Variety-dependent accumulation of glucomannan in the starchy endosperm and aleurone cell walls of rice grains and its possible genetic basis

Abstract

Plant cell wall plays important roles in the regulation of plant growth/development and affects the quality of plant-derived food and industrial materials. On the other hand, genetic variability of cell wall structure within a plant species has not been well understood. Here we show that the endosperm cell walls, including both starchy endosperm and aleurone layer, of rice grains with various genetic backgrounds are clearly classified into two groups depending on the presence/absence of β-1,4-linked glucomannan. All-or-none distribution of the glucomannan accumulation among rice varieties is very different from the varietal differences of arabinoxylan content in wheat and barley, which showed continuous distributions. Immunoelectron microscopic observation suggested that the glucomannan was synthesized in the early stage of endosperm development, but the synthesis was down-regulated during the secondary thickening process associated with the differentiation of aleurone layer. Significant amount of glucomannan in the cell walls of the glucomannan-positive varieties, i.e., 10% or more of the starchy endosperm cell walls, and its close association with the cellulose microfibril suggested possible effects on the physicochemical/biochemical properties of these cell walls. Comparative genomic analysis indicated the presence of striking differences between OsCslA12 genes of glucomannan-positive and negative rice varieties, Kitaake and Nipponbare, which seems to explain the all-or-none glucomannan cell wall trait in the rice varieties. Identification of the gene responsible for the glucomannan accumulation could lead the way to clarify the effect of the accumulation of glucomannan on the agronomic traits of rice by using genetic approaches.

Keywords: all-or-none trait; cell wall; cellulose synthase-like family; glucomannan; rice endosperm.

Figure 1. Starchy endosperm cell walls from rice varieties are classified into two groups depending on the presence/absence of mannose as a component sugar. A. Average mannose contents of starchy endosperm cell walls from mannose-less and mannose-containing varieties. Endosperm cell wall preparations from arbitrary collected 58 rice varieties (not including those varieties described in Figure 1C, except Kitahikari and Dohoku 43) were analyzed for their neutral sugar composition after acid hydrolysis. Average mannose contents were 0.22±0.28 and 11.5±2.1% for mannose-less and mannose-containing varieties, respectively. Error bars indicate standard deviation. The asterisk indicates significant difference from the mannose-less varieties by Student’s t-test (p<0.01). For the detailed distribution of mannose-containing and mannose-less varieties, see Supplementary Figure S1. B. Typical example of neutral sugar composition of starchy endosperm cell walls from mannose-containing and mannose-less rice varieties. Dotted square shows the difference in the mannose content between the two groups. C. Inheritance of mannose-containing cell wall trait among genetically related rice varieties. Solid squares indicate the mannose-containing varieties. Those varieties shown in parentheses were not used for cell wall analysis. Mannose contents were 8.0–12.4% and 0–0.4% for mannose-containing and mannose-less varieties, respectively. Symbol “*” indicates the varieties, “Kitahikari” and “Dohoku-43”, described in the text.

Figure 2. Purification of rice glucomannan. Glucomannan was purified from the starchy endosperm cell wall of var. Kitaake, a mannose-containing rice variety by stepwise extraction and endoxylanase treatment. For details, see “Materials and methods”.

Figure 3. Analysis of the partial acid hydrolysate of rice glucomannan. A. HPAEC profile of the partial acid hydrolysate of the purified glucomannan. Mixture of the partial acid hydrolysate was analyzed by HPAEC. Symbols M₂–M₉ indicate the peaks corresponding to the standard β-1,4-linked mannooligosaccharides. B. HPAEC profile of the disaccharide fraction from the partial acid hydrolysate of the glucomannan. Partial acid hydrolysate of the rice glucomannan was first fractionated on a Bio Gel P-2/P-4 column to obtain disaccharide fraction (Supplementary Figure S2). The disaccharide fraction was further analyzed by HPAEC. Symbols M-M and G-M indicate the peaks corresponding to the standard disaccharides, Man-(β1,4)Man and Glc-(β-1,4)-Man, respectively.

Figure 4. Immunoelectron microscopic analysis of glucomannan in the starchy endosperm and aleurone cell walls of maturing rice grains. A. Starchy endosperm of a glucomannan-containing variety, Fujisaka 5. Left, without mannooligosaccharides; Right, with mannooligosaccharides. B. Aleurone layer of Fujisaka 5. Left, without mannooligosaccharides; Right, with mannooligosaccharides. C. Starchy endosperm (left) and aleurone layer of a glucomannan-less variety, Reishikou. Both tissues were treated with the Man₄-antibody in the absence of mannooligosaccharides. PB, protein body; ER, endoplasmic reticulum; M, mitochondria; W, cell wall; ST, starch granule; V, vacuole. Arrowheads indicate immune labeling with gold particles.

Figure 5. Mannose contents of the cell wall preparations obtained from the starchy endosperm and outer layer of glucomannan-containing rice grains. Cell walls were prepared from the starchy endosperm and outer layer, which is mainly composed of aleurone cells and lesser amount of seed coat and pericarp (Shibuya et al. 1985), of 4 glucomannan-containing varieties, Eiko, Kitaake, Matsumae and Michikogane. Bars indicate standard deviation. The asterisk indicates significant difference from the mannose contents of outer layer cell walls by Student’s t-test (p<0.01).

Figure 6. Mannose contents of the cell wall preparations obtained from the starchy endosperm and callus tissues from glucomannan-containing and glucomannan-less rice varieties. Cell walls were prepared from the starchy endosperm and callus cells of following rice varieties. Glucomannan(GM)-containing varieties; Fujisaka 5, Ishikari, Kitaake, Michikogane. Glucomannan(GM)-less varieties; Nipponbare, Reishikou, Yukihikari. Bars indicate standard deviation. The asterisks indicate significant differences from the mannose contents of glucomannan-containing endosperm cell walls by Student’s t-test (p<0.01).

Figure 7. Comparison of OsCslA12 proteins between Nipponbare and Kitaake. A. Amino acid sequences of OsCslA12 of Nipponbare (Os09t0572500-01) and Kitaake (OsKitaake09g208700.1). Note that the amino acid sequences of these proteins are completely the same until 331 aa but the sequences after that are totally different, because of the frame shift caused by the deletion of 10 bases in the Nipponbare gene. Shadowed sequences indicate putative transmembrane domains (TMs) predicted by phobius (https://phobius.sbc.su.se/ (Accessed May 22, 2023)). Box indicates common sequences between Nippobare and Kitaake. Asterisks indicate the conserved residues in both sequences. B. Prediction of transmembrane domains in OsCslA12 proteins of Nippobare (Glucomannan (−)), Kitaake (Glucomannan (+)) and TaCslA12 of wheat, Triticum aestivum. Gray box indicates putative transmembrane domains (TMs). The amino acid (aa) and cDNA sequences near the region where the 10 bp deletion caused the frame shift in the Nipponbare OsCslA12 sequence are also shown.