Computational analysis of di-peptides correlated with the optimal temperature in G/11 xylanase

Abstract

The di-peptides positively correlated with optimal temperature in G/11 xylanase are computed to be: LA, LG, CD, GD, RY, CH; the negatively ones are: DC, YI, YP, and CP. The comparison of the structures of mesophilic and thermophilic xylanase showed that these di-peptides predominantly occur in beta-turns. These results explain the successful improvement of thermostability by introducing arginines into the xylanase surface area and therefore have implications for xylanase engineering.

Introduction

Recently there has been an increasing interest in xylanase (EC 3.2.1.8) for its wide application in industry. It breaks randomly down the internal β-1,4-glycosidic bond of xylan, the major hemicellulose component of the plant cell wall. This makes it a very useful enzyme in animal feed, food and beverage industry, bakery industry, fruit juices-clarifying, and especially in pulp and paper industry [1]. But very often, the biotechnology environment is extreme and demands robust xylanase. For example, pulp bleaching is carried out in a very low pH environment, connected with a hot, caustic treatment of wood, which limits the usage of xylanase. Therefore, thermostable xylanase is of interest [2], [3]. To elevate the optimal temperature of an enzyme, many methods have been employed, such as site-directed mutagenesis (SDM), directed evolution and computational design [4]. To elucidate the thermostability of protein, such methods as crystal structure comparison and genomic comparison have been used [5], [6], [7], [8], [9], and many factors were assumed to be responsible [5], [6], [7], [10]. However, the limited number of known protein structures available in the protein data bank (PDB) makes it difficult to study the thermostability in a single enzyme family; since different proteins use different strategies for maintaining thermostability, the genomic comparison methods may give inconsistent results [11], [12], [13]. Because of the limited number of crystal structures and the prevalence of neutral mutations occurring in proteins, different thermostability analyses often do not agree or even contradict with each other [14]. Enzyme engineering based on these conflicting results often did not lead to an expected conclusion, sometimes even led to the reverse effect [15]. Given the Anfinsen's principle that the fold of a protein is encoded in its amino acid sequence [16], and the large number of protein sequences available in sequence databases, it is necessary to study the relationship between sequence properties and thermostability, we selected xylanase in this study because a large number of sequences of this protein are available in the Swiss-prot databank. Using sequence analysis, we studied the amino acids correlated with the optimal pH in G/11 xylanase [17]; using principle components analysis, we discriminated between these two families [18], and found the di-peptides correlated with the optimal temperature in F/10 xylanase [19].

In the process of analyzing thermostability of enzyme, the optimal growth temperature of the source organism was commonly used [5], [20]. Whereas, it usually cannot reflect the physicochemical property of an enzyme itself, there is often a big difference between the T_opt of xylanase and the growth temperature of source organism, sometimes the difference reaches to 33 °C [21]. The xylanase produced by an Aspergillus strain living at 37 °C showed maximal activity at 80 °C [22]. During the construction of the analysis data set, xylanases coming from the same organism were also found to have different T_opts; this fact reflects a different thermostability. Based on the investigation of structural similarities, it is commonly held that xylanases have evolved by domain shuffling [23], the study of genomic comparison also indicated that gene transfer occurred between archaea and bacteria [24]. Based on the homologous alignment of G/11 xylanases, we also found that the relationship between sequences is closer than that of the organisms from taxonomic classification. Therefore, it is no advantage to use the optimal growth temperature of source microorganism as criterion, and the T_opt of xylanase is used instead in the present study. Because of its low molecular weight, G/11 xylanase is often regarded as an advantageous agent in pulp bleaching compared with that of F/10 xylanase. In the present study, the dipeptides correlated with the optimal temperature of xylanase were computed by using G/11 xylanase sequences from Swiss-Prot (release 43.5.0 of 07-Jun-2004).