Concerted suppression of all starch branching enzyme genes in barley produces amylose-only starch granules

Abstract

Background: Starch is stored in higher plants as granules composed of semi-crystalline amylopectin and amorphous amylose. Starch granules provide energy for the plant during dark periods and for germination of seeds and tubers. Dietary starch is also a highly glycemic carbohydrate being degraded to glucose and rapidly absorbed in the small intestine. But a portion of dietary starch, termed "resistant starch" (RS) escapes digestion and reaches the large intestine, where it is fermented by colonic bacteria producing short chain fatty acids (SCFA) which are linked to several health benefits. The RS is preferentially derived from amylose, which can be increased by suppressing amylopectin synthesis by silencing of starch branching enzymes (SBEs). However all the previous works attempting the production of high RS crops resulted in only partly increased amylose-content and/or significant yield loss.

Results: In this study we invented a new method for silencing of multiple genes. Using a chimeric RNAi hairpin we simultaneously suppressed all genes coding for starch branching enzymes (SBE I, SBE IIa, SBE IIb) in barley (Hordeum vulgare L.), resulting in production of amylose-only starch granules in the endosperm. This trait was segregating 3:1. Amylose-only starch granules were irregularly shaped and showed peculiar thermal properties and crystallinity. Transgenic lines retained high-yield possibly due to a pleiotropic upregualtion of other starch biosynthetic genes compensating the SBEs loss. For gelatinized starch, a very high content of RS (65 %) was observed, which is 2.2-fold higher than control (29%). The amylose-only grains germinated with same frequency as control grains. However, initial growth was delayed in young plants.

Conclusions: This is the first time that pure amylose has been generated with high yield in a living organism. This was achieved by a new method of simultaneous suppression of the entire complement of genes encoding starch branching enzymes. We demonstrate that amylopectin is not essential for starch granule crystallinity and integrity. However the slower initial growth of shoots from amylose-only grains may be due to an important physiological role played by amylopectin ordered crystallinity for rapid starch remobilization explaining the broad conservation in the plant kingdom of the amylopectin structure.

Figure 1

Generation and identification of amylose-only barley. (a) Chimeric RNAi hairpin construct simultaneously targeting the three different SBE genes SBEI, SBEIIa and SBEIIb. Expression was driven by the maize Ubiquitin promoter. Promoter and intron are not drawn to scale. The actual length of the intron is 1290 bp. The amplification product of the primer set Hairpin Fw and Hairpin Rev, which specifically recognizes the hairpin construct is indicated (b) Relative gene expression levels of the three SBEs isoforms (SBEI, SBEIIa and SBEIIb) assessed by RT qPCR in three individual grains, A, B and C at (20DAP) each of control and transgenic T₁ lines (three technical replicates each). SE bars are indicated. (c) SBE enzyme activity in developing endosperm of SBE RNAi 4.1 and control grains, based on the average value of 3 experiments. Bars indicate standard error (d) Size exclusion chromatography (SEC) profile of starch from control and SBE RNAi4.1 lines. Black lines show the elution profile determined measuring the total sugar content of each fraction. λ-max absorbance of the α-glucan-iodine complex in each fraction is indicated by grey dots.

Figure 2

Solubility. (a) Thermal properties and swelling power of starch from SBE RNAi4.1 and control lines. Upper lines (red, SBE RNAi4.1; blue, control) show the endothermic heat flow and lower dots and lines show the water gain of starch. Vertical line indicates melting of amylopectin. The swelling power of starch after gelatinization at 100°C is reported as the ratio of water gain of the swollen starch pellet compared to starch dry matter. (b) Solubility (1% starch granule suspensions in water) of control (filled squares) and amylose-only (open squares) starch as a function of temperature. The solubility is reported as the ratio of total carbohydrate in the supernatant to total starch.

Figure 3

SBE silencing affected grain shape, starch granule morphology, plants yield and expressions of other starch biosynthetic genes. (a) Morphology, median cross sections and thin sections (50 μm) of representative SBE RNAi and control grains. (b) Scanning electron microscopy (SEM) pictures of control (left) and SBE RNAi4.1 (right) starch granules. Scale bars represent 50 μm and 10 μm in the lower and higher magnifications respectively. (c) Semi-field trial yield calculations, from left to right: average number of spikes per plants, average number of grains per spike, average grain mass and average yield per plant for SBE RNAi4.1 (dark grey) and control lines (light grey). Average yield per plant is expressed as total grain mass in milligram and SD bars are indicated. (d) Quantitative gene expressions level measured by RT qPCR of starch biosynthetic enzymes: starch synthases (SSI, SSIIa, SSIIIa, SSIV, GBSSIa, GBSSIb) and glucan water dikinase 1 (GWDI). Data are expressed in gene expression fold change between control and SBE RNAi4.1 lines (3 biological and 9 technical replicates per gene). SE bars are indicated. Genes with significant up-regulation in the SBE RNAi 4.1 line as compared to the control line are marked with * (P < 0.05) or ** (P < 0.01).

Figure 4

Powder X-ray diffraction analysis (XRD). X-ray powder diffractograms of control barley starch granules (top) and the amylose-only starch granules from the SBE RNAi4.1 line (bottom). The diffraction peaks at 2θ typical for A-type, B-type and V_h-type crystalline polymorphs are indicated.

Figure 5

Plant height. Height average of plants of control (blue) and SBE RNAi4.1 (red) T₂ lines. Plant heights (22 individuals for SBE RNAi 4.1 and 15 individual for control) were measured at 20, 40, 60, 80, 100, 120 days after sowing. SE is indicated.

Figure 6

Concerted in vitro enzymatic degradation of starch by pancreatic alpha amylase and glucoamylase. Filled dots and empty dots indicate control starch granules and amylose granules respectively. SD bars are indicated. (a) Native starch granules. (b) Gelatinized starch. (c) Gelatinized and retrograded (re-crystallised) starch.