Eric Westerlund

The effects of various thermal processes on dietary fibre and starch content of whole grain wheat and white flour

The content of dietary fibre (DF) and starch in raw and processed materials obtained from wheat was evaluated. The processes studied were extrusion cooking, drum-drying, autoclaving, popping and steam flaking. The processing conditions were chosen to represent the limits that could be used commercially. The DF content was detennined using an enzymic-gravimetric method for soluble and insoluble DF components and a gas-liquid chromatographic method for complete characterisation of the constituent monosaccharides. Values for the total DF obtained by the two methods for a range of samples were closely correlated ( r = 0⁗99; P

Plant cell walls and monogastric diets

Abstract A brief summary is given of the recent and progressing research at the Swedish University of Agricultural Sciences in development of analytical methodology for plant cell wall components (dietary fiber) and mixed-linkage β-glucans. Applications of the methods for the chemical characterization of various types of human and monogastric animal foods will be presented, including heat-processed, starch-rich foods. In the latter group Maillard reaction products, “resistant starch” and chemically transformed starch increase the dietary fiber values. Nutritional studies on the role and properties of plant cell wall materials in connection with bile acid and cholesterol metabolism in hamster and with digestion of pig and poultry diets will be discussed.

Isolation and chemical characterization of water-soluble mixed-linked β-glucans and arabinoxylans in oat milling fractions

Abstract Dehulled oats were separated into fractions of bran, outer starchy endosperm and inner starchy endosperm, after and without prior steam-flaking. The bran had significantly higher contents of protein, fat, ash and dietary fibre and lower content of starch than the endosperm fractions. An improved procedure was developed for the isolation of pure (92–99%) water-soluble mixed-linked (1–3), (1–4)-β- d -glucans, involving fat extraction, enzymatic removal of starch and protein, and subsequent precipitation of water-soluble polysaccharides with 60% aqueous ethanol and 20% (w/v) aqueous (NH 4 ) 2 SO 4 . High-field 13 C-NMR analysis of the isolated mixed-linked β-glucans showed the polysaccharide structure to be similar in all oat fractions. The proportion of isolated (1–3)-β-linkages in the polymer was on average 28·3% (SD 0·8) according to 1 H-NMR analysis. 1 H-NMR also showed that water-soluble oat arabinoxylans were composed of a main chain of (1–4)-linked β- d -xylopyranosyl residues, substituted by terminal α- l -arabinofuranosyl residues predominantly at O-3 but also at both O-2 and O-3. The proportion of xylose residues substituted solely at O-3 was 35% higher in bran than in endosperm arabinoxylans.

Effects of baking on water-soluble non-starch polysaccharides in white bread fractions

Water-soluble non-starch polysaccharides were extracted from white flour and dough as well as from crumb, inner crust and outer crust fractions of white bread, baked at 210 °C for 22 or 35 min. Results from the two baking experiments were very similar. Extracted polysaccharides were fractionated on DEAE-cellulose and by different solubility properties in aqueous saturated ammonium sulfate. Fractions isolated were characterized by sugar- and 1 H-n.m.r. analysis, revealing the presence of arabinoxylan mixtures with widely different degrees of arabinose substitution. Arabinogalactanproteins isolated showed little variation in galactose/arabinose ratio (1·5–1·8). Highfield 1 H-n.m.r, of arabinoxylan fractions was particularly useful for estimating the ratios of terminal arabinofuranosidic residues linked to xylose residues at O -2 and O -3 or solely at O -3. Water-soluble arabinoxylans had a higher degree of arabinose substitution in bread fractions than in the flour and dough, probably due to increased solubilization of highly substituted arabinoxylans during baking.

Investigation of the distribution of methyl ester groups in pectin by high-field 13C NMR

Abstract Fully methyl-esterified, water-soluble citrus pectin was partially hydrolysed with alkali to different degrees of esterification (DE). High-field 13C-NMR spectroscopy of the products showed that ester carbonyl carbons appeared as four signals due to different influences from neigh-bouring sugar residues (free or methyl-esterified galactosyluronic acid). Based on the relative peak areas of these signals, it was deduced that alkaline de-esterification occurred randomly. The distribution of methyl ester groups was also found to be random in water-soluble native pectins of tobacco (DE 20) and papaya (DE 70).

Chemical characterization of water-soluble pectin in papaya fruit

Water-soluble pectin was extracted from ripe papaya fruit pulp and fractionated by ion-exchange chromatography and gel filtration. Sugar analysis of fractions obtained revealed that the major component was a rhamnogalacturonan with low content of arabinose and galactose residues (2–3%). About 50% of the galactopyranosyluronic acid residues were methyl-esterified and 1H and 13C NMR-analysis established that these ester groups were randomly distributed. COSY-NMR was used to distinguish between resonances for protons in free or esterified uronic acid residues.

Effects of rye bran, oat bran and soya-bean fibre on bile composition, gallstone formation, gall-bladder morphology and serum cholesterol in Syrian golden hamsters (Mesocricetus auratus)

The effects of rye bran, oat bran and soya-bean fibre on serum lipids, bile composition and gallstone formation were studied in male Syrian golden hamsters (Mesocricetus auratus). The control groups received fibre-free stone-provoking (O1 diet) or non-stone-provoking (O2 diet) diets. The serum cholesterol levels were lower for all groups fed on the diets supplemented with the dietary fibre sources compared with the control groups. The total content of bile acids in bile was higher in groups given rye-bran diets compared with the corresponding controls. The proportion of cholic acid was higher and that of chenodeoxycholic and lithocholic acid lower in the groups given rye-bran-, oat-bran- or soya-bean-fibre-supplemented diets, compared with the corresponding controls. The secondary:primary bile acid ratio was lower in the group given the rye-bran-supplemented O1 diet. The lithocholic:deoxycholic acid ratio was lower in the groups given rye-bran-, oat-bran- or soya-bean-fibre-supplemented diets than in the corresponding controls. A lower frequency of gallstones was observed only for the group receiving the rye-bran-supplemented O1 diet while the lithogenic index was lower in the groups given the rye-bran-supplemented O2 diet. A decreased epithelial volume density of the gall-bladder and an increased smooth muscular volume density were observed in animals given oat-bran- and rye-bran-supplemented O1 diets, whereas for the soya-bean-fibre-supplemented O1 diet, only the smooth muscular volume density was increased.

Structural features of an arabinan fragment isolated from the water-soluble fraction of dehulled rapeseed

A water-soluble polysaccharide fraction was prepared from dehulled rapeseed meal (winter rapeseed variety Casino). Further purification yielded two major fractions having a high content of arabinose and galactose residues, with Ara/Gal ratios of 5.4 (G1) and 1.8 (G2). The Ara/Gal ratio of the high molecular weight fraction G1 was stable over the whole gel filtration peak, indicating that the arabinose and galactose residues are part of the same polysaccharide. The high molecular weight fraction G1 was studied further by methylation analysis and several NMR techniques. Structural studies showed G1 to consist mainly of arabian fragments, which have terminal alpha-L-arabinofuranosyl groups with anomeric carbons bound (1-->5) (A) or (1-->2) (B), and 2,5-substituted arabinosyl residues with anomeric carbons bound (1-->5) (D) or (1-->2) (C) to adjacent arabinosyl residues. The A:B:C:D ratios were 2:1:1:1 according to results from NMR and methylation analysis.

The Uppsala Method for Rapid Analysis of Total Dietary Fiber

An understanding of the nutritional effects of dietary fiber has been considerably hampered by the lack of an appropriate definition, and consequently of adequate analysis methods, for this component. Based on physiological criteria, Trowell (1972) defined dietary fiber as “the remnants of the plant cell-wall that are not hydrolysed by the alimentary enzymes of man”, and this was later simplified and expanded to include “the plant polysaccharides and lignin which are resistant to hydrolysis by the enzymes of man” (Trowell et al., 1976). In 1979 we proposed that dietary fiber could be defined as the sum of non-starchy polysaccharides and Klason lignin, and, in conjunction with this chemical definition, published the first method (the Uppsala method) for the analysis and characterization of fibers (Theander and Aman, 1979). As this analytical procedure includes ‘starch’ resistant to α-amylases and, as part of the Klason lignin complex, other indigestible components such as tannins, cutins and Maillard products, this definition conforms well with the original definitions of Trowell and co-workers.

The effects on aqueous ethanol-soluble carbohydrates and protein in heat-processed whole grain wheat and white flour

Whole grain wheat and white flour were processed by extrusion cooking, drum-drying, autoclaving, popping and steam-flaking using two conditions representing the limits of commercial treatments. Processing effects on content and composition of 80% (v/v) aqueous ethanol-soluble carbohydrates were studied using enzymic and gas-liquid chromatographic methods. The total content of low-mol. wt. carbohydrates in the extracts was only slightly affected, except for the process of drum-drying in which a significant increase of maltose occurred. Trace amounts of D -glucosaccharides having end units of β-1,6-anhydro- D -glucopyranose were detected (as maltosan), after amylolytic treatment and enrichment, on charcoal-Celite of extracts from popped wheat. The content of 80 % ethanol-soluble protein was markedly decreased in all processes (due to denaturation) and this was a sensitive marker for the extent of heat treatment.