BIOTECHNOLOGICAL PRODUCTS AND PROCESS ENGINEERING

Development of hemicellulolytic enzyme mixtures for plantbiomass deconstruction on target biotechnological applications

Rosana Goldbeck & André R. L. Damásio & Thiago A. Gonçalves &

Carla B. Machado & Douglas A. A. Paixão & Lúcia D. Wolf &Fernanda Mandelli & George J. M. Rocha & Roberto Ruller & Fabio M. Squina

Received: 14 April 2014 /Revised: 7 July 2014 /Accepted: 8 July 2014# Springer-Verlag Berlin Heidelberg 2014

Abstract An essential step in the conversion of lignocellu-losic biomass to ethanol and other biorefinery productsis conversion of cell wall polysaccharides into ferment-able sugars by enzymatic hydrolysis. The objective ofthe present study was to understand the mode of action ofhemicellulolytic enzyme mixtures for pretreated sugarcanebagasse (PSB) deconstruction and wheat arabinoxylan(WA) hydrolysis on target biotechnological applications.In this study, five hemicellulolytic enzymes—two endo-1,4-xylanases (GH10 and GH11), two α-L-arabinofuranosidases(GH51 and GH54), and one β-xylosidase (GH43)—weresubmitted to combinatorial assays using the experimentaldesign strategy, in order to analyze synergistic and antagonis-tic effects of enzyme interactions on biomass degradation. Thexylooligosaccharides (XOSs) released from hydrolysis wereanalyzed by capillary electrophoresis and quantified by high-performance anion exchange chromatography with pulsedamperometric detection (HPAEC–PAD). Based on this anal-ysis, it was possible to define which enzymatic combinationsfavor xylose (X1) or XOS production and thus enable thedevelopment of target biotechnological applications. Our re-sults demonstrate that if the objective is X1 production fromWA, the best enzymatic combination is GH11+GH54+GH43, and for xylobiose (X2) production from WA, it is bestto combine GH11+GH51. However, if the goal is to produce

XOS, the five enzymes used in WA hydrolysis are important,but for PSB hydrolysis, only GH11 is sufficient. If the finalobjective is bioethanol production, GH11 is responsible forhydrolyzing 64.3 % of hemicellulose from PSB. This workprovides a basis for further studies on enzymatic mechanismsfor XOS production, and the development of more efficientand less expensive enzymatic mixtures, targeting commercial-ly viable lignocellulosic ethanol production and otherbiorefinery products.

Keywords Hemicellulolytic enzymemixtures . Sugarcanebagasse . Enzymatic hydrolysis . Biotechnological application

Introduction

Lignocellulosic biomass is a renewable energy source and isconsidered the most suitable feedstock for production ofbiofuels due to its versatility, availability, and low cost(Pauly and Keegstra 2008). For the conversion of lignocellu-losic biomass into ethanol and other liquid transportationfuels, enzymes are needed to depolymerize polysaccharidesinto fermentable sugars (Dashtaban et al. 2009). However, thehigh cost of enzymes is the major hindrance for developmentof an economically viable lignocellulosic ethanol industry(Lynd et al. 2008; Banerjee et al. 2010a).

The development of more efficient lignocellulose-degrading enzyme cocktails will require deeper and moreprecise knowledge about the specific enzymes that are in-volved in the degradation of lignocellulose (Banerjee et al.2010a, b). Currently, commercial enzyme preparations avail-able for the depolymerization of lignocellulosic materials arevaguely defined as complex mixtures that contain around 80to 200 proteins. Nevertheless, the specific roles of each ofthese components found in complex cocktails for lignocellu-lose deconstruction are poorly understood (Banerjee et al.

Electronic supplementary material The online version of this article(doi:10.1007/s00253-014-5946-6) contains supplementary material,which is available to authorized users.

R. Goldbeck :A. R. L. Damásio : T. A. Gonçalves :C. B. Machado :D. A. A. Paixão : L. D. Wolf : F. Mandelli :G. J. M. Rocha :R. Ruller : F. M. Squina (*)Brazilian Bioethanol Science and Technology Laboratory (CTBE),Brazilian Centre of Research in Energy and Materials (CNPEM),Giuseppe Máximo Scolfaro Street 10000, Campinas, SP 13083,Brazile-mail: fabio.squina@bioetanol.org.br

Appl Microbiol BiotechnolDOI 10.1007/s00253-014-5946-6

2010b). Accordingly, little knowledge was gained whenworking with partially defined complex mixtures (Kumarand Wyman 2009; Banerjee et al. 2010a). Therefore, thespecific contribution of an individual enzyme is establishedonly after working with enzymes in a purified state, as well asin comprehensive studies through combinatory designs (Kimet al. 1998; Gusakov et al. 2007; Banerjee et al. 2010a; Billardet al. 2012; Gao et al. 2011; Barr et al. 2012).

The hydrolytic efficiency of a multienzyme complex in thelignocellulose saccharification process depends on both theproperties of individual enzymes and their ratio in the multi-enzyme cocktail (Gusakov et al. 2007). There are numerousstudies on enzymatic hydrolysis of pretreated biomassusing enzyme mixtures. However, there are few studiesemploying sugarcane bagasse as the substrate. Barret al. (2012) assessed the use of enzymatic mixtures(endoglucanases, cellobiohydrolases, β-glucosidases, β-xylosidases, endoxylanases, and acetylxylan esterases) forhydrolysis of two ionic liquid pretreated biomass substrates,poplar and switchgrass. Gao et al. (2010) studied the optimumratio of six core fungal enzymes for hydrolysis of cornstover pretreated by ammonia fiber expansion. Szijártóet al. (2011) identified the key enzymes needed forefficient liquefaction of hydrothermally pretreated wheatstraw. Billard et al. (2012) analyzed the hydrolysis of asingle substrate, steam-pretreated wheat straw, by a six-component mixture, which included four cellulases, Cel7a,Cel6a, Cel7b, and Cel5a, as well as the xyloglucanase Cel74aand the xylanase Xyl11a.

Comprehensive studies of the mode of operation ratioof enzymes in cocktails can lead to positive results forrational designing of enzymatic mixtures with betterefficiencies which are also cost effective (Banerjee et al.2010b). For this purpose, in the present study, five previouslycharacterized hemicellulolytic enzymes where chosen forcombinatory designs. Two endo-1,4-xylanases (GH10 fromsoil metagenome and GH11 from Penicillium funiculosum),two α-L-arabinofuranosidases (GH51 from Bacillus subtilisand GH54 from Aspergillus niger), and one β-xylosidase(GH43 from B. subtilis) were produced at high levels, puri-fied, and combinatorial assayed using an experimental design(Plackett and Burman). The synergistic and antagonistic ef-fects of enzyme interactions for biomass (wheat arabinoxylanand pretreated sugarcane bagasse) degradation were deter-mined by the Pareto analysis. Products generated by hydroly-sis were analyzed by capillary electrophoresis and quantifiedby high-performance anion exchange chromatography withpulsed amperometric detection (HPAEC–PAD). Based on ourresults, it was possible to define which enzymatic combina-tions favored xylose production, as well as those which fa-vored the production of xylooligosaccharides, with differentdegrees of polymerization, and thus enable the developmentof target biotechnological applications.

Pretreatment is an obligatory and rate-limiting step prior toconversion of plant cell wall carbohydrates into fermentablesugars for bioethanol production. The pretreatment alter lig-nocellulosic biomass structure, turning less recalcitrant toenzymes (Mosier et al 2005). Various studies have describedthat cellulose hydrolysis is improved after hemicellulose re-moval, although differences have been reported in the degreeof lignin removal necessary for efficient cellulose conversion(Converse 1993; Yang et al. 2002; Yang andWyman 2004). Inface this, assays were also conducted to evaluate the effect ofhemicellulases as a pretreatment step to enhance sugarcanebagasse digestibility by commercial cellulolytic preparation.

Material and methods

Enzymes

Five recombinant hemicellulolytic enzymes were stud-ied: two endo-1,4-xylanases (GH10 and GH11), two α-L-arabinofuranosidases (GH51 and GH54), and one β-xylosidase (GH43). The GH11 endo-1,4-xylanase (XynC11/CAC15487) from P. funiculosum and the GH54 α-L-arabinofuranosidases (AbfB54/AAB53944) from A. nigerwere expressed and secreted in Aspergillus nidulans(Gonçalves et al. 2012). The GH10 endo-1,4-xylanases(SCXyl10/KC904514) from the sugarcane soil metagenome(Alvarez et al. 2013), the GH51 α-L-arabinofuranosidases(BsAbf51/BSU28720) from B. subtilis (Hoffmam et al.2013), and GH43 β-xylosidase (BsXyn43/BSU17580), alsofrom B. subtilis, were expressed and produced in Escherichiacoli (Santos et al. 2010). All enzymes were stored at −80 °C inthe enzyme collection of the Brazilian Bioethanol Science andTechnology Laboratory—CTBE/CNPEM, Campinas, Brazil.

Expression and purification of enzymes from E. coli

The pET28a vector (Novagen,Madison,WI,USA) was usedfor heterologous expression in E. coli of three of the enzymesstudied: GH10 (SCXyl10), GH51 (BsAbf51), and GH43(BsXyn43). E. coli BL21 (DE3) (Promega, Madison,WI,USA) was transformed with pET28a constructs, and enzymeproduction and purification were performed as described pre-viously (Santos et al. 2010; Mandelli et al. 2013).

Expression and purification of enzymes from A. nidulans

The pEXPYR shuttle vector (Segato et al. 2012) was used forthe heterologous expression of GH11 (XynC11) and GH54(AbfB54) in A. nidulans A773 (pyrG89, wA3, pyroA4), pur-chased from Fungal Genetic Stock Center (FGSC, Universityof Missouri, Kansas City, MO, USA). Cloning of the selectedgenes into pEXPYR was confirmed by sequencing, and the

Appl Microbiol Biotechnol

plasmid was transformed into A. nidulansA773 as previouslydescribed (Tilburn et al. 1983). Transformant selection, cul-ture conditions, and enzyme purification were selected asdescribed by Gonçalves et al. 2012. Pure fractions of GH11and GH54 were assayed for enzymatic activity as previouslydescribed (Squina et al. 2009) and analyzed by sodium dode-cyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE).

Determination of protein concentration and SDS-PAGE

The protein concentration was determined according to themethodology proposed by Bradford (1976) using bovine se-rum albumin (Sigma-Aldrich, St. Louis, MO, USA) as astandard. Readings were obtained in a spectrophotometer at595 nm. Electrophoresis (SDS-PAGE) was used to separatethe proteins and to estimate their molecular weight. Thisprocess was performed in 12 % (w/v) polyacrylamide gelaccording to the protocol proposed by Laemmli (1970). Gelwas stained by Coomassie blue. Samples were denatured inbuffer at 99 °C for 5 min. A mixture of high molecular weightproteins (PageRuler™, Unstained Protein Ladder, ThermoScientific,Waltham, MA, USA) was used as a molecularweight standard.

Enzymatic activity assay

The polysaccharides xylan beechwood and arabinan fromsugar beet were used to measure xylanase (GH11 andGH10) and GH54 activities, respectively. The enzymaticreaction mixtures consisted of 50 μL of the substrate(0.5 %w/v), 40 μL of citrate phosphate buffer (0.1 M) atpH 5.0, and 10 μL of the purified enzyme incubated at 50 °Cin a Thermostat® (Eppendorf, Hamburg, Germany) for10 min. Reactions were stopped by addition of 100 μL of3,5-dinitrosalicylic acid (DNS) and immediately boiled for5 min at 99 °C and cooled (Miller 1959). The solution wasanalyzed at 540 nm, in an Infinite M200®spectrophotometer(Tecan, Switzerland) to measure the release of reducingsugars. One enzyme unit was defined as the quantity ofenzyme that released reducing sugar at a rate of 1 μmol/min.To measure the GH51 and GH43 activities, 2 mM of thesubstrates 4-nitrophenyl-α-L-arabinofuranoside (pNP-ara)and p-nitrophenyl-β-xylobioside (pNP-xyl) were used, re-spectively, according to methods previously described(Koseki et al. 2007). One enzyme unit was defined as thequantity of enzyme that released p-nitrophenyl at a rate of1 μmol/min.

Substrates for hydrolysis

Two different substrates were used in the analysis: insolublewheat arabinoxylan (WA), purchased from Megazyme®(Megazyme International Ireland, Bray, County Wicklow,

Ireland), lot 120801, and sugarcane bagasse, kindly providedby the Usina Vale do Rosário (Morro Agudo, SP, Brazil).Sugarcane bagasse was pretreated with a 1:1 mixture of gla-cial acetic acid P.A. and hydrogen peroxide P.A. at 60 °C for7 h (Bragatto et al. 2013).

Cellulose, hemicellulose, and lignin contents were quanti-fied in the pretreated sugarcane bagasse (PSB) according toGouveia et al. (2009). Representative samples of 200 mg werehydrolyzed in two steps: 72 % H2SO4 for 7 min at 45 °Cfollowed by dilution to 3%H2SO4 and an additional 30min at121 °C. The samples were then quenched on ice and filtered.The cellulose and hemicellulose contents of the filtrates weredetermined by HPAEC–PAD with a Dionex ICS-3000(Thermo Fisher Scientific, Dionex Product, Sunnyvale, CA,USA) system using a CarboPac PA10 column (4×250 mm).The monosaccharide contents found in the hydrolysates wereconverted to percentage of polysaccharides. For ash contentdetermination, the sample was slowly calcined at 300 °C for1 h followed by 2 h at 800 °C in a muffle furnace. Aftercooling the crucible in a dissector, the ash mass was deter-mined on an analytical balance (adequate based on the ASTMStandards (1976)).

Enzymatic hydrolysis

Enzymatic microassays were carried out in 1.5-mL Eppendorftubes using automated pipettes (Epmotion 5075, Eppendorf,Hamburg, Germany) which sought to evaluate the per-formance of experimental designs (Plackett and Burman) toanalyze synergistic and antagonistic effects of hemicellulolyticenzymes in biomass degradation. A Plackett and Burmandesign was carried out with 12 trials and three repetitions ofthe central point, totalizing to 15 trials, in order to assess theeffects of the major variables (hemicellulolytic enzymes) forPSB degradation andWA hydrolysis. Analysis of the effects ofeach independent variable was performed with the aid of thesoftware Statistica 8.0 (StatSoft Inc., Tulsa, OK, USA).

The independent variables of the experimental design werethe five hemicellulolytic enzymes: GH10, GH11, GH43,GH51, and GH54. Process responses (dependent variables)were the amounts of xylose and xylooligosaccharides releasedand the % hemicellulose conversion. Protein concentrationsemployed in the experiments were set at 0.8 mgGH10/gsubstrate,0.4 mgGH11/gsubstrate, 1.2 mgGH43/gsubstrate, 1.2 mgGH51/gsubstrate, and 1.2 mgGH54/gsubstrate. In relation to the levels ofexperimental design, level −1 corresponds to absence of theenzyme; level +1 corresponds to the presence of the enzyme,100 % of enzyme concentration; and level 0 corresponds to50 % of the enzyme concentration (center point).

The reaction mixtures contained 20 mg of each substrate,the combination of purified enzymes (experimental designs),and sodium phosphate buffer (0.1 M, pH 5.0) to complete afinal volume of 1 mL. Samples were incubated at 50 °C for

Appl Microbiol Biotechnol

48 h under agitation (1,000 rpm). After this period, the sam-ples were centrifuged (12,000g for 15 min at 4 °C), and thesupernatant was collected for subsequent analysis. Theamount of reducing sugar in the supernatant was measuredaccording to Miller (1959), and the data was presented ashemicellulose conversion (%) according to the biomass com-position employed in the experiments.

After performing the experimental designs, the en-zymes that had significant effects were evaluated as apretreatment to enhance digestibility of sugarcane ba-gasse fibers during 48 h. Therefore, 1.6 mg/g substrate ofAccellerase® 1500 (DuPontTM Genencor Science, Palo Alto,CA, USA) was added and incubated at 50 °C foranother 24 h. Subsequently, the supernatant was sepa-rated from the residual polysaccharides, analyzed accord-ing to the DNS method, and translated to polysaccharidicbiomass conversion (%).

Capillary zone electrophoresis

Xylooligosaccharides (XOSs) released by the action ofhemicellulolytic enzymes, as well as the standards purchasedfromMegazyme® (xylose, X1; xylobiose, X2; xylotriose, X3;xylotetraose, X4; xylopentaose, X5; and xylohexaose, X6),were derivatized with 8-aminopyrene-1,3,6-trisulfonic acid(APTS) by reductive amination as described previously(Cota et al. 2011). Capillary zone electrophoresis (CZE) ofxylooligosaccharides was performed using a P/ACETM MDQsystem (Beckman Coulter, Fullerton, CA, USA) with laser-induced fluorescence detection. A fused-silica capillary col-umn (TSP050375, Polymicro Technologies, Phoenix, AZ,USA) with an internal diameter of 50 μm and length of31 cm was used for separation of the oligosaccharides.Samples were injected by application of 0.5 psi for 0.5 s.Electrophoresis conditions were 15 kV/70–100 μA with thecathode at the inlet, 0.1 M sodium phosphate, pH 2.5, as therunning buffer, and a controlled temperature of 20 °C. Thecapillary column was rinsed with 1 M NaOH followed byrunning buffer with a dip cycle to prevent carry over afterinjection. Oligomers labeled with APTS were excited at488 nm and emission was collected through a 520-nm band-pass filter (Gonçalves et al. 2012). Because of the smallvolumes of capillary electrophoresis combined with smallvariations in buffer strength, retention times varied slightlywhen comparing separate electrophoresis runs. The combinedinformation obtained from the electrophoretic behavior andco-electrophoresis with monosaccharide and oligosaccharidestandards (purchased from Megazyme®) was used to identifythe degradation products. Based on this technique, it was notpossible to identify the possible acetylation and glycosylationthat can occur and change the degree of polymerization ofproducts formed from WA and PSB hydrolysis.

Xylose and xylooligosaccharide quantification

The enzymatic products were analyzed by high-performanceanion exchange chromatography with pulsed amperometricdetection (HPAEC–PAD) to detect xylose and XOSs releasedby the hemicellulolytic enzyme mixture. Separation was per-formed using a Dionex ICS-3000 instrument (Thermo FisherScientific, Dionex Product, Sunnyvale, CA, USA) with aCarboPac PA100 column (4×250 mm) and CarboPacPA100 guard column (4×50 mm), according to a linear gra-dient of A (NaOH 500 mM) and B (NaOAc 500 mM, NaOH80mM). The gradient programwas 15% of A and 2% of B at0–10 min, and 15–50 % of A and 2–20 % of B at 10–20 min,with flow rate of 1.0 mL min−1. The integrated peak areaswere adjusted based on standards (×1 to×6).

Results

Production and purification of enzymes

The enzymes GH10, GH51, and GH43 were expressed inE. coli (Alvarez et al. 2013; Hoffmam et al. 2013) while theenzymes GH11 and GH54 were expressed in A. nidulans(Gonçalves et al. 2012) as previously described. The chroma-tography purification steps provided highly purified enzymessuitable for biochemical assays (Fig. S1 in SupplementaryMaterial). Values obtained after the activity assays at 50 °Cand pH 5.0 were 100, 716, 8.25, 1.55, and 2.43 U/mg forGH10, GH11, GH54, GH51, and GH43, respectively.

The enzymes chosen were comprehensively characterizedpreviously, showing straightforward protocols for productionand purification (Gonçalves et al. 2012; Alvarez et al. 2013;Hoffmam et al. 2013), and present maximum enzymatic activitywithin a similar pH and temperature range. The enzymaticactivity values were determined at pH 5.0 and temperature of50 °C, the conditions used in enzymatic hydrolysis (experimen-tal design). These conditions are not ideal for all enzymesstudied; however, our goal was not to develop optimized enzymecocktails but to understand the interaction of these enzymes andgeneration of products with biotechnological applications.

To determine the optimal protein concentration to be used inthe experimental design, preliminary tests were performed andkinetic curves were generated to verify the minimum proteinconcentration required to be employed in the experiments; thus,the concentrations were set at 0.8 mgGH10/gsubstrate, 0.4 mgGH11/gsubstrate, 1.2 mgGH43/gsubstrate, 1.2 mgGH51/gsubstrate, and1.2 mgGH54/gsubstrate (Fig. S2 in Supplementary Material).

Substrate chemical compositions

Chemical composition of the substrates, WA and PSB, usedfor enzymatic hydrolysis are shown in Fig. 1. Cellulose

Appl Microbiol Biotechnol

(59.9 %) was the major compound of the PSB composition,followed by hemicellulose with 35.7 %. Chemical pretreat-ment preserved about 2.1 % of lignin and 0.6 % of ash. ForWA, 80 % of the composition consisted of hemicellulose,6.5 % cellulose, and 0.2 % ash. Analysis of the chemicalcomposition of PSB was performed in triplicate while thechemical composition of WA was obtained from the specifi-cation sheet provided by Megazyme®.

Effect of hemicellulolytic enzyme mixtures

The combinatorial effects of five hemicellulolytic en-zymes (GH11, GH54, GH10, GH51, and GH43) wereassessed using the experimental design strategy (Plackettand Burman), in order to analyze synergistic and/orantagonistic effects of enzyme interaction for biomassdegradation.

Based on the Pareto chart analysis (p>0.05) (Figs. S3and S4 in Supplementary Material), GH11 was respon-sible for the release of higher concentrations of xyloseand XOSs in both biomass samples analyzed. However,if the significance level was increased to 10 %, GH54would also have a significant effect on xylose produc-tion in WA (Fig. 2a). However, the GH51 enzyme had anegative effect on xylose production, and the addition ofGH10 showed a negative effect on XOS production fromWA(Fig. 2b).

Analyzing the results obtained from PSB enzymatic hydro-lysis at the 10 % significance level, as in the case of WA,GH11 and GH54 had significant effects on xylose production(Fig. 3a). Clearly, when combined GH11 and GH54 presentheterosynergy (Van Dyk and Pletschke 2012), a phenomenon

characterized in this case by improvement of xylose releasefrom the main chain by GH11 as a consequence of the GH54debranching activity.

Production of xylose and xylooligosaccharides by enzymatichydrolysis

The products from WA and PSB enzymatic hydrolysis byhemicellulolytic enzyme mixtures were analyzed by CZEusing APTS-labeled oligosaccharides. The CZE was usedherein as a qualitative technique for oligosaccharide profiling,because it gives faster results and provides high-resolutionseparation when compared to HPLC. The results shown inFigs. 4 and 5 revealed the ability of the enzyme mixture torelease xylose (X1), xylobiose (X2), and short-chain XOSsfrom WA and PSB, respectively.

According to Fig. 4, combining GH51 and GH10 resultedin release of X2 and X3 (xylotriose) as major products fromWA hydrolysis. When GH51 and GH11 were combined, therewas a higher production of X1 and X2. However, the substi-tution of GH51 for GH54 increased xylose production. Thisdata corroborates with the screening design factor (Fig. 2a),where the GH54 enzyme showed a positive effect while GH51had a negative effect on xylose production. Experiments 4and 10 (Table 1) reinforce the effective hemicellulosebreakdown by GH54 when compared to GH51 (Table 1).The addition of a third enzyme (GH43) in both combinations,GH51+GH11 and GH54+GH11 (Fig. 4), increased the xy-lose concentration as expected, since the GH43 enzyme isa β-xylosidase.

Moreover, the enzyme combination (GH11+GH43+GH51) favored X2 and X3 production from PSB hydrolysis(Fig. 5),where as in WA hydrolysis, the substitution of GH51for GH54 increased xylose production in PSB hydrolysis.This data corroborates with the positive effects of GH54 forxylose production (Fig. 3a).

Xylobiose was the major product from PSB hydrolysis bythe combination of GH10, GH11, and GH51. The addition ofa fourth enzyme (GH54) to this combination increased xyloseproduction, as well as the addition of GH43 (Fig. 5).

Based on CZE results, we can conclude that the GH11enzyme released more xylose from WA and PSB hydro-lysis, corroborating with the screening design factors(Figs. 2 and 3). In the same way, GH54 was the bestarabinofuranosidase for arabinoxylan debranching inboth WA and PSB, increasing access of xylanases(GH11 and GH10) to arabinoxylan. Qualitatively, it ispossible to observe that the addition of GH43 was importantto hydrolyze X2 to X1, but this enzyme was not suitable tohydrolyze higher XOSs with degree of polymerizationexceeding three.

The supernatants were also analyzed by HPAEC–PAD inorder to quantify xylose and XOSs released from WA and

Fig. 1 Chemical composition of the substrates: wheat arabinoxylan (WA)and pretreated sugarcane bagasse (PSB). The chemical composition ofPSB was performed in triplicate while the chemical composition of WAwas obtained from the specification sheet provided by Megazyme®

Appl Microbiol Biotechnol

PSB enzymatic hydrolysis (Fig. 6a, b). The data showeddifferent amounts of X1/XOS for each substrate. WA hydro-lysis generated greater amounts of X1 and X2, and an in-creased quantity of XOS was released from PSB hydrolysis.These results show that substrates with a less complex struc-ture (WA) release higher amounts of X1 and X2 than XOS.Furthermore, arabinofuranosidases (GH54 and GH51) werenot necessary to reach high levels of PSB hydrolysis, wherethe addition of GH11 alone converted almost 65 % of PSB(Fig. 6b and Table 2).

The Plackett-Burman experimental design was used todefine the most important enzymes for WA and PSB hydro-lysis. Table 1 shows that the highest hemicellulose

conversions in theWA substrate were achieved in experiments3 (35.92 %), 5 (37.17 %), and 6 (34.97 %). The commonenzymes in these experiments were GH11 and GH54, sug-gesting that GH10, GH51, and GH43 are of lesser importancefor WA hydrolysis, while GH11 is responsible for morethan 50 % of the substrate conversion, as can be ob-served in experiment 11 (20.61 %). In experiment 2(Table 1) a hemicellulose conversion of 24.81 % wasobserved, indicating that the presence of enzymesGH10+GH51 resulted in an increase of 4.20 %, whileGH11 alone is responsible for 20.61 % of conversion.Similar result was observed in experiment 7 (Table 1),where the presence GH43+GH51 increased hemicellulose

Fig. 2 Pareto chart ofstandardized effects (p>0.10) of axylose and b xylooligosaccharides(XOSs) released after wheatarabinoxylan (WA) hydrolysis.GH10 is the endo-1,4-xylanasefrom sugarcane soil metagenome(SCXyl10), GH11 is the endo-1,4-xylanase from Penicilliumfuniculosum (XynC11), GH43 isthe β-xylosidase from Bacillussubtilis (BsXyn43), GH51 is theα-L-arabinofuranosidase fromB. subtilis (BsAbf51), and GH54is the α-L-arabinofuranosidasefrom Aspergillus niger (AbfB54)

Appl Microbiol Biotechnol

conversion by 8 %. Although some of the enzymes such asGH10, GH43, and GH51 are not operating at their idealconditions, they contributed to biomass degradation in a lesserdegree. However, if the goal is to produce XOS, all fiveenzymes are important, as can be visualized in Table 1, whereexperiments 13, 14, and 15 (central points) favored XOSproduction. These experiments showed the production ofmore than 5 μmol/mL of XOS from WA using only half ofthe enzyme concentration.

In the case of PSB, the contribution of GH11 waseven more prominent, where this enzyme was able tohydrolyze 64.3 % of hemicellulose, and the addition ofother enzymes had a negative effect on the final yield

(Table 2). The GH11 enzyme also favored XOS productionfrom PSB hydrolysis, as can be visualized in Table 2,where the highest values of XOS, above 3 μmol/mL, wereobtained in experiments 2, 7, and 11. In these experiments,the only common enzyme present in the enzymatic mixtureswas GH11.

Biotechnological applications

Due to the ability of the hemicellulase mixture to release XOS,X2, and X1 from PSB, we also analyzed the effect of thisenzymatic treatment prior to biomass sacharification by com-mercial cellulases (Accellerase® 1500). The pretreatment with

Fig. 3 Pareto chart ofstandardized effects (p>0.10) of(a) xylose and (b)xylooligosaccharides (XOSs) re-leased after pretreated sugarcanebagasse (PSB) hydrolysis. GH10is the endo-1,4-xylanase fromsugarcane soil metagenome(SCXyl10), GH11 is the endo-1,4-xylanase from Penicilliumfuniculosum (XynC11), GH43 isthe β-xylosidase from Bacillussubtilis (BsXyn43), GH51 is theα-L-arabinofuranosidase fromB. subtilis (BsAbf51), and GH54is the α-L-arabinofuranosidasefrom Aspergillus niger (AbfB54)

Appl Microbiol Biotechnol

enzymes (GH11) and (GH11+GH54) significantly enhancedPSB saccharification by Accellerase® 1500, wherepolysaccharidic biomass (cellulose and hemicellulose) con-version increased by 18.03 % and 10.62 %, respectively(Fig. 7), after 24 h of hydrolysis. The GH11 enzymealone is sufficient for efficient saccharification of sug-arcane bagasse. Table 2 shows that GH11 is responsiblefor 64.30 % of hemicellulose conversion from PSB, ascan be observed in experiment 11, resulting in release of 0.802and 3.702 μmol/mL of xylose and XOS, respectively.However, considering the low concentration of proteinemployed (0.008 mg/mL of GH11=0.4mgGH11/gsubstrate),these values becomes even more attractive and of greatinterest for industrial application, resulting in 100.25 and462.75 μmol/mg of xylose and XOS released, respectively,per milligram of GH11 applied.

Discussion

The strong cellulose crystalline arrangement and theprotective effects of lignin, as well as hemicellulosecontent, make enzyme access difficult and constitutean obstacle to conversion of biomass polysaccharidesinto value-added products (Canettieri et al. 2007;Rocha et al. 2012). Pretreatment of lignocellulosic bio-mass results in changes to the chemical composition andphysical structure of substrates, including redistributionof hemicellulose and lignin fractions, increase of thesurface area, and formation of pores (Bragatto et al.2012). The rate of enzymatic hydrolysis and reducingsugar yield is directly affected by biomass structure andcomposition; this affirmation is valid for hemicellulasesand cellulases as demonstrated in different xylan sources

Fig. 4 Product profilesobtained by capillaryelectrophoresis of APTS-labeledxylooligosaccharides releasedfrom wheat arabinoxylan (WA)hydrolysis. For identification ofthe degradation products, thecombined information was used asobtained from the electrophoreticbehavior and co-electrophoresiswith monosaccharide and oligo-saccharide standards (xylose, X1;xylobiose, X2; xylotriose, X3;xylotetraose, X4; xylopentaose,X5; and xylohexaose, X6) pur-chased from Megazyme®

Fig. 5 Product profilesobtained by capillaryelectrophoresis of APTS-labeledxylooligosaccharides releasedfrom pretreated sugarcane ba-gasse (PSB) hydrolysis. Foridentification of the degradationproducts, the combined informa-tion was used as obtained fromthe electrophoreticbehavior and co-electrophoresiswith monosaccharide and oligo-saccharide standards (xylose, X1;xylobiose, X2; xylotriose, X3;xylotetraose, X4; xylopentaose,X5; and xylohexaose, X6) pur-chased from Megazyme®

Appl Microbiol Biotechnol

and in pretreated sugarcane bagasse (Vasconcelos et al.2013). Vasconcelos et al. (2013) also showed that diluteacid pretreatments are responsible for production of lowhemicellulose content biomasses, which limits under-standing the conjugate effect of hemicellulase activities.

Gonçalves et al. (2012) worked with pretreated sugarcanebagasse (peroxide-HAc) followed by enzymatic hydrolysisusing endoxylanase (GH11 family) for XOS production. Inaddition, the use of peroxide-HAc is not only consideredto be environmentally friendly but is also very effectiveto cleave aromatic rings present in lignin (Khama et al.2005). Peroxide-HAc has long been recognized as aneffective reagent for lignin removal, isolating plant fi-bers for use in some techniques of the pulp and paperprocess (Brasileiro et al. 2001; Teixeira et al. 2000).Moreover, the peroxide-HAc pretreatment showed tobe an important method for studies on the physicalproperties of cellulosic insoluble substrates (Bragattoet al. 2012). In this context, in order to maintain high levelsof biomass hemicellulose, sugarcane bagasse was pretreatedwith a 1:1 mixture of glacial acetic acid P.A. and hydrogenperoxide P.A. at 60 °C for 7 h (Bragatto et al. 2013).

Five hemicellulolytic enzymes—two endo-1,4-xylanases(GH10 and GH11), two α-L-arabinofuranosidases (GH51and GH54), and one β-xylosidase (GH43)—were combina-torial assayed using the experimental design strategy, in orderto analyze synergistic and antagonistic effects of enzymeinteractions on biomass degradation. Among the enzymes

studied, the GH11 enzyme was responsible for the release ofhigher concentrations of xylose and XOSs in both biomasssamples analyzed. However, if the level of significance wasincreased to 10 % (Pareto chart), the GH54 enzyme also hadsignificant effects on xylose production from WA and PSBenzymatic hydrolysis. The addition of a third enzyme (GH43)in enzymatic combinations increased the xylose concentrationas expected, since the GH43 enzyme is a β-xylosidase(EC 3.2.1.37), which cleaves the resulting xylooligomersto free xylose (Czjzek et al. 2005). According to Gonçalveset al. (2012), to improve arabinoxylan hydrolysis, GH11should be combined with GH54. Synergism assays werecarried out at 50 °C, a temperature in which both enzymeswere fully stable and work at high efficiency, using rye andwheat arabinoxylan as substrates. Synergistic action ofGH11 and GH54 could be observed for both arabinoxylansubstrates. The degree of synergy was higher on insolublewheat arabinoxylan (1.6) than on rye arabinoxylan (1.2).Collectively, the results suggest that synergistic improvementof xylose and xylooligosaccharide production when bothenzymes were combined is correlated to the action ofGH54 on xylooligosaccharides presenting an arabinofuranoseresidue (Gonçalves et al. 2012).

It is important to point out that GH51 had a negative effectonxylose production, while the addition of GH10 showed anegative effect on XOS production from WA. This can beexplained by the fact that neither enzyme is acting in its idealconditions. According to Hoffmam et al. (2013), the

Table 1 Plackett-Burman design matrix (PB12+3 center points) carried out for the enzymatic hydrolysis of wheat arabinoxylan (WA)

Experiments Var1 Var2 Var3 Var4 Var5 Xylose (μmol/mL) XOS (μmol/mL) Hemicelluloseconversion (%)GH10 GH11 GH43 G51 GH54

1 +1 −1 +1 −1 −1 0.032 0.191 1.68

2 +1 +1 −1 +1 −1 0.638 1.532 24.81

3 −1 +1 +1 −1 +1 2.691 1.970 35.92

4 +1 −1 +1 +1 −1 0.041 0.010 2.24

5 +1 +1 −1 +1 +1 2.115 1.881 37.17

6 +1 +1 +1 −1 +1 2.200 1.852 34.97

7 −1 +1 +1 +1 −1 0.934 2.131 28.61

8 −1 −1 +1 +1 +1 0.260 0.633 6.72

9 −1 −1 −1 +1 +1 0.121 0.262 5.04

10 +1 −1 −1 −1 +1 0.329 0.440 4.06

11 −1 +1 −1 −1 −1 0.786 1.881 20.61

12 −1 −1 −1 −1 −1 0.012 0.000 0.00

13 0 0 0 0 0 2.208 5.811 27.19

14 0 0 0 0 0 2.227 5.561 26.38

15 0 0 0 0 0 2.025 5.112 25.73

Level −1 corresponds to absence of the enzyme, level +1 corresponds to the presence of the enzyme, and level 0 corresponds to 50 % of the enzymeconcentration (center point). GH10 is the endo-1,4-xylanase from sugarcane soil metagenome (SCXyl10), GH11 is the endo-1,4-xylanase fromPenicillium funiculosum (XynC11), GH43 is the β-xylosidase from Bacillus subtilis (BsXyn43), GH51 is the α-L-arabinofuranosidase from B. subtilis(BsAbf51), and GH54 is the α-L-arabinofuranosidase from Aspergillus niger (AbfB54)

Appl Microbiol Biotechnol

GH51 enzyme has an optimum pH of 6.6 and optimumtemperature of 37 °C. However, when the temperature isincreased to 50 °C, the relative activity of the GH51enzyme is maintained at 85 %. Regarding GH10, thisenzyme acts better at pH 6.0 and 45 °C, and at pH 5.0,the relative activity of GH10 is 80 % (Alvarez et al. 2013).Another hypothesis for the low performance of the GH10enzyme is substrate competition with GH11, an enzymebelonging to a family of enzymes known to act moreefficiently on insoluble substrates such as wheat arabinoxylan(Beaugrand et al. 2004).

This work provides basis for further studies on enzy-matic mechanisms for XOS production, and the devel-opment of more efficient and less expensive enzymaticmixtures, targeting commercially viable lignocellulosicethanol production and other biorefinery products. Inaddition, the production of value-added products fromplant biomass is of great interest to advance not onlythe biofuel field but also the pharmaceutical and foodindustries. XOS and xylobiose are of great interest tothe food industry because of their applications as prebi-otics and sweeteners (Vázquez et al. 2000). Moreover,the production of xylose directly from these substrates

can be used for bioethanol production, as well as forthe production of xylitol, an alternative sweetener(Winkelhausen and Kuzmanova 1998). In the food in-dustry, XOS can be used in functional foods due to thepositive effects that oligosaccharides have on gastroin-testinal microbiota, promoting several benefits to humanhealth (Gullon et al. 2010). Xylobiose can also beemployed as a sweetener due to its sweetness degreeof 30 % when compared to sucrose and its low caloriecharacteristics (Toshio et al. 1990). The preferred rangefor degree of polymerization of XOS is 2−4 for food-related applications (Loo et al. 1999). In pharmaceuti-cals, XOSs offer advantages when compared with otheroligosaccharides in terms of stability and benefic effects,such as stimulating the growth of probiotics includingLactobacillus spp., noncariogenicity, and inhibiting thegrowth of pathogenic microorganisms, providing a num-ber of benefits to the digestive and immune systems(Chapla et al. 2012).

The experimental design was of great importance for un-derstanding the action of enzymes when combined and tooptimize the enzymes necessary for elaboration of less costlyenzymatic cocktails, thus enabling the development of target

Fig. 6 HPAEC–PAD analysis ofmonosaccharide andxylooligosaccharides releasedfrom the substrates (a) wheatarabinoxylan (WA) and (b)pretreated sugarcane bagasse(PSB) after hydrolysis with theenzyme mixture. The concentra-tions were calculated using astandard curve of xyloose, X1;xylobiose, X2; xylotriose, X3;xylotetraose, X4; xylopentaose,X5; and xylohexaose, X6(Megazyme®)

Appl Microbiol Biotechnol

biotechnological applications. Collectively, our findings indi-cate the best way to direct xylose and/or XOS production.From WA, the major products were X1 using GH11, GH43,and GH54, and for X2 production from WA, it is best tocombine GH11 and GH51. However, if the goal is to produceXOS, the five enzymes are important for WA hydrolysis, butfor PSB hydrolysis, GH11 alone is sufficient. If the objectiveis bioethanol production, GH11 is responsible for hydrolyzing64.3 % of hemicellulose from PSB.

We also analyzed the effect of this enzymatic treatmentprior to biomass saccharification by commercial cellulases(Accellerase® 1500), and the GH11 enzyme alone wassufficient for efficient saccharification of sugarcane

bagasse. The xylanase GH11 probably facilitated the accessof endoglucanases/exoglucanases and β-glucosidases to cel-lulose microfibrils, which are naturally surrounded by hemi-cellulose (Souza et al. 2012). Knowledge of how GH11xylanase is able to degrade complex xylans will underpin theirindustrial exploitation, particularly in the continued quest toconvert lignocellulosic biomass into biofuels (Vardakou et al.2008). This result demonstrated the potential application ofhemicelulolytic enzymes, mainly the xylanase GH11, for bio-fuel production, as well as for XOS production according tothe chosen strategy.

Acknowledgments We are grateful to FAPESP (The State of São PauloResearch Foundation) for its financial support (2012/18859-5). ARLD isa FAPESP fellow (2013/18910-3). This work was also financially sup-ported by FAPESP 2008/58037-9 and by CNPq 475022/2011-4 and310177/2011-1 (FMS). We would like to thank the entire team of themolecular biology laboratory (CTBE/CNPEM), in particular the technicalsupport of Rodrigo F. Almeida, also Thabata M. Alvarez and Zaira B.Hoffmam for cloning the GH10 and GH51 enzymes, respectively.

References

Alvarez TM, Goldbeck R, Santos RC, Paixão DAA, Gonçalves TA,Franco Cairo JPL, Almeira RF, Pereira IO, Jackson G, Cota J,Buchli F, Citadini AP, Ruller R, Polo CC, Neto MO, MurakamiMT, Squina FM (2013) Development and biotechnological applica-tion of a novel endoxylanase family GH10 identified from sugar-cane soil metagenome. PLoS ONE 8:e70014

Table 2 Plackett-Burman design matrix (PB12+3 center points) carried out for the enzymatic hydrolysis of pretreated sugarcane bagasse (PSB)

Experiments Var1 Var2 Var3 Var4 Var5 Xylose (μmol/mL) XOS (μmol/mL) Hemicelluloseconversion (%)GH10 GH11 GH43 G51 GH54

1 +1 −1 +1 −1 −1 0.002 0.381 2.88

2 +1 +1 −1 +1 −1 0.681 3.342 46.29

3 −1 +1 +1 −1 +1 1.543 2.682 54.18

4 +1 −1 +1 +1 −1 0.091 0.491 6.13

5 +1 +1 −1 +1 +1 1.152 2.845 53.52

6 +1 +1 +1 −1 +1 1.472 2.583 51.16

7 −1 +1 +1 +1 −1 0.714 3.474 43.49

8 −1 −1 +1 +1 +1 0.190 0.357 9.23

9 −1 −1 −1 +1 +1 0.211 0.398 11.87

10 +1 −1 −1 −1 +1 0.239 0.601 12.93

11 −1 +1 −1 −1 −1 0.802 3.702 64.30

12 −1 −1 −1 −1 −1 0.022 0.081 0.00

13 0 0 0 0 0 1.281 2.825 66.02

14 0 0 0 0 0 1.235 2.797 65.28

15 0 0 0 0 0 1.352 2.895 64.93

Level −1 corresponds to absence of the enzyme, level +1 corresponds to the presence of the enzyme, and level 0 corresponds to 50 % of the enzymeconcentration (center point). GH10 is the endo-1,4-xylanase from sugarcane soil metagenome (SCXyl10), GH11 is the endo-1,4-xylanase fromPenicillium funiculosum (XynC11), GH43 is the β-xylosidase from Bacillus subtilis (BsXyn43), GH51 is the α-L-arabinofuranosidase from B. subtilis(BsAbf51), and GH54 is the α-L-arabinofuranosidase from Aspergillus niger (AbfB54)

Fig. 7 The pretreatment step with enzymes GH11, and GH11+GH54followed by addition of Accellerase® 1500 (cellulotytic enzymatic cock-tail) improved sugarcane bagasse saccharification. GH11 is the endo-1,4-xylanase from Penicillium funiculosum (XynC11) and GH54 is the α-L-arabinofuranosidase from Aspergillus niger (AbfB54)

Appl Microbiol Biotechnol

Banerjee G, Scott-Craig JS, Walton JD (2010a) Improving enzymes forbiomass conversion: a basic research perspective. Bioenerg Res 3:82–92

Banerjee G, Car S, Scott-Craig JS, Borrusch MS, Walton JD (2010b)Rapid optimization of enzyme mixtures for deconstruction of di-verse pretreatment/biomass feedstock combinations. BiotechnolBiofuels 3:22

Barr CJ, Mertens JA, Schall CA (2012) Critical cellulase andhemicellulase activities for hydrolysis of ionic liquid pretreatedbiomass. Bioresource Technol 104:480–485

Beaugrand J, Chambat G, Wong VW, Goubet F, Rémond C, Paës G,Benamrouche S, Debeire P, O'Donohue M, Chabbert B (2004)Impact and efficiency ofGH10 andGH11 thermostableendoxylanaseson wheat bran and alkali-extractable arabinoxylans. Carbohydr Res339:2529–2540

Billard H, Faraj A, Lopes Ferreira N, Menir S, Heiss-Blanquet S (2012)Optimization of a synthetic mixture composed of majorTrichoderma reesei enzymes for the hydrolysis of steam-explodedwheat straw. Biotechnol Biofuels 5:9

Bradford MM (1976) A rapid and sensitive method for quantitation ofmicrogram quantities of protein utilizing the principle of protein-dyebinding. Anal Biochem 72:245–248

Bragatto J, Segato F, Cota J, Mello DB, Oliveira MM, Buckeridge MS,Squina SM, Driemeier C (2012) Insights on how the activity of anendoglucanase is affected by physical properties of insoluble cellu-loses. J Phys Chem B 116:6128–6136

Bragatto J, Segato F, Squina FM (2013) Production of xylooligosaccharides(XOS) from delignified sugarcane bagasse by peroxide-HAc processusing recombinant xylanase from Bacillus subtilis. Ind Crop Prod 51:123–129

Brasileiro LB, Colodete JL, Piló-Veloso DA (2001) Utilização deperácidos na deslignificação e no branqueamento de polpascelulósicas. Quim Nova 6:819–829

Canettieri E, Rocha GJM, de Carvalho JR, Silva JBA (2007)Optimization of acid hydrolysis from the hemicellulosic fraction oftheEucalyptus grandis residue using response surfacemethodology.Bioresource Technol 98:422–428

Chapla D, Pandit P, Shah A (2012) Production of xylooligosaccharidesfrom corncob xylan by fungal xylanase and their utilization byprobiotics. Bioresource Technol 115:215–221

Converse AO (1993) Substrate factors limiting enzymatic hydro-lysis. In: Saddler JN (ed) Bioconversion of forest and agri-cultural plant residues. CAB International, Wallingford, pp93–106

Cota J, Alvarez TM, Citadini AP, Santos CR, de Oliveira NM, OliveiraRR, Pastore GM, Ruller R, Prade RA, Murakami MT, Squina FM(2011) Mode of operation and low-resolution structure of amulti-domain and hyperthermophilicendo-β-1,3-glucanasefrom Thermotoga petrophila . Biochem Biophys ResCommun 406:590–594

Czjzek M, Ben David A, Bravman T, Shoham G, Henrissat B, Shoham Y(2005) Enzyme-substrate complex structures of a GH39 β-xylosidase from Geobacillus stearothermophilus. J Mol Biol 353:838–846

Dashtaban M, Schraft H, Qin W (2009) Fungal bioconversion of ligno-cellulosic residues; opportunities and perspectives. Int J Biol Sci 5:578–595

Gao D, Chundawat SPS, Krishnan C, Balan V, Dale BE (2010)Mixture optimization of six core glycosyl hydrolases formaximizingsaccharification of ammonia fiber expansion (AFEX)pretreated corn stover. Bioresource Technol 101:2770–2781

Gao D, Uppugundla N, Chundawat SPS, Yu X, Hermanson S,Gowda K, Brumm P, Mead D, Balan V, Dale BE (2011)Hemicellulases and auxiliary enzymes for improved conversion oflignocellulosic biomass to monosaccharides. Biotechnol Biofuels4:5

Gonçalves TA, Damásio ARL, Segato F, Alvarez TM, Bragatto J,Brenelli LB, Citadini APS, Murakami MT, Ruller R, Paes LemeAF, Prade RA, Squina FM (2012) Functional characterization andsynergic action of fungal xylanase and arabinofuranosidase forproduction of xylooligosaccharides. Bioresource Technol 119:293–299

Gouveia ER, Nascimento RT, Souto-Maior AM, Rocha GJM (2009)Avaliação de metodologia para a caracterização química do bagaçode cana-de-açúcar. Quim Nova 32:1500–1503

Gullon B, Yanez R, Alonso JL, Parajo JC (2010) Production of oligosac-charides and sugars from rye straw: a kinetic approach. BioresourceTechnol 101:6676–6684

Gusakov AV, Salanovich TN, Antonov AI, Ustinov BB, Okunev ON,Burlingame R, Emalfarb M, Baez M, Sinitsyn AP (2007) Design ofhighly efficient cellulase mixtures for enzymatic hydrolysis of cel-lulose. Biotechnol Bioeng 97:1028–1038

Hoffmam ZH, Oliveira LC, Cota J, Alvarez TM, Diogo JA, Neto MO,Citadini APS, Leite VBP, Squina FM, Murakami MT, Ruller R(2013) Characterization of a hexamericexo-acting GH51 a-L-arabinofuranosidase from the mesophilic Bacillus subtilis. MolBiotechnol 55:260–267

Khama L, Bigot YL, Delmas M, Avignon G (2005) Delignification ofwheat strawusing a mixture of carboxylic acid and peroxoacids. IndCrops Prod 21:9–15

Kim E, Irwin DC, Walker LP, Wilson DB (1998) Factorial optimizationof a six-cellulase mixture. Biotechnol Bioeng 58:494–501

Koseki T, Mimasaka N, Hashizume K, Shiono Y, Murayama T (2007)Stimulatory effect of ferulic acid on the production of extracellularxylanolytic enzymes by Aspergillus kawachii. Biosci BiotechnolBiochem 71:1785–1787

Kumar R, Wyman CE (2009) Effect of xylanase supplementation ofcellulase on digestion of corn stover solids prepared by leadingpretreatment technologies. Bioresour Technol 100:4203–4213

Laemmli UK (1970) Cleavage of structural proteins during the assemblyof head of bacteriophage T4. Nature 227:680–685

Loo JV, Cummings J, Delzenne N, Englyst H, Franck A, Hopkins M,Kok N, Macfarlane G, Newton D, Quigley M, Roberfroid M, vanVliet T, van den Heuvel E (1999) Functional food properties ofnondigestible oligosaccharides: a consensus report from the ENDOProject (DGXII-AIRII-CT94-1095). Brit J Nutr 81:121–132

Lynd LR, Laser MS, Bransby D, Dale BE, Davison B, Hamilton R,Himmel M, Keller M, McMillan JD, Sheehan J, Wyman CE(2008) How biotech can transform biofuels. Nature Biotechnol 26:169–172

Mandelli F, Franco Cairo JPL, Citadini APS, Büchli F, Alvarez TM,Oliveira RJ, Leite VBP, PaesLeme AF, Mercadante AZ, SquinaFM (2013) The characterization of a thermostable and cambialisticsuperoxide dismutase from Thermus filiformis. Lett Appl Microbiol57:40–46

Miller GL (1959) Use of dinitrosalicylic acid reagent for determination ofreducing sugar. Anal Chem 31:426–428

Mosier N, Wyman C, Dale B, Elander R, Lee YY, HoltzappleM, LadischM (2005) Features of promising technologies for pretreatment oflignocellulosic biomass. Bioresource Technol 96:673–686

Pauly M, Keegstra K (2008) Cell-wall carbohydrates and their modifica-tion as a resource for biofuels. Plant J 54:559–568

Rocha GJM, Gonçalves AR, Oliveira BR, Olivares EG, RossellCEV (2012) Steam explosion pretreatment reproduction andalkaline delignification reactions performed on a pilot scalewith sugarcane bagasse for bioethanol production. Ind CropProd 35:274–279

Santos CR, Meza AN, Hoffmam ZB, Silva JC, Alvarez TM, Ruller R,Giesel GM, Verli H, Squina FM, Prade RA, Murakami MT (2010)Thermal-induced conformational changes in the product release areadrive the enzymatic activity of xylanases 10B: Crystal structure,conformational stability and functional characterization of the

Appl Microbiol Biotechnol

xylanase 10B from Thermotoga petrophilaRKU-1. Biochem BiophRes Co 403:214–219

Segato F, Damásio ARL, Gonçalves TA, Lucas RC, Squina FM, DeckerSR, Prade RA (2012) High-yield secretion ofmultiple client proteinsin Aspergillus. Enz Microb Technol 51:100–106

Souza AP, Leite DCC, Pattathil S, Hahn MG, Buckeridge MS (2012)Composition and structure of sugarcane cell wall polysaccharides:implications for second-generation bioethanol production. BioenergRes 6:564–579

Squina FM,Mort AJ, Decker SR, Prade RA (2009) Xylan decompositionby Aspergillus clavatus endo-xylanase. Protein Expr Purif 68:65–71

ASTM Standards (1976) Method for absolute calibration of reflectancestandards, ASTM E306-71. ASTM International. http://www.astm.org/DATABASE.CART/WITHDRAWN/E306.htm. Accessed 06June 2014

Szijártó N, Siika-aho M, Sontag-Strohm T, Viikari L (2011) Liquefactionof hydrothermally pretreated wheat straw at high-solids content bypurified Trichoderma enzymes. Bioresource Technol 102:1968–1974

Teixeira CL, Linden CJ, Shroeder AH (2000) Simultaneous saccharifi-cation and cofermentation of peracetic acid-pretreated biomass.Appl Biochem Biotechnol 84:111–127

Tilburn J, Scazzocchio C, Taylor GG, Zabicky-Zissman JH, LockingtonRA, Davies RW (1983) Transformation by integration inAspergillusnidulans. Gene 26:205–221

Toshio, NoriyoshiI, ToshiakiK, ToshiyukiN, KunimasaK (1990)Production of xylobiose. Japanese Patent JP 2119790

Van Dyk JS, Pletschke BI (2012) A review of lignocellulose bioconver-sion using enzymatic hydrolysis and synergistic cooperation be-tween enzymes-factors affecting enzymes, conversion and synergy.Biotechnol Adv 30:1458–1480

Vardakou M, DumonC MJM, Christakopoulos P, Weiner DP, Juge N,Lewis RJ, Gilbert HJ, Flint JE (2008) Understanding the structuralbasis for substrate and inhibitor recognition in eukaryotic GH11xylanases. J Mol Biol 375:1293–1305

Vasconcelos SM, SantosAMP,RochaGJM, Souto-MaiorAM (2013)Dilutedphosphoric acid pretreatment for production of fermentable sugars in asugarcane-based biorefinery. Bioresource Technol 135:46–52

Vázquez MJ, Alonso JL, Domínguez H, Parajó JC (2000)Xylooligosaccharides: manufacture and applications. Trends FoodSci Tech 11:387–393

Winkelhausen E, Kuzmanova S (1998)Microbial conversion of D-xyloseto xylitol. J Ferment Bioeng 86:1–14

Yang B, Wyman CH (2004) Effect of xylan and lignin removal by batchand flow through pretreatment on the enzymatic digestibility of cornstover cellulose. Biotechnol Bioeng 86:88–95

Yang B, Boussaid A, Mansfield SD, Gregg DJ, Saddler JN (2002) Fastand efficient alkaline peroxide treatment to enhance the enzymaticdigestibility of steam-exploded softwood substrates. BiotechnolBioeng 77:678–684

Appl Microbiol Biotechnol