Behavior of Triticum durum Desf. arabinoxylans and arabinogalactan peptides during industrial pasta processing.

Three industrial pasta processing lines for different products (macaroni, capellini and instant noodles) were sampled at three subsequent stages (semolina, extruded, and dried end products) in the process. Arabinoxylans (AX) and arabinogalactan peptides (AGP) were analyzed. Although very low endoxylanase activities were measured, the level of water-extractable AX (WE-AX) increased, probably because of mechanical forces. No change was observed in the level and structural characteristics of AGP. The WE-AX molecular weight (MW) profiles showed a very small shift toward lower MW profiles; those of AGP revealed no changes as a result of the production process. After separation of WE-AX and AGP, (1)H NMR analysis and gas chromatography of the alditol acetates obtained following hydrolysis, reduction, and acetylation revealed no changes in the arabinose substitution profile of the WE-AX samples during pasta processing. At optimal cooking times, WE-AX losses in the cooking water are small (maximally 5.9%). However, the loss of AGP is more pronounced (maximally 25.0%). Overcooking led to more losses of both components.

Endoxylanases in durum wheat semolina processing: solubilization of arabinoxylans, action of endogenous inhibitors, and effects on rheological properties.

Endoxylanases seriously affect the rheological properties of durum wheat (Triticum durum Desf.) semolina spaghetti doughs prepared with, and as evaluated, by the farinograph. Under the experimental conditions, control doughs (34.9% moisture content) made from two semolinas (semA and semB) yielded a maximal consistency of 525 and 517 farinograph units (FU), with, respectively, 19.4 and 16.4% of the total level of arabinoxylans (TOT-AX) being water-extractable (WE-AX). When 75.4 Somogyi units/50 g of semolina of the endoxylanases from Trichoderma viride (XTV), rumen microorganisms (XRM), Bacillus subtilis (XBS), and Aspergillus niger (XAN) were used, the maximal consistencies at 34.9% moisture decreased for semA to 467, 436, 448, and 417 FU, respectively. This was accompanied by increased WE-AX contents of 60.8, 71.2, 70.7, and 73.0%, respectively. Similar results were observed for semB. By reducing the total water content of doughs, it was possible to recover the maximal consistency of the original doughs. Both the decrease in maximal consistency and the amount of water to be omitted were significantly related to the decrease in molecular weight (MW) of the WE-AX and the percentage of WE-AX solubilized as a result of the enzymic action. At the same time, it was clear that endogenous endoxylanase inhibitors were present in the durum wheat semolinas and that they inhibited the endoxylanases used to different degrees. Part of the differences in effects between the different endoxylanases (decrease in maximal consistency, amount of AX solubilized, MWs of the WE-AX, and amount of water that could be omitted) could be ascribed to the differences in inhibition of the endoxylanases by endogenous inhibitors.