are significantly different from one another as determined by the Tukey–Kramer method (p < 0.05). Mean values of height and area of 3b or 4b peaks were divided by the height and area of the peak obtained from calcium fluoride. 4˚ < 2θ < 27.5˚ means the total peak area between 4˚ and 27.5˚ of 2θ.

Among the three lines with A-type crystallinity, ss2a (EM204) showed a lower RSC than the WT lines Kinmaze and Nipponbare, and the difference was statistically significant when RSC was calculated using the second and third methods (Table 4). Among the four lines with B-type crystallinity, two ss2a be2b mutant lines, ss2a be2b (-) (#1520) and ss2a be2b (+) (#1522), showed a higher RSC than the be2b single mutants be2b (-) (EM10) and be2b (+) (ssg3), and the differences were statistically significant when RSC was calculated using the third method (Table 4).

Apparent amylose content and RSC are generally negatively correlated.³⁵⁾³⁷⁾ The reason why ss2a (EM204) had a lower RSC than the WT may be that ss2a (EM204) had a higher apparent amylose content. Another possible reason is that ss2a (EM204) has a greater number of short amylopectin chains than WT lines because of the loss of SSIIa (Figs. 5A and B; Miura et al. 2018¹⁴⁾). The ss2a be2b (-) (#1520) and ss2a be2b (+) (#1522) mutants had a higher RSC (Table 4), although their apparent amylose content was higher than that of the be2b single mutants be2b (-) (EM10) and be2b (+) (ssg3) (Table 2). The amylopectin structure of the be2b single mutants and ss2a be2b double mutants is predicted to have a higher frequency of long amylopectin branches than the WT lines (Fig. 5; Table 2). The double helix structure formed by amylose and amylopectin in the ss2a be2b lines may contribute to the high RSC because the ss2a be2b lines had a higher apparent amylose content. Whether a double helix between amylose and amylopectin actually exists needs to be determined in future studies.

CONCLUSION

The double mutant rice lines used in the present study lack two major enzymes, SSIIa and BEIIb, which are the constituents of the SSI-SSIIa-BEIIb hetero-trimeric protein complex involved in the formation of the amylopectin cluster. However, the seed weight of ss2a be2b mutant lines was similar or greater than that of the parental be2b single mutants (Table 1). The apparent amylose was higher in the ss2a be2b mutant lines than in the parental be2b single mutants (Table 2). It is speculated that the loss of two major enzymes, SSIIa and BEIIb, decreased amylopectin biosynthesis and increased amylose biosynthesis, resulting in increased seed weight in the ss2a be2b mutants. The ratio of short to long amylopectin branches was lower in the ss2a be2b mutants than in the WT, and the chain length distribution pattern of the ss2a be2b mutants was similar to that of the be2b mutants. These results indicate that the amylopectin structure of the ss2a be2b mutant is greatly influenced by the loss of BEIIb (Fig. 5). The be2b single mutants analyzed in the present study possess ss2a^L, which slightly extends DP 6–12 to DP 13–24. However, the number of short branches that can become primers for SSIIa is drastically decreased in the absence of BEIIb. This may account for the similar chain length distribution pattern between the ss2a be2b and be2b mutants.

Analysis of the thermal properties of starch showed that ss2a had a lower gelatinization temperature than the WT (Table 3). By contrast, the gelatinization temperature of the ss2a be2b mutants was higher than that of the WT but lower than that of the be2b mutant lines (Table 3). This may reflect small differences in chain length distribution patterns between the ss2a be2b and be2b mutant lines caused by loss of SSIIa.

The ss2a and WT lines showed a polygonal starch granule morphology, whereas the ss2a be2b and be2b mutants had larger, rounded starch granules (Fig. 4). X-ray diffraction analysis of starch showed that the ss2a and WT lines displayed A-type crystallinity, whereas the ss2a be2b and be2b mutants displayed B-type crystallinity (Table 4). There was a negative correlation between the apparent amylose content and RSC in rice lines with A-type crystallinity (ss2a and two WT lines). However, among the four lines with B-type crystallinity, RSC was higher in the ss2a be2b mutants than in the be2b mutants, despite the fact that the apparent amylose content was higher in the ss2a be2b mutant than in the be2b mutant (Tables 2 and 4). This suggests that long amylopectin branches resulting from the loss of BEIIb formed double helixes with amylose, thereby contributing to the higher RSC.

CONFLICTS OF INTEREST

The authors declare no conflict of interests.

ACKNOWLEDGMENTS

We would like to thank emeritus Prof. Hikaru Satoh in Kyushu University for kindly providing the rice seeds EM10. We would like to thank emeritus Prof. Keiji Kainuma, hornorary member of the Agricultural Society of Japan and Prof. Tamao Hatta from the Chiba Institute of Science for advice regarding the assessment of starch crystallinity by X-ray diffraction. We also thank Ms. Yuko Nakaizumi and Ms. Rika Takahashi for technical support. This work was partly supported by the President’s Funds of Akita Prefectural University (N.F. and N.C.), Grants-in-Aid for JSPS Fellows from the Japan Society for the Promotion of Science (#15J40176 and JP18J40020 to N.C.) and the Japan Society for the Promotion of Science (#16K18571, JP18K14438, and 20K05961 to N.C. and 19H01608 to N.F.).

REFERENCES

• 1)  S. Hizukuri: Polymodal distribution of the chain lengths of amylopectins, and its significance. Carbohydr. Res., 147, 324–347 (1986).
• 2)  Y. Nakamura: Towards a better understanding of the metabolic system for amylopectine biosynthesis in plants: rice endosperm as a model tissue. Plant Cell Physiol., 43, 718–725 (2002).
• 3)  N. Abe, Y. Nakamura, and N. Fujita: Thermal properties, morphology of starch granules and crystallinity of endosperm starch in SSI and BE isozymes double mutant lines. J. Appl. Glycosci., 60, 171–176 (2013).
• 4)  I.J. Tetlow, R. Wait, Z. Lu, R. Akkasaeng, C.G. Bowsher, S. Esposito, K.H. Behjat, M.K. Morell, and M.J. Emes: Protein phosphorylation in amyloplasts regulates starch-branching enzyme activity and protein-protein interactions. Plant Cell, 16, 694–708 (2004).
• 5)  I.J. Tetlow, K.G. Beisel, S. Cameron, A. Makhmoudova, F. Liu, N.S. Bresolin, R. Wait, M.K. Morell, and M.J. Emes: Analysis of protein complexes in wheat amyloplasts reveals functional interactions among starch biosynthetic enzymes. Plant Physiol., 146, 1878–1891 (2008).
• 6)  T.A. Hennen-Bierwagen, F. Liu, R.S. Marsh, S. Kim, Q. Gan, I.J. Tetlow, M.J. Emes, M.J. James, and A.M. Myers: Starch biosynthetic enzymes from developing maize endosperm associate in multi subunit complexes. Plant Physiol., 146, 1892–1908 (2008).
• 7)  T.A. Hennen-Bierwagen, Q. Lin, F. Grimaud, V. Planchot, P.L. Keeling, M.G. James, and A.M. Myers: Proteins from multiple metabolic pathways associate with starch biosynthetic enzymes in high molecular weight complexes: a model for regulation of carbon allocation in maize amyloplasts. Plant Physiol., 149, 1541–1559 (2009).
• 8)  F. Liu, A Makhmoudova, E.A. Lee, R. Wait, M.J. Emes, and I.J. Tetlow: The amylose extender mutant of maize conditions novel protein-protein interactions between starch biosynthetic enzymes in amyloplasts. J. Exp. Bot., 60, 4423–4440 (2009).
• 9)  F. Liu, Z. Ahmed, E.A. Lee, E. Donner, Q. Liu, R. Ahmed, M.K. Morell, M.J. Emes, and I.J. Tetlow: Allelic variants of the amylose extender mutation of maize demonstrate phenotypic variation in starch structure resulting from modified protein-protein interactions. J. Exp. Bot., 63, 1167–1183 (2012).
• 10)  N. Crofts, N. Abe, N.F. Oitome, R. Matsushima, I.J. Tetlow, M.J. Emes, Y. Nakamura, and N. Fujita: Amylopectin biosynthetic enzymes from developing rice seed form enzymatically active protein complexes. J. Exp. Bot., 66, 4469–4482 (2015).
• 11)  F. Liu, N. Romanova, E.A. Lee, R. Ahmed, M. Evans, E.P. Gilbert, M.K. Morell, M.J. Emes, and I.J. Tetlow: Glucan affinity of starch synthase IIa determines binding of starch synthase I and starch-branching enzyme IIb to starch granules. Biochem. J., 448, 373–387 (2012).
• 12)  N. Crofts, Y. Nakamura, and N. Fujita: Critical and speculative review of the roles of multi-protein complexes in starch biosynthesis in cereals. Plant Sci., 262, 1–8 (2017).
• 13)  Y. Nakamura, B.P. Francisco Jr., Y. Hosaka, A. Sato, T. Sawada, A. Kubo, and N. Fujita: 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).
• 14)  S. Miura, N. Crofts, Y. Saito, Y. Hosaka, N.F. Oitome, T. Watanabe, T. Kumamaru, and N. Fujita: Starch synthase IIa-deficient mutant rice line produces endosperm starch with lower gelatinization temperature than japonica rice cultivars. Front. Plant Sci., 9, 645 (2018).
• 15)  N. Crofts, Y. Iizuka, N. Abe, S. Miura, K. Kikuchi, R. Matsushima, and N. Fujita: Rice mutants lacking starch synthase I or branching enzyme IIb activity altered starch biosynthetic protein complexes. Front. Plant Sci., 9, 1817 (2018).
• 16)  A. Nishi, Y. Nakamura, N. Tanaka, and H. Satoh: Biochemical and genetic analysis of the effects of amylose-extender mutation in rice endosperm. Plant Physiol., 127, 459–472 (2001).
• 17)  R. Matsushima, M. Maekawa, N. Fujita, and W. Sakamoto: A rapid, direct observation method to isolate mutants with defects in starch grain morphology in rice. Plant Cell Physiol., 51, 728–741 (2010).
• 18)  N. Fujita, Y. Toyosawa, Y. Utsumi, T. Higuchi, I. Hanashiro, A. Ikegami, S. Akuzawa, M. Yoshida, A. Mori, K. Inomata, R. Itoh, A. Miyao, H. Hirochika, H Satoh, and Y. Nakamura: Characterization of pullulanase (PUL)-deficient mutants of rice (Oryza sativa L.) and the function of PUL on starch biosynthesis in the developing rice endosperm. J. Exp. Bot., 60, 1009–1023 (2009).
• 19)  N. Fujita, I. Hanashiro, S. Suzuki, T. Higuchi, Y. Toyosawa, Y. Utsumi, R. Itoh, S. Aihara, and Y. Nakamura: Elongated phytoglycogen chain-length in transgenic rice endosperm expressing active starch synthase IIa affects the altered solubility and crystallinity of the storage α-glucan. J. Exp. Bot., 63, 5859–5872 (2011).
• 20)  Y. Toyosawa, Y. Kawagoe, R. Matsushima, N. Crofts, M. Ogawa, M. Fukuda, T. Kumamaru, Y. Okazaki, M. Kusano, K. Saito, K. Toyooka, M. Sato, Y. Ai, J.L. Jane, Y. Nakamura, and N. Fujita: Deficiency of starch synthase IIIa and IVb alters starch granule morphology from polyhedral to spherical in rice endosperm. Plant Physiol., 170, 1255–1270 (2016).
• 21)  M.G. O’Shea and M.K. Morell: High resolution slab gel electrophoresis of 8-amino-1,3, 6-pyrenetrisulfonic acid (APTS) tagged oligosaccharides using a DNA sequencer. Electrophoresis, 17, 681–686 (1996).
• 22)  N. Fujita, H. Hasegawa, and T. Taira: The isolation and charavterization of waxy mutant of diploid wheat (Triticum monococcum L.). Plant Sci., 160, 595–602 (2001).
• 23)  T. Hamanishi, T. Hatta, F.S. Jong, K. Kainuma, and S. Takahashi: The relative crystallinity, structure and gelatinization properties of sago starches at different growth stages. J. Appl. Glycosci., 47, 335–341 (2000).
• 24)  I. Kodama, C. Shibata, N. Fujita, K. Ishikawa, T. Takahashi, Y. Nakamura, T. Kawamoto, K. Kato, K. Sato, M. Matsunami, and Y. Akiyama: Starch properties of waxy rice cultivars influencing rice cake hardening. Jpn. J. Food Eng., 12, 157–162 (2011).
• 25)  K. Kato, Y. Suzuki, Y. Hosaka, R. Takahashi, I. Kodama, K. Sato, T. Kawamoto, T. Kumamaru, and N. Fujita: Effect of high temperature on starch biosynthetic enzymes and starch structure in japonica rice cultivar ‘Akitakomachi’ (Oryza sativa L.) endosperm and palatability of cooked rice. J. Cereal Sci., 87, 209–214 (2019).
• 26)  H. Asai, N. Abe, R. Matsushima, N. Crofts, N.F. Oitome, Y. Nakamura, and N. Fujita: Deficiencies in both starch synthase IIIa and branching enzyme IIb lead to a significant increase in amylose in SSIIa-inactive japonica rice seeds. J. Exp. Bot., 65, 5497–5507 (2014).
• 27)  Y. Itoh, N. Crofts, M. Abe, Y. Hosaka, and N. Fujita: 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).
• 28)  N. Abe, H. Asai, H. Yago, N.F. Oitome, R. Itoh, N. Crofts, Y. Nakamura, and N. Fujita: Relationships between starch synthase I and branching enzyme isozymes determined using double mutant rice lines. BMC Plant Biol., 14, 80 (2014).
• 29)  S. Miura, N. Koyama, N. Crofts, Y. Hosaka, M. Abe, and N. Fujita: Generation and starch characterization of non-transgenic BEI and BEIIb double mutant rice (Oryza sativa) with ultra-high level of resistant starch. Rice, 14, 3 (2021).
• 30)  M. Hayashi, M. Kodama, Y. Nakamura, and N. Fujita: Thermal and pasting properties, morphology of starch granules, and crystallinity of endosperm starch in the rice SSI and SSIIIa double-mutant. J. Appl. Glycosci., 62, 81–86 (2015).
• 31)  E. Bertoft: Investigation of the fine structure of alpha-dextrins derived from amylopectin and their relation to the structure of waxy-maize starch. Carbohydr. Res., 212, 229–244 (1991).
• 32)  N. Fujita: Manipulation of Rice Starch Properties for Application. in Starch Metabolism and Structure, Y. Nakamura, ed., Springer, Tokyo, pp. 335–369 (2015).
• 33)  N. Tanaka, N. Fujita, A. Nishi, H. Satoh, Y. Hosaka, M. Ugaki, S. Kawasaki, and Y. Nakamura: The structure of starch can be manipulated by changing the expression levels of starch branching enzyme IIb in rice endosperm. Plant Biotechnol. J., 2, 507–516 (2004).
• 34)  J.I. Jane, K.S. Wong, and E.M. Andrew: Branch-structure difference in starches of A-and B- type X-ray patterns revealed by their Naegeli dextrins. Carbohydr. Res., 300, 219–227 (1999).
• 35)  T. Takahashi and N. Fujita: Thermal and rheological characteristics of mutant rice starches with widespread variation of amylose content and amylopectin structure. Food Hydrocoll., 62, 83–93 (2017).
• 36)  S. Hizukuri and Z. Nikuni: X-ray diffractometric studies on Starches. II. Structure of “C”-type crystallite. Nippon Nōgeikagaku Kaishi, 7, 525–527 (1957).
• 37)  S.H. Yoo and J.L. Jane: Structural and physical characteristics of waxy and other wheat starches. Carbohydr. Polym., 49, 297–305 (2002).