Mixed-linkage Β-glucan Research Articles

Differences in enzymic properties of five recombinant xyloglucan endotransglucosylase/hydrolase (XTH) proteins of Arabidopsis thaliana

Xyloglucan endotransglucosylase/hydrolases (XTHs) are cell wall enzymes that are able to graft xyloglucan chains to oligosaccharides or to other available xyloglucan chains and/or to hydrolyse xyloglucan chains. As they are involved in the modification of the load-bearing cell-wall components, they are believed to be very important in the regulation of growth and development. Given the large number (33) of XTH genes in Arabidopsis and the overlapping expression patterns, specific enzymic properties may be expected. Five predominantly root-expressed Arabidopsis thaliana XTHs belonging to subgroup I/II were analysed here. These represent two sets of closely related genes: AtXTH12 and 13 on the one hand (trichoblast-enriched) and AtXTH17, 18, and 19 on the other (expressed in nearly all cell types in the root). They were all recombinantly produced in the yeast Pichia pastoris and partially purified by ammonium sulphate precipitation before they were subsequently all subjected to a series of identical in vitro tests. The kinetic properties of purified AtXTH13 were investigated in greater detail to rule out interference with the assays by contaminating yeast proteins. All five proteins were found to exhibit only the endotransglucosylase (XET; EC 2.4.1.207) activity towards xyloglucan and non-detectable endohydrolytic (XEH; EC 3.2.1.151) activity. Their endotransglucosylase activity was preferentially directed towards xyloglucan and, in some cases, water-soluble cellulose acetate, rather than to mixed-linkage β-glucan. Isoforms differed in optimum pH (5.0-7.5), in temperature dependence and in acceptor substrate preferences.

Synergism between cucumber α-expansin, fungal endoglucanase and pectin lyase

Several recombinant fungal enzymes (endoglucanase and pectinase) were studied for their interactions with α-expansin in cell wall extension and polysaccharide degradation. Both Cel12A and Cel5A were able to hydrolyze cellulose CMC-Na and mixed-linkage β-glucan. In contrast to Cel5A, Cel12A could also hydrolyze xyloglucan and induce wall extension of cucumber hypocotyls in an in vitro assay. Combining α-expansin, even at high concentrations, with Cel12A did not enhance the maximum/final wall extension rate induced by Cel12A alone. These results strongly suggest that modification/degradation of the xyloglucan molecule/network is the key for cell wall extension, and α-expansin and Cel12A may share the same acting site in the substrate. Pectinase (Pel1, a pectin lyase) enhanced α-expansin-induced wall extension in a concentration-dependent manner, suggesting that the pectin network may normally regulate accessibility of expansin to the xyloglucan–cellulose complex. α-Expansin enhanced Cel12A's hydrolytic activity on cellulose CMC-Na but not on xyloglucan and β-glucan. Expansin did not affect Cel5A's hydrolytic activity. Interestingly, expansin also enhanced Pel1's activity on degrading high esterified pectin. A potential explanation for why expansin could synergistically interact with only certain enzymes on specific polysaccharides is discussed. Additional results also suggested that cell wall swelling may not be a significant event during the action of expansin and hydrolases.

Barley and oat cultivars with diverse carbohydrate composition alter ileal and total tract nutrient digestibility and fermentation metabolites in weaned piglets

An experiment was conducted to evaluate the effects of cereal carbohydrate form (isolated v. cereal matrix) and level, especially mixed-linked β-glucan (hereafter referred to as β-glucan) and starch amylase/amylopectin ratio on nutrient digestibility and fermentation parameters in the intestines of weaned pigs. Four hulless barley cultivars containing varying β-glucan levels (41 to 84 g/kg) were compared with hulled barley, supplemented or not with a β-glucan concentrate (BBG; 270 g/kg β-glucan) and two oat cultivars for digestibility and fermentation metabolites. Seventy-two weaned piglets (BW = 12.8 ± 1.9 kg) were assigned to one of nine diets composed of 815 g/kg cereal, 60 g/kg whey, 90 g/kg soy protein isolate and 35 g/kg minerals. After 15 days, the pigs were killed, and digesta collected from ileum and colon were analyzed for proximate nutrients, short-chain fatty acids (SCFAs), lactic acid (LA) and ammonia. Ileal and total tract digestibility of proximate nutrients and non-starch polysaccharides (NSPs) were determined using HCl-insoluble ash as a marker. Organic matter (OM) ileal digestibility was greater (P < 0.05) for diets based on hulless barley (77% ± 1.1% on average), as compared with hulled barley (64% ± 1.4%) and oat (58% ± 1.5%). Similar trends were found for total tract OM digestibility, varying from 90% ± 0.3% for hulless barley to 67% ± 0.4% for oat, on average. NSP digestibility differed (P < 0.05) within and between cereal types, ranging from 20% (hulled barley plus 163 g/kg BBG or ∼40 g/kg β-glucan) to 51% (SB94893 hulless barley cultivar with high β-glucan and high amylose ratio) at the ileum and from 44% (hulled barley) to 84% (SB94893 cultivar) at the total tract level. No dietary effect (P > 0.05) was found for SCFA concentration in ileal contents, whereas in colonic contents, SCFA was lower in pigs fed oat (P < 0.001). LA concentration was greater (P < 0.001) in the colon of pigs fed hulless barley than in pigs fed hulled barley and oat. Expressed per kg carbohydrate (NSP + starch) fermented, the ammonia concentration at the colon was lowest for hulled barley diets (supplemented with β-glucan) and the highest for oat diets. In conclusion, the interaction of both form and level of β-glucan impacted nutrient digestibility and fermentation. Hulless barleys with high soluble NSP such as β-glucan and resistant starch yielded, in general higher SCFA and LA and lower ammonia. Hulless barleys may, therefore, have potential for use in feeding strategies designed to improve gut health in pigs.

A combined nuclear magnetic resonance and molecular dynamics study of the two structural motifs for mixed-linkage β-glucans: methyl β-cellobioside and methyl β-laminarabioside

The conformational and hydration properties of the two disaccharides methyl β-cellobioside and methyl β-laminarabioside were investigated by NMR spectroscopy and explicit solvation molecular dynamics simulations using the carbohydrate solution force field (CSFF). Adiabatic maps produced with this force field displayed 4 minima A: ( Φ = 300°, Ψ = 280°), B: ( Φ = 280°, Ψ = 210°), C: ( Φ = 260°, Ψ = 60°), and D: ( Φ = 60°, Ψ = 260°) for methyl β-cellobioside and 3 minima A: ( Φ = 290°, Ψ = 130°), B: ( Φ = 270°, Ψ = 290°), and C: ( Φ = 60°, Ψ = 120°) for methyl β-laminarabioside. Molecular dynamics simulations were initiated from all minima. For each disaccharide, the simulation started from the A minimum was conducted for 50 ns, while the other minima were explored for 10 ns. The simulations revealed two stable minima for both compounds. For methyl β-cellobioside, the simulation minima in aqueous solution were shifted from their adiabatic map counterparts, while the simulation minima for methyl β-laminarabioside coincided with the corresponding adiabatic map minima. To validate the simulation results, NMR-derived NOEs and coupling constants across the glycoside linkage, 3 J HC and 3 J CH, were compared with values calculated from the MD trajectories. For each disaccharide, the best agreement was obtained for the simulations started at the A minimum. For both compounds, inter-ring water bridges in combination with the direct hydrogen bonds between the same groups were found to be determining factors for the overall solution structure of the disaccharides which differed from solid-state structures. Comparison with helical parameters showed that the preferred glycosidic dihedral configurations in the methyl β-cellobioside simulation were not highly compatible with the structure of cellulose, but that curdlan helix structures agreed relatively well with the methyl β-laminarabioside simulation. Polymers generated using glycosidic dihedral angles from the simulations revealed secondary structure motifs that that may help to elucidate polymer associations and small-molecule binding.

Impact of mixed-linkage (1→3, 1→4) β-glucans on physical properties of acid-set skim milk gels

The effects of concentration and molecular structure of barley β-glucans on physical properties of skim milk gels acidified at 42 °C by glucono-δ-lactone (GDL) and reaching a final pH of 4.8 were investigated. Incorporation of β-glucans (0.5–2.0%, w/w) into skim milk aqueous dispersions (10%, 12% and 14%, w/w) resulted in weakening of the protein network upon acidification as shown by reduction of the storage modulus ( G′) values of the composite gels, compared with control (free of β-glucan) formulations. However, dynamic rheometry and large deformation measurements (texture profile analysis) revealed that this weakening effect could be reduced or eliminated by using β-glucans with certain molecular features that either promote development of a high viscosity medium (high molecular weight) or formation of a secondary gel network structure (low molecular weight and/or high trimer/tetramer ratio) by the polysaccharide itself. Despite the fact that inclusion of β-glucans at low concentrations increased the levels of spontaneous syneresis, using high β-glucan concentrations or polysaccharide preparations with the aforementioned structural characteristics resulted in milk gels with significantly lower serum separation compared to control formulations.

Cereal β-glucans in diet and health

The native mixed linkage β-glucan of cereals is classified as a soluble dietary fibre, with rheological properties generally similar to guar gum and other random coil polysaccharides. The ability of oat and barley products to attenuate postprandial glycemic and insulinemic response is related to content of (1→3)(1→4)- β- d-glucan ( β-glucan) and viscosity. A role of viscosity of β-glucan has not been directly demonstrated for lowering of serum cholesterol levels, and not all studies report a statistically significant lowering. The wide range of effectiveness reported may partially be explained by the properties of the β-glucan in the diets used, as well as the dose.

Interaction between barley cultivars and growth temperature on starch degradation properties measured in vitro

Eight 2-row barley cultivars (waxy, normal and high amylose) were grown in controlled environment chambers at 9, 12, 15, 18 and 21 °C until maturity. The flour samples were analysed for total starch content, mixed-linked β-glucan content (total, soluble and insoluble), content of protein, fat and neutral detergent fibre (NDF) in addition to the amount of total, free and lipid complexed amylose. A non-ruminant in vitro method was used to analyse the degradation of starch (5–10–15–30–60–90–120–240 min incubation times) from the samples grown at different temperatures. A partial least square (PLS) regression model was made to highlight the flour characteristics that influenced the starch degradation the most. The waxy starches were degraded faster and to a higher extent than the normal and high amylose starches at all incubation times. At short incubation times the high amylose was not degraded as rapidly as the normal ones, but as incubation time increased, they changed range, and at 240 min incubation time the normal genotypes were degraded to a lesser extent than the high amylose. The amylose–amylopectin ratio was most important regarding starch degradation. The growth temperature influenced the starch degradation considerably; the degradation was negatively correlated to the growth temperature. The PLS regression model showed that the content of total and insoluble β-glucans became increasingly important as the incubation time increased.

Isolation of cellotriosyl blocks from barley β-glucan with endo-1,4-β-glucanase from Trichoderma reesei

Mixed-linkage β-glucan from barley, with a cellotriosyl to cellotetraosyl ratio of 2.9, was hydrolysed with two endo-1,4-β-glucanases (cellulases) and one non-cellulolytic β-glucanase isolated from Trichoderma reesei. The hydrolysates were precipitated in 90% ethanol and the fragments obtained were further treated with lichenase, followed by analysis of the oligosaccharides released. One of the endo-1,4-β glucanases, Cel 5A (EG II), selectively degraded cellotetraosyl units in the polymer and left the cellotriosyl units unhydrolysed. Blocks of cellotriosyl units were isolated on a larger scale using this enzyme. The isolated blocks were fractionated into four fractions using gel permeation chromatography on Biogel P-6 and the structures of the blocks were analysed by 1H NMR spectroscopy. The fractions essentially contained cellotriosyl units of different sizes with a 3-linked glucose residue at the non-reducing end and a reducing end linked to two 4-linked glucose residues. The results thus indicated that the enzyme could hydrolyse a 4-linked glucose residue next to the 3-linked residue at the non-reducing terminal but left two 4-linked glucose residues at the reducing end. The isolated blocks of cellotriosyl units had a molecular weight distribution that fits a theoretical model based on random blocks of triosyl units in the mixed-linkage β-glucan.

A comparative study on structure–function relations of mixed-linkage (1→3), (1→4) linear β- d-glucans

The effects of fine structure and molecular size on the rheological properties of six mixed-linkage (1→3), (1→4)-β- d-glucans (β-glucans) in the solution and gel state were studied. Molecular size characterization was carried out with high-performance size exclusion chromatography combined with a refractive index detector. Samples were divided into two groups according to the values of apparent molecular weight ( M w) of the peak fraction of the main eluting peak calculated as ∼200×10 3 for an oat, a barley, and a wheat β-glucan and ∼100×10 3 for an oat and a barley β-glucan, and a lichenan sample. All polysaccharides analyzed by 2D NMR spectroscopy and high-performance anion-exchange chromatography of the cellulosic oligomers released by the action of lichenase showed the typical fine structure of mixed-linkage linear (1→3), (1→4)-β- d-glucan. Following lichenase digestion of β-glucans, the molar ratios of tri- to tetrasaccharides (DP3/DP4) were found to follow the order of lichenan (24.5)>wheat (3.7)>barley (2.8–3.0)>oat (2.1). Differences in critical concentration (c ∗ ∗), viscosity, viscoelastic and shear thinning properties among samples were dependent mainly on differences in molecular size of the polymeric chains as well as on the β-glucan fine structure. All β-glucan isolates were able to form gels, as probed by dynamic rheometry; with decreasing molecular size and increasing DP3/DP4 ratio, the gelation time decreased and the gelation rate (I E =[ d log G′/ dt] max ) increased. Differential scanning calorimetry (DSC) showed that cereal β-glucan gels exhibit rather broad endothermic gel→sol transitions at 55–80 °C, while lichenan gels give a sharper transition, implying a more cooperative process. The DSC kinetic data showed similar responses to that from dynamic rheometry; the rate of development of the endotherm increased with increasing DP3/DP4 ratio of the polysaccharide. Furthermore, the storage modulus ( G′) and the apparent melting enthalpy values (plateau Δ H) increased with decreasing molecular size and with increasing DP3/DP4 ratio. The melting temperature of the gel network, as determined by DSC and dynamic rheometry, was found to increase with the molecular size and the DP3/DP4 ratio of β-glucans; the T m for lichenan was ∼89 °C and for cereal β-glucans varied in the narrow range of ∼65–72 °C. Large deformation mechanical tests (compression mode) up to failure revealed an increase in strength and a decrease in brittleness of mixed-linkage β-glucan gels with increasing DP3/DP4 ratio and molecular size of the polysaccharide.

Cell Wall Architecture of the Elongating Maize Coleoptile

The primary walls of grasses are composed of cellulose microfibrils, glucuronoarabinoxylans (GAXs), and mixed-linkage beta-glucans, together with smaller amounts of xyloglucans, glucomannans, pectins, and a network of polyphenolic substances. Chemical imaging by Fourier transform infrared microspectroscopy revealed large differences in the distributions of many chemical species between different tissues of the maize (Zea mays) coleoptile. This was confirmed by chemical analyses of isolated outer epidermal tissues compared with mesophyll-enriched preparations. Glucomannans and esterified uronic acids were more abundant in the epidermis, whereas beta-glucans were more abundant in the mesophyll cells. The localization of beta-glucan was confirmed by immunocytochemistry in the electron microscope and quantitative biochemical assays. We used field emission scanning electron microscopy, infrared microspectroscopy, and biochemical characterization of sequentially extracted polymers to further characterize the cell wall architecture of the epidermis. Oxidation of the phenolic network followed by dilute NaOH extraction widened the pores of the wall substantially and permitted observation by scanning electron microscopy of up to six distinct microfibrillar lamellae. Sequential chemical extraction of specific polysaccharides together with enzymic digestion of beta-glucans allowed us to distinguish two distinct domains in the grass primary wall. First, a beta-glucan-enriched domain, coextensive with GAXs of low degrees of arabinosyl substitution and glucomannans, is tightly associated around microfibrils. Second, a GAX that is more highly substituted with arabinosyl residues and additional glucomannan provides an interstitial domain that interconnects the beta-glucan-coated microfibrils. Implications for current models that attempt to explain the biochemical and biophysical mechanism of wall loosening during cell growth are discussed.

Chemical Composition and Microstructure of Two Naked Waxy Barleys

Two naked waxy barley cultivars, Bz 489-30 and Prowashonupana were analysed in order to study the chemical composition and the microstructure. Prowashonupana contained more protein, fat and ash, and less starch than Bz 489-30. The content of mixed-linked β-glucan was almost three times higher in Prowashonupana (14·9%) than in Bz 489-30 (5·6%), but extractability at 40 °C was much lower. Microscopy studies showed that kernels and endosperm cells of Prowashonupana were irregular, and that cell walls were thicker with a higher content of mixed-linked β-glucan than in Bz 489-30. In both barleys the sub-aleurone layer contained black starch granules, indicating a higher amylose content in this structure compared to the rest of the kernel. Starch was isolated from both barleys using two different procedures and analysed for amylose content. The amylose content was higher, while volume % of A-granules was lower and mean diameter of A-granules was smaller in starch isolated from Prowashonupana than from Bz 489-30. Both composition and yield of starch were however influenced by the isolation procedure used. The procedure which gave higher starch yield resulted in a higher amylose content, probably because of greater recovery of starch from outer parts of the kernel with a higher amylose content.

An endoglucanase, EglA, from the hyperthermophilic archaeon Pyrococcus furiosus hydrolyzes beta-1,4 bonds in mixed-linkage (1-->3),(1-->4)-beta-D-glucans and cellulose.

The eglA gene, encoding a thermostable endoglucanase from the hyperthermophilic archaeon Pyrococcus furiosus, was cloned and expressed in Escherichia coli. The nucleotide sequence of the gene predicts a 319-amino-acid protein with a calculated molecular mass of 35.9 kDa. The endoglucanase has a 19-amino-acid signal peptide but not cellulose-binding domain. The P. furiosus endoglucanase has significant amino acid sequence similarities, including the conserved catalytic nucleophile and proton donor, with endoglucanases from glucosyl hydrolase family 12. The purified recombinant enzyme hydrolyzed beta-1,4 but not beta-1,3 glucosidic linkages and had the highest specific activity on cellopentaose (degree of polymerization [DP] = 5) and cellohexaose (DP = 6) oligosaccharides. To a lesser extent, EglA also hydrolyzed shorter cellodextrins (DP < 5) as well as the amorphous portions of polysaccharides which contain only beta-1,4 bonds such as carboxymethyl cellulose, microcrystalline cellulose, Whatman paper, and cotton linter. The highest specific activity toward polysaccharides occurred with mixed-linkage beta-glucans such as barley beta-glucan and lichenan. Kinetics studies with cellooliogsaccharides and p-nitrophenyl-cellooligosaccharides indicated that the enzyme had three glucose binding subsites (-I, -II, and -III) for the nonreducing end and two glucose binding subsites (+I and +II) for the reducing end from the scissile glycosidic linkage. The enzyme had temperature and pH optima of 100 degreesC and 6.0, respectively; a half-life of 40 h at 95 degreesC; and a denaturing temperature of 112 degreesC as determined by differential scanning calorimetry. The discovery of a thermostable enzyme with this substrate specificity has implications for both the evolution of enzymes involved in polysaccharide hydrolysis and the occurrence of growth substrates in hydrothermal vent environments.

Measurement of Fermentation Products and Substrate Disappearance During Incubation of Dietary Fibre Sources with Human Faecal Flora

A staticin vitrofermentation system using fresh human faeces has been developed. Fermentability may be quantified by determining metabolites and substrate disappearance. Methods to measure hydrogen production by gas chromatography (GC), short chain fatty acids in the fermentation supernatants by GC, and neutral sugar content in the residues by high performance anion exchange chromatography have been developed. These methods have been examined by conducting experiments using dietary fibre sources of variable chemical composition. The results obtained from the different fermentation parameters measured were consistent and indicated the degradation patterns of the substrates tested. The poor fermentability of pea hulls was found to be due to a considerable quantity of highly crystalline cellulose. Oat bran fermentation was characterized by a fast degradation of mixed-linked β-glucans and starch, whereas in apple pomace and celery cell walls pectic substances were easily metabolized by the microflora. It was therefore concluded that the degradation of soluble polysaccharides is mainly determined by their chemical composition. Conversely, the fermentability of insoluble fibres is primarily related to their susceptibility to the faecal flora.In vitrofermentation experiments performed under well defined conditions offer an appropriate means to measure fermentability of dietary fibre rich substances.

Compositional analysis and viscosity measurements of commercial oat brans

Chemical analysis of commercial oat brans (n=10) collected from five different countries showed wide variations in the contents of starch (40–61%), crude protein (9·6–21%), crude fat (8·2–12%), ash (2·0–4·1%), dietary fibre (10·7–19·4%) and mixed-linkage β-glucan (4·7–8·3%). A significant correlation (r=0·83; P<0·005) between the contents of total and extractable mixed-linkage β-glucan was found. Reproducible increases in viscosity were noted for all samples when the finely ground samples were dispersed in phosphate buffer and viscosity development was followed for 100 min. Addition of mixed-linkage β-glucanase greatly reduced the viscosity. Addition of amyloglucosidase, with no detectable mixed-linkage β-glucanase activity, increased viscosity, which could be related to an increased solubilisation of mixed-linkage β-glucan. Highly significant correlations (r>0·90) were found between the contents of both total and extractable mixed-linkage β-glucan and viscosity after dispersion. © 1998 SCI.

Compositional analysis and viscosity measurements of commercial oat brans

Chemical analysis of commercial oat brans (n=10) collected from five different countries showed wide variations in the contents of starch (40–61%), crude protein (9·6–21%), crude fat (8·2–12%), ash (2·0–4·1%), dietary fibre (10·7–19·4%) and mixed-linkage β-glucan (4·7–8·3%). A significant correlation (r=0·83; P 0·90) were found between the contents of both total and extractable mixed-linkage β-glucan and viscosity after dispersion. © 1998 SCI.

Neutral-sugar determination by HPAEC-PAD to characterise polysaccharide degradation by colonic bacteria

In vitro methods are widely used to characterise dietary fibre fermentability in the human hind gut. The activity of faecal bacteria can be followed by measuring fermentation products and substrate degradation. This paper describes a method for establishing neutral-sugar profiles by HPAEC-PAD during in vitro fermentation studies. Freeze-dried fermentation residues are hydrolysed with H2SO4, and the monomers are separated on an anion-exchange column using a reverse NaOH gradient. Sulphate ions are removed from hydrolysates on-line with a special ion-exchange guard column. 2-D-deoxyribose is used as internal standard for quantification. A method for separating all neutral sugars that typically occur in dietary fibre is given. Possibilities for enhancing the resolution for other purposes are suggested. Compared to GC the described method is particularly attractive for routine analyses. Data relating to the decreasing levels of neutral sugars in fermentation residues of oat bran fractions that differ in their mixed-linked β-glucan content are given and degradation patterns of the polysaccharides present in these substrates are discussed.

Solubilisation of Mixed Linkage (1→3), (1→4)β-D-glucans From Barley: Effects of Cooking and Digestion

The extractability behaviour of mixed linkage β-glucans from barley has been assessed during cooking and under different regimens used for fibre isolation. Whilst β-glucan extractability increased to around 50% during cooking, from around 30% in the raw flour, methods used for NSP isolation resulted in even greater extractability (>60%). Enzymic rather than chemical treatments were more important in accounting for the increased extractability. Endogenous proteases from the small intestine were shown to be capable of enhancing the extractability of the β-glucan to a level similar to that found in ileal effluent from patients fed an acute barley-based diet. This is consistent with the presence of mixed linkage β-glucans as proteoglycan complexes in barley. The extractability of β-glucans measuredin vivowas significantly higher (P<0·001) than that measured in the original barley but without protease treatment. Thus, not only do «unextractable» β-glucans from barley have the potential to behave as a source of soluble NSP but the fibre matrix can be modified during transit in the upper gut. Hence the amount of soluble fibre present during gut transit can differ significantly from that measured in foods. This has implications for assessing the mechanism(s) involved in determining the response to fibre in the diet. It is important to take account of modifications to the fibre matrix when assessing the dietary response to a fibre source.

INCUBATION OF BARLEY KERNEL SECTIONS WITH PURIFIED CELL WALL DEGRADING ENZYMES

The effects of purified Trichoderma reesei endo-xylanase, XYN II and endoglucanases, EG I and EG II on the cell walls of embedded barley cross sections were studied. The cell walls were stained with Calcofluor which is specific for (1–3)(1–4)-β-D-glucans (mixed-linked β-glucans) in histochemical applications. Whereas endosperm cell walls remained highly fluorescent after incubation with buffer, the aleurone cell walls were only weakly fluorescent. Both EG I and EG II were effective in degrading the outer endosperm cell walls, but had little or no effect on the inner endosperm and aleurone cell walls. Corresponding sections incubated with XYN II showed highly intensive fluorescence of aleurone cell walls suggesting that the surface of alaurone cell walls contains a high level of xylan which is removed by xylanase treatment exposing a layer rich in mixed linked β-glucans. This indicates that arabinoxylans and β-glucans in the aleurone cell walls of barley exist as separate, macroscopic layers. Xylanase did not have any Influence on the microstructure of endosperm cell walls which are rich in mixed linked β-glucans. However, incubation with EG II after xylanase treatment caused the aleurone cell walls to become very thin.

Water unextractable polysaccharides from three milling fractions of rye grain

Water unextractable material from bran, an intermediate milling fraction and sieved flour of rye grain were sequentially extracted at room temperature with saturated barium hydroxide, water, 4 M potassium hydroxide and water followed by extraction with 2 m potassium hydroxide in a boiling water bath, giving repeatable recoveries of extracts and polysaccharide residue compositions in collected fractions. Total recoveries of polysaccharide residues in extracts and residue from the different water unextractable materials were 78–88%. Extracts in which 90–93% of the carbohydrates were arabinose and xylose residues were obtained by extraction with saturated barium hydroxide. Subsequent extraction with water yielded a fraction in which 64–68% of the carbohydrates were glucose residues. The extraction with hot alkali resulted in extracts in which 85–89% of the carbohydrates were arabinose and xylose residues. The ara/xyl ratio in the collected fractions ranged from 0.1–1.3, with the lowest ratios in fractions that precipitated after neutralisation of the 4 m potassium hydroxide extract and the highest ratios in the unextractable residues. Structural characterisation with 1H-NMR spectroscopy revealed varying substitution patterns for arabinoxylans in the different extracts and that glucose residues in the extracts essentially originated from mixed-linked β-glucan. The proportion of disubstituted xylose residues was lower in barium hydroxide extracts compared to the other main extracts. A highly branched heteroxylan was extracted with hot alkali. The polysaccharides found in the corresponding extracts for all the starting materials had generally similar structural features, but the yield differed considerably.

Effects of environment and cultivar on total mixed-linkage β-glucan content in eastern and western Canadian barleys (Hordeum vulgare L.)

Two-row and six-row barley (Hordeum vulgare L.) cultivars, originating from eastern and western Canada were grown in six environments across Canada. The grain was studied for the effects of environment and cultivar on the total mixed-linkage β-glucan content (TBG, g hg−1 DM). Environment significantly affected TBG content in barley. TBG content differed among the 32 eastern or 43 western Canadian cultivars but the mean TBG for the eastern cultivars was not different from the mean of western cultivars. All eastern cultivars were hulled-feed types, and their TBG ranged from 3.31 in Micmac to 4.09 in Birka for two-row barleys, and from 3.78 in OAC Kippen to 4.41 in Maskot for six-row barleys. Western hulled cultivars which included feed and malt types, contained TBG ranging from 3.37 in Manley to 4.14 in Betzes for two-row barleys, and from 3.71 in Heartland to 4.42 in Tankard for six-row barleys. The two-row hulless cultivar, Condor, contained more TBG (4.69) than the other two hulless cultivars, CDC Richard (3.91) and Scout (3.85). The six-row hulless cultivars, CDC Buck and Tupper, were equal in TBG (4.35). The mean TBG was lower for the two- than six-row barleys (3.7 vs. 4.0), and a broader range for TBG in the two- (3.31–4.55) than six-row barleys (3.71–4.42) suggesed the greater possibility of producing low and high TBG barleys from the two-row cultivars. Canadian cultivars contained between 3.31 and 4.55 g TBG hg−1 DM, and there is a need to develop feed-type barleys containing a third less than the minimum TBG, and food-type barleys containing a third more than the maximum TBG. Key words: Barley, Hordeum vulgare cultivars, β-glucan