1 Department of Chemical and Biochemical Engineering, Technical University of Denmark2 Center for BioProcess Engineering, Department of Chemical and Biochemical Engineering, Technical University of Denmark
The work presented in this PhD thesis has provided a better understanding of the enzyme kinetics and quantitative phenomena of the hydrolysis of xylan substrates by selected pure enzyme preparations. Furthermore, the options for producing specific substituted xylooligosaccharides from selected substrates by specific xylanase treatment have been examined. The kinetics of the enzymatic degradation of water-extractable wheat arabinoxylan (WE-AX) during designed treatments with selected monocomponent enzymes was investigated by monitoring the release of xylose and arabinose. The results of different combinations of α-Larabinofuranosidases (EC 184.108.40.206), one derived from Aspergillus niger (AFAn) and one from Bifidobacterium adolescentis (AFBa), a β-xylosidase (EC 220.127.116.11) from Trichoderma reesei, and a D11F/R122D variant of an endo-1,4- -xylanase (EC 18.104.22.168) from Bacillus subtilis (BsXmut) were examined. The selected arabinofuranosidases catalyze liberation of arabinofuranosyl residues linked 1→3 to singly (AFAn) or doubly (AFBa) substituted xylopyranosyl in arabinoxylan, respectively. AFBa catalyzed the release of more arabinose, i.e. had a higher rate constant than AFAn, when added to arabinoxylan at equimolar levels. With respect to the xylose release, AFAn exhibited a better synergistic effect than AFBa with β-xylosidase. The differences in the synergistic effect could be related to the different mode of action for the two arabinofuranosidases: AFAn enhanced the probability of more unsubstituted xyloses at (or near) the non-reducing ends for β-xylosidase to attack. AFBa catalyzed the removal of 1→3 linked arabinofuranosyl, but the β-xylosidase still could not work on the xylan backbone, because there was a α-1→2 linked arabinofuranosyl blocking the binding site. However, the synergistic effects between -xylosidase and the α-L-arabinofuranosidases on the xylose release were low as compared to the effect of xylanase addition with β-xylosidase, which increased the xylose release by ~25 times in 30 minutes. At equimolar addition levels of the four enzymes, the xylanase activity was thus rate-limiting for the -xylosidase catalyzed depolymerization to release xylose from arabinoxylan. Thus, the provision of more (unsubstituted) non-reducing ends resulting from xylanase action was more efficient to boost the -xylosidase activity than provision of more (randomly) unsubstituted xyloses in the arabinoxylan backbone. The kinetics and substrate selectivity of the B. subtilis wildtype xylanase, BsX, which is sensitive to inhibition by TAXI, and the engineered variant, BsXmut, which is much less inhibited by TAXI, was examined in order to elucidate the influence of the structural point mutations. Three dimensional structures of both xylanases were superimposed to elucidate the structural basis for differences in their hydrolytic properties. The comparison showed that the D11F mutation appeared to cause a slight narrowing of the entrance to the active site cleft because the phynylalanine was more bulky than the aspartic acid. The two xylanases were incubated individually with WEAX, water-unextractable arabinoxylan (WUAX), birchwood xylan, and wheat bran, respectively. At equimolar addition, the activity of BsXmut was lower than that of BsX with respect to both the initial rate and the product yields obtained after prolonged reaction on the xylan substrates. The lower activity could be related to steric hindrance caused by the D11F mutation. The calculated substrate selectivity factors indicated that BsX and BsXmut both had higher catalytic rate on WUAX than on WEAX. Addition of a 100:1 (TAXI:xylanase) molar ratio of the inhibitor confirmed the significantly decreased inhibition of BsXmut by TAXI. Addition of TAXI also influenced the xylanases’ selectivity factor differently. In order to assess the heterogenous structure of the substrate matrix and the change occurring during the xylanolytic reaction, the possibilities for using high-performance size exclusion chromatography (HPSEC) as a quantitative method to assess xylo-oligosaccharide profiles was examined. HPSEC is a widely used method for the qualitative profiling of oligosaccharide mixtures. A novel method employing HPSEC for the quantitative analytical profiling of the progress of enzymatic hydrolysis of different xylan substrates was developed. The method relies on dividing the HPSEC elution profiles into fixed time intervals and utilizing the linear refractive index response (area under the curve) of defined standard compounds. In order to obtain optimal high-performance size exclusion chromatography profiles, the method was designed using 0.1 M CH3COONa in both the mobile phase and as the sample solution. This was based on the systematic evaluation of the influence of the mobile phase, including the type, ionic strength and pH, on the refractive index detector response. A time study of the enzyme catalyzed hydrolysis of birchwood xylan and wheat bran by BsX was used as an example to demonstrate the workability of the new HPSEC method for obtaining progress curves describing the evolution in the product profile during enzyme catalysis. Flaxseed mucilage (FM) has recently been reported to contain an interesting structure, notable a mixture of highly doubly substituted arabinoxylan as well as rhamnogalacturonan I (RGI) with unusual side group substitutions. This substrate was therefore evaluated as a potential substrate for the production of xylo-oligosaccharides catalyzed by BsX. Treatment of FM with BsX resulted in limited depolymerization, but when BsX and FM were incubated together on WE-AX, WUAX and birchwood xylan, significant amounts of xylose were released. Moreover, arabinose was released from both WE-AX and WU-AX. Since no xylose or arabinose was released by BsX addition alone on these substrates, nor without FM or BsX addition, the results indicate the presence of endogenous β-D-xylosidase and α-Larabinofuranosidase activities in FM. FM also exhibited activity on both p-nitrophenyl α-Larabinofuranoside (pNPA) and p-nitrophenyl β-D-xylopyranoside (pNPX). The potential of producing glucurono-xylo-oligosaccharides (GXOS) from wheat bran via specific treatment with BsXmut was investigated. After the enzyme catalyzed hydrolysis by BsXmut, the GXOS were isolated by anion exchange chromatography and the fractions obtained were analyzed for the presence of uronic acid, and by High Performance Anion Exchange Chromatography (HPAEC) and LC/MS for structural verification. Since phosphate also co-eluted during the anion exchange chromatography, the amount of phosphate in the fractions was also determined. LC/MS analysis showed that GXOS was isolated from wheat bran but an even larger amount of RGI was present in the obtained samples together with phosphate. Therefore, further purification has to be made in order to obtain GXOS.