Understanding xylanase functionality in wheat bread making using protein engineering and 1H NMR as tools

Arabinoxylan (AX) is the primary non-starch carbohydrate in cereal cell walls and is as such present in wheat flour and meal. It constitutes an important part of our daily dietary fibre intake. Because of its properties, AX directly influences the quality of products made from wheat in general and that of bread in particular. Endo-β-(1,4)-D-xylanases, enzymes able to hydrolyze internal β-(1,4)-linkages in AX, are commercially available to modify the AX population and its properties to improve dough handling and bread quality. To further optimise these enzymes and take full advantage of their quality improving capacity, one must thoroughly understand their functionality in bread making and the way the biochemical properties of the enzymes affect their mode of action. In the last two decades, much progress has been made, tying xylanase functionality to inhibition sensitivity, substrate specificity and substrate selectivity. Solubilization of water-unextractable AX (WU-AX) to water-extractable AX (WE-AX) with a high molecular weight was mainly associated with an increased AX induced viscosity of the dough aqueous phase. In addition, solubilization of WU-AX results in the release of water, which becomes available for gluten hydration and film formation. The importance of hydration for the xylanase functionality in bread making is however not clear. A lot of research has, however, focussed on comparing xylanases of different microbial origin that differ not only in a biochemical property of interest, but also in other aspects. Using such an approach, one-on-one correlations between specific biochemical parameters and xylanase functionality in bread making cannot be made. In addition, while in depth studies on the relation between enzyme structure and biochemical properties using protein engineering were performed, amongst others at the Laboratory of Food Chemistry and Biochemistry, the knowledge gained there has not been put to use in bread making. The aim of this PhD study was hence to generate new insights in xylanase functionality in bread making by making use of knowledge on the molecular structure of selected xylanases in combination with their biochemical properties and protein engineering. Using site-directed mutagenesis mutant xylanases differing only in their substrate selectivity on the one hand and in inhibition sensitivity on the other hand were produced. It was shown that increasing the substrate selectivity of a xylanase, being its preference for hydrolysis of water-unextractable substrates compared to water-extractable ones, by modifying its secondary binding site (SBS) can enhance its functionality in bread making. Due to an increased WU-AX solubilization during fermentation, more WU-AX was converted into solubilized AX (S-AX). This resulted in an increased specific loaf volume and a lower optimal dosage when weak wheat flour was used. Xylanases differing only in their inhibition sensitivity were useful for investigating the activity profile of a xylanase during bread making. Xylanases with an increased inhibition sensitivity were active during a shorter time interval. For xylanases sensitive to one or more inhibitors, the solubilization rate was reduced to zero after the first minutes of fermentation. While solubilization was also strongly reduced after mixing for inhibition insensitive xylanases, still a significant amount of WU-AX was solubilized during the beginning of fermentation. The formation of reducing end xylose was also strongly reduced at the beginning of fermentation, but still some activity was measurable for all xylanases after mixing. For inhibition insensitive xylanases, hydrolysis could be detected until the end of fermentation. Furthermore, increasing the inhibition sensitivity of a xylanase increased the dosage that was needed to obtain an optimal effect on specific loaf volume. When dosage optimization was performed, no correlation was observed between the inhibition sensitivity of xylanases and their bread improving capacity. This led to the hypothesis that enzyme activity during mixing is quite important and that water release by WU-AX rather than viscosity build-up determines xylanase functionality. Indeed, viscosity differences could not be pinpointed as driving mechanism for differences in bread making performance between wild-type and mutant enzymes. This is in contrast to the previously postulated model which states that the effects on bread making were mainly associated to changes in viscosity of the dough liquor phase due to hydrolysis of AX. From rheological measurements, water redistribution in the dough as a result of xylanase activity on WU-AX was suggested. This was confirmed by investigating model systems with increasing complexity using time domain (TD) proton nuclear magnetic resonance (1H NMR) relaxometry. The measurements showed that WU-AX absorbs a lot of water, which makes that the other flour components are less hydrated. Addition of a xylanase with a WU-AX solubilizing activity reduces the impact of WU-AX on the water dynamics in dough. This amount of water was immediately bound by other wheat flour components. For a three component system containing wheat starch, gluten and WU-AX and in a wheat flour-water model system, hydrolysis of WU-AX reduced the impact on water distribution, resulting in less withdrawal of water from other wheat flour components. This resulted for the wheat flour-water model system in a more structured 1H NMR profile. Above findings were confirmed when using three different xylanases, extracted from commercial xylanase formulations, in bread making on 10, 100 and 400 g scale. When changing the bread making scale without changing the bread making procedure, only differences in temperature-time profile during baking were observed. Xylanases with a higher temperature optimum remained active longer and could further increase the specific loaf volume. When wheat dough was fermented up to a constant dough height, as was done in bread making on pilot scale (400 g), also the properties of the fermenting dough were different. It was observed that dough samples supplemented with an inhibition insensitive xylanase were drier and less extensible after mixing, while dough manageability was higher. During fermentation, dough became more sticky compared to when inhibition sensitive xylanases were used. The temperature optimum of the xylanase seemed an important parameter for its bread improving capacity. The disadvantage of xylanase activity after mixing in this bread making procedure without punching during fermentation was a less homogeneous bread crumb structure. The findings in this PhD dissertation underpin the importance of the capacity of a xylanase to solubilize WU-AX as an important parameter in selecting xylanases for bread making purposes. WU-AX absorbs a lot of water, which makes that the other flour components are less hydrated. Solubilization of WU-AX reduces the impact of this AX population on the water dynamics of dough and leads to an improved dough and bread quality. These observations are also advantageous for enzyme producers to select xylanases with a high dough and bread improving capacity and for other cereal based processes in which AX is an important determining component.status: publishe