Acidic xylanases possess the unique features necessary for the tolerance of acidic environments, which may have great potentials for industrial purposes. However, factors controlling the pH-dependent stability of xylanases are only partially known. Here we proposed a residue interaction networks based method to analyze the differences of residue interactions between 6 pairs of experimentally verified acidic and neutral xylanases. They had very close numbers of aromatic amino acids, however extremely significant more (p < 0.001) π-π stacking interactions existed in acidic xylanases, which has not been reported before. Whereas the interactions between Tyrosine-Phenylalanine (Tyr-Phe) and Phenylalanine-Phenylalanine (Phe-Phe) were the main contributors. An equation quantitatively described the relationship between the optimal pH and the number of π-π stacking interactions was proposed. The predicted optimal pHs for three xylanases was 4.13, 6.7 and 6.1, while the experimental values of the optimum pHs were 4.6, 6.5 and 6.5, with an absolute error of 0.47, 0.2 and 0.4 pH unit, respectively. By counting the aromatic residue pairs forming π-π stacking in the 3D structure of an acidic (PDB ID: 1BK1, with an optimal pH of 2) and a neutral (PDB ID:1XXN, with an optimal pH of 6.5) xylanase, we found significant differences existed in the positions ranging from 145 to 166 in forming π-π stacking. Two phenylalanines at position 149 and 157 in the acidic xylanase, which involved in 7 π-π stacking interactions, played an important role in the stability of xylanase at low pH environment, which was further proved by a mutation experiment. A mutated xylanase with Phe149 → Ala149 and Phe157 → Ala157 was expressed and purified, resulting the optimal pH shifted from 2 to 4.5. The interaction networks based method paved a new way in underlying and engineering the acid-stability of xylanase, as well as the characteristics of other enzymes.
Keywords: Acid-stability; Enzyme engineering; Residue interaction networks; Xylanase; π-π stacking.
Copyright © 2018 Elsevier B.V. All rights reserved.