Biochemical and genetic analysis of the effects of amylose-extender mutation in rice endosperm

Abstract

Biochemical analysis of amylose-extender (ae) mutant of rice (Oryza sativa) revealed that the mutation in the gene for starch-branching enzyme IIb (BEIIb) specifically altered the structure of amylopectin in the endosperm by reducing short chains with degree of polymerization of 17 or less, with the greatest decrease in chains with degree of polymerization of 8 to 12. The extent of such change was correlated with the gelatinization properties of the starch granules, as determined in terms of solubility in urea solution. The ae mutation caused a dramatic reduction in the activity of BEIIb. The activity of soluble starch synthase I (SSI) in the ae mutant was significantly lower than in the wild type, suggesting that the mutation had a pleiotropic effect on the SSI activity. In contrast, the activities of BEI, BEIIa, ADP-Glc pyrophosphorylase, isoamylase, isoamylase, pullulanase, and Suc synthase were not affected by the mutation. Therefore, it is stressed that the function of BEIIb cannot be complemented by BEIIa and BEI. These results strongly suggest that BEIIb plays a specific role in the transfer of short chains, which might then be extended by SS to form the A and B(1) chains of amylopectin cluster in rice endosperm.

Figure 1

Effects of the concentration of urea on gelatinization of starch granules from the ae mutant EM10 and the wild-type cv Kinmaze and from lines with the ae and Ae genes on the waxy background. After incubation of starch granules in solutions of urea for 24 h, swelling was examined by measuring the volume of the swollen starch sediment. □, Kinmaze (wild type, AeAe/WxWx); ●, EM21 (wx mutant, AeAe/wxwx); ▵, EM10 (ae mutant, aeae/WxWx); ♦, AMF44 (double-recessive mutant, aeae/wxwx). Results are means ± sd of results from three replicate experiments.

Figure 2

Distribution of chain length of total α-polysaccharides from wild-type, ae, wx, and aeae/wxwx rice endosperm as determined by HPAEC-PAD. Amylopectin was debranched with isoamylase from P. amyloderamosa, reduced with sodium borohydride, and then fractionated on a Carbopac PA1 column. The α-1,4-glucan chains were eluted with a gradient of sodium hydroxide and sodium acetate and monitored with a PAD. A, The distribution of α-1,4-glucan chains in amylopectin from the wild type (AeAe/WxWx) and the ae mutant (aeae/WxWx). B, The difference in amylopectin chain lengths between the ae mutant and the wild type. The columns show the area of each peak for the ae mutant minus the area of the corresponding peak with the same DP for the wild type. C, The distribution of α-1,4-glucan chains in amylopectin from the wx mutant (AeAe/wxwx) and the double-recessive mutant (aeae/wxwx). D, The difference in amylopectin chain lengths between the double mutant and the wx mutant. The results show representatives from three experiments that gave similar results.

Figure 3

Gene dosage effects of the Ae allele on the level and activity of BEIIb, chain length distribution, gelatinization properties, and grain morphology. A, Western-blotting analysis of BEIIb in mature rice kernels. Protein was extracted from 20 mg of rice powder. The immunoblot was developed with antiserum raised against BEIIb from rice endosperm (Nakamura et al., 1992) at a dilution of 1:500. B, Native PAGE/activity stainings of BEs (left) and endogenous phosphorylase (right) in the endosperm of four genotypes. The migration and identification of each band corresponding to three BE isoforms (BEI, BEIIa, and BEIIb) and phosphorylase were according to our previous report (Yamanouchi and Nakamura, 1992). The volumes of crude enzyme extract applied were 0.67 and 6.7 μL for BE and phosphorylase, respectively. Note that the phosphorylase band was not detected under the lower protein concentration (0.67 μL of crude extract). C, Differences in the distribution of α-1,4-glucan chains among the four genotypes. The columns show the peak areas for each glucan chain from each genotype minus that from the wild-type cv Kinmaze. The sd was given from three separate experiments. D, Effects of 4 m urea on the swelling of starch granules from the endosperm of four genotypes. Ten milligrams of rice powder in an Eppendorf tube was mixed with 0.5 mL of 4 m urea, and shaken for 24 h at 25°C. After centrifugation, samples were allowed to stand for 1 h. E, Kernels from the four genotypes. a through d, Results for the four genotypes, AeAeAe, AeAeae, Aeaeae, and aeaeae, respectively.

Figure 4

Effects of the ae mutation on other starch-synthesizing enzymes. A, Native PAGE/activity staining of SS in the endosperm of ae mutant EM10 and the wild-type cv Kinmaze. Each lane contained 7.5 μL multiplied by the number given above the lane of the crude enzyme extract. The migration of each band corresponding to two SS isoforms (SSI and SSIII) was according to Abel et al. (1996). B, SDS-PAGE of total protein in rice mature kernel. C, Western-blotting analysis of BEI in mature rice kernels. The immunoblot was developed with antiserum raised against BEI from rice endosperm (Nakamura et al., 1992) at a dilution of 1:1,000.

Figure 5

Northern-blot analyses of BEI, BEIIa, BEIIb, SSI, and SSIII transcripts in ae mutant EM10 and the wild-type cv Kinmaze. Total RNA from developing grains, stages I, II, III, and IV, were blotted and probed with specific RNA probes. The lower portion of the figure shows the ethidium bromide-stained RNA gel.