3.1. Production of AnLPMO14g
The total RNA of A. niger was successfully extracted and reverse transcribed into cDNA. The pPIC9k recombinant plasmid containing An14g02670 was successfully constructed by molecular cloning. High-copy positive clones were obtained by electroporation of Pichia pastoris strains, G418 screening and PCR validation. After induction culture medium cultured positive clones for 4-5 days, the results of SDS-PAGE verification were shown in Fig. S1. Negative control strains containing only empty plasmids showed no significant banding after induction, indicating that P. pastoris had no such purpose gene. The corresponding band size of AnLPMO14g was larger than the theoretical value (36073 Da), probably because of the glycosylation (including N-glycosylation and O-glycosylation) of the gene during expression (Bey et al., 2013). The results showed that AnLPMO14g was successfully expressed, and the purified protein was obtained by purification and dialysis culture supernatant.
3.2. Substrate specificity of AnLPMO14g
The purified sample was applied to different substrates, and the results of determining the reducing sugar concentration after the reaction was completed were shown in Fig. 1. The results of the action of AnLPMO14g on different substrates indicated that AnLPMO14g was able to produce a certain amount of reducing sugars on various substrates, but the yield of reducing sugars when it acting on different substrates was different. The most reducing sugar was produced by hydrolysis of Avicel®, followed by straw and xylan; CMC, filter paper and corn cob produced less reducing sugar. It suggested that the structure and composition of the substrate had a great influence on hydrolysis activity of AnLPMO14g. Avicel® (microcrystalline cellulose) is a purified, partially depolymerized cellulose composed of β-D-glucopyranose units joined by β-1,4 glycosidic linkages in a linear arrangement (Rashid et al., 2017); rice straw contains about 30% Cellulose and 20% hemicelluloses and xylan is structurally composed predominantly of a β-D-(1,4)-linked xylopyranosyl residue backbone (Chen et al. 2011; Corradini et al. 2018). This suggested that AnLPMO14g might mainly oxidatively cleave the β-D-1,4 glycosidic bonds and β-D-1,4 xylosidic bonds.
3.3. Product analysis of cellulose catalyzed by AnLPMO14g
LPMOs have different substrate specificities and vary oxidation region selectivity. LPMOs may only oxidize the C1 or C4 carbon in the scissile glycosidic bond, whereas others produce mixtures of C1- and C4- oxidation products(Fig. S2a)(Kjaergaard et al. 2014; Walton et al. 2016). The reaction products of Avicel® by AnLPMO14g were analyzed by MALDI-TOF/TOF and the results were shown in Fig. S2b. In the figure, m/z 689, 851, 1013, 1175, 1337 and 1499 represent sodium addendum of cello-oligosaccharide with degree of polymerization (DP) of 4-9 respectively, which confirmed that AnLPMO14g could be able to degrade cellulose. m/z 849, 1011, 1173 and 1497 represent the sodium admixture of 1,5-δ-lactone formed by the oxidation of cello-oligosaccharide with DP of 5, 6, 7 and 9, respectively. According to the previous report, 1,5-δ-lactone was instability and usually converted into aldonic acid (Vu et al. 2014; Forsberg et al. 2014). Therefore, m/z 867, 1029 and 1191 in the figure could be found and they represent the sodium admixture of aldonic acid formed by 1,5-δ-lactone corresponding to cello-pentose, hexose and heptasaccharide oxidation. m/z 1375 represents the double sodium aldonic acid admixture of cello-oligosaccharide with DP8. There are characteristic peaks of the aldonic acid sodium adjuncts (M+16 and M+38) that combine one and two sodium in this study. The peaks corresponding to M-2 are more likely to be the unstable lactone existed in the form of aldonic acid rather than 4-ketoaldose (Kim et al. 2017). Therefore, C1 position of AnLPMO14g oxidized fractured cellulose substrate can be inferred and its scheme of chemical reaction in the form of C1 oxidation and C4 oxidation in Fig. S2a.
3.4. Synergism activity of AnLPMO14g with cellulase
Avicel®, CMC, straw powder, filter paper and corncob were used as the substrate to conduct synergism activity study and the results were shown in Fig. 2a. It could be seen that the addition of AnLPMO14g significantly increased the yield of reducing sugar compared with the cellulase alone. The yield of reducing sugar obtained by the combination of the two was 92.66% higher than that of the cellulase alone when Avicel® was used as substrate, and 25.7% higher than the sum of AnLPMO14g and cellulase. The DS of AnLPMO14g with cellulase for the degradation of Avicel® was 1.257, indicating that AnLPMO14g and cellulase were not simply superimposed on the degradation of Avicel®. This AA9 family of polysaccharide monooxygenases could further promote the degradation of cellulose substrates by traditional cellulases, which was consistent with the properties of LPMO (Zhang et al. 2019). When they acted together on filter paper, the yield of reducing sugar was 43.5% higher than that of the cellulase alone, but the DS of AnLPMO14g with cellulase on filter paper degradation was up to 1.375. As for CMC, the addition of AnLPMO14g increased the yield of reducing sugar by 7.51% compared to the cellulase alone, which was significantly lower than that of Avicel®. Although Avicel®, filter paper and CMC were all cellulosic substrates, AnLPMO14g had different effects on these substrates, which indicated that the structure of the substrate had a great influence on the synergism activity. The synergism activity of AnLPMO14g with cellulase on filter paper and Avicel® were much higher than that on CMC, which was probably because the presence of carboxymethyl groups in the CMC substrate hindered binding of the cellulase to the substrate.
When AnLPMO14g and cellulase acted together on straw powder, the reducing sugar yield was 141.42% higher than that of cellulase alone, and the DS was 1.322. While the results when they acting on corncob was 55.37% higher than that of cellulase alone, and the DS was 1.24. Therefore, different composition and structure of the lignocellulose substrate significantly influenced the activity of AnLPMO14g and the synergistic activity between AnLPMO14g and cellulase. Similar results were reported when studying on other LPMO, for example, Basotra et al. found that the improved level of the hydrolysis was different when AA9 LPMO from Malbranchea cinnamomea and commercial cellulase CellicCTec2 acted together on different pretreated biomass samples i.e., acid and alkali treated rice straw (AcRS and AlRS) and bagasse (AcBG and AlBG) (Basotra et al. 2019). Using cocktails of cellulase and AA9 LPMO can greatly improve the degradation rate of lignocellulosic biomass substrate.
3.5. Major factors affecting the synergistic activity
All factors were studied with Avicel® as the substrate. Fig. 2b showed the effect of the dosages of AnLPMO14g and cellulase on the yield of reducing sugar. It was worth noting that when the dosage of cellulase was 0 or 0.045 FPU/g, the yield of reducing sugar increased with the dosage of AnLPMO14g increased, but when it reached 3.6 FPU/g, the yield of reducing sugar showed a decreasing trend with the increase of AnLPMO14g. The possible reason was that the two enzymes might compete for the substrate at high enzyme dosage, making the synergy impossible to function properly. This compete can be alleviated by increasing the substrate concentration, but the compete still appeared when the enzyme dosage increased. Similar effects could be seen in Kim’s research which studied the synergy between AA9 enzymes from Chaetomium globosa and cellulase, and high AA9 and cellulase dosages were found to cause the competition for substrates (Kim et al. 2015). To further analyze the synergism activity on cellulose, DS of different dosages of AnLPMO14g and cellulase were calculated in this study and results were shown in Table 1. The synergy degree was greater than 1 when the dosages of AnLPMO14g and cellulase were low, and highest DS of 1.182 was obtained when the dosages of AnLPMO14g and cellulase were 0.9 mg/g and 0.9 FPU/g, respectively. But when the dosages of AnLPMO14g and cellulase increased, DS decreased and most of them were less than 1. The possible reason is the competition between AnLPMO14g and cellulase on the substrate at high enzyme dosages and low substrate concentration. In addition, DS is also influenced by different substrate from Fig. 2a. Therefore, in order to maximize the synergistic effect of LPMO and cellulase in the degradation of cellulose or lignocellulose substrates, it is necessary to control the ratio and the dosages of the cocktail of LPMO and cellulase.
Results of the effect of pH on synergy of AnLPMO14g and cellulase were shown in Fig. 2c. The activity of AnLPMO14g or cellulase alone was highly dependent on pH. Reducing sugar production at pH 3.0 was extremely low in all three systems. When AnLPMO14g acted alone, the reducing sugar production was the highest at pH 4.0; as for cellulase, the highest yield of reducing sugar was obtained at pH 5.0. While the yield of reducing sugar was highest at pH 5.0 when AnLPMO14g and cellulase acted together. And the synergism activity was the highest when pH was 5.0 and the amount of reducing sugar produced by AnLPMO14g and cellulase was 1.99 times of that by cellulase alone.
The effect of metal ions on reducing sugar production and synergy activity were shown in Fig. 2d. For the system of AnLPMO14g acting alone, the yield of reducing sugar was higher when Cu2+ and Al3+ existed, and the promotion of Cu2+ was the most obvious. The existence of K+, Ca2+, Li2+ and Ni2+ inhibited the activity of AnLPMO14g on cellulose, and the inhibitory effect of Ni2+ was the most obvious. As for the system of cellulase acting alone, K+, Ca2+ and Li2+ promoted the cellulase activity while Cu2+ and Al3+ inhibited cellulase activity significantly. For the systems of AnLPMO14g and cellulase acting together, the highest yield of reducing sugar was obtained when no metal ions addition, which was followed by the addition of Cu2+ and Al3+. The addition of other metal ions had no significant effect and Ca2+ would reduce the yield of reducing sugars. As a heavy metal, Cu2+ would cause the inactivation of certain proteins, which might be the reason why it had obvious inhibitory effect on cellulase. However, the addition of AnLPMO14g could greatly promote the activity of cellulase to increase the yield of reducing sugar in the existence of Cu2+. The active center of LPMOs was a flat surface with Cu2+ binding sites. On the one hand, AnLPMO14g could bind part of Cu2+ to reduce the inhibition of Cu2+ on cellulase; on the other hand, AnLPMO14g could also degrade cellulose through cleaving the cellulose glycosidic chain by the oxidation mechanism of divalent metal ions (Li et al. 2012). It could also be seen that Ni2+ had an inhibitory effect on both AnLPMO14g and cellulase. Therefore, the expressed crude enzyme must be purified by strict dialysis processing to remove small molecules such as imidazole and metal ions, thereby eliminating the inhibitory influence of Ni2+.
Studies have shown that LPMO of AA9 family needs electrons provided by external electron donor to reduce Cu(Ⅱ) in the active center to Cu(Ⅰ) when it oxidatively cleaves the glycosidic bond (Hemsworth et al. 2015). According to the way they provide electrons and their molecular structure characteristics, exogenous electron donors can be classified into three types: a number of reducing small molecules, some oxidoreductases with flavin as a prosthetic group and some light-sensitive organic or inorganic substances (Walton et al. 2016). As the reducing small molecules can directly provide electrons with high efficiency and simple application, they were used in this work and their effect on the activity of AnLPMO14g as well as the synergistic activity between AnLPMO14g and cellulase was shown in Fig. 3. All of them could promote the activity of AnLPMO14g and the synergism activity of AnLPMO14g with cellulase. Among them, the promotion effect of ascorbic acid was the highest. Simmons et al. reported similar results that the cleavages of MLG, glucomannan and xyloglucan by LsAA9A were sensitive to reducing agent potential, with ascorbate as reductant yielding much higher amount of product (Simmons et al. 2017). The effect of gallic acid and pyrogallic acid was not significant. The reason might be that the latter two had weak reduction ability, and they were not stable enough to provide the electrons and reducing power required for normal functioning of AnLPMO14g.
3.6 Sequence alignment and structural modeling
To investigate the potential function of AnLPMO14g as the common activity of AA9 family, sequence alignment and homology modeling were performed. From Fig.4a, the sequence alignment of AnLPMO14g exhibited high similarity to those of the referred AA9s with known structures, especially to HjLPMO9A (sequence identity of 58.19 %). In addition, amino acid residues that were reported to be highly conserved in other AA9s (His1, His86, and Tyr175) were also contained in AnLPMO14g (Fig. 4b). These conserved residues are divalent metal-binding residues that are known to be important for enzyme activity (Quinlan et al. 2011; Hansson et al. 2017; Leggio et al. 2018).
Using HjLPMO9A as the template, the predicted structure of AnLPMO14g by homology modeling was revealed to possess an immunoglobulin-like β-sandwich fold which consisted of two β-sheets formed by eight β-strands and the conserved metal-binding sites (HIC1, His86, and Tyr175) were observed on the surface. In addition, the two nitrogen atoms on the N-terminal histidine (2.0 Å and 2.3 Å from copper ion, respectively) and the nitrogen atom on histidine at 86 site (2.1 Å from copper ion) formed the so-called “histidine brace” (Fig. 4c) (Hansson et al., 2017), which was first revealed and named by Quinlan and similar to the particulate methane monooxygenase (Lieberman et al. 2005; Cao et al. 2018). Oxygen atoms of a tyrosine (2.9 Å) and a above water molecule (2.4 Å) occupied the apical positions and completed the immediate coordination sphere (Fig. 4d) (Hansson et al. 2017). additionally, the metal binding motif and the putative cellulose-binding surface of AnLPMO14g were superimposed with those of other AA9 homologs (Fig. 4b). The three metal-binding sites in AnLPMO14g were structurally conserved among AA9 homologs.
Notably, N-terminal histidine that binds to copper is modified by methylation, and histidine methylation has been reported in the PDB, with the code of HIC. Although the phenomenon that N-terminal histidine methylation had been ignored in earlier crystallographic researches (Harris et al. 2010; Karkehabadi et al. 2008), current studies have shown that this modification has little effect on the catalytic properties of enzymes but may help to protect LPMOs from the autocatalytic oxidation damage (Petrovi et al. 2018). furthermore, three regions with high B-factors in the structure of AnLPMO14g were observed, which was named as chain A(P23-P29), chain B(L120-G127) and chain C(Y212-I221). From Fig.4a, the chain A in AnLPMO14g were highly conserved in the structures among AA9 homologs, which may help the enzyme to maintain catalytic activity at high temperature. Interestingly, the current studies found that in all cases the active site residues predominantly occurred in the low B-factor region, while the residues within the binding pocket tend to be present in the high B-factor region, which provides a novel thinking for improving the catalytic efficiency of LPMOs by site-directed mutation (Guo et al. 2012; Alvarezi et al. 2014; Gaspar et al. 2012; Yang et al. 2005).