GH30-7 Endoxylanase C from the Filamentous Fungus Talaromyces cellulolyticus

Abstract

Glycoside hydrolase family 30 subfamily 7 (GH30-7) enzymes include various types of xylanases, such as glucuronoxylanase, endoxylanase, xylobiohydrolase, and reducing-end xylose-releasing exoxylanase. Here, we characterized the mode of action and gene expression of the GH30-7 endoxylanase from the cellulolytic fungus Talaromyces cellulolyticus (TcXyn30C). TcXyn30C has a modular structure consisting of a GH30-7 catalytic domain and a C-terminal cellulose binding module 1, whose cellulose-binding ability has been confirmed. Sequence alignment of GH30-7 xylanases exhibited that TcXyn30C has a conserved Phe residue at the position corresponding to a conserved Arg residue in GH30-7 glucuronoxylanases, which is required for the recognition of the 4-O-methyl-α-d-glucuronic acid (MeGlcA) substituent. TcXyn30C degraded both glucuronoxylan and arabinoxylan with similar kinetic constants and mainly produced linear xylooligosaccharides (XOSs) with 2 to 3 degrees of polymerization, in an endo manner. Notably, the hydrolysis of glucuronoxylan caused an accumulation of 2²-(MeGlcA)-xylobiose (U^4m2X). The production of this acidic XOS is likely to proceed via multistep reactions by putative glucuronoxylanase activity that produces 2²-(MeGlcA)-XOSs (X _n U^4m2X, n ≥ 0) in the initial stages of the hydrolysis and by specific release of U^4m2X from a mixture containing X _n U^4m2X. Our results suggest that the unique endoxylanase activity of TcXyn30C may be applicable to the production of linear and acidic XOSs. The gene xyn30C was located adjacent to the putative GH62 arabinofuranosidase gene (abf62C) in the T. cellulolyticus genome. The expression of both genes was induced by cellulose. The results suggest that TcXyn30C may be involved in xylan removal in the hydrolysis of lignocellulose by the T. cellulolyticus cellulolytic system.IMPORTANCE Xylooligosaccharides (XOSs), which are composed of xylose units with a β-1,4 linkage, have recently gained interest as prebiotics in the food and feed industry. Apart from linear XOSs, branched XOSs decorated with a substituent such as methyl glucuronic acid and arabinose also have potential applications. Endoxylanase is a promising tool in producing XOSs from xylan. The structural variety of XOSs generated depends on the substrate specificity of the enzyme as well as the distribution of the substituents in xylan. Thus, the exploration of endoxylanases with novel specificities is expected to be useful in the provision of a series of XOSs. In this study, the endoxylanase TcXyn30C from Talaromyces cellulolyticus was characterized as a unique glycoside hydrolase belonging to the family GH30-7, which specifically releases 2²-(4-O-methyl-α-d-glucuronosyl)-xylobiose from hardwood xylan. This study provides new insights into the production of linear and branched XOSs by GH30-7 endoxylanase.

Keywords: Talaromyces cellulolyticus; endoxylanase; glycoside hydrolase family 30; lignocellulose; xylan; xylooligosaccharide.

FIG 1

(A) Schematic illustration of the domain structure of the TcXyn30C protein. SP, signal peptide; CD, catalytic domain. (B) SDS-PAGE analysis of purified TcXyn30C protein. Lane 1, molecular mass standards; lane 2, purified TcXyn30C (10 μg protein). The solid black arrow indicates the position of TcXyn30C.

FIG 2

Multiple sequence alignment of GH30-7 and GH30-8 xylanases. Primary structures of TcXyn30C, TcXyn30A, and TcXyn30B from T. cellulolyticus, XYLD from Bispora sp. strain MEY-1, XYN VI from T. reesei, PpXynC from P. purpurogenum, CaXyn30A-CD from C. acetobutylicum, and BsXynC from B. subtilis were used for sequence alignment. The features shown are as follows: predicted signal sequence of TcXyn30C (highlighted in black), Arg residue conserved in GH30-7 glucuronoxylanases and endoxylanases (highlighted in green), the β2-α2 loop possibly contributing to xylobiohydrolase activity (blue characters), catalytic Glu residues (highlighted in gray), Cys-pair (boxed by a red line), α6-helix in GH30-8 xylanases (blue highlight box), Gln residue conserved in GH30-7 Rex (boxed by a black line), β8A and β8B strands (yellow arrows), the linker region between the catalytic domain and CBM1 domain of TcXyn30C (bold characters), CBM1 domain (highlighted in red), and C-terminal position of TcXyn30C-ΔCBM1 (an arrow). The amino acid sequence identity with TcXyn30C is shown at the end of each primary structure.

FIG 3

Time-course analysis of the hydrolysis of beech wood glucuronoxylan and wheat arabinoxylan by TcXyn30C. (A) Linear XOSs released from beech wood glucuronoxylan. (B) Linear XOSs released from wheat arabinoxylan. (C) Linear XOSs released from wheat arabinoxylan by the combination of TcXyn30C and TcAbf62A. (A to C) Hydrolysis was performed at 45°C in a mixture (pH 4.0) consisting of 10 mg ml⁻¹ xylan substrate and 20 μg ml⁻¹ TcXyn30C alone (A and B) or a mixture of 20 μg ml⁻¹ TcXyn30C and 20 μg ml⁻¹ TcAbf62A (C).

FIG 4

HPAEC-PAD profiles of the hydrolysate of beech wood glucuronoxylan and wheat arabinoxylan produced by TcXyn30C. Hydrolysis was performed at 45°C using a mixture (pH 4.0) consisting of 10 mg ml⁻¹ xylan substrate and 20 μg ml⁻¹ TcXyn30C. (A) Profiles of the hydrolysate from beech wood glucuronoxylan. (B) Profiles of the hydrolysate from wheat arabinoxylan. Peak A, arabino-XOS possessing the single-substituted l-arabinofuranose residue; peaks B and C, arabino-XOS possessing the double-substituted l-arabinofuranose residues (see Fig. S4 in the supplemental material). nC, nano coulomb.

FIG 5

(A) ESI(−) single-stage mass spectrum of a beech wood glucuronoxylan hydrolysate from TcXyn30C after 24 h of reaction. (B) ESI(−) tandem mass spectrum of the main acidic XOS X₂MeGlcA at m/z 471 with the pairs of diagnostic product ions highlighted using colored circles. R. Int., relative intensity. Inset shows the deduced structure.

FIG 6

Comparison of HPAEC-PAD profiles of acidic XOSs generated by TcXyn30B and TcXyn30C. The hydrolysis reaction using TcXyn30B was performed at 40°C for 1 h in a mixture (pH 4.0) consisting of 10 μg ml⁻¹ TcXyn30B and 10 mg ml⁻¹ beech wood glucuronoxylan. The reaction using TcXyn30C was performed at 45°C for 15 min in a mixture (pH 4.0) containing 5 μg ml⁻¹ TcXyn30C and 10 mg ml⁻¹ beech wood glucuronoxylan. The hydrolysate obtained by TcXyn30C was analyzed at twice the amount of that obtained by TcXyn30B and compared in the chromatogram.

FIG 7

Time-course analysis of the hydrolysis of the X_nU^4m2X (n = 0 to 12) mixture by TcXyn30C. Hydrolysis was performed at 45°C using a mixture (pH 4.0) consisting of 1 μg ml⁻¹ TcXyn30C and 9.5 mg ml⁻¹ X_nU^4m2X. (A) HPAEC-PAD profiles. (B) Linear XOSs and U^4m2X released by the hydrolysis. X₁–X₁₀ indicates the sum of the concentrations of linear XOSs (DP, 2 to 10) and xylose. The concentrations of X₇, X₈, X₉, and X₁₀ were estimated from the peak area of HPAEC-PAD profiles based on a calibration curve obtained using an X₆ standard.

FIG 8

(A) The locus of xyn30C and abf62C in the T. cellulolyticus genome. The accession numbers of the scaffold and each gene are shown in parentheses. (B) Expression of xyn30C and abf62C genes. T. cellulolyticus Y-94 was cultured in medium containing 50 g liter⁻¹ cellulose for 24, 72, and 120 h. Expression is shown relative to that of gpdA, which was an internal control. Gray bars, xyn30C; white bars, abf62C. The data presented are the means of the results from three independent experiments.

FIG 9

Molecular phylogenetic analysis of the amino acid sequence of GH30-7 and GH30-8 enzymes using the neighbor-joining method (47). In addition to the known GH30-7 and GH30-8 xylanases (indicated by asterisks), amino acid sequences of top 20 E values collected using BLASTp with the amino acid sequence of TcXyn30C were used for the analysis. Enzymes were divided into the Arg-conserved group that possesses a conserved residue corresponding to Arg-46 of TcXyn30B, the Phe-conserved group having a conserved Phe at the Arg-46 position, the ReX group, and the GH30-8 group. The optimal tree with the sum of branch length of 6.29216595 is shown. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were computed using the Poisson correction method (48) and are in the units of the number of amino acid substitutions per site. All positions containing gaps and missing data were eliminated. There were a total of 349 positions in the final data set. Evolutionary analyses were conducted in MEGA7 (49).

FIG 10

Suggested multistep reaction of TcXyn30C against glucuronoxylan. (i) Production of X_nU^4m2X mixture from glucuronoxylan. (ii) Hydrolysis of the X_nU^4m2X mixture to U^4m2X and linear XOSs. (iii) Hydrolysis of linear XOSs to shorter XOSs, mainly into X₂ and X₃.