Determinants for the improved thermostability of a mesophilic family 11 xylanase predicted by computational methods

Abstract

Background: Xylanases have drawn much attention owing to possessing great potential in various industrial applications. However, the applicability of xylanases, exemplified by the production of bioethanol and xylooligosaccharides (XOSs), was bottlenecked by their low stabilities at higher temperatures. The main purpose of this work was to improve the thermostability of AuXyn11A, a mesophilic glycoside hydrolase (GH) family 11 xylanase from Aspergillus usamii E001, by N-terminus replacement.

Results: A hybrid xylanase with high thermostability, named AEXynM, was predicted by computational methods, and constructed by substituting the N-terminal 33 amino acids of AuXyn11A with the corresponding 38 ones of EvXyn11TS, a hyperthermostable family 11 xylanase. Two AuXyn11A- and AEXynM-encoding genes, Auxyn11A and AExynM, were then highly expressed in Pichia pastoris GS115, respectively. The specific activities of two recombinant xylanases (reAuXyn11A and reAEXynM) were 10,437 and 9,529 U mg-1. The temperature optimum and stability of reAEXynM reached 70 and 75°C, respectively, much higher than those (50 and 45°C) of reAuXyn11A. The melting temperature (Tm) of reAEXynM, measured using the Protein Thermal Shift (PTS) method, increased by 34.0°C as compared with that of reAuXyn11A. Analyzed by HPLC, xylobiose and xylotriose as the major hydrolytic products were excised from corncob xylan by reAEXynM. Additionally, three single mutant genes from AExynM (AExynMC5T, AExynMP9S, and AExynMH14N) were constructed by site-directed mutagenesis as designed theoretically, and expressed in P. pastoris GS115, respectively. The thermostabilities of three recombinant mutants clearly decreased as compared with that of reAEXynM, which demonstrated that the three amino acids (Cys5, Pro9, and His14) in the replaced N-terminus contributed mainly to the high thermostability of AEXynM.

Conclusions: This work highly enhanced the thermostability of AuXyn11A by N-terminus replacement, and further verified, by site-directed mutagenesis, that Cys5, Pro9, and His14 contributed mainly to the improved thermostability. It will provide an effective strategy for improving the thermostabilities of other enzymes.

Figure 1

Computational prediction of AEXynM by comparison of B-factor values. (A) The B-factor values of amino acid residues of AuXyn11A (dashed line) and EvXyn11A^TS (solid line) were calculated after a 15 ns MD simulation process at a temperature of 300 K. (B) The homology alignment of N-terminal sequences between AEXynM and AuXyn11A. The site of N-terminus replacement is marked with a bold arrow. MD, molecular dynamics.

Figure 2

Calculation and distribution of the RMSD values. (A) The curves of RMSD values of AuXyn11A (dashed line), AEXynM (dotted line), and EvXyn11^TS (solid line), respectively, after MD simulation processes at 500 K for 10 ns. (B) The distributions of RMSD values of AuXyn11A (dashed line) and AEXynM (dotted line), respectively. MD, molecular dynamics; RMSD, root mean square deviation.

Figure 3

SDS-PAGE analysis of the recombinant xylanases. Lane M, standard protein molecular mass markers; lanes 1, 3, 5, 6, and 7, cultured supernatants of P. pastoris GSAu4-2, GSAEM4-8, GSAEM^C5T4-1, GSAEM^P9S4-3, and GSAEM^H14N4-5, respectively; and lanes 2 and 4, purified reAuXyn11A and reAEXynM with apparent molecular masses of 22.8 and 24.7 kDa, respectively.

Figure 4

Temperature optimum and stability of the recombinant xylanases. (A) The temperature optima were measured under the standard assay conditions, but temperatures ranged from 35 to 60°C for reAuXyn11A as well as from 45 to 80°C for reAEXynM and its recombinant mutants. (B) To estimate the temperature stabilities, reAuXyn11A and reAEXynM^C5T were incubated from 35 to 60°C and other recombinant xylanases from 45 to 80°C, respectively, for 1.0 hour. ■, reAuXyn11A; ▲, reAEXynM; □, reAEXynM^C5T; △, reAEXynM^P9S; ○, reAEXynM^H14N.

Figure 5

Derivative melting curves of reAuXyn11A (dashed line) and reAEXynM (solid line). The emission intensity of the fluorescence dye was recorded from 40 to 99°C at an elevated rate of 1°C min^-1.

Figure 6

HPLC analysis of the hydrolytic products released from corncob xylan by reAEXynM at pH 4.6 and 60°C for 3.0 hours. The positions of xylose (X1), xylobiose (X2), xylotriose (X3), xylotetraose (X4), xylopentaose (X5), and xylohexaose (X6) are shown by arrows.

Figure 7

Analysis of the three-dimensional structure of AEXynM. (A) The three-dimensional structure of AEXynM predicted by MODELLER 9.9. Two invariant catalytic residues, Glu⁸⁹ and Glu¹⁸⁰, reside at the center of an active region. The amino acid residues (Cys⁵, Pro⁹, and His¹⁴) mainly responsible for the high thermostability of AEXynM are located in β-strands A1 and B1, respectively. (B) One disulfide bridge (Cys⁵-Cys³²) is illustrated in the locally magnified three-dimensional structure. (C) The residues (Pro⁹, Phe²¹, and Trp²²) represented with spheres compose a hydrophobic interaction cluster. The hydrogen bond between His¹⁴ and Phe¹⁷ is illustrated with a dashed line.