The water extractable (WE) arabinoxylans from two rye flours differing in baking quality were studied following sequential extraction with water at 4, 40, and 100 degrees C. Ammonium sulfate fractionation of the resulting WE fractions and subsequent analysis revealed substantial differences in the structure of the isolated materials. Furthermore, it allowed us to identify the factors contributing to arabinoxylan water extractability. Our results provide compelling evidence for the existence of separate polymers in rye WE arabinoxylans with different substitution degrees, ranging from quantitatively dominating, lowly substituted populations (arabinose to xylose ratio, Ara/Xyl approximately 0.5) to comparatively less abundant, highly substituted analogues (Ara/Xyl approximately 1.3). Generally, arabinoxylan water extractability was governed by the relative proportion of lowly and highly branched structures. A gradually increasing proportion of highly substituted populations was observed from cold to hot WE fractions. This was associated with the lower proportion of monosubstituted xylopyranosyl residues in the backbone, the higher proportion of disubstituted xylopyranosyl residues, and the higher level of substitution with feruloyl residues. Notable differences in the ratio of phenolic compounds to arabinose residues were observed between corresponding polymers isolated from rye flours of high and low baking quality, whereas the differences in their molecular weights were much less pronounced.

The alkali extractable (AE) arabinoxylans from two rye flours differing in baking quality were studied following sequential extraction of water-unextractable and starch-free rye flour residue with saturated barium hydroxide solution, water and 1 M sodium hydroxide solution (Ba, BaH, and Na, respectively), and further fractionation of isolated fractions by ammonium sulfate precipitation. (1)H NMR and sugar analyses of AE subfractions provided evidence for the presence of lowly branched arabinoxylans (average arabinose-to-xylose ratio, Ara/Xyl approximately 0.5), containing mainly un- and monosubstituted xylopyranosyl residues (Xylp) in the chain. The proportion of this subfraction decreased from 50% in the Ba fraction to 35 and 17% in the Na and BaH fractions, respectively. Other subfractions, rich in both mono- and disubstituted Xylp, represented arabinoxylan populations with intermediate (Ara/Xyl approximately 0.8) and high substitution degree (Ara/Xyl approximately 1.1). The Ba and Na fractions contained phenolic compounds, whereas they were absent in the BaH fraction. The higher ratio of such phenolic compounds to arabinose (PhC/Ara) found in AE arabinoxylans from rye flour of inferior baking quality was one of the most pronounced differences between arabinoxylan populations from rye flours with high and low baking quality. The arabinoxylans from rye flour of high baking quality present in Ba and Na fractions had slightly higher apparent molecular weights (MWs) when compared to those from rye flour with low baking quality. The arabinoxylans present in the BaH fractions, characterized by the highest MWs, had similar MWs.

No abstract

Recent studies have indicated that some structural features of arabinoxylans, the major cell wall polysaccharides, might be potential quality markers in the selection of rye breeding materials. To specify the most appropriate characteristics, the differences in the structure of cell wall components were studied in two ryes with high and low breadmaking qualities. Two cell wall fractions were isolated from the outer layers of the grain (pooled shorts and bran fractions) by a consecutive water extraction with alpha-amylase (WE-A) and proteinase K (WE-P). Polysaccharides predominated in the WE-A fraction (approximately 64%, mainly arabinoxylans). By contrast, the WE-P fraction contained mostly protein (approximately 63%), and its level of polysaccharides was relatively low (approximately 18%). The 1H NMR and sugar analysis of the ammonium sulfate precipitated subfractions revealed that the WE-A was built of four arabinoxylan populations with marked structural differences (arabinose-to-xylose ratios, Ara/Xyl, of approximately 0.3, 0.5, 0.8, and 1.2). Instead, the arabinoxylans present in the WE-P were generally enriched in disubstituted xylopyranosyl residues. The ratio of phenolic components to arabinose residues in the WE-P fraction (indicated by 1H NMR) and the proportion of polymers with the highest molecular weights in the WE-A fraction (revealed by HPSEC) distinguished well two ryes with diverse breadmaking qualities. Much less obvious differences between both ryes were observed in the ratio of amide I to amide II band intensities of FTIR spectra for the WE-P and in the level of phenolic acids and ferulic acid dehydrodimers for both cell wall preparations.