A C1/C4-Oxidizing AA10 Lytic Polysaccharide Monooxygenase from Paenibacillus xylaniclasticus Strain TW1

Abstract

Lytic polysaccharide monooxygenases (LPMO) are key enzymes for the efficient degradation of lignocellulose biomass with cellulases. A lignocellulose-degradative strain, Paenibacillus xylaniclasticus TW1, has LPMO-encoding PxAA10A gene. Neither the C1/C4-oxidizing selectivity nor the enzyme activity of PxAA10A has ever been characterized. In this study, the C1/C4-oxidizing selectivity of PxAA10A and the boosting effect for cellulose degradation with a cellulase cocktail were investigated. The full-length PxAA10A (rPxAA10A) and the catalytic domain (rPxAA10A-CD) were heterologously expressed in Escherichia coli and purified. To identify the C1/C4-oxidizing selectivity of PxAA10A, cellohexaose was used as a substrate with the use of rPxAA10A-CD, and the products were analyzed by MALDI-TOF/MS. As a result, aldonic acid cellotetraose and cellotetraose, the products from C1-oxidization and C4-oxidization, respectively, were detected. These results indicate that PxAA10A is a C1/C4-oxidizing LPMO. It was also found that the addition of rPxAA10A into a cellulase cocktail enhanced the cellulose-degradation efficiency.

Abbreviations

AA, auxiliary activity; BMC, ball-milled cellulose; CD, catalytic domain; CBH, cellobiohydrolase; GH, glycoside hydrolase; LPMO, lytic polysaccharide monooxygenase.

Microbial cellulolytic enzymes are promising tools for the degradation and saccharification of plant biomass to obtain sustainable energy sources. Lytic polysaccharide monooxygenases (LPMOs), members of the auxiliary activity (AA) families, are oxidoreductases that require a copper ion in the catalytic center and promote the degradation of plant biomass.¹⁾ So far, eight AA families have been identified and their substrate specificities are diverse: cellulose, xylan, chitin and starch.²⁾,³⁾ LPMOs were characterized into three types, depending on the oxidized position: C1-oxidizing LPMOs oxidize the C1 carbon of pyranose and generate aldonic acid; C4-oxidizing LPMOs oxidize the C4 carbon of pyranose and generate 4-keto aldose; C1/C4-oxidizing LPMOs randomly oxidize the C1/C4 carbons of pyranose.⁴⁾ Paenibacillus xylaniclasticus strain TW1 is a Gram-positive and facultative bacterium that has many genes encoding cellulolytic/xylanolytic enzymes.⁵⁾,⁶⁾ P. xylaniclasticus TW1 has an LPMO gene encoding PxAA10A that belongs to the AA10 family. However, both C1/C4 selectivity for the oxidized position and the enzyme activity of PxAA10A remain to be characterized.

The open reading frame of the PxAA10A gene consists of 1,845 nucleotides encoding a protein of 615 amino acids (accession number WP_246027971). It was also predicted that PxAA10A is composed of an AA10 catalytic domain, a carbohydrate-binding module family 5, two FnIII domains, and a CBM3 (Fig. 1a). BlastN homology analysis indicated that the DNA sequence of PxAA10A has 97.4 % identity to an AA10 enzyme from P. curdlanolyticus strain B-6 (KM387441, PcAA10A). SignalP-5.0 predicted that the PxAA10A-encoding protein includes a 39-amino-acid signal peptide at the N-terminus. The amino acid sequence analysis showed that PxAA10A has 100 % identity to PcAA10A from P. curdlanolyticus strain B-6, 68 % identity to the LPMO from P. woosongensis (WP_230873677) and 67 % identity to the LPMO from P. ihumii (WP_074048692). The C1/C4-oxidization type of these other Paenibacillus LPMOs has not been determined.

Fig. 1. Domain structure of PxAA10A and phylogenetic analysis with AA10 enzymes.

　(a) Schematic diagram of PxAA10A and PxAA10A-CD. The numbers of amino acids are indicated. PxAA10A was composed of 160 amino acids AA10 (1-160), 43 amino acids CBM5 (186-228), 2 Fn III domains (242-331 and 336-426), and 80 amino acids CBM3 (434-513). (b) Phylogenetic tree of PxAA10A and other characterized AA10 enzymes. The phylogenetic tree was built using the neighbor-joining method in Mega X. The names from the CAZy database and the accession numbers of each LPMO were described. Clade A includes many chitin-active LPMOs (EC 3.14.99.53) and Clade B includes many cellulose-active LPMOs (EC 3.14.99.54 and EC 3.14.99.56). The arrow points to PxAA10A.

To determine the LPMO activity as well as the C1/C4 selectivity of PxAA10A from P. xylaniclasticus TW1, the recombinant proteins of the full-length PxAA10A (rPxAA10A, 576 amino acids except for 39 amino acids of the signal peptide) and the AA10 catalytic domain (rPxAA10A-CD, 164 amino acids) were obtained as follows. The DNA fragments of PxAA10A (1,731 bp) and PxAA10A-CD (495 bp) were obtained from genomic DNA of P. xylaniclasticus TW1 by PCR using combinations of primers #1 and #2 and of #3 and #4, respectively (Table S1; see J. Appl. Glycosci. Web site). Both DNA fragments were amplified again using primers #5 and #6 and primers #7 and #8, respectively. The linear pMAL-c2 for in vivo cloning was amplified using primers #9 and #10. The DNA fragments of PxAA10A and PxAA10A-CD were inserted into pMAL-c2 by in vivo cloning using Escherichia coli strain ME9806.⁷⁾ Transformed E. coli strain JM109 cells having pMAL-c2-PxAA10A or pMAL-c2-PxAA10A-CD were grown overnight at 37 °C in LB medium supplemented with ampicillin (50 μg/mL). After several hours of incubation, cells were collected by centrifugation at 12,000 rpm and disrupted with sonication in 20 mM Tris-HCl buffer (pH 7.4) including 200 mM NaCl and 100 μM CuCl₂. The recombinant proteins were purified from the cell-free extracts with the pMAL Protein Fusion and Purification System (New England BioLabs Inc., Ipswich, MA, USA) according to the manufacturer's instructions. The molecular sizes of the recombinant proteins with maltose-binding proteins of rPxAA10A and rPxAA10A-CD were estimated by SDS-PAGE to be 104 and 61.1 kDa, respectively, corresponding to the predicted molecular sizes.

To analyze the LPMO activity of PxAA10A as well as the oxidized form, thin-layer chromatography (TLC) and MALDI-TOF/MS analysis were performed. Avicel (1 %), ball-milled cellulose (BMC) (1 %), chitin (1 %), carboxymethyl-chitin (1 %), cellohexaose (1 mg/mL) (Seikagaku Corporation, Tokyo, Japan) and chitooligosaccharides (1 mg/mL) (Tokyo Chemical Industry Co., Ltd., Tokyo, Japan) were used as substrates. Each substrate was incubated with 0.4 µM rPxAA10A-CD and 100 µM ascorbic acid as the electron donor in 50 mM potassium phosphate buffer (pH 7.0) at 37 °C for 3 days. Oxidized products were separated by TLC on a DC-Fertigplatten SIL G-25 plate (Macherey-Nagel GmbH, Dueren, Germany) developed with a butanol-acetic acid-H₂O solvent mixture (1:2:1, vol/vol/vol). The products were visualized by incubation at 130 °C for several minutes after soaking in a solution of methanol (95 %) and sulfuric acid (5 %). The TLC analysis showed that rPxAA10A-CD did not have detectable activity against Avicel, BMC, chitin, carboxymethyl-chitin and chitooligosaccharides (data not shown). When cellohexaose was used as the substrate, the product at the position corresponding to cellotetraose (G4) was clearly generated after the enzyme reaction for 3 days (Fig. 2a). Then, the reaction product at the G4 position was scraped from the TLC plate and mixed with an equal volume of 10 mg/mL α-Cyano-4-hydroxycinnamic acid in 50 % acetonitrile and 0.1 % trifluoroacetic acid for analysis by a 4,800 Plus MALDI-TOF/TOF Analyzer (Applied Biosystems, Foster City, CA, USA). As a result, aldonic acid cellotetraose (705.41 [Glc3Glc1A+Na]⁺) and cellotetraose (689.46 [Glc4+Na]⁺) were detected (Fig. 2b). These products would be aldonic acid cellotetraose from the C1-oxidization and cellotetraose from the C4-oxidization. Aldonic acid cellotetraose is an oxidized oligosaccharide whose the glucose residue at the reducing end of cellotetraose is oxidized to glucuronic acid (Fig. 2c). Generally, LPMOs use reduced copper and external electrons to oxidize and hydroxylate the C1- or C4-carbons. The oxidized C1- or C4-carbon becomes unstable and the glycosyl bond is cleaved. Then, the oxidized ends are converted to either aldonic acid (C1-oxidized product) or gemdiol (C4-oxidized product) by hydroxylation.⁸⁾ In the reaction with PxAA10A, aldonic acid cellotetraose as well as cellobiose were produced by the C1-oxidization. Whereas, cellotetraose as well as gemdiol were produced by the C4-oxidization (Fig. 2c). These results suggest that rPxAA10A recognized cellohexaose and cleaved it into cellotetraose and cellobiose. Most bacterial LPMOs have been classified into C1 or C1/4-oxidizing types. C4-oxidizing LPMO from bacteria has yet to be discovered.⁴⁾

Fig. 2. Enzyme assay of rPxAA10A and rPxAA10A-CD.

　(a) TLC of hydrolysis products of rPxAA10A-CD released from cellohexaose. Each reaction mixture was incubated in 50 mM potassium phosphate buffer (pH 7.0) at 37 °C for 3 days. Lane M, standard sugar (G1: glucose, G2: cellobiose, G3: cellotriose, G4: cellotetraose, G5: cellopentaose, G6: cellohexaose). No enzyme, the reaction without PxAA10A-CD. The black arrow points to the reaction products corresponding to the G4 (cellotetraose) position. Maltose, observed at the G2 position, was included in the enzyme solution for protein purification (indicated by the white arrow). (b) Mass spectra of aldonic acid cellotetraose and cellotetraose. Each products was identified by major peak of [M+Na]⁺ adducts. Glc4 represents cellotetraose from a C4-oxidizing reaction, and Glc3Glc1A represents aldonic acid cellotetraose. (c) Schematic diagram of the products from C1- and C4-oxidization. Cellohexaose was separated into cellotetraose (Glc4) and gemidiol on C4-oxidization, aldonic acid cellotetraose (Glc3Glc1A) and cellobiose on C1-oxidization. In this study, cellotetraose (Glc4) and aldonic acid cellotetraose (Glc3Glc1A) were detected by MALDI-TOF/MS analysis. (d) Synergy effects between Meicelase and rPxAA10A or rPxAA10A-CD. The concentration of rPxAA10A and that of rPxAA10A-CD were 0.4 µM, and Meicelase was added at 0.4 mg/mL. Each reaction mixture was incubated in 50 mM potassium phosphate buffer (pH 7.0) at 37 °C for 3 days. After the enzymatic reactions, reducing sugars were quantified using the 3,5-dinitrosalicylic acid method measured by the absorbance at 570 nm. The synergy effect was calculated by dividing the value of the simultaneous reaction by the sum of the values of LPMO and Meicelase. Values are the means of triplicate experiments. ±, standard deviation. An asterisk indicates a significant difference (p < 0.01). ND, not detected.

Phylogenetic analysis with 32 AA10 enzymes, whose oxidization forms have been determined, was carried out (Fig. 1b). The phylogenetic tree indicated that they were divided into two groups: Clade A and Clade B. Most LPMOs in clade A are chitin-active (EC 1.14.99.53), and most LPMOs in clade B are cellulose-active (EC 1.14.99.54 and EC 1.14.99.56). Some LPMOs, however, are active on both chitin and cellulose. An LPMO (BlAA10A) in Clade A (the chitin-active group) was active against cellulose.⁹⁾ PcAA10A was also active against chitin and cellulose.¹⁰⁾ On the other hand, PxAA10A was active against cellulose but not against chitin (data not shown). Given that the amino acid sequence of PxAA10A is homologous to that of PcAA10A, the substrate specificity of PxAA10A was presumed to be the same as that of PcAA10A. One possibility is that the maltose-binding protein added to the N-terminus of rPxAA10A may affect this enzyme activity. However, some chitin-active LPMOs are active against cellulose, and it is plausible that PxAA10A is active against cellulose.

It is known that some bacterial LPMOs boosted cellulosic substrate degradation.¹¹⁾ For efficient degradation of cellulosic biomass, the boosting ability of LPMOs is required. Therefore, the synergy effects between Meicelase (Meiji Seika Pharma Co., Ltd., Tokyo, Japan) and rPxAA10A or rPxAA10A-CD were examined by incubation with 1 % Avicel as the substrate and 100 µM ascorbic acid as the electron donor. Meicelase is an enzyme cocktail containing a cellobiohydrolase (CBH) from Trichoderma viride.¹²⁾ In the reaction with rPxAA10A solely, reducing sugars were not detected. When either rPxAA10A or rPxAA10A-CD was added, the amount of reducing sugar released by Meicelase increased by about 10 %, compared to that of the reaction by Meicelase solely (Fig. 2d). The effect for rPxAA10A was higher than that for rPxAA10A-CD. rPxAA10A has a CBM3 that binds to cellulose, suggesting that CBM3 may enhance the synergy effect. However, few differences in reducing sugar production were detected between rPxAA10A-CD and rPxAA10A. Also, it was not evaluated how CBM3 affected the enzyme stability as well as the optimal pH of PxAA10A. In addition, a CBM5 and Fn III domains of PxAA10A may also affect the LPMO activity. Therefore, further analysis is required to clarify that CBM3 enhances the enzyme activity of PxAA10A against cellulosic substrate. This study indicated that PxAA10A boosted the cellulose degradation by using it in combination with a commercially available cellulase cocktail, Meicelase. PxAA10A may be utilized for cellulosic biomass degradation.

CONFLICTS OF INTEREST

The authors declare that they have no conflict of interest.