Comparative secretome of white‐rot fungi reveals co‐regulated carbohydrate‐active enzymes associated with selective ligninolysis of ramie stalks

P. eryngii and P. ostreatus protein expression patterns exhibit increased oxidoreductase potential and diminished cellulolytic capability. Co‐regulated carbohydrate‐active enzymes were associated with selective ligninolysis of ramie stalks.


Introduction
Eight million tons of agricultural by-products rich in lignocellulose are produced in China every year (Venanzi et al., 2018). These by-products can result in environmental pollution and a serious waste of resources for their disposal (Fasolato et al., 2016). At present, methods for comprehensive utilization of these lignocellulosic stalks have been developed, which have greatly improved the degree of utilization of stalks (Kim, 2018). One example of such improved utilization is the use of stalks as feed stocks for producing ethanol (Jiang et al., 2017). However, the lignin content of native lignocellulosic biomass is highly resistant to enzymatic hydrolysis, resulting in low cellulose conversion. Therefore, lignin must be selectively degraded in an effective pretreatment process to improve enzymatic saccharification (Guo et al., 2018). To date, there has been extensive research into the best pretreatment methods, including physical, chemical or biological degradation of lignin (Kim, 2018). Biological pretreatment has garnered considerable attention in recent years because it is a mild, safe and environmental-friendly way to remove lignin from lignocellulosic materials (Paudel et al., 2017).
Although white-rot fungi are the most promising microorganisms used for lignin degradation, they differ in their gross morphological patterns of decay (da Silva et al., 2017). Phanerochaete chrysosporium simultaneously degrades cellulose, hemicellulose and lignin, whereas other typical white-rot fungi, such as Pleurotus ostreatus, Coriolus versicolor and Ceriporiopsis subvermispora, degrade lignin before degrading cellulose (Fernandez-Fueyo et al., 2012). Pretreatment of wheat straw with C. subvermispora showed that after 35 days, the degradation ratio of lignin in wheat straw was 55% and the conversion efficiency of cellulose was 66.61% (Cianchetta et al., 2014). In addition, after pretreatment of rice straw by P. ostreatus for 60 days, the lignin degradation ratio was 52%, while 83% of cellulose was retained, indicating that P. ostreatus also had a selective ability to degrade lignin (Taniguchi et al., 2005). To investigate the mechanistic basis for such selective ligninolysis, Fernandez-Fueyo et al. conducted a comparative transcriptomic analysis of C. subvermispora and P. chrysosporium (Fernandez-Fueyo et al., 2012). Several genes, including those encoding manganese peroxidase (MnP) and laccases, were found to be differentially expressed between the C.subvermispora and P. chrysosporium transcriptomes. However, that study only compared transcriptomic differences between two species of white-rot fungi. With the completion of the genome sequences of a variety of white-rot fungi and advances in quantitative proteomics technologies, analyses of the expression, mechanisms and regulation of enzymes involved in the lignin degradation pathways from a systems biology point of view will be emphasized in the near future (Kameshwar and Qin, 2016).
Ramie (Boehmeria nivea L. Gaud) is an important natural fiber crop that is widely planted in China, India and other Southeast Asian and Pacific Rim countries (Luan et al., 2018). Large quantities of ramie fibres are extracted from ramie stalks every year (Meng et al., 2018). The massive ramie stalks are a promising renewable source of lignocellulose and can be used as animal feed, fertilizer, fuel ethanol and a substrate for mushroom cultivation (Xie et al., 2017). In our previous research, we used ramie stalks to cultivate Pleurotus eryngii and reached a biological conversion efficiency of 71% (Xie et al., 2016a). In addition, we obtained a global profile of the secretome of P. eryngii cultivated in ramie stalk medium. Diverse enzymes, including cellulases, hemicellulases, pectinase, ligninase, proteases, peptidases and phosphatases, implicated in lignocellulose degradation were detected (Xie et al., 2016a).
The lignin content of ramie stalks is approximately 18.1% by weight and directly affects enzymatic saccharification during the process of ethanol production (Xie et al., 2016b). Therefore, it has become necessary to screen for white-rot fungi that can carry out selective ligninolysis to use for biological pretreatment. Thus, we screened for fungal species that could selectively degrade lignin from ramie stalks. Further, to better understand the molecular mechanism of this selective ligninolysis of ramie stalks at the level of protein expression, iTRAQ labelling combined with liquid chromatography/tandem mass spectrometry (LC-MS/MS) was used to investigate the differential expression of lignin degradation-related proteins between P. eryngii and P. ostreatus, which selectively degrade lignin, and P. chrysosporium and I. lacteus, which non-selectively degrade lignin. A proposed molecular mechanism of selective degradation of lignin by P. eryngii (CICC50126) and P. ostreatus (bio-67015) was also discussed.

Results and discussion
Lignocellulose degradation of ramie stalks by P. chrysosporium, I. lacteus, P. eryngii or P. ostreatus Lignocellulose, including cellulose, hemicellulose and lignin, is the most abundant renewable biomass in the world. Bioconversion of lignocellulosic biomass plays important roles in exploiting value added products (Zhu et al., 2017). Lignin makes lignocellulose recalcitrant to degradation because of its highly irregular and heterogeneous biopolymer (Schutyser et al., 2018). Currently, white-rot fungi are considered as the most important lignin-degrading organisms (Knop et al., 2015). Degradation of lignin by white-rot fungi can occur in two ways: (i) fungi secrete enzymes that selectively remove lignin polymers and leave cellulose intact; or (ii) fungi degrade lignin and cellulose simultaneously (van Kuijk et al., 2015).
Degradation of lignin is the limiting step for enzymatic hydrolysis of cellulose in ramie stalks. The lignin-selective degradation index can be measured by determining lignin degradation with simultaneous conservation of cellulose. In the present research, ramie stalks were treated with P. chrysosporium, I. lacteus, P. eryngii and P. ostreatus for 7, 14, 21 or 28 days. Figure 1 showed that degradation of cellulose in ramie stalks treated with P. eryngii and P. ostreatus is lower than that with I. lacteus or P. chrysosporium, while the lignin degradation caused by P. eryngii is greater than that caused by other fungi. The lignin-selective degradation indices for P. chrysosporium, I. lacteus, P. eryngii and P. ostreatus at 21 days were 0.66, 0.89, 3.86 and 2.22, respectively. These results indicated that P. eryngii and P. ostreatus had more selective ability to degrade lignin in ramie stalks, while P. chrysosporium and I. lacteus degrade large amounts of cellulose, hemicellulose and lignin due to their non-selective pattern of ligninolysis. Previous study reported similar degradative capability of P. chrysosporium and P. ostreatus. P. chrysosporium was found to cause higher amount of lignin and carbohydrate fraction loss. P. ostreatus have shown lower cellulose and hemicellulose loss, while lignin removal reached 40% (Taniguchi et al., 2005;Shrivastava et al., 2011).
Enzymatic hydrolysis of ramie stalks treated with P. chrysosporium, I. lacteus, P. eryngii or P. ostreatus Lignin is a complex biopolymer which is considered as a main barrier against enzymatic degradation. The unproductive binding of cellulase and lignin is considered to To evaluate the effect of selective ligninolysis on the subsequent enzymatic hydrolysis of cellulose, the reducing sugar yield of ramie stalks pretreated with P. chrysosporium, I. lacteus, P. eryngii or P. ostreatus was determined after 48-h hydrolysis with cellulase, using untreated ramie stalks as control. Figure 2 showed that the yield of reducing sugars of P. eryngii was 388 mg g −1 after addition of 40 FPU g −1 cellulase when treated for 21 days, which was significantly higher than that of I. lacteus (179 mg g −1 ) or P. chrysosporium (159 mg g −1 ) but was not statistically significant different from that of P. ostreatus (373 mg g −1 ). As expected, the control ramie stalks that were not pretreated had a very low hydrolysis ratio (20.5 mg g −1 ). Previous studies reported the release of 330 mg g −1 of sugar from rice straw fermented with P. ostreatus and release of 402 mg g −1 sugar from Prosopis Juliflora fermented with Pycnoporus cinnabarinus. These results indicated that pretreatment of substrates with white-rot fungi can improve enzymatic saccharification compared with those untreated substrates (Consolacion et al., 2016). Our study showed that P. ostreatus and P. eryngii were suitable for pretreatment of ramie stalks prior to hydrolysis by cellulases due to its selective ligninolysis ability. This may be attributed to the increase of the effective surface area, the decrease of polymerization degree, the separation of lignin carbohydrate bond and the modification of lignin, so as to facilitate the accessibility of enzymatic hydrolysis substrate (Juan and Estrella, 2019).
Enzyme activities of P. chrysosporium, I. lacteus, P. eryngii and P. ostreatus during lignin degradation in ramie stalks The difference in selective ligninolysis by P. chrysosporium, I. lacteus, P. eryngii and P. ostreatus could be attributed to the different enzymatic activities observed during growth on ramie stalks. In this study, carboxymethyl cellulose enzyme (CMCase), MnP and laccase activities were detected during lignin degradation of ramie stalks by P. chrysosporium, I. lacteus, P. eryngii and P. ostreatus. Figure 3 showed that the CMCase activity of P. chrysosporium was significantly higher than that of the other three fungi. Peak CMCase activities of P. chrysosporium and I. Fig. 1. The ratios of weight loss, conservation of cellulose, lignin degradation and lignin-selective degradation index after treatment of P. chrysosporium, I. lacteus, P. eryngii and P. ostreatus in 7, 14, 21 and 28 days. From all treatments at 21 or 28 days, the weight loss and ligninselective degradation index of P. chrysosporium had significant difference with P. eryngii and P. ostreatus, while conservation of cellulose and lignin degradation ratios of P. eryngii and P. ostreatus were significantly higher than I. lacteus and P. chrysosporium (P < 0.01). lacteus were 0.384 and 0.239 IU ml −1 , respectively, at 28 days, while the CMCase activities of P.ostreatus and P. eryngii were lower than 0.1 IU ml −1 at that time point. Not surprisingly, ramie stalks pretreated with P. chrysosporium exhibited considerable cellulose degradation. The peak MnP activity value of P. chrysosporium (19.76 U l −1 ) was appeared in at 28 days, while the peak MnP activities of P. eryngii (55.32 U l −1 ) and P. ostreatus (41.37 U l −1 ) were detected at 21 days. Figure 3 showed that laccase activity of P. ostreatus and P. eryngii also peaked at 28 days, while P. chrysosporium exhibited no laccase activity. Interestingly, laccase activity of P. eryngii remained high during the entire incubation period (694 U l −1 at 7 days, 522 U l −1 at 14 days, 630 U l −1 at 21 days and 749 U l −1 at 28 days) as compared with the other three fungal species tested here. Thus, the lower cellulase activities and higher laccase activities resulted in greater selective degradation of lignin and retention of cellulose upon treatment of ramie stalks with P. ostreatus or P. eryngii.
Global analysis of the P. chrysosporium, I. lacteus, P. eryngii and P. ostreatus secretomes The difference in the degradation pattern of substrate by these four white-rot fungi could be attributed to the different enzyme profile observed during growth on ramie stalks. In order to understand the molecular mechanism of selective ligninolysis of ramie stalks, the secretomes of P. chrysosporium, I. lacteus, P. eryngii and P. ostreatus were analysed by iTRAQ labelling combined with LC-MS/MS. As showed in Fig. 1D, the selective degradation index of lignin tends to be stable at 21 days for all of the four white-rot fungi. So the diversity and relative abundance of the secreted proteins at 21 days of growth on ramie stalk medium were compared. A total of 222, 249, 269 and 260 extracellular proteins were identified in the P. chrysosporium, I. lacteus, P. eryngii and P. ostreatus secretomes, respectively (Tables S1-S5). As shown in Fig. 4, 43, 70, 90 and 81 proteins were only detected in the secretomes of P. chrysosporium, I. lacteus, P. eryngii and P. ostreatus, respectively. Interestingly, 179 proteins were expressed by all four of these fungal species, revealing a similar effect of ramie stalks on the secretion of proteins by each species. In addition, 20% of the expressed proteins were varied in the secretomes of these four fungi. This reveals that the types of extracellular enzymes produced by white-rot fungi were mainly influenced by the culture medium and also were related to the unique strains. The proteins identified in the secretomes of P. chrysosporium, I. lacteus, P. eryngii and P. ostreatus grown on ramie stalk medium were functionally classified as carbohydrate-active enzymes (CAZymes), sugar transporters, intracellular metabolism proteins, cytochrome P450s, proteases, phosphatases and proteins with other functions. The overall proteomic profiles of the P. eryngii and P. ostreatus secretomes were similar (Tables S4 and S5). The highest diversity of CAZyme proteins was found in the P. eryngii and P. ostreatus secretomes, while sugar transporter, proteases and phosphatases were similarly abundant among these four fungi (Tables S1-S5).
Differentially expressed CAZyme proteins in the P. chrysosporium, I. lacteus, P. eryngii and P. ostreatus secretomes An iTRAQ-labelling quantitative proteomics approach was used to analyse the differentially expressed proteins in the secretomes of P. chrysosporium, I. lacteus, P. The reducing sugar yield of four white-rot fungi treated ramie stalks after 48-h enzymatic hydrolysis with cellulase. The yield of reducing sugars of P. eryngii has no statistically significant difference with P. ostreatus, while was significantly higher than I. lacteus and P. chrysosporium (P < 0.01). eryngii and P. ostreatus. Our results showed special emphasis on CAZymes in the secretomes of these white-rot fungi. In this study, by querying differentially expressed proteins against the Carbohydrate-Active Enzyme database (CAZy), 58 proteins were identified as CAZymes. Among these CAZymes, 19 glycoside hydrolases (GHs), 10 glucosyl transferases (GTs), 25 auxiliary activities (AAs) and 4 carbohydrate-binding modules (CBMs) were identified, while no carbohydrate esterases (CEs) or polysaccharide lyases (PLs) were found (Table 1). Closer inspection of particular CAZyme families revealed clear differences in the number and distribution of CAZymes among these four white-rot fungal species (Fig. 5). For instance, GH17, GH95, CBM1 and GT90 were absent from the P. eryngii and P. ostreatus secretomes, while laccases were absent from the P. chrysosporium secretome. To highlight comparisons of the cellulolytic system, compared with the P. chrysosporium and I. lacteus secretomes, Table 1 showed that GH2, GH3, GH7, GH20, GH31, GH63, CBM21, GT1, GT2, GT24, GT39, cellobiose dehydrogenase and betagalactosidase were less strongly expressed in P. eryngii and P. ostreatus secretomes. Based on these results, it was suggested that P. chrysosporium and I. lacteus degrade cellulose, hemicellulose and pectins using a large machinery of GHs. Due to the specific hydrolytic action of these enzymes, large polysaccharide fragments of ramie stalks were mostly released and lost when treated by P. chrysosporium and I. lacteus.
Lignin has a highly branched three-dimensional polyphenolic structure and can ultimately be converted into CO 2 by microorganisms to maintain the carbon balance in nature (Houtman et al., 2018). Oxidoreductases play an important part in the biological degradation of lignin by microorganisms (Janusz et al., 2017;Martinez et al., 2017).Classical fungal oxidoreductases comprise ligninolytic peroxidases and multicopper oxidases, mainly laccases with different redox potentials, that can act on Fig. 3. CMCase, MnP and laccase activities of P. chrysosporium and I. lacteus had significant difference with P. eryngii and P. ostreatus during ramie stalks lignin degradation process at 7, 14 or 21 days (P < 0.01). lignin-derived products (Kersten and Cullen, 2014). The ligninolytic system of white-rot fungi includes extracellular lignin peroxidase, MnP, alkyl-aryl etherase and laccase. In addition, heme peroxidases, glucose oxidase, isoamyl alcohol oxidase, glutathione reductase, glutathione Stransferase, copper-radical oxidase, cellobiosedehydrogenase (CDH) and glucose/methanol/choline oxidase/dehydrogenase were also found to provide the hydrogen peroxide required by ligninolytic peroxidases. Here, Table 1 showed that a total of 25 lignin-degrading enzymes providing auxiliary oxidoreductases were identified in the secretomes of P. chrysosporium, I. lacteus, P. eryngii and P. ostreatus. Among of these proteins, 8 laccases, 5 manganese peroxidases, 2 peroxidase, 1 diphenol oxidase, 3 glyoxal oxidases, 4 glucosemethanol-choline(GMC) oxidoreductases, 1 glyoxylate dehydrogenase and 1 phenol oxidase were identified as differentially expressed proteins. As expected, laccases were absent in the secretome of P. chrysosporium. MnP was the most highly expressed auxiliary oxidoreductases in the secretomes of P. eryngii and P. ostreatus, while it was less highly expressed in the secretomes of P. chrysosporium and I. lacteus. As shown in Table 1, compared with P. chrysosporium and I. lacteus, glyoxal oxidase, GMC oxidoreductase, laccase, MnP, glyoxylate dehydrogenase and phenol oxidase were more highly expressed in the P. eryngii and P. ostreatus secretomes cultivated on ramie stalk medium. Although these ligninases were identified, at present, the complete degradation of natural lignin by a single enzyme cannot be realized. Enzymes related to the degradation of lignin can be divided into two types: (i) enzymes directly acting on lignin: including laccase, other related polyphenol oxidase, peroxidase and cellobiose dehydrogenase. (ii) Auxiliary enzymes collaboratively degrade lignin with enzymes directly acting on lignin: including aryl alcohol oxidase, vanillin alcohol oxidase and glyoxal oxidase (Anthony et al., 2013). Previous research has shown that a single laccase or manganese peroxidase does not have ideal activity for lignin degradation and that two enzymes can improve the lignin degradation ratio and that their synergistic coefficient can reach 1.464 (Janusz et al., 2013). Cellobiose dehydrogenase contains a flavin and a heme group. It can oxidize lignin by generating hydroxyl groups in Fenton-type reaction in the presence of H 2 O 2 and chelated iron ions. It was proposed to act on the reduction of quinones, which can be used by ligninolytic enzymes or in support of a MnP reaction. Glyoxal oxidases are copper-radical oxidases with a broad specificity able to oxidize simple aldehydes to the corresponding carboxylic acids (Whittaker et al., 1996;Zamocky et al., 2006). However, the synergistic potentials of other sets of lignin-degrading enzymes are still unclear. As we described above, seven ligninases were expressed at the same time in these secretomes, and although there might be synergistic interactions between these proteins, more experiments are needed to examine this question.
Differentially expressed sugar transporters, cytochrome P450 monooxygenases and other proteins in the P. chrysosporium, I. lacteus, P. eryngii and P. ostreatus secretomes In addition to secreting extracellular enzymes to degrade lignocellulose, the ability to metabolize wood polysaccharides depends on the transport and intracellular conversion of sugars. ABC transporters, major facilitator superfamily (MFS) transporters, phosphonate ABC transporter substrate-binding proteins and metal ABC transporter permeases were also identified in the P. chrysosporium, I. lacteus,P. eryngii and P. ostreatus secretomes in this study (Table 2). But more interestingly, the ABC transporter and the MFS ransporter were more highly expressed in the P. chrysosporium and I. lacteus secretomes. This group comprises cellobiose transporters, supporting that the transporter proteins identified in the P. chrysosporium and I. lacteus secretomes could have a role in cellobiose and/or cellodextrin uptake. White-rot fungi encode a large repertoire of P450 genes, which are known to be involved in the oxidation of phenolic and non-phenolic aromatic compounds. The current proteomic analysis revealed that 7 cytochrome P450 monooxygenases were highly expressed in the P. eryngii and P. ostreatus secretomes. In addition, peptidases, proteinases and kinases were also quantified (Table 3). Carboxypeptidase and polyphosphate kinase were highly expressed in the P. eryngii and P. ostreatus secretomes. The secretion of proteases during lignocellulose degradation has previously been reported, which suggests that the proteases detected in cellulolytic cultures are correlated to the activation of cellulase activity and the cleavage of CDH functional domains (Baldrian and Valaskova, 2008). But the roles of peptidases, proteases and kinases in lignocellulose degradation are still not completely clear.
The current work describes for the first time the composition of the secretome of P. chrysosporium, I. lacteus, P. eryngii and P. ostreatus growing on ramie stalks. It was suggested that the knowledge on the different expressions of CAZyme was sufficient to discern the mechanisms implicated in selective ligninolysis of ramie stalks. Consistent with secretome analysis, cellulase activity was higher in P. chrysosporium and I. lacteus, while lignin-degrading enzymes were higher in P. eryngii and P. ostreatus. The secretome of P. eryngii and P. ostreatus can be used for the pretreatment of straw lignocellulosic materials or to improve the efficiency of enzymatic saccharification through optimized enzyme-cocktails.

Microorganisms
Phanerochaete chrysosporium (CICC40934) and P. eryngii (CICC50126) were obtained from the China Center of Industrial Culture Collection. I. lacteus (CGMCC5.0809) was obtained from the China General Microbiological Culture Collection Center. P. ostreatus (bio-67015) was obtained from the inquiry network for microbial strains of China. These four species were precultured on potato dextrose agar plates at 28°C for 7 days and were then used for ligninolysis of ramie stalks.

Lignocellulosic degradation of ramie stalks
Air-dried ramie stalks from the Institute of Bast Fiber Crops at the Chinese Academy of Agricultural Sciences were cut into small chips (400-800 μm mesh). Three grams of ramie stalks was added into 7 ml H 2 O, 0.1% Tween-80, 4 mmol l −1 veratryl alcohol, 0.2 mmol l −1 Mn 2+ in 300-ml Erlenmeyer flasks and sterilized at 121°C for 20 min. P. chrysosporium was cultured under 37°C, and pH value of substrate was adjusted to 4.5. I. lacteus, P. eryngii and P. ostreatus were cultured under 28°C, and pH value of substrates was adjusted to 5.5. Four pieces (approximately 40 mm 2 each) of each precultured white-rot fungus were inoculated separately into the substrates. These cultures were statically incubated for 7, 14, 21 or 28 days. All of the experiments were carried out in triplicate. The percentages of cellulose, hemicellulose, and lignin in the treated and untreated ramie stalks were determined according to methods of the Laboratory Analytical Procedure of the National Renewable Energy Laboratory (Sluiter et al., 2008). The lignin-selective degradation coefficient was calculated as follows:

Enzymatic hydrolysis
Samples of treated or untreated ramie stalks (0.16 g) were suspended in citric acid buffer (pH4.8) and hydrolysed with 10-50 filter paper units (FPU) g −1 cellulase for 48 h at 50°C. FPU g −1 is the activity of the cellulase added per gram of stalk. After 48 h, the hydrolysis mixture was separated by centrifugation at 10 000 rpm for 5 min. Reducing sugar was measured using the 3,5-dinitrosalicylic acid (DNS) reagent. The ratio of reducing sugar (mg g −1 ) was calculated as follows: Reducing sugar ratio ðmg=gÞ ¼ Amount of reducing sugar after enzymatic hydrolysis Amount of dry hemp woody core

Enzyme assay
Ramie stalks colonized by mycelia were collected at different stages of the solid-state fermentation (7-28 days) for time-course studies of enzyme activities. Extracellular enzymes were extracted as follows: 3 g wet weight of substrate was added to 50 ml of 50 mM citric acid buffer (pH 4.8) and stirred at 125 rpm for 5 h on an ice bath. The mixture was separated by centrifugation at 10 000 rpm for 5 min, and the supernatant was used for determination of enzyme activity. Carboxymethyl cellulase (CMCase), laccase and manganese peroxidase (MnP) activities were assayed as described in previous works (Kuhar et al., 2016;Vijayaraghavan et al., 2016;Jiang et al., 2018). CMCase assays were performed in 50 mM citrate/ phosphate buffer, at 50°C for 30 min using 1% w/v carboxymethyl cellulose (CMC) as the substrate. After 30 min, the reaction was stopped by adding 3 ml of DNS reagent, the mixture was boiled for 15 min cooled on ice, and optical density at 550 nm was determined. One unit of CMCase activity was defined as μmol glucose equivalents liberated min −1 .
Laccase activity was measured using 2,2'-azino-bis (3ethylbenzothiazoline-6-sulfonic acid) (ABTS) as substrate. The rate of ABTS oxidation was determined at 420 nm. The reaction mixture contained 600 μl 0.1 M sodium acetate buffer (pH 5.0), 300 μl ABTS (5 mM), 300 μl culture filtrate and 1400 μl distilled water. The mixture was then incubated for 2 min at 30°C, and the absorbance was measured immediately at 1-min intervals. One unit of laccase activity was defined as the conversion of 1 μmol of ABTS per minute.
MnP activity was measured as 2,6-dimethoxyphenol (2,6-DMP;ϵ = 27500 M −1 cm −1 ) oxidation. The reaction mixture contained 20 mM 2,6-DMP, 20 mM MnSO 4 ÁH 2 O, 4 mM H 2 O 2 and enzyme solution. The reaction was started by adding H 2 O 2 and monitored by the absorbance at 469 nm. One unit (U) of enzyme activity was defined as the amount of MnP that oxidizes 1 μmol of 2,6-DMP per minute at 30°C. The activities of laccase and MnP were expressed in units per litre. Enzyme activity results were the average of 20 replicates.

Protein extraction, digestion, and iTRAQ labelling
Secretome analysis was performed using 10 culture bottles for each fungal species pooled together and labelled as sample A to minimize biological variation. Sample B was prepared in the same way as sample A, and each sample (A and B) was processed separately. Extracellular proteins were extracted from 10 g wet weight substrate incubated at 28°C for 21 days with 200 ml of 50 mM sodium acetate buffer (pH 5.0) and were stirred at 180 rpm for 1 h on an ice bath. The supernatants were collected by filtration through cheese cloth, centrifuged at 8000 g for 30 min at 4°C and passed through a 0.22-μm cellulose acetate filter. The supernatants were then mixed with an equal volume of cold 40% trichloroacetic acid (TCA) and incubated on ice overnight. The precipitate was collected, washed with 20% TCA and washed with acetone twice. Proteins were then lyophilized.
Eight secretome samples from four white-rot fungal species were analysed using iTRAQ technology according to the manufacturer's instructions (AB Sciex Inc., Framingham, MA). Protein samples (100 μg) were dissolved on ice in 100 μl buffer A containing 50 μl 8 M urea, 0.1 mg sodium dodecyl sulfate, 9 μl 500 mM triethyl ammonium bicarbonate and 40 μl ultrapure water. Then, 5 μl of 200 mM tris (2-carboxyethyl) phosphine was added and the mixture was oscillated at 700 rpm at 55°C for 1 h. Reduction, alkylation and trypsin digestion of proteins were performed as described in the literature  (Qin et al., 2013;Wang et al., 2013). Tryptic peptides were labelled using the iTRAQ 8Plex Multiplex Kit (Applied Biosystems, Foster City, CA, USA) according to the manufacturer's protocol. Peptides from different fungal species were labelled as follows: 113 and 117 for P. chrysosporium, 114 and 118 for I. lacteus, 115 and 119 for P. ostreatus, and 116 and 121 for P.eryngii. After labelling, the eight labelled groups of peptides were mixed, lyophilized and re-suspended in 2% H 3 PO 4 . The peptides were then purified by strong cation exchange chromatography and lyophilized as described in the literature (Guest, 2018).

Mass spectrometric data search and analysis
Protein identification and quantification were performed as previously described in the literature (Qin et al., 2013;Wang et al., 2013). All MS analyses were performed on a LTQ Orbitrap Velos (Thermo Fisher Scientific, Bremen, Germany) connecting to an EASY-nLC system via a nanospray source. The iTRAQ-labelled peptides were dissolved in mobile phase A (99.9% H 2 O and 0.1% formic acid), and the samples (about 1 μg) were fractionated on a high-performance liquid chromatography (HPLC) precolumn(2 cm, ID 100 μm, 5 μm, C18) followed by an XBridge BEH130 Nano Ease column (15 cm, ID 75 μm, 3.5 μm, C18) with a flow rate of 300 nl min −1 . The HPLC separation was performed with mobile phase A (0.1% acetic acid) and mobile phase B (98% acetonitrile, 0.5% acetic acid). The mobile phase B gradient was set as follows: 5% to 17% for 5 min, 17% to 25% for 90 min, 25% to 60% for 10 min, 60% to 80% for 5 min, and 80% buffer B for 10 min. The MS was operated in data-dependent mode. MS/MS for each duty cycle was determined from a survey scan, and one full scan was analysed using the 25 most intense precursor ions for collision-induced dissociation fragmentation (collision energy 35%). The intense precursor ions were detected in the Orbitrap analyser and selected for higher energy collision-induced dissociation (HCD)-MS 3 fragmentation.

Protein identification
Protein identification was performed using Maxquant (version 1.2.2.5) as described in the literature (Qin et al., 2013;Wang et al., 2013). The LC-MS/MS data were queried against Uniprot. Our defined parameters were set as follows: (i) fixed modification, cysteine carbamidomethylation and iTRAQ modifications (N-terminus and lysine residues); (ii) variable modification, methionine oxidation; (iii) precursor mass tolerance, AE20 ppm; 0.5-dalton product ion mass tolerance; and (iv) trypsin digestion, up to two missed cleavages. Based on a target-decoy approach, low-confidence peptides with a global false-discovery rate (FDR) ≥ 1% were removed from further protein analysis. Relative protein abundance ratios between pairs of white-rot fungal species were calculated from iTRAQ reporter ion intensities derived from HCD spectra. The median ratio in Maxquant was used to weight and normalize the quantitative ratios. Analysis of variance (ANOVA) was used to detect significant differences among groups, and a permutation-based FDR value < 0.05 was considered significant.

Statistical analysis
All of the experimental values presented in figures and tables are the mean AE standard deviation calculated using Excel 2007. Multiple comparison tests were performed using the t-test with the Bonferroni correction (significance levels = 0.05).

Supporting information
Additional supporting information may be found online in the Supporting Information section at the end of the article.
Table S1. All of the identified proteins from the four whiterot fungi cultivated on ramie stalks at 21 days. Table S2. All of the identified proteins from P. chrysosporium cultivated on ramie stalks at 21 days. Table S3. All of the identified proteins from Irpex lacteus cultivated on ramie stalks at 21 days. Table S4. All of the identified proteins from P. eryngii cultivated on ramie stalks at 21 days. Table S5. All of the identified proteins from P. ostreatus cultivated on ramie stalks at 21 days.